Split (1)

train · ~9.11M rows (showing the first 93.5k)

identifier stringlengths 11 32	pdf_url stringlengths 17 4.62k ⌀	lang stringclasses 120 values	error stringclasses 1 value	title stringlengths 2 500 ⌀	source_name stringlengths 1 435 ⌀	publication_year float64 1.9k 2.02k	license stringclasses 3 values	word_count int64 0 1.64M	text stringlengths 1 9.75M
https://openalex.org/W3038972297	https://sportpedagogy.org.ua/index.php/ppcs/article/download/1509/803	English	null	Timeless evolution of walking and pace strategy of women’s race walking	Pedagogy of physical culture and sports	2,020	cc-by	4,753	© Panagiotα Fitili, Vasilios Giovanis, 2020 doi:10.15561/26649837.2020.0601 Abstract The purpose of this research was to study the timeline evolution of walking, as well as the Pacing Strategy Profiles of high-level women in the 20 km of race walking. The practical example of applying the theoretical basis was made during the Women’s Greek Championship (Megara 2016), in which 12 athletes aged 19 to 40 participated (28.50 ± 7.20). The certified distance of the 20km route was divided into 10 sections of 2 km each. The same happened with the times (intermediate, final) corresponding to the individual sections (2 km) of the route. The athletes were divided into 4 groups: the first 3, those who finished 15% slower than the first, those who finished 15% - 30% slower, and those who finished more than 30% slower than the winner. Finally became comparison of the first 6 and last 6 athletes’ groups. The individual pace strategies that describe the tactics of the athletes in this race have been calculated. It was found that the winners of the race used Even Pacing Strategy, maintaining a steady speed on most of the route. As the level of women athletes became lower, Variable Pacing Strategy was used, while the athletes who finished last did not seem to be able to maintain any particular pacing strategy. It is suggested that athletes should follow Even Pacing Strategy during the race in order to improve their performance. f p race walking, even pacing strategy, variable pacing strategy, performance. Introduction1 on: (a) Leisure as free time, (b) Leisure as an experience, (c) Leisure as an activity. It is therefore understood, that recreation is a broader concept that involves active activity, whereas entertainment involves passive activity [5]. 4) Nordic walking is walking with specially designed walking sticks, to improve fitness [6]. 5) Race walking is a part of the classical athletics and one of the Olympic Games, in which athletes move as fast as possible without running in routes of 20 km and 50 km. of Physical Culture and Sports PEDAGOGY ORIGINAL ARTICLE Timeless evolution of walking and pace strategy of women’s race walking ORIGINAL ARTICLE race walking Panagiotα FitiliABCDE, Vasilios GiovanisABCDE National and Kapodistrian University of Athens, Greece Authors’ Contribution: A – Study design; B – Data collection; C – Statistical analysis; D – Manuscript Preparation; E – Funds Collection. Abstract Purpose: The purpose of this research was to study the timeline evolution of walking, as well as the Pacing Strategy Profiles of high-level women in the 20 km of race walking. Material: The practical example of applying the theoretical basis was made during the Women’s Greek Championship (Megara 2016), in which 12 athletes aged 19 to 40 participated (28.50 ± 7.20). The certified distance of the 20km route was divided into 10 sections of 2 km each. The same happened with the times (intermediate, final) corresponding to the individual sections (2 km) of the route. The athletes were divided into 4 groups: the first 3, those who finished 15% slower than the first, those who finished 15% - 30% slower, and those who finished more than 30% slower than the winner. Finally became comparison of the first 6 and last 6 athletes’ groups. Results: The individual pace strategies that describe the tactics of the athletes in this race have been calculated. It was found that the winners of the race used Even Pacing Strategy, maintaining a steady speed on most of the route. As the level of women athletes became lower, Variable Pacing Strategy was used, while the athletes who finished last did not seem to be able to maintain any particular pacing strategy. Conclusions: It is suggested that athletes should follow Even Pacing Strategy during the race in order to improve their performance. Keywords: race walking, even pacing strategy, variable pacing strategy, performance. Introduction1 The most important athlete of race walking was the Polish Robert Korzeniowski. He has had gain four gold Olympic medals (3 in 50 km and 1 in 20 km), and four medals in World Championships (3 gold and 1 bronze, all in 50 km). The basic rules of race walking:l Various activities that are repeated on a daily basis put the human body on a motor alert, whether it is for social, working or sporting purposes. Beginning of walking from stillness is considered to be one of the most common activities of a person in its daily living. Depending on the purpose and the method, we can divide it as follows: Race walking is carried out on flat public roads. Athletes walk 20 km, repeating 10 times routes of 2 km, or 4 times routes of 5 km. There are rules that differentiate walking from running and athletes of race walking have to know them and follow them. According to Regulations, article 230 IAAF Track and Field Regulations 2012-13 “Race walking is a sequence of steps. The race walker contacts the ground in a way that, no visible (in the human eye) loss occurs. 1) Simple walk or walk is considered as the natural way of moving the human body into space from its infancy [1]. 2) The healing walk, which is the acceptance of exercise as a practical form of rehabilitation, promotion and preservation of health, as it was known by Byzantine medicine [2]. 3) Recreational walk or walk as a recreation. Mannel & Reid, [3] and Sylvester, [4] define three dimensions based on: (a) Leisure as free time, (b) Leisure as an experience, (c) Leisure as an activity. It is therefore understood, that recreation is a broader concept that involves active activity, whereas entertainment involves passive activity [5]. The forward leg should be stretched (not bent at the knee) from the point of first contact with the ground to the vertical upright position”. Judges of race walking watch the athletes during the race and exclude them, if their foot is not stretched on the ground or, if they both lose contact with it. When the judge finds an infringement, he makes his first remark. If the offense is repeated, then the athlete’s warning exclusion is placed on a sign visible to all participants in the race. If three different judges charge the athlete in violation, then exclusion follows. Panagiotα FitiliABCDE, Vasilios GiovanisABCDE National and Kapodistrian University of Athens, Greece Panagiotα FitiliABCDE, Vasilios GiovanisABCDE National and Kapodistrian University of Athens, Greece Panagiotα FitiliABCDE, Vasilios GiovanisABCDE National and Kapodistrian University of Athens, Greece Authors’ Contribution: A – Study design; B – Data collection; C – Statistical analysis; D – Manuscript Preparation; E – Funds Collection. Research Design i Initially, the certified distance of the race track (St- 20km) divided into 10 sections of 2km each was recorded. The same happened with the times (intermediate, final) corresponding to the individual sections (2 km) of the route. Based on the data of the individual track distances and the respective times of the athletes, the individual pace strategies were found that describe the athletes’ tactics in this race. Initially, the certified distance of the race track (St- 20km) divided into 10 sections of 2km each was recorded. i The appliances that were used to perform the measurements and to evaluate the data were: a) One video camera (Sony, Full HD 1080, 50 Hz). The camera’s resolution was 0.02s and it was firmly positioned at the start-stop (where the athletes completed the 2km cycle). The video recording of the athletes’ passes at 2km and recording the electronic timer, [ ] Hypothesis The following research questions will be investigated in the present study: The following research questions will be investigated in the present study: p y a. Will analysing the strategy of the race help coaches and athletes? g (b) The protocols in which the distances of the race and the kinematic parameters were written. b. Do athletes or groups of athletes differ in terms of pace strategy? Statistical Analysis p gy Purpose p gy Purpose The analysis included: The purpose of this research was to study the timeless evolution of walking, as well as the Pacing Strategy Profiles of high-level women in the 20 km of race walking. The importance of the research was significant as follows: the above information would be able to extend theoretical knowledge, so that the methodology of analysing the data of races, that have been or will be conducted in the future, is applied in practice. The purpose of this research was to study the timeless evolution of walking, as well as the Pacing Strategy Profiles of high-level women in the 20 km of race walking. The importance of the research was significant as follows: h b i f i ld b bl d h i l 1. Descriptive statistics: mean (M), standard deviation (SD) and coefficient of variation (V). ( )i ( ) 2. Material and Methods Participants 4. After a detailed explanation of all the terms used for the statistical processing of the work, follows a reference to the t-test. The t - test method investigates the difference between the mean values of a variable at two time points. In other words, it examines whether the difference of two averages is due to random factors. A prerequisite for the above hypothesis to be valid is that the index t to be greater than or equal to the criterion (tc) value of the t-test. The criterion value is derived from the special t-student price table by selecting any level of significance and any degrees of freedom. In this work the t - test calculations were performed with 5% statistical significance and two -sided control, with degrees of freedom N - 1, where N is the sample population. 4. After a detailed explanation of all the terms used for the statistical processing of the work, follows a reference to the t-test. The t - test method investigates the difference between the mean values of a variable at two time points. The practical example of applying the theoretical basis was made during the Women’s Greek Championship (Megara 2016), in which 12 athletes of race walking aged 19 to 40 participated (28.50 ± 7.20). Athletes had experience in endurance training in race walking for at least 5 years [15, 16]. A prerequisite for their participation in the study was their ability to have reached the qualifying thresholds for the Race Walking National Championships. Which means that these athletes had a high level of training experience and endurance [17]. The race course was certified and measured by SEGAS at 20km. Took place on the Megara beach on a public road, and consisted of a 2km circular route, which athletes were required to walk 10 times. Pace strategy in race walking Pace strategy in race walking The observation that athletes’ speed during a race varies caused interest as far as the pace strategy they should follow is concerning. This strategy is a key factor in the success of athletes in sports events [7]. Pace strategy is the ability to regulate the speed of an athlete’s movement in order to reach the end of the race in a shorter The great schools of race walking were created after the war. First the Soviet School, later the Italian, the Spain, the Mexican, the Poland, the German and last the Chinese. The great schools of race walking were created after the war. First the Soviet School, later the Italian, the Spain, the Mexican, the Poland, the German and last the Chinese. 278 2020 06 2020 06 time [8, 9]. The tempo or pace strategy relates to racing: (a) up to 40 sec (sprint), (b)from 40 sec up to a few minutes (short distance), (c) medium and long distance and overtime, which last for hours [10]. Aschenbrenner, Erdmann, Giovanis, & Lipinska, [11] had investigated the tactics and technique of race walking at the 2004 Athens Olympics. Ruchlewicz et al., [12] studied the tactic of race walking, based on measurements made in athletes on a floor meter. It is well known that athletes should not run at high speed early in their race. Often the elite athletes run the second part of a distance faster than the first part [13, 14]. Research Design i Research Design i Pace strategy analysis for the top 3 athletes, those who finished 15% slower than the first, those who finished 15% to 30% slower than that and, finally, those who finished more than 30% slower than the winner [18]. 3. Relation of the times (intermediate and final) of the 12 athletes, as well as the first 6 and last 6 athletes, in relation to the distances of the sections of the track. Results The present study was a targeted review work with a practical example of applying the theoretical basis. The following results expand the theoretical knowledge of women’s pace strategy in 20km of race walking. So that the methodology of analysing the data of races that have been or will be conducted in the future is applied in practice. Independent variables: High-level women athletes in the 20 km race track, initially divided into 4 groups. The first 3, those who finished 15% slower than the first, those who finished 15% - 30% slower and, finally, those who finished more than 30% slower than the winner. Then, they are splitted into two groups: those finishing in the top 6 and bottom 6. Also, the predefined distances of the sections of the route. Figure 1 shows the time course of the athletes in the predetermined intermediate sections of the 20km route of the race, in relation to their ranking. The Table 1 shows the finish times (t) of the leader in each group of athletes of 20 km of race walking, and their relationship (r) with the distance travelled (s) in individual sections of the route. i Dependent variables: The performance of top-level women athletes in the race of 20 km. The individual times of the athletes in the predetermined sections of the track, as well as their pace strategy. For the first athlete (Figure 2) we can see that, there are no significant fluctuations in the intermediate times between her passes. In other words, it is observed that 279 of Physical Culture and Sports PEDAGOGY of Physi and Spo PEDAGOGY she follows Even Pacing Strategy, which concerns the uniform distribution of the expenditure of her forces during the struggle [19]. she follows Even Pacing Strategy, which concerns the uniform distribution of the expenditure of her forces during the struggle [19]. The athletes, who finished in times up to 15% slower than the 1st winner (Figure 3), had fluctuations in the intensity of their effort or rhythm during the race. In the team that finished in times up to 15-30% slower than the 1st winner (Figure 4) you can see that the fatigue comes much faster than in the previous group. In other words, their intermediate times, along the route, vary from passage to passage, either increasing or decreasing significantly. These athletes display Variable Pacing Strategy [19, 7].i Table 1. Results Finish times (t) of the leader in each group of athletes of 20 km of the race walking and their relation (r) with the distance travelled (s) in individual sections of the route. Athletes Time(M) r = distance (s) - time (t) Winner 1.29.35 0,44 Group< 15% 1.37.14 0,50 Group 15 - 30% 1.43.36 0,49 Group 30% 1.59.54 0,47 Group 100% (12) 1.47.83 0,55 As for the athletes who finished in times over 30% slower than the first winner (Figure 5), their common tactics are obvious up to the 14th kilometre. But from then on, with the effect of fatigue, we notice a big difference in times of this group in the last kilometres. The athletes don’t seem to have used a specific pace strategy; they just tried to keep a steady pace all the way, which they didn’t manage. The difference in capacity between this group and the previous ones is obvious. Since we are talking about athletes who finished in the last positions of the race In other words, the first winner tried to keep constant the time in the individual passes, throughout the race, in order to run at the same pace. Figure 1. An intermediate time in the respective sections (F.i) of women’s 20km of race walking in relation to their classification position. Figure 1. An intermediate time in the respective sections (F.i) of women’s 20km of race walking in relation to their classification position. mediate time in the respective sections (F.i) of women’s 20km of race walking in relation to their tion. Figure 2. The individual times per 2km of the winners in the 20km women’s race walking. Figure 2. The individual times per 2km of the winners in the 20km women’s race walking. Figure 2. The individual times per 2km of the winners in the 20km women’s race walking. 280 2020 06 2020 06 Figure 3. The individual times per 2km of the group<15% of the athletes in the 20km of race walking Figure 4. The individual times per 2km of the group 15-30%, of the athletes in the 20km of race walking. Figure 3. The individual times per 2km of the group<15% of the athletes in the 20km of race walking Figure 3. The individual times per 2km of the group<15% of the athletes in the 20km of race walking 281 Figure 3. Results The individual times per 2km of the group<15% of the athletes in the 20km of race walking Figure 4. The individual times per 2km of the group 15-30%, of the athletes in the 20km of race walking. Figure 5. The individual times per 2km of the group> 30% of the athletes in the 20km of race walking. Figure 3. The individual times per 2km of the group<15% of the athletes in the 20km of race walking The individual times per 2km of the group<15% of the athletes in the 20km of race walking Figure 4. The individual times per 2km of the group 15-30%, of the athletes in the 20km of race walking. gure 4. The individual times per 2km of the group 15-30%, of the athletes in the 20km of race walking. Figure 4. The individual times per 2km of the group 15-30%, of the athletes in the 20km of race wal Figure 5. The individual times per 2km of the group> 30% of the athletes in the 20km of race walking. Figure 5. The individual times per 2km of the group> 30% of the athletes in the 20km of race walking. 281 of Physical Culture and Sports PEDAGOGY (Figure 6), and that was the reason that the fatigue came earlier than the previous groups [19].i athletes M1-6 (9.69 ± 0.48 min) and that of 6 last athletes M7-12 (11.92 ± 0.97 min), with a correlation r = 0.97. Showed the difference in performance between the two groups of athletes (Figure 7) t = 6,255> tc = 2,179 with bilateral control ( p <0.05). It was found that none of the Correlation coefficients (r) in relation to the performance of the women’s teams were significant (Table 2). The t-test between the team of the first 6 Figure 6. The individual times per 2km of the four groups in the 20km women’s race walking. Figure 6. The individual times per 2km of the four groups in the 20km women’s race walking. 282 Table 2. The correlation coefficients (r) in relation to the performance of the groups of walkers. r The 3 winners <15% slower 15-30% slower >30% slower 12 athletes The 3 winners Χ 0,94 0,90 0,98 0,95 < 15% slower Χ 0,97 0,92 0,98 15-30% slower Χ 0,89 0,98 > 30% slower Χ 0,95 12 athletes Χ Figure 7. Results Comparison of the mean values (M) of the performance between the group of the first 6 athletes M1-6 (9.69 ± 0.48) and the last 6 athletes M7-12 (11.92 ± 0.97), with correlation r = 0.97 where the t-test showed the difference in performance between the two groups of athletes t = 6.25>tc = 2,179 with bilateral control (p <0.05). Table 2. The correlation coefficients (r) in relation to the performance of the groups of walkers. Table 2. The correlation coefficients (r) in relation to the performance of the groups of walkers. le 2. The correlation coefficients (r) in relation to the performance of the groups of walkers. r The 3 winners <15% slower 15-30% slower >30% slower 12 athletes The 3 winners Χ 0,94 0,90 0,98 0,95 < 15% slower Χ 0,97 0,92 0,98 15-30% slower Χ 0,89 0,98 > 30% slower Χ 0,95 12 athletes Χ Figure 7. Comparison of the mean values (M) of the performance between the group of the first 6 athletes M1-6 (9.69 ± 0.48) and the last 6 athletes M7-12 (11.92 ± 0.97), with correlation r = 0.97 where the t-test showed the difference in performance between the two groups of athletes t = 6.25>tc = 2,179 with bilateral control (p <0.05). 282 2020 06 2020 06 slower than the winner, show Variable Pacing Strategy. These athletes had greater fluctuations in the intensity of their effort, or rhythm, during the race. The pace strategy in race walking, as in all long-distance roads, must be accompanied by the specialized technique [19]. It was found that, the optimal deviation of the speed from the average speed improves the final performance of the athletes, which was expected according to previous studies conducted in this subject [8, 9]. athletes who belonged to the first group, who were the winners, started the race faster than their personal record. The athletes of the last group started the race faster than their own performance [19]. The fact that a large percentage of athletes that did not finish in the first places, start faster than the ones that followed. Leads us to the conclusion that, these athletes started the race seduced by how relaxed they felt at that moment, not having the perception of impending fatigue. Conclusions As we can observe by studying the profiles of the pace strategy at 20km of race walking in the practical example of application of the theoretical basis, there are differences from athlete to athlete. The winners (the first three athletes) seem to follow the Even Pacing Strategy, which deals with the uniform distribution of the cost expenditure of their forces during the race. In other words, these athletes tried to maintain a constant passage time, in each circular route of 2km and a small difference of their individual speeds between their passes [19]. The lower this deviation, the better the performance of the athletes. Leads us to the conclusion that, the optimal tactic at this distance is the constant passes of the athletes at a speed equal to the average speed. With the help of this study, we obtained the following important information: that the winners of the race used Even Pace Strategy, maintaining a constant speed for most of the route at the 20 km of race walking. We also concluded that, the groups of athletes tested differ from each other in terms of pace strategy. As the level of the athletes decreases, the Variable Pace Strategy was used, while the athletes, who finished in the last positions, do not seem to have managed to maintain any particular pace strategy. It is therefore recommended that athletes should design with their coaches the model of the pace strategy they wish to follow. Not be carried away by the momentary latent sense of relaxation, they have at the beginning of the race. The application by the coaches of the relevant training protocols to their athletes, will significantly contribute to the improvement of their final performance. After all, the tactics of walking, like all long-distance roads, must be accompanied by specialized technique and speed distribution [19]. These findings are in line with previous studies that have analysed the path of athletes in the marathon. They found that the change in speed was less for the best runners compared to the slower athletes [20, 19]. Athletes, who finished up to 30% Conflict of interest Authors declare no conflict of interest. Conflict of interest Technique analysis and the effects of 283 of Physical Culture and Sports PEDAGOGY of Phy and S PEDAGOGY PEDAGOGY fatigue. New Studies in Athletics, 2008; 23(4): 17–25. i 18. Hanley B. An analysis of pacing profiles of world- class racewalkers. International Journal of Sports Physiology and Performance, 2013; 8(4), 435–441. https://doi.org/10.1123/ijspp.8.4.435 16. Radovanovic D. Specific alterations of physiological parameters in competitive race walkers. Aplied Physiology, 2011; 98 (4): 449–455. https://doi.org/10.1556/APhysiol.98.2011.4.8l 19. Hanley B, Stellingwerff T, Hettinga FJ. Successful Pacing Profiles of Olympic and IAAF World Championship Middle-Distance Runners Across Qualifying Rounds and Finals. International Journal of Sports Physiology and Performance 2019;14:894–901. https://doi.org/10.1123/ijspp.2018-0742 17. Chwala W. Influence of space-time quantities on the oscillations of the total body mass center (BMC) as a function of the increasing speed of sports walk in the group of master class walkers. In: Urbanik Cz, Mastalerz A. (ed). Biomechanika sportu i rechabilitacji [Biomechanics of sport and rehabilitation]. Warsaw; AWF; 2009. P. 29–43. (in Polish) 20. Giovanis V, Erdmann WS. Kinematic Analysis of Runners in the 2011 Olympus Marathon. Research Journal of Physical Education Sciences, 2013; 1(1): 7–12. Conflict of interest 9. Foster C, Hettinga F, Lampen J, Dodge C, Bobbert M, Porcari J. Effect of competitive distance on energy expenditure during simulated competition. International Journal of Sports Medicine, 2004; 25 (3): 198–204. 1. Benakis Α. The technique of racing walking. [dissertation]. School of Physical Education and Sports Science, National and Kapodistrian University of Athens; 1988. (in Greek) 10. Thompson KG. Pacing: Individual strategies for optimal performance. Human Kinetics; 2014. 2. Stavrakakis N, Albanidis E. The therapeutic use of sport during the byzantine period. Archives of Hellenic Medicine, 2015; 32(1):96–101 (in Greek) 11. Aschenbrenner P, Erdmann WS, Giovanis V, Lipinska P. Investigations on technique and tactics of race walking during Olympic Games Athens 2004 - first announcement. University of Salzburg; 2006. 3. Mannell R, Reid D. Work and leisure. In: Leisure Studies, Prospects for the Twenty-first century, 1999. P. 151–163. 4. Sylvester C. The western idea of work and leisure: traditions, transformations and the future. In: Leisure Studies, Prospects for the Twenty-first century; 1999. P. 17–31. 12. Ruchlewicz,T, Staszkiewicz R, Chwala W, Laska J. Biomechanical parameters of race walking on the example of tests of an international champion class walkers. In: Urbanik Cz (ed.). Zagadnienia biomechaniki sportu –technika ruchu [Biomechanics of sport issues - movement technique]. Warsaw; 2003. P.46–57 (in Polish) 5. Giovanis V. Skiing technique. Athens: Elvekalt Publications; 2006 (in Greek) 6. Panou HA, Giovanis VF. Review on interventional “nordic walking” exercise programs for improving life quality for older adults. Journal of Educational Research and Studies, 2016; 4 (3): 26–29. 13. Erdmann WS, Giovanis V. Investigations on kinematics of giant slalom’s tactics in alpine skiing. In: Miyashita M., Fukunaga T. (ed.) Proceedings, XVI Congress of the International Society of Biomechanics, Tokyo: University of Tokyo; 1997. P. 79. 7. Abbiss CR, Laursen PB. Describing and understanding pacing strategies during athletic competition. Journal of Sports Medicine, 2008; 38(3): 239–252. https://doi.org/10.2165/00007256-200838030-00004 14. Lipińska P. Wielkości kinematyczne i geometria trasy a taktyka biegu w maratonie [Kinematic quantities and route geometry and tactics of running a marathon]. [dissertation]. Gdańsk: AWFiS; 2006 (in Polish) p g 8. Foster C, Snyder A, Thompson NN, Green MA, Foley M, Schrager M. Effect of pacing strategy on cycle time trial performance. Medicine and Science in Sports and Exercise, 1993; 25(3): 383–388. https://doi.org/10.1249/00005768-199303000-00014 15. Hanley B, Bissas A, Drake A. Biomechanical analysis of elite race walkers. Information about the authors: Vasilios Giovanis; (Corresponding Author); Ph.D.; http://orcid.org/0000-0003-2511-8286; vgiovan@phed.uoa.gr; School of Physical Education and Sport Science, National and Kapodistrian University of Athens; 41 Ethnikis Antistassis Str., Daphne 17237, Athens, Greece. Panagiotα Fitili; Msc ; https://orcid.org/0000-0001-5682-3047; fytili@hotmail.com; School of Physical Education and Sport Science, National and Kapodistrian University of Athens; 41 Ethnikis Antistassis Str., Daphne 17237, Athens, Greece. Cite this article as: Fitili P, Giovanis V. Timeless evolution of walking and pace strategy of women’s race walking. Pedagogy of Physical Culture and Sports, 2020;24(6):278-284. https://doi org/10 15561/26649837 2020 0601 Cite this article as: Fitili P, Giovanis V. Timeless evolution of walking and pace strategy of women’s race walking. Pedagogy of Physical Culture and Sports, 2020;24(6):278-284. p , ; ( ) https://doi.org/10.15561/26649837.2020.0601 p https://doi.org/10.15561/26649837.2020.0601 This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (http://creativecommons.org/licenses/by/4.0/deed.en). Received: 01.05.2020 Accepted: 12.06.2020; Published: 30.12.2020 Received: 01.05.2020 Accepted: 12.06.2020; Published: 30.12.2020 284 284
https://openalex.org/W2903648471	https://bmchealthservres.biomedcentral.com/track/pdf/10.1186/s12913-018-3781-7	English	null	Identifying MSM-competent physicians in China: a national online cross-sectional survey among physicians who see male HIV/STI patients	BMC health services research	2,018	cc-by	7,471	(2018) 18:964 (2018) 18:964 Zhao et al. BMC Health Services Research (2018) 18:9 https://doi.org/10.1186/s12913-018-3781-7 Open Access Identifying MSM-competent physicians in China: a national online cross-sectional survey among physicians who see male HIV/STI patients Peipei Zhao1, Bolin Cao2, Cedric H. Bien-Gund3, Weiming Tang3,4, Jason J. Ong5,6, Yi Ding1, Weiying Chen1, Joseph D. Tucker3,4,5* and Zhenzhou Luo1* Abstract Background: Men who have sex with men (MSM) are at high risk of human immunodeficiency virus (HIV) infection and sexually transmitted infection (STI) in China. Inadequate clinical services and poor clinical competency among physicians are major barriers to improving the sexual health of MSM. This study aims to understand physician clinical competency in providing MSM health services in China. Methods: We conducted an online cross-sectional survey among Chinese physicians who have seen male patients for STI complaints in the past year. We obtained information on individual demographics, clinical practice, attitudes toward MSM, and interest in contributing to MSM clinical services. We defined an MSM-competent physician as one who asked male patients about sexual orientation, sexual practices, and recommended HIV/ STI testing during a clinic visit. We conducted multivariable logistic regression to identify factors associated with MSM competency. Results: In total, 501 physicians completed the survey. The most common subspecialties were dermatovenereology (33.1%), urology (30.1%), and general medicine (14.4%). Roughly half (n = 267, 53.3%) reported seeing MSM in the past 12 months. Among physicians who saw MSM in the past 12 months, 60.3% (n = 161) met criteria as MSM-competent physicians, and most (n = 234, 87.6%) MSM-competent physicians reported positive or neutral attitudes towards MSM. Over 60% of all physicians were willing to participate in activities for improving MSM services, such as training and being on a list of physicians willing to serve MSM. MSM-competent physicians showed no sociodemographic differences compared with non MSM-competent physicians. MSM-competent physicians were more willing to have their medical institution named on a public clinic list capable of serving MSM (aOR: 1.70, 95%CI: 1.01–2.86) and being on a public physician list capable of serving MSM (aOR: 1.77, 95%CI: 1.03–3.03). Conclusions: MSM-competent physicians included a broad range of individuals that practiced in diverse clinical settings. Most physicians were interested in improving and expanding MSM clinical services, despite having neutral attitudes toward same-sex behavior. Future interventions should focus on developing MSM clinical competency and expanding services that meet the needs of MSM. Keywords: Men who have sex with men, Clinical services, HIV care continuum, HIV/STI testing * Correspondence: jdtucker@med.unc.edu; paulluo9909@163.com 3University of North Carolina Project-China, No. 2 Lujing Road, Guangzhou 510095, China 1Shenzhen Nanshan Center for Chronic Disease Control, Shenzhen, China Full list of author information is available at the end of the article Background prevention, adequate diagnostics and treatment. It pro- vides health care that is sensitive to the special needs of MSM and encourages them to obtain health care ser- vices and disclosure sexual orientation to health pro- viders [24]. Health providers, including physicians and nurses, need education and training to be able to deliver competent services. However, there is a lack of compe- tency training among China’s health providers to provide more appropriate routine care and remove barriers for MSM to get appropriate services [20]. g Globally, men who have sex with men (MSM) are at ele- vated risk of HIV acquisition, other sexually transmitted infections (STIs), substance abuse, and mental health disorders [1–4]. In many low- and middle-income coun- tries (LMICs), MSM face significant barriers to accessing health care [5, 6]. MSM in LMICs report stigma, dis- crimination, and ignorance of health services as major barriers to health care provision [7, 8]. In addition, MSM have reported negative interactions with physi- cians, such as providers who did not ask or disapproved of same-sex practices, did not provide sexual health counseling or HIV/STI testing services, or revealed their sexual orientation to others [9]. Physicians’ clinical competency to serve MSM has been overlooked in HIV and STI prevention programs. In order to inform the development of MSM-competent clinical care, we conducted a survey among Chinese physicians who have provided STI care to men. Given that it is diffi- cult to evaluate the physicians’ competency in a compre- hensive way, our study focused on their clinical behaviors. In this study, we adopted the definition of MSM-compe- tent physicians from other original research as those who self-reported having asked their last male STI patient about sex with other men, anal sex, and recommended both HIV and syphilis testing [25]. We also assessed the physicians’ workplace information, their attitudes towards MSM and their willingness in future training to explore factors that correlate with competency. HIV prevalence among MSM in China has increased from 6.0% in 2010, to 8.0% in 2015 [10]. It was estimated that the lifetime HIV testing for MSM is 47%, and that 18% of them didn’t retrieve their results [11]. The infec- tion risk of STIs was significantly higher in MSM com- pared to the general population [12]. The rapid spread of HIV and STIs among MSM has led to renewed efforts to improve health services for MSM [13]. Background However, Chinese policy towards MSM is ambiguous (no approval, no disapproval and no promotion) and same-sex mar- riage is illegal. In China, there is a lack of laws to protect MSM from discrimination and unfair treatment. Experi- enced and anticipated health discrimination is widely re- ported by MSM and is also a barrier for MSM to use health services (including HIV testing) in China [14, 15]. In one cross-sectional study of MSM in China, only 16.3% had disclosed their sexual orientation to any health care professional [16], lower than in high-income countries [17, 18]. Incompetent health care providers also delay the testing and services of HIV and STIs for MSM. Provider-initiated HIV testing services is an ef- fective way to increase the uptake of HIV testing [19]. But in China, STI care providers offered HIV testing to only 28.4% of their patients [20]. In addition to provid- ing HIV/STI testing services, physicians serving MSM are responsible for obtaining a thorough sexual history, discussing sexual practices, partners, and conducting an appropriate physical examination [15, 21]. But these ser- vices were not common practice among Chinese physi- cians [20]. In qualitative studies, MSM in China have indicated the importance of knowledgeable, confidential, gay-friendly physicians in accessing sexual health ser- vices [22]. Furthermore, physicians need the material re- sources necessary to provide appropriate care for MSM, such as condoms, lubricants, and sexual health informa- tion for patients [23]. © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Zhao et al. BMC Health Services Research (2018) 18:964 Zhao et al. BMC Health Services Research (2018) 18:964 Page 2 of 9 Page 2 of 9 Study design and participants We conducted a cross-sectional, nationwide online sur- vey of physicians who had seen male STI patients in China in the past twelve months. Physicians were re- cruited through the Xingren medical mobile phone ap- plication (app), where licensed physicians can register as online physicians on the app. The app allows registered physicians to communicate with patients, manage cases, and share medical knowledge and experiences with other physicians. Verified physicians in the app can get paid from their online encounters with patients by providing medical advice to patients. App-based counseling typic- ally involves obtaining a clinical history, but without providing physical examination. The survey link with an invitation message was sent to a subset of Xingren verified physicians who registered in dermatovenereology, urology, proctology, internal medi- cine, pediatrics, infectious disease, and general practice specialties (n = 120,126). Based on the profile of the reg- istered physicians in the app, they were located all over mainland China. Physicians were included from both hospitals and primary care centers. In order to be eli- gible for the survey, physicians needed to live in China, have used the app within the last six months, and have seen at least one male STI patient offline in the past 12 months. The physicians registered in the Xingren app all had practice licenses in China. Male STI patients were MSM-competent health services are seen as a key component in HIV and STI prevention programs, which offers MSM comprehensive sexual health services in- cluding health promotion, counselling, peer support, Zhao et al. BMC Health Services Research (2018) 18:964 Page 3 of 9 Page 3 of 9 male STI patient about sex with other men, anal sex, and recommended both HIV and syphilis testing [25]. Although this is not a comprehensive benchmark for MSM clinical competence, it is consistent with previous literature on MSM health care [28–30]. In our analysis about MSM –competent physicians, we excluded those who have not seen MSM patients in the past 12 months. defined as those who were diagnosed with at least one type of STI or had symptoms of STIs when they saw physicians. STIs include HIV, syphilis, chlamydia, gonor- rhoea, human papillomavirus, herpes, trichomonas, and mycoplasma [26]. We pre-specified the recruitment sample as 500 par- ticipants. Based on previous research in China [27] we recruited a sample size of 500 physicians. Interest in contributing to MSM clinical services g Physicians’ interest in contributing to quality improve- ment programs for MSM clinical services were also measured with three items. Physicians were asked whether they were interested in having their medical institution named on a public clinic list capable of serving MSM. In addition, physicians were asked about being included on a public physician list cap- able of serving MSM. They were also asked if they were interested in obtaining further training focused on clinical services for MSM. Attitudes toward MSM Attitudes toward MSM Physician attitudes toward MSM were measured by ask- ing participants about their agreement with five state- ments. The 5-item scale was adapted from Herek’s Attitudes Toward Lesbians and Gay Men Scale [31], which has been validated in China [32]. Cronbach’s alpha value of the Chinese scale is 0.90 and confirmatory factor analysis demonstrates that it has good validity. The scale includes the following statements: “Male ho- mosexuals are disgusting;” “Male homosexuality is a per- version;” “Homosexual behavior between two men is just plain wrong;” “Male homosexuality is merely a different kind of lifestyle that should not be condemned;” and “Homosexual should be segregated by society (residen- tial segregation or occupational segregation).” The sur- vey used a scale of 1 to 5 (1 = strongly disagree; 2 = disagree somewhat; 3 = neither agree nor disagree; 4 = agree somewhat; 5 = strongly agree).For each statement, 1 and 2 indicated positive attitudes; 3 indicate neutral at- titudes; 4 and 5 indicate negative attitudes. Then items were summed to create a total score. Scores range from 5 (extremely positive attitude) to 25 (extremely negative attitude), with a value of 15 considered neutral. Those who scored less than 15 presented positive attitudes and more than 15 presented negative attitudes. Survey Eligible physicians completed a survey that included demographic and workplace information, clinical prac- tice information, attitudes toward MSM, and interest in contributing to MSM clinic services. Demographic and workplace information Demographic and workplace information We asked physicians their age, sex (male/female), high- est educational degree (Associate/Bachelor/Master/ PhD), and medical specialty. We also asked about the medical institution where physicians worked, including the level of care (primary, secondary or tertiary) and type of medical institution (public or private). The higher level of care means larger service coverage, more health functions and more medical practitioners. Higher level medical institution provided more comprehensive ser- vices. We asked for clinic information on medical equip- ment including the availability of proctoscopes or anoscopes (Yes/No), condoms and lubricants (Yes/No), and HIV/STI prevention pamphlets and educational ma- terials (Yes/No). Study design and participants This sample size was also limited by funding constraints as we pro- vided an incentive for physicians to participate. We im- plemented a recruitment strategy in two steps. First, we sent the participation link to physicians whose unique Xingren identification number ended in 1, 2, or 3 (n = 6689). In order to meet the pre-specified sample size, we then sent the participation link to an additional 1400 physicians who were selected randomly from the rest of the eligible physicians. We used Randbetween function in Microsoft Excel to ensure that a random sample was invited. Physicians initiated the survey by clicking on the participation link. Participants were asked to sign an electronic consent form before they started the survey. We provided each participant with a small financial in- centive for completion of the survey (~$4.50 USD). Physician clinical practice In their last clinical encounter with a male STI patient, 75.4% (n = 378) of physicians reported asking about sex with other men. Over half (n = 277, 55.3%) asked about anal sex, and 90.8% (n = 455) recommended both HIV and syphilis testing. Results A total of 8089 physicians received the participation link, and 1556 physicians clicked the survey link and entered the survey, of which 699 provided informed consent. Among physicians who consented, 186 did not meet eli- gibility criteria, and another 12 surveys were invalidated due to insufficient information. A total of 501 physicians were included in the final analysis. Demographic characteristics and institutional information On average, physicians in our survey were 37.6 ± 8.2 years old (Table 1). Three-quarters of them were male (n = 367, 75.0%), and over half of the physicians had ob- tained a master’s degree or higher (n = 276, 53.3%). One third of physicians reported specialization in dermatove- nerology (n = 166, 33.1%), followed by urology (n = 151, 30.1%), and internal medicine (n = 51, 10.2%). Almost two-thirds of the physicians worked in tertiary care hospitals (n = 322, 64.3%) and 89.6% (n = 449) worked in public medical institutions. Most physicians reported that proctoscopes or anoscopes were available in their medical facilities (n = 403, 80.4%), 75.2% (n = 377) re- ported STI prevention pamphlets or educational materials were available, and 51.9% (n = 260) reported free condom and lubricants were available on site. Clinical practice information In the multivariable logistic regression, we ad- justed for demographic characteristics (age, sex, and education) and reported adjusted odds ratios (aOR). Statistical significance was determined at P < 0.05 in the model. Clinical practice information Statistical analysis IBM SPSS Statistics 19 was used for all analyses. Descriptive statistics were used to describe physicians’ socio-demographic information, institutional information, interests in contributing to the MSM clin- ical services and attitudes towards MSM. The primary outcome of the study is MSM-competent physicians, de- fined by physicians’ clinical practice with their last STI patients. Bivariate logistic regression was used to Physicians were asked three items about their clinical practice during the last time they saw a male STI pa- tient. These items included whether they asked about sex with other men, whether they asked about anal sex, and whether they recommended both HIV and syphilis testing. In this study, we defined MSM-competent physi- cians as those who self-reported having asked their last Zhao et al. BMC Health Services Research (2018) 18:964 Page 4 of 9 Table 1 Demographic characteristics of physicians who saw at least one male STI patient, 2017 (N = 501) Characteristics Total n % Age (years) Mean: 37.6 ± 8.2; Min: 23; Max: 76 ≤30 110 22.0% 31–40 244 48.7% > 40 147 29.3% Sex Male 376 75.0% Female 125 25.0% Education Associate’s degreea 36 7.2% Bachelor’s degree 198 39.5% Master’s degree 216 43.1% PhD degree 51 10.2% Specialty Dermatovenerology 166 33.1% Urology 151 30.1% General medicineb 72 14.4% Proctology 41 8.2% Othersc 37 7.4% Infectious disease 34 6.8% Level of care Primary 34 6.8% Secondary 145 28.9% Tertiary 322 64.3% Type of Medical institute Public 449 89.6% Private 52 10.4% Proctoscope or anoscope available Yes 403 80.4% No 98 19.6% Free condom and lubricants available Yes 260 51.9% No 241 48.1% STI prevention pamphlets or educational materials available Yes 377 75.2% No 124 24.8% Had seen MSM STI patients in the last 12 months Yes 267 53.3% No 234 46.7% Asked about sex with other men when seeing last patient Yes 378 75.4% N 123 24 6% Table 1 Demographic characteristics of physicians who saw at least one male STI patient, 2017 (N = 501) examine factors associated with being MSM-competent physicians. Demographic characteristics, workplace in- formation, attitudes towards MSM and interested in contributing to MSM services are examined as inde- pendent variables. The results were reported as odds ra- tios (OR) with corresponding 95% confidence intervals (95%CI). Variables that were significant in the bivariate logistic regression were included in the multivariable analysis. Discussion This study explores clinical competency and attitudes to- ward MSM among an online sample of physicians in China. This study contributes to the limited literature on clinical services for MSM in China by examining clinical competency in serving MSM patients and attitudes among physicians who see male STI patients. We found that most physicians reported asking about sexual orien- tation, recommending STI testing, and asking about anal intercourse in their last clinical encounters. We also found that most physicians were willing to attend MSM-focused clinical training and openly identify as physicians willing to see MSM. Physician attitudes to MSM , ; dSTI testing means both HIV testing and Syphilis testing third (n = 77, 29.2%) had neutral attitudes, and 11.6% (n = 33) had negative attitudes toward MSM. Physician attitudes to MSM Asked about sex with other men when seeing last patient When presented with individual statements regarding MSM (Table 2), most physicians indicated they had neu- tral attitudes toward MSM. Nonetheless, the total scores of the scale showed most (n = 158, 59.2%) physicians surveyed had a positive attitude towards MSM. Nearly a Page 5 of 9 Zhao et al. BMC Health Services Research (2018) 18:964 Page 5 of 9 third (n = 77, 29.2%) had neutral attitudes, and 11.6% ( 33) h d i i d d MSM Table 1 Demographic characteristics of physicians who saw at least one male STI patient, 2017 (N = 501) (Continued) Characteristics Total n % Ask about anal sex when seeing last patient Yes 277 55.3% No 224 44.7% Recommended STI testingd when seeing last patient Yes 455 90.8% No 46 9.2% aAssociate’s degree is usually earned in two years or more and can be attained at community colleges, technical colleges, vocational schools, and some colleges; bGeneral medicine includes internal medicine, pediatrics and general practice; cOthers include Reproductive Medicine, Andrology, Emergence clinic, Hematology, Professional Health, AIDS Prevention Office; dSTI testing means both HIV testing and Syphilis testing Table 1 Demographic characteristics of physicians who saw at least one male STI patient, 2017 (N = 501) (Continued) Characteristics Total n % Ask about anal sex when seeing last patient Yes 277 55.3% No 224 44.7% Recommended STI testingd when seeing last patient Yes 455 90.8% No 46 9.2% a Table 1 Demographic characteristics of physicians who saw at least one male STI patient, 2017 (N = 501) (Continued) institution, there was no difference between MSM- competent and non-MSM-competent physicians. We did not observe any correlation between attitudes toward MSM and MSM-competency among physicians in our sample. Medical institutions where free condoms and lubricants were available were more likely to have MSM-competent physicians (aOR = 2.01, 95%CI: 1.21–3.34). This was true for medical institutions with STI pamphlets and educa- tional material to have more MSM-competent physicians working there. Physicians who were more interested in contributing to activities to improve MSM clinical services were more likely to be MSM-competent. Activities in- cluded having their medical institution names on a public clinic list capable of serving MSM (aOR: 1.70, 95%CI: 1.01–2.86), and being on a public physician list capable of serving MSM (aOR: 1.77, 95%CI: 1.03–3.03). Physician interest in contributing to improving medical services for MSM When asked about interest in contributing to MSM clinical services, 59.5% (n = 298) reported they were interested in having their medical institution names on a public clinic list capable of serving MSM, 61.3% (n = 307) reported they were interested being on a public physician list capable of seeing MSM, and 68.1% (n = 341) of physicians were interested in fur- ther training focused on clinical services for MSM. Factors associated with being an MSM-competent physician Over 60% of physicians who reported seeing MSM STI patients met basic criteria for MSM competency. Our re- sults showed that the percentage of asking about same-sex sexual practices was higher than a previous study in China, which reported only 11% of the physicians asked their patients occasionally or most of the time [20]. The percentage of physicians asking about sexual practices is lower than that reported in high-income country studies focused on physicians who see patients with sexual health concerns [33, 34]. In our study, nearly all physicians Among the 267 physicians who saw MSM STI patients, 60.3% (n = 161) met criteria as MSM-competent physicians. Among those who did not meet criteria, most (n = 98, 92.5%) failed to ask about anal sex with their patients. There were statistically significant differences between MSM-competent group with the non MSM- competent group in terms of age, sex, education, and spe- cialty (Table 3). As for the level and type of the medical Table 2 Physicians’ attitudes towards male homosexuals in China, 2017 (N = 267) Positive Attitudes Neutral Attitudes Negative Attitudes Statements n (%) n (%) n (%) Overall 158(59.2%) 78(29.2%) 31(11.6%) Male homosexuals are disgusting 60(22.5%) 180(67.4%) 37(10.1%) Male homosexuality is a perversion 77(28.8%) 166(62.2%) 24(9.0%) Homosexual behavior between two men is just plain wrong 73(27.3%) 155(58.1%) 39(14.6%) Male homosexuality is merely a different kind of lifestyle that should not be condemned 113(42.3%) 124(46.5%) 30(11.2%) Male homosexuals should be segregated by society (residential segregation; occupational segregation) 147(55.1%) 113(42.3%) 7(2.6%) Table 2 Physicians’ attitudes towards male homosexuals in China, 2017 (N = 267) Zhao et al. Factors associated with being an MSM-competent physician BMC Health Services Research (2018) 18:964 Page 6 of 9 Table 3 Factors associated with being a MSM-competent physician in China, 2017 (N = 267) Overalla (n = 267) MSM-competent physicians (n = 161) Non MSM-competent physicians (n = 106) OR (95%CI) aORb (95%CI) n (%) n (%) n (%) Age (years) ≤30 55(20.6%) 27(16.8%) 28(26.4%) Ref 30–44 127(47.6%) 79(49.1%) 48(45.3%) 1.71(0.90–3.23) > 44 85(31.8%) 55(34.2%) 30(28.3%) 1.90(0.95–3.79) Sex Male 195(73.0%) 121(75.2%) 74(69.8%) 1.31(0.76–2.26) Female 72(27.0%) 40(24.8%) 32(30.2%) Ref Education Associate’s degreec 15(5.6%) 10(6.2%) 5(4.7%) 2.00(0.56–7.10) Bachelor’s degree 98(36.7%) 62(38.5%) 36(34.0%) 1.72(0.78–3.79) Master’s degree 120(44.9%) 72(44.7%) 48(45.3%) 1.50(0.70–3.22) PhD degree 34(12.7%) 17(10.6%) 17(16.0%) Ref Specialty Dermatovenerology 109(40.8%) 65(40.4%) 44(42.3%) 1.85(0.47–7.26) Urology 63(23.6%) 41(25.5%) 22(20.8%) 2.33(0.57–9.57) Proctology 27(10.1%) 19(11.8%) 8(7.5%) 2.97(0.63–14.03) General medicined 38(14.2%) 16(9.9%) 22(20.8%) 0.91(0.21–3.93) Infectious Disease 21(7.9%) 16(9.9%) 5(4.7%) 4.00(0.77–20.92) Otherse 9(3.4%) 4(2.5%) 5(4.7%) Ref Level of care Primary 13(4.9%) 8(5.0%) 5(4.7%) Ref Secondary 65(24.3%) 33(20.5%) 32(30.2%) 0.65(0.19–2.18) Tertiary 189(70.8%) 120(74.5%) 69(65.1%) 1.09(0.34–3.45) Type of Medical institute Public 238(89.1%) 145(90.1%) 93(87.7%) 1.27(0.58–2.76) Private 29(10.9%) 16(9.9%) 13(12.3%) Ref Proctoscope or anoscope available Yes 219(82.0%) 138(85.7%) 81(76.4%) 1.85(0.99–3.47) No 48(18.0%) 23(14.3%) 25(23.6%) Ref Free condom and lubricants available Yes 150(56.2%) 101(62.7%) 49(46.2%) 1.96(1.19–3.22) 2.01(1.21–3.34) No 117(43.8%) 60(37.3%) 57(53.8%) Ref Ref STI prevention pamphlets or educational materials available Yes 208(77.9%) 138(85.7%) 70(66.0%) 3.09(1.70–5.61)* 3.10(1.68–5.73)* No 59(22.1%) 23(14.3%) 36(34.0%) Ref Ref Interested in having medical institution name on the public clinic list capable of serving MSM Yes 174(65.2%) 113(70.2%) 61(57.5%) 1.74(1.04–2.90)* 1.70(1.01–2.86)* No 93(34.8%) 48(29.8%) 45(42.5%) Ref Ref Interested in being on a public physician list capable of serving MSM Yes 182(68.2%) 118(73.3%) 64(60.4%) 1.80(1.07–3.04)* 1.77(1.03–3.03)* No 85(31.8%) 43(26.7%) 42(39.6%) Ref Ref Interested in further training focused on clinical services for MSM Table 3 Factors associated with being a MSM-competent physician in China, 2017 (N = 267) Zhao et al. Factors associated with being an MSM-competent physician Although there are no comparable data for HIV testing among HIV/STI patients, research showed that a lower percentage of US physicians reported offering HIV testing. Among general practitioners in the US whose patients were general population, 41.7% of the physicians offered HIV testing [35]. Among HIV care providers whose patients were partly HIV infected, 60% routinely of- fered HIV testing to their patients [36]. Further studies should be conducted to investigate outcomes when physi- cians are seeing general patients and explore factors asso- ciated with patients accepting HIV testing offered by physicians in China. reported recommending HIV/syphilis testing for their last STI patient. Although there are no comparable data for HIV testing among HIV/STI patients, research showed that a lower percentage of US physicians reported offering HIV testing. Among general practitioners in the US whose patients were general population, 41.7% of the physicians offered HIV testing [35]. Among HIV care providers whose patients were partly HIV infected, 60% routinely of- fered HIV testing to their patients [36]. Further studies should be conducted to investigate outcomes when physi- cians are seeing general patients and explore factors asso- ciated with patients accepting HIV testing offered by physicians in China. MSM-competent physicians practiced in a wide range of clinical settings and medical subspecialties. Although some studies have found higher quality STI care in sub- specialist clinics [37, 38], we did not observe significant differences in MSM competency between subspecialists and general practitioners. Additionally, MSM-competent physicians were associated with institutional factors (providing condom, lubricants, STI prevention pam- phlets or educational materials), rather than individual factors. Recommendations of educational materials are recommended in guidelines for HIV prevention and LGBT patients, reinforcing the office-based education effect [39]. Guidelines for care of LGBT patients state that LGBT patients often find clues to determine what information they are comfortable to share with health care providers [40]. The educational and prevention material may have a potential effect on the inter- action between physicians and STI patients. In addition, these materials may be the results of the in- stitutional support for sexual health issues and risk population, which have a positive effect on physicians’ competency. Provision of educational and prevention material can play a useful role in an MSM compe- tency improvement program. We also found that most physicians were willing to re- ceive further training to better serve MSM. Factors associated with being an MSM-competent physician BMC Health Services Research (2018) 18:964 Page 7 of 9 Table 3 Factors associated with being a MSM-competent physician in China, 2017 (N = 267) (Continued) Overalla (n = 267) MSM-competent physicians (n = 161) Non MSM-competent physicians (n = 106) OR (95%CI) aORb (95%CI) n (%) n (%) n (%) Yes 194(72.7%) 123(76.4%) 71(67.0%) 1.60(0.93–2.75) No 73(27.3%) 38(23.6%) 35(33.0%) Ref Physicians’ attitudes towards male homosexual Positive attitude 157(58.8%) 90(63.2%) 67(55.9%) Ref Neutral attitude 77(28.8%) 50(25.5%) 27(31.3%) 1.38(0.78–2.43) Negative attitude 33(12.4%) 21(11.3%) 12(28.8%) 1.30(0.60–2.83) aPhysicians who have not seen MSM patients in the past 12 months were excluded; baOR controlled for age, sex, and education; cAssociate’s degree is usually earned in two years or more and can be attained at community colleges, technical colleges, vocational schools, and some colleges; dGeneral medicine includes internal medicine, pediatrics and general practice; eOthers include Reproductive Medicine, Andrology, Emergence clinic, Hematology, Professional Health, AIDS Prevention Office; * p < 0.05; p ≤0.01; * p ≤0.001 Table 3 Factors associated with being a MSM-competent physician in China, 2017 (N = 267) (Continued) ciated with being a MSM-competent physician in China, 2017 (N = 267) (Continued) aOR controlled for age, sex, and education; cAssociate’s degree is usually earned in two years or more and can be attained at community colleges, technical colleges, vocational schools, and some colleges; dGeneral medicine includes internal medicine, pediatrics and general practice; eOthers include Reproductive Medicine, Andrology, Emergence clinic, Hematology, Professional Health, AIDS Prevention Office; * p < 0.05; p ≤0.01; * p ≤0.001 Physician clinical competency and attitudes are both critical toward engaging MSM in care. In our study, most physicians reported positive attitudes towards MSM, although based on individual statement alone, most physicians reported neutral attitudes. Our finding contrasts with previous Chinese research from 1101 physicians showing negative physician attitudes toward MSM and widespread discrimination [41]. The more in- clusive societal attitudes towards LGBT population showed from a recent national survey may influence health providers’ attitudes [42]. We found no correlation between the physicians’ MSM competency and attitudes towards MSM, indicating the disjunction between clin- ical practice and personal attitudes. However, it is the practical competency and non-judgmental attitude that work together to optimize MSM’s well-being. Physicians in China need both training to improve clinical practice and reduce stigma to better serve the MSM population. reported recommending HIV/syphilis testing for their last STI patient. Competing interests h h d l h g The authors declare that they have no competing interests. Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Abbreviations b l h App: Mobile phone application; HIV: human immunodeficiency virus; LMICs: Low- and middle-income countries; MSM: Men who have sex with men; STI: Sexually transmitted infection 5. Centers for Disease Control and Prevention. HIV and Young Men Who Have Sex With Men. 2014. 6. Wirtz AL, Kamba D, Jumbe V, Trapence G, Gubin R, Umar E, et al. A qualitative assessment of health seeking practices among and provision practices for men who have sex with men in Malawi. BMC Int Health Hum Rights. 2014;14:20. https://doi.org/10.1186/1472-698X-14-20. Author details 1 1Shenzhen Nanshan Center for Chronic Disease Control, Shenzhen, China. 2School of Media and Communication, Shenzhen University, Nanshan Block Huaming Road, 7, Shenzhen 518052, China. 3University of North Carolina Project-China, No. 2 Lujing Road, Guangzhou 510095, China. 4School of Medicine, University of North Carolina at Chapel Hill, North, Carolina, USA. 5Faculty of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine, London, UK. 6Central Clinical School, Monash University, Melbourne, Australia. Factors associated with being an MSM-competent physician Training and education initiatives to improve knowledge, attitudes, and clinical competencies among physicians for serving MSM patients is essential to improving the health of sexual minorities [43]. In LMICs, increased capacity from training in providing non-stigmatizing and non-discriminatory services to MSM patients can im- prove clinical outcomes [44, 45]. The high interest in our study suggests that subsequent MSM training may be feasible. Further research is needed to find the effect of training interventions or programs for physicians to improve their clinical practice and atti- tudes towards MSM. The study has several limitations. First, the study was based on physician self-report, introducing the potential for social desirability and recall bias [46]. Physicians were asked about their last patients to re- duce the likelihood of the recall bias, but responses Zhao et al. BMC Health Services Research (2018) 18:964 Zhao et al. BMC Health Services Research (2018) 18:964 Page 8 of 9 related to competent clinical practice and willingness to contribute to MSM services might be over re- ported due to the social desirability bias. However, self-report surveys have been widely used in the stud- ies with physicians [33, 34, 37, 47], allowing a reason- able comparison to the literature. Second, we used a narrow definition of MSM-competency that focuses on physicians’ clinical practice. Another limitation of the definition is the lack of patients’ voice and experi- ence of the competency. Third, we did not assess whether the last clinical encounter was a new patient or an established patient, which may also influence physician behaviors. For established patient, physicians may have known the sexual history, so they did not need to ask these questions again in their last en- counter. In this case, it would not necessarily indicate they are not competent. Forth, physicians were re- cruited from a medical app online, and there may be selection bias because all survey participants were app users. Fifth, this was a cross-sectional study, so no causal relationship can be established. Sixth, we didn’t obtain the information on how the STIs were con- firmed by physicians. There was a potential for inval- idity of STI diagnosis. References 1 Wh l 1. Wheeler J, Anfinson K, Valvert D, Lungo S. Is violence associated with increased risk behavior among MSM? Evidence from a population-based survey conducted across nine cities in Central America. Glob Health Action. 2014;7:24814. 2. Zahn R, Grosso A, Scheibe A, Bekker LG, Ketende S, Dausab F, et al. Human rights violations among men who have sex with men in southern Africa: comparisons between legal contexts. PLoS One. 2016;11:e0147156. 3. Mgopa LR, Mbwambo J, Likindikoki S, Pallangyo P. Violence and depression among men who have sex with men in Tanzania. BMC Psychiatry. 2017;17:296. 4. Sandfort TGM, Knox JR, Alcala C, El-bassel N, Kuo I, Smith LR. Substance use and Hiv risk among men who have sex with men in Africa: a systematic review. Jaids J Acquir Immune Defic Syndr. 2017;76:e34–46. https://doi.org/ 10.1097/qai.0000000000001462. Authors’ contributions JT and ZL designed the study and survey. PZ and BC conducted the study, collected and interpreted the data. PZ, BC and CB together drafted the manuscript. JT, ZL, WT, JO, YD and WC helped to conduct the survey and participated in critical revision of the manuscript. All authors read and approved the final manuscript. Acknowledgements The authors would like to thank all the participants and express the gratitude to the staff members of Social Entrepreneurship to Spur Health (SESH) for their help in the crowdsourcing. We would also thank Xingren App Company for broadcasting the recruitment advertisement via their platform. 7. Risher K, Adams D, Sithole B, Ketende S, Kennedy C, Mnisi Z, et al. Sexual stigma and discrimination as barriers to seeking appropriate healthcare among men who have sex with men in Swaziland. J Int AIDS Soc. 2013; 16(Suppl 2):18715. 7. Risher K, Adams D, Sithole B, Ketende S, Kennedy C, Mnisi Z, et al. Sexual stigma and discrimination as barriers to seeking appropriate healthcare among men who have sex with men in Swaziland. J Int AIDS Soc. 2013; 16(Suppl 2):18715. 8. Kennedy CE, Baral SD, Fielding-Miller R, Adams D, Dludlu P, Sithole B, et al. “They are human beings, they are Swazi”: intersecting stigmas and the positive health, dignity and prevention needs of HIV-positive men who have sex with men in Swaziland. J Int AIDS Soc. 2013;16(Suppl 3):18749. 9. Kushwaha S, Lalani Y, Maina G, Ogunbajo A, Wilton L, Agyarko-Poku T, et al. “But the moment they find out that you are MSM…”: a qualitative 8. Kennedy CE, Baral SD, Fielding-Miller R, Adams D, Dludlu P, Sithole B, et al. “They are human beings, they are Swazi”: intersecting stigmas and the positive health, dignity and prevention needs of HIV-positive men who have sex with men in Swaziland. J Int AIDS Soc. 2013;16(Suppl 3):18749. 9 Kushwaha S Lalani Y Maina G Ogunbajo A Wilton L Agyarko Poku T et al Consent for publication Not applicable. Consent for publication Not applicable. Ethics approval and consent to participate The institutional review board of Nanshan Center of Chronic Disease Control, Shenzhen, China provided approval for this study. The IRB code of the study is II20170016. All participants signed the electronic consent form before starting the survey. Conclusions This study expands the limited literature on MSM service providers in China, providing a current perspective of MSM-competent physicians. We observed a diverse sam- ple of physicians from a range of subspecialties and med- ical institutions in China meeting criteria for competent clinical care for MSM. Most physicians were interested in the training and improving MSM clinical services. More research on physicians’ clinical competency is strongly rec- ommended to provide a more accurate and comprehen- sive description of MSM-competency. Further research is needed to explore novel methods of engaging physicians in MSM care and developing interventions aimed at phy- sicians in order to improve MSM clinical competency. Received: 13 July 2018 Accepted: 29 November 2018 8. Kennedy CE, Baral SD, Fielding-Miller R, Adams D, Dludlu P, Sithole B, et al. “They are human beings, they are Swazi”: intersecting stigmas and the positive health, dignity and prevention needs of HIV-positive men who have sex with men in Swaziland. J Int AIDS Soc. 2013;16(Suppl 3):18749. 9. Kushwaha S, Lalani Y, Maina G, Ogunbajo A, Wilton L, Agyarko-Poku T, et al. “But the moment they find out that you are MSM…”: a qualitative Funding Thi g This research was supported by Shenzhen Healthcare Research Project (No. SZFZ2017024), the National Institutes of Health (NIAID 1R01AI114310), and the UNC-South China STD Research Training Center (FIC 1D43TW009532). The funders played no role in any part of the study. 9. Kushwaha S, Lalani Y, Maina G, Ogunbajo A, Wilton L, Agyarko-Poku T, et al. “But the moment they find out that you are MSM…”: a qualitative Page 9 of 9 Page 9 of 9 Zhao et al. BMC Health Services Research (2018) 18:964 Page 9 of 9 investigation of HIV prevention experiences among men who have sex with men (MSM) in Ghana’s health care system. BMC Public Health. 2017;17:770. https://doi.org/10.1186/s12889-017-4799-1. 32. Yu Y, Xiao S, Xiang Y. Application and testing the reliability and validity of a modified version of herek’s attitudes toward lesbians and gay men scale in China. J Homosex. 2011;58:263–74. 33. Lanier Y, Castellanos T, Barrow RY, Jordan WC, Caine V, Sutton MY. Brief sexual histories and routine HIV/STD testing by medical providers. AIDS Patient Care STDs. 2014;28:113–20. https://doi.org/10.1089/apc.2013.0328. 10. Tang S, Tang W, Meyers K, Chan P, Chen Z, Tucker JD. HIV epidemiology and responses among men who have sex with men and transgender individuals in China: a scoping review. BMC Infect Dis. 2016;16:588. https:// doi.org/10.1186/s12879-016-1904-5. 34. Wimberly YH, Hogben M, Moore-Ruffin J, Moore SE, Fry-Johnson Y. Sexual history-taking among primary care physicians. J Natl Med Assoc. 2006;98:1924–9 11. Zou H, Hu N, Xin Q, Beck J. Hiv testing among men who have sex with men in China: a systematic review and meta-analysis. AIDS Behav. 2012;16: 1717–28. 35. McNaghten AD, Valverde EE, Blair JM, Johnson CH, Freedman MS, Sullivan PS. Routine HIV testing among providers of HIV Care in the United States, 2009. PLoS One. 2013;8:e51231. 12. Chow EPF, Tucker JD, Wong FY, Nehl EJ, Wang Y, Zhuang X, et al. Disparities and risks of sexually transmissible infections among men who have sex with men in China: a meta-analysis and data synthesis. PLoS One. 2014;9:e89959. https://doi.org/10.1371/journal.pone.0089959. 36. Zielinski M, Leung SJ, Akkaya-hocagil T, Rowe KA, Ortega-peluso C, Smith LC. Correlates of Routine HIV Testing Practices. A Survey of New York State Primary Care Physicians. 2011;2015(68 suppl 1):S21–9. 2014;9:e89959. https://doi.org/10.1371/journal.pone.0089959. 37. Asch SM, Sa’adah MG, Lopez R, Kokkinis A, Richwald GA, Rhew DC. Comparing quality of care for sexually transmitted diseases in specialized and general clinics. Public Health Rep. 2002;117:157–63. 13. Funding Thi 19. Kennedy CE, Fonner VA, Sweat MD, Okero FA, Baggaley R, O’Reilly KR. Provider-initiated HIV testing and counseling in low- and middle-income countries: a systematic review. AIDS Behav. 2013;17:1571–90. 44. Gu J, Lau JTF, Wang Z, Wu AMS, Tan X. Perceived empathy of service providers mediates the association between perceived discrimination and behavioral intention to take up HIV antibody testing again among men who have sex with men. PLoS One. 2015;10:e0117376. https://doi.org/10. 1371/journal.pone.0117376. 20. Tucker JD, Walensky RP, Yang LG, Yang B, Bangsberg DR, Chen XS, et al. Expanding provider-initiated HIV testing at STI clinics in China. AIDS Care. 2012;24:1316–9. https://doi.org/10.1080/09540121.2012.661835. 21. Makadon HJ, Mayer HK, Potter J, Goldhammer H. The Fenway guide to lesbian , gay , bisexual , and Transgender Health; 2015. 45. Micheni M, Kombo BK, Secor A, Simoni JM, Operario D, van der Elst EM, et al. Health provider views on improving antiretroviral therapy adherence among men who have sex with men in coastal Kenya. AIDS Patient Care STDs. 2017;31:113–21. https://doi.org/10.1089/apc.2016.0213. 22. Bien CH, Muessig KE, Lee R, Lo EJ, Yang LG, Yang B, et al. HIV and syphilis testing preferences among men who have sex with men in South China: a qualitative analysis to inform sexual health services. PLoS One. 2015;10: e0124161. https://doi.org/10.1371/journal.pone.0124161. 46. Montaño DE, Phillips WR. Cancer screening by primary care physicians: a comparison of rates obtained from physician self-report, patient survey, and chart audit. Am J Public Health. 1995;85:795–800. https://doi.org/10.2105/AJPH.85.6.795. 46. Montaño DE, Phillips WR. Cancer screening by primary care physicians: a comparison of rates obtained from physician self-report, patient survey, and chart audit. Am J Public Health. 1995;85:795–800. https://doi.org/10.2105/AJPH.85.6.795. 23. Taegtmeyer M, Davies A, Mwangome M, van der Elst EM, Graham SM, Price MA, et al. Challenges in providing counselling to MSM in highly stigmatized contexts: results of a qualitative study from Kenya. PLoS One. 2013;8:e64527. 47. St. Lawrence JS, Montaño DE, Kasprzyk D, Phillips WR, Armstrong K, Leichliter JS. STD screening, testing, case reporting, and clinical and partner notification practices: a national survey of US physicians. Am J Public Health. 2002;92:1784–8. 24. European Centre for Disease Prevention and Control. HIV and STI prevention among men who have sex with men. European Centre for Disease Prevention and Control; 2015. https://doi.org/10.2900/66666. 25. Watson J, Tang W, Pan SW, Wu DD, Zhao P, Cao B, et al. Out of the closet, into the clinic: opportunities for expanding MSM-competent Services in China. Funding Thi Wu Z, Xu J, Liu E, Mao Y, Xiao Y, Sun X, et al. HIV and syphilis prevalence among men who have sex with men: a cross-sectional survey of 61 cities in China. Clin Infect Dis. 2013;57:298–309. 38. Voeten HA, Otido JM, O’Hara HB, Kuperus AG, Borsboom GJ, Ndinya-Achola JO, et al. Quality of sexually transmitted disease case management in Nairobi, Kenya: a comparison among different types of healthcare facilities. Sex Transm Dis. 2001;28:633–42 http://ovidsp.ovid.com/ovidweb.cgi?T= JS&PAGE=reference&D=med4&NEWS=N&AN=11677385. 14. Liu Y, Sun X, Qian H-Z, Yin L, Yan Z, Wang L, et al. Qualitative assessment of barriers and facilitators of access to HIV testing among men who have sex with men in China. AIDS Patient Care STDs. 2015;29:481–9. https://doi.org/ 10.1089/apc.2015.0083. 15. Feng Y, Wu Z, Detels R. Evolution of men who have sex with men community and experienced stigma among men who have sex with men in Chengdu, China. J Acquir Immune Defic Syndr. 2010;53(Suppl 1):S98–103. https://doi.org/10.1097/QAI.0b013e3181c7df71. 39. Workowski KA, Bolan GA. Sexually Transmitted Diseases Treatment Guidelines. MMWR Recomm reports. 2015;64(RR3):1–137. https://doi.org/10. 1097/00008480-200308000-00006. 40. Gay and Lesbian Medical Association. Guidelines for Care of Lesbian, Gay, Bisexual, and transgender patients. Environment. 2006. http://www.qahc.org. au/files/shared/docs/GLMA_guide.pdf. 16. Tang W, Mao J, Tang S, Liu C, Mollan K, Cao B, et al. Disclosure of sexual orientation to health professionals in China: results from an online cross- sectional study. J Int AIDS Soc. 2017;20:21416. 41. Li L, Liang L-J, Wu Z, Lin C, Wen Y. Individual attitudes and perceived social norms: reports on HIV/ AIDS-related stigma among service providers in China. Int J Psychol. 2009;44:443–50. 17. Stupiansky NW, Liau A, Rosenberger J, Rosenthal SL, Tu W, Xiao S, et al. Young Men’s disclosure of same sex behaviors to healthcare providers and the impact on health: results from a US National Sample of young men who have sex with men. AIDS Patient Care STDs. 2017;31:342–7. https://doi. org/10.1089/apc.2017.0011. 42. United Nation Development Programme. Being LGBTI in China: a National Survey on social attitudes towards sexual orientation, gender identity and gender expression. 2016. http://www.cn.undp.org/content/china/en/home/ library/democratic_governance/being-lgbt-in-china.html. 18. Callander D, Bourne C, Pell C, Finlayson R, Forssman B, Baker D, et al. Recording the sexual orientation of male patients attending general practice. Fam Pract. 2015;32:35–40. 43. Sekoni AO, Gale NK, Manga-Atangana B, Bhadhuri A, Jolly K. The effects of educational curricula and training on LGBT-specific health issues for healthcare students and professionals: a mixed-method systematic review. J Int AIDS Soc. 2017;20:21624. https://doi.org/10.7448/IAS.20.1.21624. Funding Thi Sex Transm Dis. 2018;45:527–33. https://doi.org/10.1097/OLQ. 0000000000000808. 26. Wang Q, Liu Q, Xu J. Sexual Transmitted Disease Clinical Diagnosis and Prevention Guideline. China CDC: National Center for STD control; 2014. 26. Wang Q, Liu Q, Xu J. Sexual Transmitted Disease Clinical Diagnosis and Prevention Guideline. China CDC: National Center for STD control; 2014. 27. Xu Z, Zhao Y, Xu B, Zhu D, Li H. Research on online medical consultation from the Physician’s perspective. Wirel Internet Technol. 2015;23:119–20. 27. Xu Z, Zhao Y, Xu B, Zhu D, Li H. Research on online medical consultation from the Physician’s perspective. Wirel Internet Technol. 2015;23:119–20. 28. Bourne C. Clinical guideline for sexual health Care of men who have sex with men and transgender; 2011. 29. Famliy Health International. Guidelines and Standards for Counseling High- Risk Groups IN Clinic Settings. 2011. 29. Famliy Health International. Guidelines and Standards for Counseling High- Risk Groups IN Clinic Settings. 2011. 30. Guidance P, Collaborative FOR. Implementing comprehensive HIV-STI programmes MSM. 2014. http://www.unfpa.org/sites/default/files/pub-pdf/ MSMIT_for_Web.pdf. 30. Guidance P, Collaborative FOR. Implementing comprehensive HIV-STI programmes MSM. 2014. http://www.unfpa.org/sites/default/files/pub-pdf/ MSMIT_for_Web.pdf. 31. Herek GM. Heterosexuals’ attitudes toward lesbians and gay men: correlates and gender differences. J Sex Res. 1988;25:451–77.
https://openalex.org/W2626111192	https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0179343&type=printable	English	null	National survey of outcomes and practices in acute respiratory distress syndrome in Singapore	PloS one	2,017	cc-by	4,849	Introduction In the past 20 years, our understanding of acute respiratory distress syndrome (ARDS) management has improved, but the worldwide incidence and current outcomes are unclear. The reported incidence is highly variable, and no studies specifically characterise ARDS epi- demiology in Asia. This observation study aims to determine the incidence, mortality and management practices of ARDS in a high income South East Asian country. Editor: Chiara Lazzeri, Azienda Ospedaliero Universitaria Careggi, ITALY Received: February 2, 2017 Accepted: May 26, 2017 Published: June 16, 2017 Copyright: © 2017 Siddiqui et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Editor: Chiara Lazzeri, Azienda Ospedaliero Universitaria Careggi, ITALY Received: February 2, 2017 Accepted: May 26, 2017 Published: June 16, 2017 Copyright: © 2017 Siddiqui et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. OPEN ACCESS OPEN ACCESS Citation: Siddiqui S, Puthucheary Z, Phua J, Ho B, Tan J, Chuin S, et al. (2017) National survey of outcomes and practices in acute respiratory distress syndrome in Singapore. PLoS ONE 12(6): e0179343. https://doi.org/10.1371/journal. pone.0179343 Methods We conducted a prospective, population based observational study in 6 public hospitals. During a one month period, we identified all ARDS patients admitted to public hospital inten- sive care units (ICU) in Singapore, according to the Berlin definition. Demographic informa- tion, clinical management data and ICU outcome data was collected. Data Availability Statement: Our ethics approval only approves use of the data for the purposes of this study, and only grants access to the data by specified investigators because it contains identifiable data. Therefore, the full data set cannot be made freely available without breaching our IRB guidelines. It can be made available upon request. Requestor would first need IRB approval from Singapore’s Domain Specific Institutional Review Board (DSRB). A version of the data file with all identifiers removed is included as Supporting Information, and original data can be obtained from RESEARCH ARTICLE * shahlasi@yahoo.com National survey of outcomes and practices in acute respiratory distress syndrome in Shahla Siddiqui1, Zudin Puthucheary2,3, Jason Phua2, Benjamin Ho4, Jonathan Tan4, 5 5 6 6 2 Shahla Siddiqui1, Zudin Puthucheary2,3, Jason Phua2, Benjamin Ho4, Jonathan Tan4, Siau Chuin5, Noelle Louise Lim5, Chai Rick Soh6, Chian Min Loo6, Addy Y. H. Tan2, Amartya Mukhopadhyay2, Faheem Ahmed Khan7, Azman Johan1, Aik Hau Tan6, Graeme MacLaren2, Juvel Taculod2, Blesilda Ramos1, Tun Aung Han8, Matthew E. Cove2 Shahla Siddiqui1, Zudin Puthucheary2,3, Jason Phua2, Benjamin Ho4, Jonathan Tan4, Siau Chuin5, Noelle Louise Lim5, Chai Rick Soh6, Chian Min Loo6, Addy Y. H. Tan2, 2 1 6 q , y , , j , , Siau Chuin5, Noelle Louise Lim5, Chai Rick Soh6, Chian Min Loo6, Addy Y. H. Tan2, 2 7 1 6 Amartya Mukhopadhyay2, Faheem Ahmed Khan7, Azman Johan1, Aik Hau Tan6, Graeme MacLaren2, Juvel Taculod2, Blesilda Ramos1, Tun Aung Han8, Matthew E. Cove2 1 Khoo Teck Puat Hospital, Yishun, Singapore, Singapore, 2 Departments of Medicine, Anaesthesia and Surgery, National University Hospital, National University Health System, Singapore, Singapore, 3 Centre for Human Health and Performance, University College London, London, United Kingdom, 4 Departments of Medicine and Anaesthesia, Tan Tock Seng Hospital, Singapore, Singapore, 5 Department of Medicine and Anaesthesia, Changi General Hospital, Singapore, Singapore, 6 Department of Medicine and Anaesthesia, Singapore General Hospital, Singapore, Singapore, 7 Department of Critical Care, Ng Teng Fong General Hospital, Jurong Health, Singapore, Singapore, 8 School of Nursing, Ngee Ann Polytechnic, Singapore, Singapore a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 ARDS practices and outcomes in a high income Asian country Materials and methods Study design and setting We conducted a national, prospective observational study over a period of 4 weeks, starting on May 15th 2015, to determine the incidence and outcomes of ARDS in Singapore. Healthcare in Singapore is provided by 6 regional health sectors, each with one or more public hospitals. In this study, we included the major adult public hospital in each of these 6 sectors. The hospitals ranged in size from 590 to 1500 beds and represented a total of 132 ICU beds. Two public hos- pitals were not included in the study, a women and children’s hospital and a psychiatric hospi- tal. Principle investigators at each of the 6 hospitals were identified and contacted through the Society of Intensive Care Medicine in Singapore. Conclusion authors after permission from DSRB at dsrb@nhg. com.sg. authors after permission from DSRB at dsrb@nhg. com.sg. The incidence of ARDS in a developed S.E Asia country is comparable to reported rates in European studies. Funding: The authors acknowledge the following as the total funding sources for this study: 1. SICM NICER grant: logistical, non-monetary, support from the Society of Intensive Care Medicine Singapore. This was in the form of Ngee Ann Polytechnic students (8) who collected the data for the study for one month. 2. NMRC (National medical research council) grant for Dr, Matthew Cove (partial support for this study): This was in the shape of salary support for all his research related activity. (NMRC/TA/0015/2013) (MEC). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. There was no additional external funding received for this study. Results A total of 904 adult patients were admitted to ICU during the study period and 15 patients met ARDS criteria. The unadjusted incidence of ARDS was 4.5 cases per 100,000 popula- tion, accounting for 1.25% of all ICU patients. Most patients were male (75%), Chinese (62%), had pneumonia (73%), and were admitted to a Medical ICU (56%). Management strategies varied across all ICUs. In-hospital mortality was 40% and median length of ICU stay was 7 days. 1 / 9 PLOS ONE \| https://doi.org/10.1371/journal.pone.0179343 June 16, 2017 Introduction The term “Acute Respiratory Distress Syndrome” (ARDS), was first used in 1967 to describe a series of mechanically ventilated patients who developed refractory hypoxia, diffuse pulmo- nary infiltrates and reduced compliance [1]. Since then, our understanding of the underlying pathophysiology has greatly improved, particularly the role of the ventilator [2]. However, the world-wide incidence remains poorly characterized. The wide range quoted, 5–64 per 100,000 persons per year [3,4], may reflect regional variation in health care practices and different interpretations of ARDS definitions [4,5]. Despite this regional variation, there are very little data characterising the epidemiology of ARDS in Asian countries. Furthermore, the impact of introducing a new ARDS definition in 2012 has not been evaluated in large nationwide studies. As a result, the incidence of ARDS following the newer definition is unknown [6]. Competing interests: The authors have declared that no competing interests exist. Determining the incidence and outcomes for ARDS is necessary for appropriate intensive care resource planning. ARDS management is complex and consumes considerable health resources. In 1999, adult ARDS patients in the USA cost $57,000 per admission and accounted for 2.7 million hospital days [4,7]. In addition, more that 50% of ARDS patients are unable to return to work at 12 months, which may burden societal resources [8]. The current cost of managing ARDS is unknown, but may vary between centres, particularly with the recent increase in the reported use of extracorporeal membrane oxygenation (ECMO) for respiratory failure [9]. In this prospective multicentre study we aim to describe the incidence, outcomes, cost and management of ARDS by conducting a nationwide study in a small, high income, Asian coun- try. We hypothesised that these characteristics would be no different to those observed in other high income countries. ARDS practices and outcomes in a high income Asian country in the study if they were admitted to ICU during the study period, received ventilator support, were at least 21 years old, and fit the new Berlin criteria for ARDS: i.e. an arterial partial pres- sure of oxygen (PaO2) to fraction of inspired oxygen (FiO2) ratio of 300 or less, bilateral infil- trates on chest x-ray (CXR) and a positive end expiratory pressure (PEEP) of 5 cmH2O or more. Onset of ARDS was defined as the first day the patient met all ARDS criteria. Clinical and physiological data was collected for the first 24 hours of admission to determine APACHE II score. Ventilation parameters were collected at the time ARDS criteria were met, as well as ventilation parameters corresponding to the highest FiO2 ratio. Hospital mortality and cost data were also collected. Outcome measures and statistical analysis The primary aim was to determine the incidence of adult ARDS expressed as cases per 100,000 inhabitants per year in the population aged 21 years and older. The catchment population was determined using published resident population figures from the government of Singapore (3,902,690) [11]. Age adjusted incidence rate was calculated using age specific resident popula- tion data. The median age in Singapore is 39.6 years, 22% are below the age of 21 (0.9 million) and 11.8% are aged 65 years, or older, just under half the residents are male (49%). Ethnically, the population is mostly composed of persons of Chinese descent (74.3%), with those of Malay and Indian descent accounting for 13.3% and 9.1% respectively [11]. Data are reported as per- centage, mean ± standard deviation and median with interquartile range (IQR) where appro- priate. Differences between continuous variables were analysed using Students T test or Wilcoxon rank-sum test, for normal or parametric distributions respectively. Distributions of categorical data were compared using Pearson’s χ2 test. All data analysis was performed using STATATM version 13 (StataCorp, Texas, USA), a p-value < 0.05 was considered significant. PLOS ONE \| https://doi.org/10.1371/journal.pone.0179343 June 16, 2017 Patient identification and data collection After obtaining Institutional Review Board approval to collect data with waiver of consent from Singapore’s National Healthcare Group Domain Specific Review Board (DSRB 2015/ 00365), all mechanically ventilated patients admitted during the study period were screened daily for development of ARDS using the Berlin definition criteria [10]. Patients were included PLOS ONE \| https://doi.org/10.1371/journal.pone.0179343 June 16, 2017 2 / 9 ARDS practices and outcomes in a high income Asian country Table 1. Demographics and general clinical data. Unless otherwise indicated median is presented and value ranges in brackets represent interquartile range (IQR). Variable Total (15) Moderate ARDS (7) Severe ARDS (8) p-value Age 59 (52–70) 62 (48–71) 59 (51–65) 0.52 Male Gender, n (%) 11 (73) 5 (70) 6 (75) 0.88 Race Chinese, n (%) 9(63) 6 (86) 3 (37.5) 0.073 Malay, n (%) 4 (26) n/o 4 (50) Indian, n (%) 1 (7) n/o 1 (12.5) Other, n (%) 1(7) 1 (14) n/o Body Mass Index (IQR) 22.8 (20.3–26.4) 25 (19–27) 23 (20–22) 0.75 APACHE II score (IQR) 33 (25–40) 33 (22–38) 33 (28–40) 0.64 PaO2:FiO2 ratio at onset (IQR) 93 (80–115) 115 (102–115) 83 (48–90) PEEP (cmH2O) at onset (IQR) 10 (8–10) 10 (8–12) 10 (5–10) 0.63 Causes of ARDS§ Pneumonia, n (%) 11 (69) 6 (86) 5 (71) 0.21 Trauma, n (%) 1 (7) 1 (14) n/o Sepsis, non-pulmonary, n (%) 2 (14) n/o 2 (29) Ventilation ﬁrst 24 hours VT, ml/kg PBW‡ (IQR) 7.5 (6.2–9) 6.5 (6.2–8.3) 8.15 (6.9–9.5) 0.27 >8ml/kg PBW‡, n (%) 6 (38) 2 (28) 4 (50) 0.40 FiO2 (IQR) 0.8 (0.6–1.0) 0.65 (0.6–0.8) 0.9 (0.8–1.0) 0.23 Peak insp. pressure cmH2O‡ (IQR) 32 (30–36) 31 (30–32) 36 (32–36) 0.14 Plateau pressure cmH2O‡ (IQR) 29 (27–32) 33.5 (29–38)† 29 (27–30) 0.33 PEEP cmH2O‡ (IQR) 15 (10–16) 14 (12–20) 15 (10–15) 0.35 Driving pressure cmH2O¶ (IQR) 20 (7–22) 9.5 (1–18)† 22 (20–22) 0.12 Rescue strategies Prone positioning, n (%) 2 (13) n/o 2 (25) 0.16 NMB, n (%) 1 (7) n/o 1 (13) 0.33 Corticosteroids, n (%) 1 (7) 1 (14) n/o 0.27 ECMO Referral, n (%) 1 (7) n/o 1 (13) 0.33 nical data. Unless otherwise indicated median is presented and value ranges in brackets represent interquartile Table 1. Demographics and general clinical data. Unless otherwise indicated median is presented and value ran range (IQR). n/o = not observed. § Data missing for one patient. PBW = Predicted body weight. ‡Value at ARDS onset. ¶ Driving pressure is deﬁned as difference between Plateau pressure and PEEP. Peak insp. pressure = peak inspiratory pressure. † Plateau pressure only available for 2 patients. NMB = neuromuscular blockers. https://doi.org/10.1371/journal.pone.0179343.t001 Results From May 15th to June 15th 2015, 1200 ICU admissions occurred, of which 904 patients (73%) required mechanical ventilation for at least 24 hours. Fifteen patients had ARDS (1.25% of ICU admissions), 7 with moderate ARDS and 8 with severe ARDS. No patients met criteria for mild ARDS, but oxygenation data is missing for one patient. The median age was 59 (IQR 52– 70), and nearly three quarters (73%) of the patients were male (Table 1). Most of the patients were ethic Chinese (63%), but a quarter were Malay and all Malay patients developed severe ARDS (Table 1). Median body mass index on admission was 22.8 (IQR 20.3–26.4) and the median Apache II score was 33 (IQR 25–40). Most patients were diagnosed as having ARDS due to pneumonia (79%). All patients with ARDS secondary to non-pulmonary sepsis devel- oped severe ARDS. At the onset of ARDS, median P/F ratio was 93 (IQR 80–115) and median PEEP was 10 cmH2O (IQR 8–10), increasing to 15 cmH2O (IQR 10–16) in the first 24 hours. Median tidal volume was 7.5 ml/kg predicated body weight (PBW), but more than one third of the patients (38%) received a tidal volume exceeding 8 ml/kg PBW. Median plateau pressure (Pplat) was 29 cmH2O with a corresponding median driving pressure (Pplat—PEEP) of 20 cmH2O. Only 2 patients were nursed in the prone position, 1 patient received neuromuscular blockers and 1 patient was referred for Extracorporeal Membrane Oxygenation (Table 1). Total in-hospital mortality was 40% (Table 2). The overall, unadjusted, incidence of ARDS was 4.5 per 100,000 population per year and the age-adjusted incidence 5.8/100,000 population per year (Table 2). The median cost, inclusive of subvention, was US$51,730 (IQR US$31,530 –US$90,890). A STROBE checklist and a copy of the anonymised baseline data is attached as S1 and S2 Files. Original data may be requested from the domain specific review board. 3 / 9 PLOS ONE \| https://doi.org/10.1371/journal.pone.0179343 June 16, 2017 PLOS ONE \| https://doi.org/10.1371/journal.pone.0179343 June 16, 2017 ARDS practices and outcomes in a high income Asian country Table 2. Summary of the main studies reporting ARDS epidemiology published since 2000. Current study LUNG-SAFE [6] ALIEN Study [5] Li et al [17] Linko et al [13] ICCTG¶ [31] Rubenfeld et al [4] Bersten et al [14] Country Singapore Multiple# Spain USA Finland Ireland USA Australia Study design Prospective Prospective Prospective Retrospective Prospective Prospective Prospective Prospective Study duration 1 month 4 weeks 1 year 8 years 8 weeks 10 weeks 12 months 2 months Study Period May-Jun 2015 Winter 2014 Nov 2008-Oct 2009 Jan 2001—Dec 2008 Apr-Jun 2007 Aug—Oct 2006 Apr 1999—Jul 2000 Oct—Nov 1999 Number of hospitals 6 435 17 2 18 N/A 21 N/A Number of ICUs 13 459 354 N/A 25 14 N/A 21 Catchment population 3,902,690 N/A 3,546,629 124,277 4,164,980 N/A 1,740,000 2,941,137 Number of ICU beds 132 N/A 354 164 N/A N/A 430 253 ICU beds/100,000 population 5.0 N/A 8.2 132 N/A N/A 24.7 8.6 ICU admissions 1,200 29,144 11,363 8,034 2,670 1,029 N/A 1,977 Number receiving MV 904 12,906 3,462 N/A 958 728 6235 N/A Total ARDS cases 15 2,813 255 514 32 196 828 148 PaO2:FiO2 93 (80– 115)† 161 (158– 163)‡ 114 (104– 124)‡ N/A 200 (138– 275) 170 (160– 180)‡§ N/A 176 (166– 186)‡ ICU incidence of ARDS per ICU bed* 0.10 0.42 0.06 0.03 N/A N/A 0.14 0.27 Unadjusted ARDS cases/100,000 population 4.5 N/A 5.9 N/A N/A N/A 58.7 N/A Adjusted ARDS cases/ 100,000 population 5.8 N/A 7.2 38.3 (in 2008) 5.0 N/A 64.0 28 Mortality 40% 35.3% 47.8% 16% (in 2008) 47% 32.3% 41.1% 34% Table 2. Summary of the main studies reporting ARDS epidemiology published since 2000. ¶ Irish Critical Care Trails Group. ¶ Irish Critical Care Trails Group. # Data from 48 countries. * Number of ARDS cases per ICU bed, normalized to 4 weeks. † Median (IQR). * Number of ARDS cases per ICU bed, normalized to 4 weeks. † M di (IQR) ‡ Mean (95% Conﬁdence interval). § Converted from kPa. N/A = Not available in reported data. [15,16]. Third, seasonal variations in climate may affect the incidence of respiratory illnesses associated with ARDS, though regional variation was preserved in studies covering a year or more [4,5,17]. Fourth, the proportion of ICU beds/100,000 population in the United States is three times the number of beds in Europe and Singapore (Table 2). Discussion This is the first nationwide study of ARDS in Southeast Asia. The age-adjusted incidence is comparable to European studies (Table 2), where the reported incidence ranges from 4.9– 13.5/100,000 person-years [3,5,6,12,13]. However, similar to European studies, this incidence differs from studies in Australia, New Zealand and the USA [4,14]. The reasons for this dispar- ity remain unclear, but are likely influenced by a variety of factors. First, the incidence may be affected by interpretation of the definitions, as well as evolution of the definitions over time [4]. Second, interpretation of chest roentgenogram can vary signif- icantly, with intensivists agreeing less than half the time on the presence of bilateral infiltrates PLOS ONE \| https://doi.org/10.1371/journal.pone.0179343 June 16, 2017 4 / 9 PLOS ONE \| https://doi.org/10.1371/journal.pone.0179343 June 16, 2017 ARDS practices and outcomes in a high income Asian country susceptibility, socioeconomic differences, or a combination is unclear and warrants further investigation. Race plays an important, complex, role in the presentation of illness [21]. Another surprising observation was the low utilisation rate of rescue interventions. However, we did not collect information regarding contraindications to prone positioning or neuromus- cular blockade. In this study we were particularly interested to determine if the cost of manag- ing ARDS patients has increased, since ECMO is now used more frequently. However, only one patient received ECMO during the study period, and our median cost was US$51,730, which is similar to reported costs before adult ECMO use increased [22]. In our study we found that median ventilation settings were consistent with lung protective ventilation (Table 1), but we also identified levels of non-compliance, where over one third of patients received tidal volumes which are not considered protective. Similar findings have been reported in large registry studies in multiple countries [23–28]. Furthermore, driving pressure, a surrogate measure of over distention, was higher than that considered physiological and safe [29]. It is clear from both our study, and others [23–28], that achieving ventilation parameters consistent with lung protective settings in all ARDS patients remains a challenge for intensivists. It is therefore not surprising that the mortality in our study is similar to that reported elsewhere (Table 2). Our study has several limitations. First, we did not detect any patients mild ARDS during the study period. There is evidence that mild ARDS is under-recognised. In the recent Lung- safe study recognition rates of only 51.3% were seen for mild ARDS, compared with rates of 78.5% for severe ARDS [6]. It is possible that the reliance of the wider multi-disciplinary team to identify ARDS may have introduced susceptibility to clinician under-recognition of mild ARDS. Second, 10 private intensive care units did not participate. These hospitals account for 24% of total hospital admissions, but the population served by the private sector are predomi- nantly the non-resident, expatriate community, who are not included in resident population counts. Third, we collected data for only one month. Under ideal circumstances, we would have tracked the incidence over 12 months or more, but the resources were not available. S2 File. Baseline anonymised data. (XLS) Sev- eral other ARDS studies have been conducted over periods of 10 weeks or less [6,13,30,31], and this may partly explain the variability in published incidence rates. Conclusion The adjusted ARDS incidence rate of 5.8/100,000 population in Singapore is similar to inci- dence rates reported in Europe. Compliance with lung protective ventilation was suboptimal and may reflect the international difficulties in translating research into practise. S1 File. STROBE checklist. (PDF) S1 File. STROBE checklist. (PDF) PLOS ONE \| https://doi.org/10.1371/journal.pone.0179343 June 16, 2017 Countries with high ICU bed availability admit a larger proportion of hospitalised patients to ICU, who tend to be older and have more chronic medical conditions [18]. These characteristics are reported risk factors for ARDS [19]. Our ARDS incidence per ICU bed was much lower than the Asian components of the recent Lung-safe study, where an incidence of 0.27 cases/ICU bed over 4 weeks was seen [6]. However, data collection for Lung-safe occurred during winter, possibly inflating the inci- dence. Although Singapore has no distinct seasons, an increase in incidence of respiratory infections correlates with winter in either hemisphere [20] and we cannot exclude a higher incidence if data collection had been timed differently. A surprising observation was the fact that all Malay patients developed severe ARDS, although the overall distribution of race between moderate and severe ARDS patients was not significantly different (Table 1). Whether this represents random variation, genetic 5 / 9 PLOS ONE \| https://doi.org/10.1371/journal.pone.0179343 June 16, 2017 Writing – review & editing: JP ZP GM. Writing – review & editing: JP ZP GM. Acknowledgments We sincerely thank the Ngee Ann Polytechnic School nursing students for their help with data collection and Society of Intensive Care Medicine in Singapore for their endorsement of this study. We would also like to acknowledge all the respiratory therapists and nurses at the partic- ipating centres for their help with coordination of the study and data collection. 6 / 9 PLOS ONE \| https://doi.org/10.1371/journal.pone.0179343 June 16, 2017 ARDS practices and outcomes in a high income Asian country Software: MC. Software: MC. Supervision: SS MC GM JP ZP. Validation: SS MC. Visualization: SS MC JP ZP. Visualization: SS MC JP ZP. Writing – original draft: SS MC. Writing – review & editing: JP ZP GM. Formal analysis: SS MC. Funding acquisition: SS MC JP. Investigation: SS MC. Methodology: SS MC JP. Project administration: SS JP BH ZP J. Tan CS NLL CRS CML AYT AM FAK AJ AHT GM J. Taculod BR TAH MC. Resources: SS JP BH ZP J. Tan CS NLL CRS CML AYT AM FAK AJ AHT GM J. Taculod BR TAH MC. References Linko R, Okkonen M, Pettila¨ V, Perttila¨ J, Parviainen I, Ruokonen E, et al. Acute respiratory failure in intensive care units. FINNALI: a prospective cohort study. Intensive Care Med. 2009; 35: 1352–1361. https://doi.org/10.1007/s00134-009-1519-z PMID: 19526218 14. Bersten AD, Edibam C, Hunt T, Moran J, Australian ANZICSCTG. Incidence and mortality of acute lung injury and the acute respiratory distress syndrome in three Australian States. Am J Respir Crit Care Med. 2002; 165: 443–448. https://doi.org/10.1164/ajrccm.165.4.2101124 PMID: 11850334 15. Rubenfeld GD, Caldwell E, Granton J, Hudson LD, Matthay MA. Interobserver variability in applying a radiographic definition for ARDS. Chest. 1999; 116: 1347–1353 PMID: 10559098 16. Meade MO, Cook RJ, Guyatt GH, Groll R, Kachura JR, Bedard M, et al. Interobserver variation in inter- preting chest radiographs for the diagnosis of acute respiratory distress syndrome. Am J Respir Crit Care Med. 2000; 161: 85–90. https://doi.org/10.1164/ajrccm.161.1.9809003 PMID: 10619802 17. Li G, Malinchoc M, Cartin-Ceba R, Venkata CV, Kor DJ, Peters SG, et al. Eight-year trend of acute respiratory distress syndrome: a population-based study in Olmsted County, Minnesota. Am J Respir Crit Care Med. 2011; 183: 59–66. https://doi.org/10.1164/rccm.201003-0436OC PMID: 20693377 18. Wunsch H, Angus DC, Harrison DA, Linde-Zwirble WT, Rowan KM. Comparison of medical admissions to intensive care units in the United States and United Kingdom. Am J Respir Crit Care Med. 2011; 183: 1666–1673. https://doi.org/10.1164/rccm.201012-1961OC PMID: 21471089 19. TenHoor T, Mannino DM, Moss M. Risk factors for ARDS in the United States: analysis of the 1993 National Mortality Followback Study. Chest. 2001; 119: 1179–1184. PMID: 11296187 20. Shek LP-C, Lee B-W. Epidemiology and seasonality of respiratory tract virus infections in the tropics. Paediatric Respiratory Reviews. 2003; 4: 105–111. PMID: 12758047 21. LaVeist TA. Disentangling Race and Socioeconomic Status: A Key to Understanding Health Inequali- ties. Journal of Urban Health: Bulletin of the New York Academy of Medicine. 2005; 82: iii26–iii34. 22. Angus DC, Clermont G, Linde-Zwirble WT, Musthafa AA, Dremsizov TT, Lidicker J, et al. Healthcare costs and long-term outcomes after acute respiratory distress syndrome: A phase III trial of inhaled nitric oxide. Critical Care Medicine. 2006; 34: 2883–2890. https://doi.org/10.1097/01.CCM.0000248727. 29055.25 PMID: 17075373 23. Needham DM, Colantuoni E, Mendez-Tellez PA, Dinglas VD, Sevransky JE, Dennison Himmelfarb CR, et al. Lung protective mechanical ventilation and two year survival in patients with acute lung injury: pro- spective cohort study. BMJ. 2012; 344: e2124. https://doi.org/10.1136/bmj.e2124 PMID: 22491953 24. References 1. Ashbaugh DG, Bigelow DB, Petty TL, Levine BE. Acute respiratory distress in adults. Lancet. 1967; 2: 319–323. PMID: 4143721 2. Tremblay LN, Slutsky AS. Ventilator-induced lung injury: from the bench to the bedside. Intensive Care Med. 2006; 32: 24–33. https://doi.org/10.1007/s00134-005-2817-8 PMID: 16231069 3. Valta P, Uusaro A, Nunes S, Ruokonen E, Takala J. Acute respiratory distress syndrome: frequency, clinical course, and costs of care. Critical Care Medicine. 1999; 27: 2367–2374. PMID: 10579250 4. Rubenfeld GD, Caldwell E, Peabody E, Weaver J, Martin DP, Neff M, et al. Incidence and outcomes of acute lung injury. N Engl J Med. 2005; 353: 1685–1693. https://doi.org/10.1056/NEJMoa050333 PMID: 16236739 5. Villar J, Blanco J, Anon JM, Santos-Bouza A, Blanch L, Ambros A, et al. The ALIEN study: incidence and outcome of acute respiratory distress syndrome in the era of lung protective ventilation. Intensive Care Med. 2011; 37: 1932–1941. https://doi.org/10.1007/s00134-011-2380-4 PMID: 21997128 6. Bellani G, Laffey JG, Pham T, Fan E, Brochard L, Esteban A, et al. Epidemiology, Patterns of Care, and Mortality for Patients With Acute Respiratory Distress Syndrome in Intensive Care Units in 50 Countries. JAMA. 2016; 315: 788–800. https://doi.org/10.1001/jama.2016.0291 PMID: 26903337 7. Angus DC, Clermont G, Linde-Zwirble WT, Musthafa AA, Dremsizov TT, Lidicker J, et al. Healthcare costs and long-term outcomes after acute respiratory distress syndrome: A phase III trial of inhaled nitric oxide. Critical Care Medicine. 2006; 34: 2883–2890. https://doi.org/10.1097/01.CCM.0000248727. 29055.25 PMID: 17075373 8. Herridge MS, Cheung AM, Tansey CM, Matte-Martyn A, Diaz-Granados N, Al-Saidi F, et al. One-year outcomes in survivors of the acute respiratory distress syndrome. N Engl J Med. 2003; 348: 683–693. https://doi.org/10.1056/NEJMoa022450 PMID: 12594312 9. Extracorporeal Life Support Organization (ELSO) ECLS Registry Report, International Summary. Ann Arbor, MI; January 2015 10. The ARDS Definitions Task Force. Acute Respiratory Distress Syndrome The Berlin Definition. JAMA. 2012; 307: 2526–2533. https://doi.org/10.1001/jama.2012.5669 PMID: 22797452 7 / 9 PLOS ONE \| https://doi.org/10.1371/journal.pone.0179343 June 16, 2017 ARDS practices and outcomes in a high income Asian country 11. Department of Statistics Singapore. Population trends 2015. Available: https://www.singstat.gov.sg/ docs/default-source/default-document-library/publications/publications_and_papers/population_and_ population_structure/population2015.pdf Accessed May 2016. 12. Luhr OR, Antonsen K, Karlsson M, Aardal S, Thorsteinsson A, Frostell CG, et al. Incidence and mortal- ity after acute respiratory failure and acute respiratory distress syndrome in Sweden, Denmark, and Ice- land. The ARF Study Group. Am J Respir Crit Care Med. 1999; 159: 1849–1861. https://doi.org/10. 1164/ajrccm.159.6.9808136 PMID: 10351930 13. 31. Irish Critical Care Trials Group. Acute lung injury and the acute respiratory distress syndrome in Ireland: a prospective audit of epidemiology and management. Crit Care. 2008; 12: R30. https://doi.org/10. 1186/cc6808 PMID: 18312626 PLOS ONE \| https://doi.org/10.1371/journal.pone.0179343 June 16, 2017 ARDS practices and outcomes in a high income Asian country References Needham DM, Yang T, Dinglas VD, Mendez-Tellez PA, Shanholtz C, Sevransky JE, et al. Timing of low tidal volume ventilation and intensive care unit mortality in acute respiratory distress syndrome. A pro- spective cohort study. Am J Respir Crit Care Med. 2015; 191: 177–185. https://doi.org/10.1164/rccm. 201409-1598OC PMID: 25478681 25. Freebairn R, McHugh G, Hickling K. Extracorporeal membrane oxygenation for ARDS due to 2009 influ- enza A(H1N1). JAMA. 2010. p. 941. 26. Kumar A, Zarychanski R, Pinto R, Cook DJ, Marshall J, Lacroix J, et al. Critically ill patients with 2009 influenza A(H1N1) infection in Canada. JAMA. 2009; 302: 1872–1879. https://doi.org/10.1001/jama. 2009.1496 PMID: 19822627 27. Young MP, Manning HL, Wilson DL, Mette SA, Riker RR, Leiter JC, et al. Ventilation of patients with acute lung injury and acute respiratory distress syndrome: has new evidence changed clinical practice? Critical Care Medicine. 2004; 32: 1260–1265. PMID: 15187503 28. Rubenfeld GD, Cooper C, Carter G, Thompson BT, Hudson LD. Barriers to providing lung-protective ventilation to patients with acute lung injury. Critical Care Medicine. 2004; 32: 1289–1293. PMID: 15187508 29. Amato MBP, Meade MO, Slutsky AS, Brochard L, Costa ELV, Schoenfeld DA, et al. Driving Pressure and Survival in the Acute Respiratory Distress Syndrome. N Engl J Med. 2015; 372: 747–755. https:// doi.org/10.1056/NEJMsa1410639 PMID: 25693014 30. Bersten AD, Edibam C, Hunt T, Moran J, Australian ANZICSCTG. Incidence and mortality of acute lung injury and the acute respiratory distress syndrome in three Australian States. Am J Respir Crit Care Med. 2002; 165: 443–448. https://doi.org/10.1164/ajrccm.165.4.2101124 PMID: 11850334 8 / 9 PLOS ONE \| https://doi.org/10.1371/journal.pone.0179343 June 16, 2017 ARDS practices and outcomes in a high income Asian country ARDS practices and outcomes in a high income Asian country 9 / 9 PLOS ONE \| https://doi.org/10.1371/journal.pone.0179343 June 16, 2017
https://openalex.org/W4378778767	https://zenodo.org/records/7989842/files/portada%20gloria%20(1).pdf	unk	null	Portada	Zenodo (CERN European Organization for Nuclear Research)	2,023	cc-by	5	“UNA VIDA PROPIA” Sevilla,2022 Sevilla,2022
https://openalex.org/W3110977717	https://nottingham-repository.worktribe.com/preview/5125586/s12884-020-03469-8.pdf	English	null	Developing a complex intervention to support pregnant women with mild to moderate anxiety: application of the Medical Research Council framework	BMC pregnancy and childbirth	2,020	cc-by	9,979	© The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 https://doi.org/10.1186/s12884-020-03469-8 Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 https://doi.org/10.1186/s12884-020-03469-8 (2020) 20:777 Developing a complex intervention to support pregnant women with mild to moderate anxiety: application of the Medical Research Council framework Kerry Evans1* , Helen Spiby1,2 and C. Jane Morrell1 Kerry Evans1* , Helen Spiby1,2 and C. Jane Morrell1 Abstract Background: To design and develop an intervention to support women with symptoms of mild to moderate anxiety in pregnancy. Methods: The development followed the MRC framework for complex interventions, utilising psychological theory, review level evidence and professional and public involvement. Two systematic reviews were completed which helped identify potentially beneficial intervention components. The theory underpinning the components was explored to consider the potential benefit for women with mild to moderate anxiety symptoms in pregnancy. Methods of delivering the intervention within maternity services were explored. The intervention comprised: group discussions, one to one support and assisted self-help resources. Midwives were identified as ideally placed to facilitate the intervention supported by midwifery support workers. A bespoke training package was provided by subject experts to prepare the facilitators. Results: The absence of established interventions and a paucity of evidence based approaches for pregnant women with symptoms of mild to moderate anxiety indicated the need for a rigorous and systematic approach to the intervention design. This approach led to the development of an intervention feasible for implementation in maternity care systems tailored to the needs of pregnant women. The involvement of a multi-professional advisory team and active engagement of service users helped to consider the acceptability of the intervention for women and the feasibility of delivering the intervention in the context of maternity care. Conclusion: The MRC Framework provided useful overarching guidance to develop a midwife facilitated intervention for women with symptoms of anxiety in pregnancy. The framework assisted the development of a robust rationale for each intervention component and considered the processes of evaluation and implementation into maternity care systems. and continuing up to 10 years after birth were estimated at £35,000 for the mother and child [9]. Symptoms of anx- iety are experienced by many pregnant women; prevalence of antenatal anxiety disorders has been reported between 13 and 15% in the UK and US [63]. There were 657,076 live births in England and Wales in 2018 and it is there- fore likely that around 90,000 women experience symp- toms of anxiety in pregnancy each year. Background Anxiety disorders are reported as the sixth leading cause of disability globally, with women accounting for 65% of disability adjusted life years. Costs of additional use of public services, productivity losses and quality adjusted life year lost for women with anxiety in the perinatal period Anxiety disorders in pregnancy usually present with similar symptoms to anxiety disorders at other times UK Full list of author information is available at the end of the article * Correspondence: Kerry.evans1@nottingham.ac.uk 1University of Nottingham, 12th Floor Tower Building, Nottingham NG7 2RD, Identifying the evidence base The MRC state that the development of a complex intervention should begin by identifying the relevant, existing evidence base [24]. Existing reviews which have evaluated the effectiveness of interventions on anxiety outcomes in pregnancy have focused on depression, mind-body or pharmacological interventions or included women with severe anxiety. Therefore two systematic re- views were completed to identify the evidence base for non-pharmacological interventions delivered to women with mild to moderate anxiety in pregnancy [30, 31]. The systematic reviews asked the following questions: The aim of interventions is to provide suitable, timely support and treatment to prevent an escalation of symptoms and improve women’s ability to cope [53]. Perinatal mental health is a priority area identi- fied in the National Health Service long term plan [2] which aims to provide an additional 24,000 women each year with access to specialist perinatal mental healthcare. Priority areas include increasing access to evidence-based care including psychological therapies and mental health assessment. All women identified with mild to moderate anxiety should be offered a range of support tailored to their needs [26]. How- ever, services to support women’s mental health are not always readily available and need to be strength- ened [53]. Many women stop taking anxiety medication in pregnancy, due to uncertainty surrounding the risk of teratogenicity [6] and non-pharmacological interventions are recommended as the initial treatment option [63]. There are no existing systematic reviews which evaluate interventions to improve mild to moderate anxiety in pregnancy. New interventions need to be developed in response to the theoretical and evidence base (Medical Research Council, MRC, [24]). How effective are non-pharmacological interventions in reducing the symptoms of mild to moderate anxiety in pregnancy? How acceptable and beneficial are non- pharmacological interventions for reducing the symptoms of mild to moderate anxiety? The two systematic reviews concluded that interven- tions, specifically designed to support pregnant women with mild to moderate anxiety have mainly been evalu- ated in small scale studies. Studies evaluated different intervention designs for different populations and overall results were inconclusive regarding intervention effect- iveness. Although no particular design which could be directly recommended for clinical practice was identi- fied, the synthesised review findings helped identify components likely to increase the effectiveness and acceptability of the intervention. This paper reports the stages of an intervention devel- opment utilising the MRC framework for developing complex interventions [24]. Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 Page 2 of 12 Page 2 of 12 Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 Methods [63]. However, concerns about pregnancy may present as the predominant feature [10]. Although mild anxiety in pregnancy is a normal adaptive process, symptoms become problematic when they consume a large propor- tion of a woman’s time, when a woman is unable to focus on other tasks and when symptoms interfere with everyday life Wenzel [81]. Anxiety disorders can result in significant disability for sufferers and possible negative effects on the fetus [63]. Elevated and prolonged anxiety has been associated with pre-term birth, fetal growth restriction and behavioural problems in developing children [28, 33, 50]. Antenatal anxiety has been re- ported to have a negative impact on women’s confi- dence in mothering, satisfaction with their infants and predict post-traumatic stress disorder and postnatal depression [36, 38, 42]. The MRC described complex interventions as: 1) includ- ing several interacting components; 2) sensitive to the context in which they are delivered; 3) having a causal chain linking the intervention to outcomes; 4) having a range of possible outcomes [24]. It was considered that a new intervention would need to operate within different maternity settings and be delivered to different popula- tions of pregnant women. The choice of intervention components should include consideration of how the mechanisms of change would function within the con- text of maternity care structures and propose ways the mechanisms would influence women’s symptoms of anx- iety. Therefore, the intervention was considered as ‘com- plex’ and the stages of the intervention development followed the general principles outlined by the MRC theoretical and modelling phases for complex interven- tions [24] (Fig. 1). In the UK, midwives provide care for every preg- nant woman and are ideally placed to identify men- tal health concerns and support emotional wellbeing [53]. Maternity care previously focused on physical wellbeing; greater support for the major psycho- logical transition women experience in pregnancy and motherhood is required [1]. Psychological inter- ventions may be beneficial in reducing symptoms of anxiety but need to be evaluated in pregnant popula- tions to strengthen the evidence base. Identifying the evidence base The aim of the intervention was to support women with mild-moderate symptoms of anxiety in pregnancy. There was some evidence of benefit for group inter- ventions, women valued the opportunity to share experi- ences, reducing feelings of isolation, and accessing group Page 3 of 12 Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 Fig. 1 Key elements of the development and evaluation process (Based on MRC, [24]) Fig. 1 Key elements of the development and evaluation process (Based on MRC, [24]) Identifying appropriate theory Identifying appropriate theory support [8, 12, 17, 29, 32, 37, 39, 72, 79, 85]. Some women felt they benefitted from having an individual discussion with their healthcare professionals (HCP) [19, 23]. Women were motivated to self-select into interven- tion studies however, some had concerns about disclos- ing anxiety symptoms and joining groups. There was some evidence of benefit for multi-session interventions, and women identified group sessions as helpful once groups became established. Studies which reported an improvement in anxiety scores included group mindful- ness [39], mindfulness based cognitive therapy [32], mo- tivational interviewing [12], relaxation [8, 21, 77, 78] or CBT interventions [54]. Women welcomed interventions which presented options for managing their symptoms and included peer or professional support [31]. The theory underpinning the potentially beneficial inter- vention components as identified in the two reviews were explored (Table 1). This process strengthened the rationale for the final intervention design and helped to define the process of change in relation to anxiety symp- toms in pregnancy [57]. The development of complex interventions requires researchers to develop an aware- ness of the relevant theory underpinning intervention components to increase the likelihood of the effective- ness of the intervention design [24, 34]. A description of the intervention’s underlying theoretical basis should in- clude specific theories, theoretical positions, and frame- works as well as empirical evidence which may have been conducted in different settings or countries [56]. Identifying the evidence base Table 1 Summary of the findings from the systematic reviews and the theory underpinning the intervention components Women’s views on intervention components Theory Group and individual interventions Interventions delivered to groups of pregnant women • Able to share experiences • Accessed group support • Reduced feelings of isolation • Helped to normalise women’s experiences • Social support °Experiential knowledge °Social learning °Social comparison °Peer support Interventions delivered to individuals • Received support from HCPs • Provided reassurance and guidance • Therapeutic relationships °Collaborative role theory °Relational continuity °Social influence Intervention components Mind-body • Provided options and coping strategies for managing anxiety symptoms • Learned breathing and relaxation techniques • Learned to recognise and adapt to anxious thoughts • Felt more positive about the future • Awareness, self-regulation and adapted behaviour • Relaxation response Psychological • Developed an understanding of the causes of anxiety in their lives and self-awareness of their thought patterns. • Helped women respond in a more positive way to situations and feelings, before negative thought patterns could escalate. • Cognitive behavioural mechanisms Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 Page 4 of 12 Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 Therapeutic relation theory Collaborative therapeutic relationships enable pregnant women to feel physically and psychologically supported which facilitates confidence building and self-efficacy [20]. Continuity of carer from a midwife known to the woman throughout pregnancy and the intrapartum period has been associated with improved health out- comes for women and babies [71]. Benefits include an increased sense of trust, choice and control. Social influ- ence theory recognises that the HCP’s may be seen as a source of social power due to their access to informa- tion, resources and services. While this may be benefi- cial, it is also associated with negative outcomes if individuals are influenced or coerced into compliance to gain access to services or information. Excessive infor- mation seeking and reassurance seeking are common features of anxiety disorders and can have a negative im- pact on outcomes and the practitioner–service user rela- tionship [68]. A pregnant woman with health anxiety may continually or excessively seek reassurance about fetal growth, the progress of their pregnancy and about the birth [10]. HCPs need to be aware of possible service user motivations for seeking reassurance about their health and wellbeing and suggest strategies, such as CBT, to help modify negative behavioural patterns [83]. Psycho-education on the nature of fear/anxiety. Cognitive restructuring to challenge the truth of anxious thoughts and develop alternative thoughts to better reflect their experience [16]. Behavioural exposure components of CBT require fur- ther consideration in the context of pregnancy. There are very few studies evaluating exposure-based CBT due to concerns around potential harm to the fetus [4, 48]. Social support theory and develop coping strategies [52]. The relaxation re- sponse is thought to counteract the stress response of anx- iety. Physiological mechanisms and adjustments are activated when an individual engages in repetitive mental or physical activity and is able to passively ignore anxious thoughts [51]. Social support may have a positive effect on wellbeing, such as providing: 1. compassion, reassurance and a sense of self-worth; 2. access to new contacts and infor- mation to help develop problem solving skills; 3. redu- cing feelings of uncertainty and develop a sense of control; 4. providing instrumental support to reduce the frequency and duration of stressors; and 5. influencing positive health behaviours [40]. Social support pathways include components of experiential knowledge; social learning theory; social comparison theory and the helper-therapy principle [70]. Individuals resolve their problems through sharing their experiences of mental illness with others who are experiencing similar situa- tions [14] and can benefit by learning from others who have succeeded in managing their symptoms [75]. Multi-component approach Many of the interventions identified in the systematic re- views had multiple components: psycho-education; re- laxation; peer support; and professional support. This multi-component approach was reflected in the inter- connected theoretical approaches which underpinned existing intervention components. For example, CBT techniques are often incorporated within therapeutic re- lationship approaches and can be accessed as a resource within peer support models. A theoretical model was developed to map the poten- tial mechanisms and their usefulness in meeting the needs of pregnant women with symptoms of mild to moderate anxiety (Fig. 2). Exploring the theoretical base highlighted that positive change can occur though: 1. de- veloping collaborative relationships with women which aim to promote women’s choice and control over their care. 2. receiving support from HCPs who both under- stand women’s individual needs and can also help them access services; 3. accessing support and learning from other women who have experienced / are experiencing similar feelings or situations; 4. developing strategies to help women develop an awareness of their thought pro- cesses and learn techniques to improve the way they Cognitive-behavioural mechanisms In the treatment of anxiety disorders, the aim of cogni- tive behavioural therapy (CBT) is to reduce anxious feel- ings by undoing prior learning or by providing new, more adaptive learning experiences and changing cogni- tive and behavioural responses to anxiety [84]. Increas- ing an individual’s awareness of unwanted emotions and behaviours is thought to generate a number of alterna- tive responses. This helps the individual to decide on a course of action and monitor the outcome to re-enforce positive coping strategies [16]. CBT for anxiety disorders may include components of: Mind-body approaches Awareness of mind and body experiences enables an indi- vidual to direct their attention to their breathing or an- other object of focus, to prevent elaborative ruminative thought processing [35, 66]. Mind-body interventions like yoga, guided imagery, mindfulness or hypnotherapy may be effective for reducing anxiety as they are thought to in- duce mental relaxation and alter negative thinking related to anxiety ([52]. Mind-body approaches are intended to modify an individual’s perceptions of stressful events which can lead to improvements in adapted behaviour Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 Page 5 of 12 Fig. 2 Theoretical model outlining the mechanisms which are considered to result in an improvement in anxiety symptoms for pregnant women outlining the mechanisms which are considered to result in an improvement in anxiety symptoms for pregnant women Fig. 2 Theoretical model outlining the mechanisms which are considered to result in an improvement in anxiety sympto stories and inspire hope [25, 55, 69]. Women who feel isolated in pregnancy or have poor social support may benefit from peer group approaches, however some women may not feel confident to share their situations or feelings within a group. Women may have additional pregnancy related or mental health concerns which they would prefer to discuss individually with a midwife who can provide maternity expertise and support referrals or signposting to other specialist services such as Increasing Access to Psychological Therapies (IAPT). The options for the delivery of the intervention components, consid- ering the feasibility of employing midwife facilitators and facilitator training requirements were mapped (Fig. 3). The advisory group raise concerns that the training to deliver CBT and mindfulness-based interventions was intensive, with training usually taking 1 year or more to complete. Also, at the time of the study development readily accessible training courses were not focused on delivering interventions to pregnant women. Recent studies have highlighted the effectiveness of interper- sonal psychotherapy and CBT interventions to prevent postnatal depression which can be delivered by nurses, midwives and health visitors in antenatal care settings and require brief initial training [45, 46], and a brief midwife-led CBT intervention for maternal anxiety is in progress [82]. For this intervention, it was considered that the therapeutic intervention components (mind body and cognitive behavioural approaches) could be de- livered through supported use of self-help resources. Guided self-help has been reported as an effective cope with anxiety. Mind-body approaches Mind-body and/or CBT approaches were considered as appropriate components of the inter- vention design. Intervention protocol intervention for depression and anxiety in general popu- lations [73] and has been used as a stand-alone interven- tion or alongside group interventions for pregnant women with anxiety, stress and depression [30]. Poten- tial self-help resources were identified evaluated using IAPT criteria [43]. Intervention protocol Following the evaluation of the evidence base, exploring the theoretical base and consultations with stakeholder groups, a protocol was developed for the intervention ([56], Fig. 4, Tables 2 and 3). p Following the evaluation of the evidence base, exploring the theoretical base and consultations with stakeholder groups, a protocol was developed for the intervention ([56], Fig. 4, Tables 2 and 3). ( g ) The MRC [24] state that the future implementation of the intervention needs to be considered at an early stage of development. This should ask questions about whether implementation would be possible, who the key stakeholders are and what information they may need to implement changes in practice. De Silva et al. [27] pro- posed that the current MRC guidance could be strength- ened by incorporating Theory of Change (TOC) into the design and evaluation of complex interventions To help identify the intervention processes and success indica- tors a TOC map was developed (Fig. 5). TOC defines how and why an initiative works, providing a pragmatic framework to describe how the intervention affects change [27, 80]. Each pre-condition for the intervention is evidence based and measured through an indicator. The TOC can help reduce future implementation fail- ures as weak links in the causal pathway can be tested, revised and strengthened. The TOC map set out to an- swer a series of questions which asked how the interven- tion could be integrated into routine practice and identifying how the intervention could be empirically tested in future definitive research [13, 27]. Modelling process and outcomes For this study, potential intervention components and processes were tested through consultations with a study advisory group and a maternity research public involve- ment group. The advisory group consisted of the head of nursing and midwifery research at the local NHS trust, a community psychiatric nurse, a midwife manager, a ser- vice user, consultant clinical psychologist and mental health training providers. Service users provided insight into how the intervention would be accessed and used and ensured the intervention was relevant to the needs of pregnant women [44, 58]. Both groups supported the proposed intervention components and helped to iden- tify methods of delivery for the intervention which con- sidered: the context and methods for introducing the intervention, assessing eligibility, method of delivery and facilitation of peer groups; and delivery of the thera- peutic components. Rather than having two midwife facilitators, service managers identified that a midwifery support worker (MSW) could provide support to the midwife during the groups and co-facilitate the interven- tion. A bespoke training framework was developed for midwives and MSWs which referred to existing perinatal competency frameworks [64, 65]. Experienced mental health training providers developed a three day training workshop which included a range of educational and learning approaches e.g. role play, lectures and the com- pletion of an information and reflective workbook. Additional considerations and motivations informing the intervention design In response to the increased focus on the role of the midwife to support the psychological and emotional wellbeing of women in pregnancy [53], the development work explored ways in which women could be supported by midwives within midwives’ current scope of practice [67]. It was considered that a midwife could facilitate peer groups, acting as a resource to the women. Midwife facilitation may be more appropriate when groups are establishing, however the role of the professional in peer groups should not interfere with the potential benefits derived when group members help others in the group [18]. In maternity care, the role of the HCP in breast- feeding support groups has been reported to “normalise or counteract extreme views and help women to distin- guish between fact, anecdote and myth” ([41], page 143). In a group based antenatal care study (Andersson et al.[3], women welcomed midwives who were less structured in their approach to group facilitation. They appreciated midwives contributing their expertise in antenatal care and helping to address topics women found difficult to introduce. To maximise the benefit of social learning mechanisms, women may benefit from hearing the experiences of other women who have been through similar experiences who can share their success Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 Page 6 of 12 Fig. 3 Methods of delivery for the intervention components Fig. 3 Methods of delivery for the intervention components Discussion 4 Intervention components, intended impact and outcomes Facilitator training (uptake and participants’ evaluation) impact and outcomes of the intervention components and methods of delivery. evaluation) Acceptability and uptake (uptake and attendance rates for each intervention component across the various care settings and service user groups) Integration in maternity care systems and additional supportive services (intervention fidelity, referral rates in perinatal mental health services, time taken for screening, delivery and facilitation) Although non-pharmacological interventions are rec- ommended as the initial treatment option, psychological interventions developed specifically for pregnant women with mental health concerns have demonstrated promis- ing results but have not been rigorously evaluated in large studies. Furthermore the theoretical base to im- prove symptoms of anxiety has not been developed spe- cifically for a pregnant population [30, 47]. The MRC framework [24] was used as the overarching guidance to assist the development of a psycho-social intervention for pregnant women. The framework was particularly useful in clarifying the intervention components, linking the evidence base and theory with the intended out- comes to provide a robust rationale for each component and defining the mechanisms of change. Questions re- garding the eventual implementation of the intervention were addressed through application of the MRC guid- ance and mapping the TOC, helping to consider the intervention and study processes and highlighted the value of stakeholder engagement to increase the inter- vention feasibility and acceptability. For the proposed intervention, the TOC was developed in collaboration with stakeholders and the study advisory group, in- formed from the evidence base and the views of women and healthcare professionals working in perinatal mental health or maternity care. This enabled key assumptions and barriers to be identified and define the methods of measurement for patient-level and service level factors, for example: Acceptability and uptake (uptake and attendance rates for each intervention component across the various care settings and service user groups) g g p Integration in maternity care systems and additional supportive services (intervention fidelity, referral rates in perinatal mental health services, time taken for screening, delivery and facilitation) Levati et al. [49] conducted a scoping review of strategies used optimise complex interventions prior to definitive testing. Discussion The adoption of the MRC framework provided useful guidance to inform the development of a novel evidence-based intervention underpinned by the theoret- ical base to improve symptoms of mild to moderate anx- iety in pregnant women. The theory and evidence base were synthesised to identify potential intervention com- ponents. The modelling phase clarified the intended Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 Page 7 of 12 One to one pre-group meeting with the midwife facilitator One-to-one support from the midwife following group sessions Group sessions facilitated by the midwife and midwifery support worker Self-help resources based on 1. cognitive based skills, 2. mindfulness meditation, 3. relaxation skills Intervention components Intended impact Develop collaborative relationships to promote women’s choice and control over their care. Receive support from midwives who understand women’s individual needs and can help them access services Social support mechanisms enable women to access support and learning from other women who are experiencing similar feelings or situations Provide strategies to help women develop an awareness of their thought processes and learn techniques to improve the way they cope with anxiety Intended outcomes Increased sense of trust, choice and control, resulting in improved confidence and self- efficacy Reduced sense of isolation. Normalisation of feelings. Supportive network to develop coping strategies and provide a stress- buffering mechanism Relaxation to counteract the stress response. Passively ignore anxious thoughts. Reduce anxious feelings by undoing prior learning and provide adaptive learning experiences Fig. 4 Intervention components, intended impact and outcomes Increased sense of trust, choice and control, resulting in improved confidence and self- efficacy Reduced sense of isolation. Normalisation of feelings. Supportive network to develop coping strategies and provide a stress- buffering mechanism Reduced sense of isolation. Normalisation of feelings. Supportive network to develop coping strategies and provide a stress- buffering mechanism Social support mechanisms enable women to access support and learning from other women who are experiencing similar feelings or situations Group sessions facilitated by the midwife and midwifery support worker Provide strategies to help women develop an awareness of their thought processes and learn techniques to improve the way they cope with anxiety Relaxation to counteract the stress response. Passively ignore anxious thoughts. Reduce anxious feelings by undoing prior learning and provide adaptive learning experiences Self-help resources based on 1. cognitive based skills, 2. mindfulness meditation, 3. relaxation skills Fig. Discussion A range of different guidelines were employed by authors in the development stages, with the MRC framework used in 17 of the 27 included stud- ies with a few studies adopting the mapping framework [7] and the MOST framework [22]. The current MRC framework [24], while stressing the importance of con- text, lacked specific guidance on methods to define and describe the context of the intervention. It was import- ant for the proposed intervention that, in addition to the local maternity care structures, wider policy recommen- dations for the intervention development were defined and considered. This was particularly relevant as the intervention development was being conducted during the publication of new national maternity care policy and would need to be operational in both existing and future maternity care contexts (National Maternity [62]). In addition, the involvement of midwives to facilitate the intervention was motivated from the wider midwifery care literature which stressed the need to strengthen the Page 8 of 12 Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 Table 2 Foundation and rationale for the final intervention design Description Foundation and rationale Sample population Nulliparous women in the second trimester of pregnancy. Advisory group and service user group: focus on nulliparous women for preliminary testing (facilitate data analysis and more likely to have and ability to participate). Participant eligibility Inclusion criteria: 1. Nulliparous pregnant women 2. Self- report mild-moderate anxiety Exclusion criteria: 1.Receiving treatment for a severe and enduring mental health condition. 2. Complex social factors (NICE [60]). Current clinical policy: women with severe mental health concerns and complex social factors have established referral pathways to specialist services. Eligibility screening method: Consider using validated anxiety measurement tools (NICE [61], Sinesi et al. [76], Nath et al. [59]). Inclusion screening The anxiety measurement tool will be administered by the community midwife to indicate women who meet the cut-off score for mild to moderate. Systematic review: rationale for inclusion screening should be discussed within a supportive context. Advisory group suggested: midwives may require training of anxiety tool administration. Service user feedback: inclusion screening would be acceptable; the midwife should be aware of concerns women may have about disclosing symptoms. Intervention facilitator The intervention will be facilitated by midwives and co-facilitated by MSWs. They will receive training to deliver the intervention. One midwife and one support worker will facilitate each group. Discussion Systematic review: delivered by psychiatrists, psychologists, midwives, instructors, self-help and volunteers. Advisory group suggested: women may be more willing to seek support from midwives than mental health professionals. Service user feedback: supported midwife facilitation Consultations with trainers: two facilitators optimal for group interventions. Service Manager feedback: Suggestions to include support workers as co-facilitators. Intervention components Delivered in three components: Component 1: one to one pre-group meeting with the midwife facilitator. Systematic review: some women had concerns about disclosing symptoms and feared the judgment of others (in groups). Initial meetings with facilitators helped women feel more confident to join the group. Advisory group: one to one meetings provide opportunity to discuss concerns and answer questions. Component 2: Four sessions facilitated by a midwife and MSW. Sessions will take place fortnightly and will be held in community healthcare centres. Each session will last for 90 min (either early evening or weekends). Systematic review: group discussion sessions were highlighted as an important and valued component Advisory group: self-help resources with discussion sessions supported as an option. CBT may not be feasible for the study due to the intensive training required for delivery. Advisory group: support for community locations Service user feedback: groups may help normalise experiences and build social support. Service user feedback: offer outside daytime working hours. Component 3: Choice of self-help resources for completion between sessions: Systematic review: some participants reported self-help interventions as challenging but also helpful Advisory group: self-help resources supported as an option Service user feedback: considered useful, women should be able to choose from different formats. Page 9 of 12 Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 Table 3 Key assumptions, process interventions and indicators relating to the Theory of Change for the proposed intervention Assumptions a 1. Midwives and midwifery support workers are motivated to apply to be trained and participate as intervention facilitators; Maternity mangers are willing to release midwives and midwifery supporters time to complete training and facilitate the intervention; Intervention facilitators are supported by specialist PMH teams and professional midwifery advocates 2. Community midwives are confident and competent to delivery anxiety screening tools; Community midwives feel confident to discuss perinatal mental health with women and create the right context for women to disclose their symptoms and access supportive services 3. Table 3 Key assumptions, process interventions and indicators relating to the Theory of Change for the prop Table 3 Key assumptions, process interventions and indicators relating to the Theory of Change for the proposed intervention Assumptions a 1. Midwives and midwifery support workers are motivated to apply to be trained and participate as intervention facilitators; Maternity mangers are willing to release midwives and midwifery supporters time to complete training and facilitate the intervention; Intervention facilitators are supported by specialist PMH teams and professional midwifery advocates 2. Community midwives are confident and competent to delivery anxiety screening tools; Community midwives feel confident to discuss perinatal mental health with women and create the right context for women to disclose their symptoms and access supportive services 3. Specialist perinatal mental health teams and psychological services support the intervention as a service for women with sub- threshold symptoms of anxiety 3. Specialist perinatal mental health teams and psychological services support the intervention as a service for women with sub- threshold symptoms of anxiety y p y Specialist perinatal mental health teams and psychological services are willing to support intervention facilitators by providing training in supporting women with anxiety and provide advice and referral pathways for women who are identified as having more severe symptoms or requiring more specialist support Specialist perinatal mental health teams and psychological services are willing to support intervention facilitators by providing training in supporting women with anxiety and provide advice and referral pathways for women who are identified as having more severe symptoms or requiring more specialist support n are willing to disclose their symptoms and women with mild to moderate symptoms of anxiety are willing to attend and with the intervention 4. Women are willing to disclose their symptoms and women with mild to moderate symptoms of anxiety are willing to attend and engage with the intervention develop more severe symptoms or are identified by intervention facilitators are requiring specialist support are willing to eferral to specialist PMH services for assessment and treatment Women who develop more severe symptoms or are identified by intervention facilitators are requiring specialist support are willing to be accept a referral to specialist PMH services for assessment and treatment 1. Intervention co-ordinator trained to monitor the intervention fidelity, measure outcomes and support facilitators across maternity systems ntion co-ordinator trained to monitor the intervention fidelity, measure outcomes and support facilitators across maternity 3. Discussion Specialist perinatal mental health teams and psychological services support the intervention as a service for women with sub- threshold symptoms of anxiety Specialist perinatal mental health teams and psychological services are willing to support intervention facilitators by providing training in supporting women with anxiety and provide advice and referral pathways for women who are identified as having more severe symptoms or requiring more specialist support 4. Women are willing to disclose their symptoms and women with mild to moderate symptoms of anxiety are willing to attend and engage with the intervention Women who develop more severe symptoms or are identified by intervention facilitators are requiring specialist support are willing to be accept a referral to specialist PMH services for assessment and treatment Interventions iv 2. Recruitment and training of facilitators 1. Intervention co-ordinator trained to monitor the intervention fidelity, measure outcomes and support facilitators across maternity systems 3. Training of community midwives to effectively screen for symptoms of anxiety and refer women with mild to moderate anxiety to intervention facilitators Intervention facilitators to raise awareness of the intervention in local community teams 4. Establish a multi-disciplinary stakeholder team to support the implementation of the intervention 5. Women who develop more severe symptoms or are identified by intervention facilitators are requiring specialist support are referred to specialist PMH services for assessment and treatment Indicators id 2. Facilitators assessment of the usefulness of training and preparedness to facilitate the intervention 3. 80% of community midwives are aware of the intervention and know how to refer women to intervention facilitators; 80% of women who are identified with mild to moderate symptoms of anxiety and are eligible for participation are referred to intervention facilitators 1. Intervention fidelity assessment reaches pre-agreed standards; Facilitators feel well supported in their roles; The intervention is imple- mented across maternity care systems 4. Women attend 75% of intervention sessions; Rates of appropriate referrals to specialist services 5. Women report an improvement in generalised and pregnancy-specific anxiety scores (pre-agreed % in improvement); Women’s evaluation of the acceptability and usefulness of the intervention; Improvement in infant outcomes; Improvement in perinatal mental health in the postnatal period (3, 6 and 12 months) Table 3 Key assumptions, process interventions and indicators relating to the Theory of Change for the prop Training of community midwives to effectively screen for symptoms of anxiety and refer women with mild to moderate anxiety to intervention facilitators . Training of community midwives to effectively screen for symptoms of anxiety and refer ntervention facilitators ntervention facilitators to raise awareness of the intervention in local community teams 4. Establish a multi-disciplinary stakeholder team to support the implementation of the intervention 5. Women who develop more severe symptoms or are identified by intervention facilitators are requiring specialist support eferred to specialist PMH services for assessment and treatment 2. Facilitators assessment of the usefulness of training and preparedness to facilitate the intervention 3. 80% of community midwives are aware of the intervention and know how to refer women to intervention facilitators; 80% of women who are identified with mild to moderate symptoms of anxiety and are eligible for participation are referred to intervention facilitators 1. Intervention fidelity assessment reaches pre-agreed standards; Facilitators feel well supported in their roles; The intervention is imple- mented across maternity care systems 4. Women attend 75% of intervention sessions; Rates of appropriate referrals to specialist services 5. Women report an improvement in generalised and pregnancy-specific anxiety scores (pre-agreed % in improvement); Women’s evaluation of the acceptability and usefulness of the intervention; Improvement in infant outcomes; Improvement in perinatal mental health in the postnatal period (3, 6 and 12 months) Fig. 5 Theory of Change Map for an intervention to support women with symptoms of mild to moderate anxiety in pregnancy Fig. 5 Theory of Change Map for an intervention to support women with symptoms of mild to moderate anxiety in pregnancy Page 10 of 12 Page 10 of 12 Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 role of the midwife in promoting women’s mental and emotional wellbeing. Thus, developing an intervention which could be delivered within midwives’ scope of practice, with minimal additional resources and which could be integrated into midwifery services was of par- ticular importance. Bleijenberg et al. [11] identify that the ways the context interact with the intervention are not always addressed by existing intervention develop- ment guidance. Abbreviations HCP H l h Abbreviations HCP: Healthcare Professional; IAPT : Increasing Access to Psychological Therapies; MRC: Medical Research Council; MMHA: Maternal Mental Health Alliance; MSW: Midwifery Support Worker; NIHR : National Institute for Health and Care Excellence; NHS: National Health Service; TOC: Theory of Change In addition to the MRC framework, the CReDECI 2 reporting guidance for comprehensive reporting of the development, piloting, and evaluation of complex in- terventions in healthcare (EQUATOR network, [56], provided further useful considerations. When report- ing methodological aspects of future evaluations, in- formation about intervention modelling should be clearly defined. This should include the target setting, macro level conditions (i.e. legal and political aspects of midwifery scope of practice, education of midwives and support workers), the meso level (i.e. system level maternity networks, supportive services) and the mi- cro level (i.e. midwifery care team composition and caseload). Further development is required to ensure the cultural appropriateness of the intervention and identify ways the intervention can be adapted to meet the needs of women with complex social factors. Ef- fective recruitment strategies need to be developed to address potential disparity of the intervention within current maternity care structures. Service users, local healthcare and community groups should be involved in designing the protocol and materials for cultural relevancy and in promoting the study in different communities [15, 24, 74]. Atif et al. [5] developed an antenatal psychological intervention for women in Pakistan following the MRC framework. The authors conducted qualitative research with women and care providers to consider the needs of the target popula- tion based on consideration of their problems, demo- graphic and contextual factors. The findings were combined with the established theoretical and evidence-based approaches to inform and adapt the intervention design. Table 3 Key assumptions, process interventions and indicators relating to the Theory of Change for the prop Information regarding the implementa- tion context, the recipients, and the providers can help optimise the ability to operationalise the intervention be- fore proceeding to the next phase of evaluation. Our ex- perience with developing the proposed intervention supports the recommendations by Bleijenberg et al. [11] that additional elements are incorporated into the MRC Framework development phase, particularly problem iden- tification and definition; determination of recipients’ and providers’ needs; and examination of current practice and context. Such information will assist future evaluations of the intervention and consider the relevance for other popu- lations or settings. Consent for publication Not applicable. Consent for publication Not applicable. Availability of data and materials y Not applicable, no primary data was collected. Ethics approval and consent to participate No ethical approval was needed as this paper does not report primary research. The study involved secondary analysis of data. (Health Research Authority decision tool: https://www.hra.nhs.uk/approvals-amendments/ what-approvals-do-i-need/ Authors’ contributions d h KE prepared the manuscript as part of Doctoral study, supervision throughout was completed by HS and JM. All authors read and approved the final manuscript. Conclusions The MRC Framework provided useful overarching guid- ance to develop a midwife facilitated intervention for women with symptoms of anxiety in pregnancy. The frame- work enabled a thorough consideration of the theoretical and evidence base and highlighted the importance of stake- holder engagement to model the intervention processes. This resulted in clear rationale for the intervention compo- nents and considered the processes of evaluation and im- plementation into maternity care systems. The intervention development was strengthened by mapping the theory of change for implementation which considered the local con- text, maternity care processes and empirical performance indicators. Inclusion of these additional processes in addition to the MRC recommendations may assist future researchers with an interest in developing the evidence- base for women with anxiety in pregnancy and facilitate the evaluation, adaption, and development of interventions. Abbreviations HCP: Healthcare Professional; IAPT : Increasing Access to Psychological Therapies; MRC: Medical Research Council; MMHA: Maternal Mental Health Alliance; MSW: Midwifery Support Worker; NIHR : National Institute for Health and Care Excellence; NHS: National Health Service; TOC: Theory of Change Funding g The project was funded as part of a Doctoral Fellowship from Wellbeing of Women and the Royal College of Midwives awarded to Kerry Evans. The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Competing interests The authors declare that they have no competing interests. Author details 1University of Nottingham, 12th Floor Tower Building, Nottingham NG7 2RD, UK. 2School of Nursing and Midwifery, University of Queensland, Brisbane, Australia. References 1. Alderdice F, Lynn F. Stress in pregnancy: identifying and supporting women. Br J Midwifery. 2009;17(9):552–9. 1. Alderdice F, Lynn F. Stress in pregnancy: identifying and supporting women. Br J Midwifery. 2009;17(9):552–9. 24. Craig P, Dieppe P, Macintyre S, Michie S, Nazareth I, Petticrew M. Developing and evaluating complex interventions: the new Medical Research Council guidance. BMJ (Clinical Research Ed). 2008;337:a1655. 3. Andersson E, Christensson K, Hildingsson I. Parents' experiences and perceptions of group-based antenatal care in four clinics in Sweden. Midwifery. 2012;28(4):502–8. https://doi.org/10.1016/j.midw.2011.07.006. Epub 2011 Sep 17. 3. Andersson E, Christensson K, Hildingsson I. Parents' experiences and perceptions of group-based antenatal care in four clinics in Sweden. Midwifery. 2012;28(4):502–8. https://doi.org/10.1016/j.midw.2011.07.006. Epub 2011 Sep 17. 25. Davidson L, Bellamy C, Guy K, Miller R. Peer support among persons with severe mental illnesses: a review of evidence and experience. World Psychiatry : Official Journal of the World Psychiatric Association. 2012;11(2): 123–8. 4. Arch J, Dimidjian S, Chessick C. Are exposure-based cognitive behavioral therapies safe during pregnancy? Archives of Womens Mental Health. 2012; 15(6):445–57. https://doi.org/10.1007/s00737-012-0308-9 Epub 2012 Sep 16. PMID: 22983422. 26. Department of Health. Maternal Mental Health pathway. London: Department of Health; 2012. 27. De Silva MJ, Breuer E, Lee L, et al. Theory of change: a theory-driven approach to enhance the Medical Research Council's framework for complex interventions. Trials. 2014;15:267 https://doi.org/10.1186/1745-6215- 15-267. 5. Atif N, Nazir H, Zafar S, Chaudhri R, Atiq M, Mullany L, Rowther A, Malik A, Surkan P, Rahman A. Development of a psychological intervention to address anxiety during pregnancy in a low-income country. Front Psychiatry. 2020;10:927. https://doi.org/10.3389/fpsyt.2019.00927. 28. Ding XX, Wu Y-L, Xu S-J, Zhu R-P, Jia X-M, Zhang S-F, Huang K, Zhu P, et al. Maternal anxiety during pregnancy and adverse birth outcomes: a systematic review and meta-analysis of prospective cohort studies. J Affect Disord. 2014;159:103–10. https://doi.org/10.1016/j.jad.2014.02.027. 6. Baldwin D, Anderson I, Nutt D, Bandelow B, Bond A, Davidson J, Wittchen H-U. Evidence-based guidelines for the pharmacological treatment of anxiety disorders: recommendations from the British Association for Psychopharmacology. J Psychopharmacol (Oxford, England). 2005;19(6):567–96. 29. Dunn C, Hanieh E, Roberts R, Powrie R. Mindful pregnancy and childbirth: effects of a mindfulness-based intervention on women’s psychological distress and well-being in the perinatal period. Archives Women’s Mental Health. 2012;15(2):139–43. y p gy y p g 7. Bartholomew L, Parcel G, Kok G. Intervention mapping: a process for developing theory and evidence-based health education programs. Heal Educ Behav. 1998;25(5):545–63. Author details 1 Page 11 of 12 Page 11 of 12 Page 11 of 12 Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 Received: 30 June 2020 Accepted: 1 December 2020 References 30. Evans K, Morrell J, Spiby H. Systematic review and meta-analysis of non- pharmacological interventions to reduce the symptoms of mild to moderate anxiety in pregnant women. J Advanced Nurs Early View. 2017;00: 1–21. 8. Bastani F, Hidarnia A, Kazemnejad A, Vafaei M, Kashanian M. A randomized controlled trial of the effects of applied pelaxation training on reducing anxiety and perceived stress in pregnant women. J Midwifery Women’s Health. 2005;50(4):e36–40. 31. Evans, K., Spiby, H., Morrell, J (2019) Non-pharmacological interventions to reduce symptoms of mild to moderate anxiety in pregnant women. Systematic Review and Narrative Synthesis of women’s views on the acceptability of and satisfaction with interventions Archives of Women’s Mental Health doi: https://doi.org/10.1007/s00737-018-0936-9. 9. Bauer A, Parsonage M, Knapp M, Iemmi V, Adelaja B, Hogg S. The costs of perinatal mental health problems. Personal social services research unit. London: London School of Economics; 2014. 10. Bayrampour H, Ali E, McNeil DA, Benzies K, MacQueen G, Tough S. Pregnancy- related anxiety: a concept analysis. Int J Nurs Stud. 2016;55:115–30. 32. Faramarzi M, Yazdani S, Barat S. A RCT of psychotherapy in women with nausea and vomiting of pregnancy. Hum Reprod. 2015;30(12):2764–73. 11. Bleijenberg N, Man-van Ginkela J, Trappenburga J, Ettemab R, Sinob C, Heima N, Hafsteindóttira N, Richards D, Schuurmansa M. Increasing value and reducing waste by optimizing the development of complex interventions: enriching the development phase of the Medical Research Council (MRC) framework. Int J Nurs Stud. 2018;79:86–93. g p g y p 33. Field T. Prenatal anxiety effects: a review. Infant Behav Dev. 2017;49:120–8 https://doi.org/10.1016/j.infbeh.2017.08.008. 34. Garber J, Weersing V. Comorbidity of anxiety and depression in youth: implications for treatment and prevention. Clinical Psychol. 2010;17(4):293–306 12. Bogaerts A, Devlieger R, Nuyts E, Witters I, Gyselaers W, Van den Bergh B. Effects of lifestyle intervention in obese pregnant women on gestational weight gain and mental health: a randomized controlled trial. Int J Obes. 2012;37(6):814–21. 35. Gard T, Noggle J, Park C, Vago D, Wilson A. Potential self-regulatory mechanisms of yoga for psychological health. Front Hum Neurosci. 2014;8:770. 35. Gard T, Noggle J, Park C, Vago D, Wilson A. Potential self-regulatory mechanisms of yoga for psychological health. Front Hum Neurosci. 2014;8:770. 36. Göbel A, Stuhrmann L, Harder S, Schulte-Markwort M, Mudra S. The association between maternal-fetal bonding and prenatal anxiety: an explanatory analysis and systematic review. J affect Disord, 15;239:313-327. 2018. https://doi.org/10.1016/j.jad.2018.07.024. 13. References Bonell C, Fletcher A, Morton M, Lorenc T, Moore L. Realist randomized controlled trials: a new approach to evaluating complex public health interventions. Soc Sci Med. 2012;75:2299–306. 37. Goodman J, Guarino A, Chenausky K, Klein L, Prager J, Petersen R, et al. CALM pregnancy: results of a pilot study of mindfulness-based cognitive therapy for perinatal anxiety. Archives Women’s Mental Health. 2014;17(5): 373–87. 14. Borkman TJ. Understanding self-help/mutual aid: Experiential learning in the commons. New Brunswick: Rutgers University Press; 1999. 15. Borrelli B. The Assessment, Monitoring, and Enhancement of Treatment Fidelity In Public Health Clinical Trials. J Public Health Dent. 2011;71(s1):S52– S63. https://doi.org/10.1111/j.1752-7325.2011.00233.x. 38. Grigoriadis S, Graves L, Peer M, et al. A systematic review and meta-analysis of the effects of antenatal anxiety on postpartum outcomes. Arch Womens Ment Health. 2019;22:543–56 https://doi.org/10.1007/s00737-018-0930-2. 16. Brewin C. Theoretical foundations of cognitive-behavior therapy for anxiety and depression. Annu Rev Psychol. 1996;47:33–57. 39. Guardino C, Dunkel Schetter C, Bower J, Lu M, Smalley S. Randomised controlled pilot trial of mindfulness training for stress reduction during pregnancy. Psychol Health. 2014;29(3):334–49. 17. Breustedt S, Puckering C. A qualitative evaluation of women’s experiences of the mellow bumps antenatal intervention. Br J Midwifery. 2013;21:187–94. 40. Heaney C, Israel B. Social networks and social support. In: Glanz K, Rimer B, Viswanath K, editors. Health behavior and Health education. 4th ed. San Francisco: Wiley; 2008. 18. Brown L, Lucksted A. Theoretical foundations of mental health self-help. In: Brown LD, Wituk S, editors. Mental Health self-help: consumer and family initiatives. New York: Springer-Verlang; 2010. p. 1–359. 41. Hoddinott P, Chalmers M, Pill R. One-to-one or group-based peer support for breastfeeding? Women’s perceptions of a breastfeeding peer coaching intervention. Birth. 2006;33(2):139–46. 19. Brugha TS, Smith J, Austin J, Bankart J, Patterson M, Lovett C, Morgan Z, Morrell CJ, Slade P. Can community midwives prevent antenatal depression? An external pilot study to test the feasibility of a cluster randomized controlled universal prevention trial. Psychol Med. 2016;46(2): 345–56. https://doi.org/10.1017/S003329171500183X. 42. Iles J, Slade P, Spiby H. Posttraumatic stress symptoms and postpartum depression in couples after childbirth: the role of partner support and attachment. J Anxiety Disorders. 2011;25(4):520–30. 20. Carolan M, Hodnett E. ‘With woman’ philosophy: examining the evidence, answering the questions. Nursing Inquiry. 2007;14(2):140–52. 43. Improving Access to Psychological Therapies. Good practice guidance on the use of self-help materials within IAPT services. London: Improving Access to Psychological Therpies; 2010. 21. Received: 30 June 2020 Accepted: 1 December 2020 22. Collins L, Murphy S, Nair V, Strecher V. A strategy for optimizing and evaluating behavioral interventions. Ann Behav Med A Publ Soc Behav Med. 2005;30(1):65–73. 23. Côté-Arsenault D, Krowchuk H, Schwartz K, McCoy T. Evidence-based intervention with women pregnant after perinatal loss. Am J Maternal Child Nurs. 2014;39(3):177–86. References Best Pract Res Clin Obstet Gynaecol. 2016; 36:57–67. https://doi.org/10.1016/j.bpobgyn.2016.05.011. 47. Lavender DT. Improving quality of care during labour and childbirth and in the immediate postnatal period. Best Pract Res Clin Obstet Gynaecol. 2016; 36:57–67. https://doi.org/10.1016/j.bpobgyn.2016.05.011. 72. Satyapriya M, Nagarathna R, Padmalatha V, Nagendra H. Effect of integrated yoga on anxiety, depression & well being in normal pregnancy. Complement Ther Clin Pract. 2013;19(4):230–6. 48. Lemon E, Vanderkruik R, Dimidjian S. Treatment of anxiety during pregnancy: room to grow. Archives of Women’s Mental Health. 2015;18: 569–70. 48. Lemon E, Vanderkruik R, Dimidjian S. Treatment of anxiety during pregnancy: room to grow. Archives of Women’s Mental Health. 2015;18: 569–70. 73. Seekles W, van Straten A, Beekman A, van Marwijk H, Cuijpers P. Effectiveness of guided self-help for depression and anxiety disorders in primary care: a pragmatic randomized controlled trial. Psychiatry Res. 2011; 187(1–2):113–20. 49. Levati S, Campbell P, Frost R, et al. Optimisation of complex health interventions prior to a randomised controlled trial: a scoping review of strategies used. Pilot Feasibility Stud. 2016;2:17 https://doi.org/10.1186/ s40814-016-0058-y. 74. Sheikh A, Halani L, Bhopal R, Netuveli G, Partridge MR, Car J, Griffiths C, Levy M. Facilitating the recruitment of minority ethnic people into research: qualitative case study of South Asians and asthma. PLoS Med. 2009;6(10): e1000148. https://doi.org/10.1371/journal.pmed.1000148. 50. Lewis A, Austin E, Galbally M. Prenatal maternal mental health and fetal growth restriction: a systematic review. J Dev Orig Health Dis. 2016;7(4):416– 28. https://doi.org/10.1017/S2040174416000076. 51. Manzoni G, Pagnini F, Castelnuovo G, Molinari E. Relaxation training for anxiety: a ten-years systematic review with meta-analysis. BMC Psychiatry. 2008;8:41. 75. Simoni J, Franks J, Lehavot K, Yard S. Peer interventions to promote health: conceptual considerations. Am J Orthopsychiatry. 2011;81(3):351–9. 76. Sinesi A, Maxwell M, O'Carroll R, Cheyne H. Anxiety scales used in pregnancy: systematic review. BJPsych Open. 2019;5(1):e5. https://doi.org/10. 1192/bjo.2018.75. 52. Marc I, Toureche N, Ernst E, Hodnett E, Blanchet C, Dodin S, et al. Mind- body interventions during pregnancy for preventing or treating women ‘s anxiety. The Cochrane database of systematic reviews, 6(7), CD007559.Ic review with meta-analysis. BMC Psychiatry. 2011;8:41. 77. Teixeira J, Martin D, Prendiville O, Glover V. The effects of acute relaxation on indices of anxiety during pregnancy. J Psychosom Obstet Gynaecol. 2005;26(4):271–6. 53. Maternal Mental Health Alliance, NSPCC & Royal College of Midwives. Specialist Mental Health Midwives: What they do and why they matter. 2013. Available from: https://www.rcm.org.uk/media/2370/specialist-mental- healthmidwives-what-they-do-and-why-they-matter.pdf. Accessed Dec 2020. 78. References Ventura T, Gomes M, Carreira T. Cortisol and anxiety response to a relaxing intervention on pregnant women awaiting amniocentesis. Psychoneuroendocrinology. 2012;37(1):148–56. 54. Milgrom J, Holt C, Holt C, Ross J, Ericksen J, Gemmill A. Feasibility study and pilot randomised trial of an antenatal depression treatment with infant follow-up. Archives of Women’s Mental Health. 2015;18(5):717–30. 79. Vieten C, Astin J. Effects of a mindfulness-based intervention during pregnancy on prenatal stress and mood: results of a pilot study. Archives of Women’s Mental Health. 2008;11(1):67–74. 55. Miyamoto Y, Sono T. Lessons from Peer support among individuals with Mental Health Diffi- culties : a Review of the literature; 2012. p. 22–9. 80. Weiss, C. Nothing As Practical As Good Theory: Exploring Theory-Based Evaluation For Comprehensive Community Initiatives For Children And Families. In New approaches to evaluating community initiatives volume 1 concepts, methods and contexts. Edited by Connell, J., Kubisch, A., Schorr, L. , Weiss, C. Washington, D. New York: The Aspen Institute; 1995. 56. Möhler R, Köpke S, Meyer G. Criteria for reporting the development and evaluation of complex interventions in healthcare: revised guideline (CReDECI 2). Trials. 2015;16:204 https://doi.org/10.1186/s13063-015-0709-y. 57. Moore G, Audrey S, Barker M, Bond L, Bonell C, Hardeman W, et al. Process evaluation of complex interventions: Medical Research Council guidance. Br Med J. 2015;350:h1258. 81. Wenzel A, Stuart S. (Collaborator). Anxiety in childbearing women: Diagnosis and treatment. American Psychological Association; 2011. https://doi.org/10.1037/12302-000. 58. Moss N, Daru J, Lanz D, Thangaratinam S, Khan K. Involving pregnant women, mothers and members of the public to improve the quality of women’s health research. BJOG. 2017;124:362–5. 82. Wilkinson E, O’Mahen H, Fearon P, et al. Adapting and testing a brief intervention to reduce maternal anxiety during pregnancy (ACORN): study protocol for a randomised controlled trial. Trials. 2016;17:156 https://doi.org/10.1186/s13063-016-1274-8. 59. Nath S, Ryan EG, Trevillion K, et al. Prevalence and identification of anxiety disorders in pregnancy: the diagnostic accuracy of the two-item Generalised Anxiety Disorder scale (GAD-2). BMJ Open. 2018;8:e023766. https://doi.org/10.1136/bmjopen-2018-023766. 83. Williams, C. Overcoming anxiety, stress, and panic : a five areas approach (third edit). Florida: CRC Press; 2012. 84. Williams C, Garland A. A cognitive – behavioural therapy assessment model for use in everyday clinical practice. Adv Psychiatr Treat. 2002;8:172–9. 84. Williams C, Garland A. A cognitive – behavioural therapy assessment model for use in everyday clinical practice. Adv Psychiatr Treat. 2002;8:172–9. 60. National Institute for Health and Clinical Excellence (NICE). References Pregnancy and complex social factors. London: National Institute for Health and Care Excellence; 2010. 85. Woolhouse H, Mercuri K, Judd F, Brown S. Antenatal mindfulness intervention to reduce depression, anxiety and stress: a pilot randomised controlled trial of the MindBabyBody program in an Australian tertiary maternity hospital. BMC Pregnancy Childbirth. 2014;14(1):369. 61. National Institute for Health and Care Excellence. Antenatal and postnatal mental health: clinical management and service guidance. London: National Institute for Health and Care Excellence; 2014. 62. National Maternity Review. Better births. Improving outcomes of maternity services in England. London: National Health Service England; 2016. References Chang M-Y, Chen C-H, Huang K-F. Effects of music therapy on psychological health of women during pregnancy. J Clin Nurs. 2008;17(100):2580–7. 21. Chang M-Y, Chen C-H, Huang K-F. Effects of music therapy on psychological health of women during pregnancy. J Clin Nurs. 2008;17(100):2580–7. Page 12 of 12 Page 12 of 12 Page 12 of 12 Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 44. INVOLVE. Briefing notes for researchers: involving the public in NHS, public health and social care research. Eastleigh: INVOLVE; 2012. 44. INVOLVE. Briefing notes for researchers: involving the public in NHS, public health and social care research. Eastleigh: INVOLVE; 2012. Mental Health problems in the perinatal period pilot. London: National Health Service England; 2016. Mental Health problems in the perinatal period pilot. London: National Health Service England; 2016. Mental Health problems in the perinatal period pilot. London: National Health Service England; 2016. 66. NurrieStearns M, NurrieStearns R. Yoga for emotional trauma : meditations and practices for healing pain and suffering. California: New Harbinger Publications; 2013. 45. Jensen J, Ciolino J, Diebold A, Segovia M, Degillio A, Solano-Martinez J, Tandon S. Comparing the effectiveness of clinicians and paraprofessionals to reduce disparities in perinatal depression via the mothers and babies course: protocol for a cluster-randomized controlled trial. JMIR Res Protoc. 2018;7(11):e11624. https://doi.org/10.2196/11624. 67. Nursing and Midwifery Council. (2013) Standards for competence for registered midwives. NMC. 46. Johnson J, Wiltsey-Stirman S, Sikorskii A, Miller T, King A, Blume J, Pham X, Moore Simas TA, Poleshuck E, Weinberg R, Zlotnick C. Protocol for the ROSE sustainment (ROSES) study, a sequential multiple assignment randomized trial to determine the minimum necessary intervention to maintain a postpartum depression prevention program in prenatal clinics serving low- income women. Implementation Science. 2018;13(1):115 https://doi.org/10. 1186/s13012-018-0807-9. 68. Osborne D, Williams C. Excessive reassurance-seeking. Adv Psychiatr Treat. 2013;19(6):420–1. 69. Repper J, Carter T. A review of the literature on peer support in mental health services. J Ment Health. 2011;20(4):392–411. 70. Salzer M, Shear S. Identifying consumer-provider benefits in evaluations of consumer-delivered services. Psychiatric Rehabilitation J. 2002;25(3):281–8. 71. Sandall, J., Soltani, H., Gates, S., Shennan, A., & Devane, D. (2016) Midwife-led continuity models versus other models of care for childbearing women. Cochrane database of systematic reviews, Cochrane pregnancy and childbirth group. CD004667 (online). 47. Lavender DT. Improving quality of care during labour and childbirth and in the immediate postnatal period. Evans et al. BMC Pregnancy and Childbirth (2020) 20:777 Publisher’s Note 63. National Institute for Health and Care Excellence (updated 2018) Antenatal and postnatal mental health: clinical management and service guidance. NICE clinical guideline, NICE London. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. 64. NHS Education for Scotland. Perinatal Mental Health curricular framework. Edinburgh: National Health Service Scotland; 2006. 65. NHS England & The Tavistock and Portman NHS Foundation Trust. The competency framework for professionals working with women who have 65. NHS England & The Tavistock and Portman NHS Foundation Trust. The competency framework for professionals working with women who have
https://openalex.org/W1488900563	https://periodicoscientificos.ufmt.br/ojs/index.php/res/article/download/2144/pdf	Portuguese	null	Crescimento e Estagnação: Os Limites da Política Econômica Brasileira do Século XXI	Revista de Estudos Sociais	2,014	cc-by	11,193	Crescimento e estagnação: os limites da política econômica brasileira do século XXI Desta forma, a economia brasileira voltou a depender da entrada de capitais para a manutenção da estabilidade de preços e do arcabouço de políticas macroeconômicas que favorecem este movimento, a sacrificar ainda mais o crescimento do produto. , p Palavras-chave: Economia Brasileira; Política Macroeconômica; Ciclo Econômico. Abstract: The Brazilian economy experienced an expansive cycle along the 2000s, marked by the increase of investment, job creation and current income. After the 2008 economic crisis, despite its epicenter in the US mortgage debt markets, its spread to several economies in the world, the Brazilian economy entered a phase of low growth, falling investment, but with maintenance level of employment. This work aims to show that the recent Brazilian economy cycle is closely related to foreign markets, both in its expansion phase, as the recessive. The high current account balances decreased dependence on external capital to maintain price stability, which reduced the interest rate and increased aggregate investment. The increase in domestic income contributed to the reduction in the trade balance, but the 2008 crisis intensified this movement, reducing the growth rates of the national economy. Thus, the Brazilian economy returned to rely on capital inflows to maintain price stability and on macroeconomic policy framework that favor this movement, further sacrificing output growth. Key words: Brazilian Economy; Macroeconomic policy; Economic Cycle. JEL: E12; F32; F43. 1 Economista. Doutorando pela Universidade Estadual de Campinas. Professor da Universidade Federal de Mato Grosso 2 Ver Modenesi (2005). 1 Economista. Doutorando pela Universidade Estadual de Campinas. Professor da Universidade Federal de Mato Grosso 2 Crescimento e estagnação: os limites da política econômica brasileira do século XXI Growth and Stagnation: The Limits of Brazilian Economic Policy in 21ST Century Leonardo Flauzino de Souza1 Resumo: A economia brasileira vivenciou um ciclo expansivo ao longo da década de 2000, marcado pela expansão do investimento, geração de empregos e aumento da renda corrente. Após a crise econômica de 2008, que, apesar de ter seu epicentro nos mercados de dívida hipotecária norte-americana, se alastrou para várias economias do mundo, a economia brasileira entrou em uma fase de baixo crescimento, queda dos investimentos, mas com manutenção do nível de emprego. Este trabalho tem por objetivo mostrar que o recente ciclo da economia brasileira está intimamente relacionado aos mercados externos, tanto na sua fase expansiva, quanto na recessiva. Os elevados saldos em transações correntes diminuíram a dependência em relação ao capital externo para manter a estabilidade de preços, o que permitiu reduzir a taxa de juros e desvencilhar o investimento agregado. O próprio aumento da renda interna contribuiu para a redução do saldo comercial, mas a crise de 2008 intensificou este movimento, reduzindo as taxas de crescimento da economia nacional. Desta forma, a economia brasileira voltou a depender da entrada de capitais para a manutenção da estabilidade de preços e do arcabouço de políticas macroeconômicas que favorecem este movimento, a sacrificar ainda mais o crescimento do produto. P l h E i B il i P líti M ô i Ci l E ô i Resumo: A economia brasileira vivenciou um ciclo expansivo ao longo da década de 2000, marcado pela expansão do investimento, geração de empregos e aumento da renda corrente. Após a crise econômica de 2008, que, apesar de ter seu epicentro nos mercados de dívida hipotecária norte-americana, se alastrou para várias economias do mundo, a economia brasileira entrou em uma fase de baixo crescimento, queda dos investimentos, mas com manutenção do nível de emprego. Este trabalho tem por objetivo mostrar que o recente ciclo da economia brasileira está intimamente relacionado aos mercados externos, tanto na sua fase expansiva, quanto na recessiva. , p , q Os elevados saldos em transações correntes diminuíram a dependência em relação ao capital externo para manter a estabilidade de preços, o que permitiu reduzir a taxa de juros e desvencilhar o investimento agregado. O próprio aumento da renda interna contribuiu para a redução do saldo comercial, mas a crise de 2008 intensificou este movimento, reduzindo as taxas de crescimento da economia nacional. Revista de Estudos Sociais \| Ano 2014, N. 32 Vol 16, Pag. 205 JEL: E12; F32; F43. 1. INTRODUÇÃO Após 1994, com o advento do Plano Real, o Brasil alcançou uma relativa estabilidade de preços ao adotar um regime macroeconômico de metas cambiais2, o qual consistia em manter a taxa de câmbio relativamente estável, permitindo sua flutuação dentro de patamares previamente estabelecidos. Pelo lado fiscal, para Bacha (1994), o sucesso do regime estava atrelado à Revista de Estudos Sociais \| Ano 2014, N. 32 Vol 16, Pag. 205 resolução do déficit público potencial3, para o qual era necessário reduzir substancialmente os gastos públicos ou elevar a tributação. No entanto, a manutenção da paridade cambial em dado patamar que não causasse o recrudescimento da inflação levava a déficits correntes persistentes no balanço de pagamentos. As massivas entradas de capitais atraídos pelas altas taxas de juros de curto prazo dos títulos da dívida pública brasileira financiavam os déficits correntes. Em síntese, o regime macroeconômico brasileiro do fim do século XX estava atrelado à taxa de câmbio administrada, contração fiscal e retração monetária como medidas para manter a estabilidade de preços recém-alcançada. O referido contexto histórico-institucional remete às abordagens teóricas de Triffin (1960), que afirmou ser impossível conciliar um regime de câmbio fixo (ou administrado) com mobilidade de capital e autonomia da política monetária4. Em outras palavras, para obter a estabilidade inflacionária era supostamente necessário ancorar os preços internos aos externos através de uma taxa de câmbio administrada em um patamar baixo (em reais por dólares), ou seja, manter a sobrevalorização da moeda doméstica, o que implicava em elevados déficits correntes financiados pelos movimentos de capitais a sacrifício da autonomia da política monetária. Assim, as taxas de juros internas variariam ao sabor das taxas externas ponderadas por um duplo risco de default: a capacidade do Governo em honrar suas dívidas e do Banco Central em manter a taxa de câmbio estável. Tais características criavam um incentivo adicional à contração fiscal e monetária. Em um cenário de contração da liquidez internacional estimulado pelas crises da Ásia em 1997 e da Rússia em 1998, exacerbação de um movimento fly to quality e as saídas de capital, essencialmente de curto prazo, a manutenção da âncora cambial se tornou infactível. Assim, em 1999, o regime macroeconômico em questão ruiu e foi substituído pelo regime de metas de inflação com câmbio flutuante e metas de superávit primário. 3 Segundo Bacha (1994), o país sofria do “Efeito Tanzi às Avessas”, na qual o déficit fiscal do governo é massivamente financiado pelo imposto inflacionário, uma vez que a arrecadação de impostos está indexada à inflação e as despesas públicas não, o contrário do “Efeito Oliveira-Tanzi” tradicional. Desta forma, uma redução abrupta da inflação realizaria o até então déficit fiscal potencial, levando a uma forte expansão monetária e retomada da inflação. 4 O referido trilema ou trindade impossível é alvo de estudos de Eichengreen (1996), Serrano (2002) e Obstfeld, Shambaugh & Taylor (2004). 5 Por modelo canônico entenda-se o modelo desenvolvido por Mundell (1963) e Fleming (1962). É importante ressaltar que os condicionantes desenvolvidos por Triffin (1960) em seu trilema estão em consonância com o modelo em questão. 6 É importante ressalvar que Mundell (1963) afirmou que suas “conclusões eram mais preto e branco do que tons de cinza” (p. 485, tradução própria) e que a validade do modelo estava relacionada ao cumprimento hipóteses fortes como perfeita substitutibilidade entre os ativos do sistema e a inexistência e movimentos especulativos e mercados futuros. Nenhuma destas condições pode ser observada na realidade. Ano 2014, N. 32, V. 16, Pag. 206 \| Revista de Estudos Sociais 7 Segundo Godley (1999), o processo oposto ao mencionado, retração da liquidez pública ou superávit nominal, se caracteriza como um processo insustentável ao longo do tempo, uma vez que os déficits nominais se transformam em ganhos privados. Para o autor, analisando séries históricas para a economia dos EUA, os superávits nominais públicos apenas foram obtidos em momentos em que o setor privado se tornou amplamente deficitário devido a um processo crônico de endividamento. 8 Ver Farhi (2006). 1. INTRODUÇÃO Segundo Carneiro (2003), a passagem para um regime com taxas de câmbio flutuante permitiria, segundo o modelo canônico5, a retomada da autonomia da política monetária. Entretanto, para o autor, as características específicas da economia brasileira não permitem inferir que a retomada da autonomia da política monetária passaria, exclusivamente, pela presença de um regime de câmbio flutuante, sendo necessária a adoção de outras políticas, como a de controle de capitais6. Ano 2014, N. 32, V. 16, Pag. 206 \| Revista de Estudos Sociais Ano 2014, N. 32, V. 16, Pag. 206 \| Revista de Estudos Sociais Os dados apresentados pela Tabela 1 ilustram o comportamento supracitado. Os superávits primários (exceto 1997) evidenciam o esforço de contração fiscal por parte do governo, no entanto, a Estado não deixou de expandir liquidez pública pela via do endividamento, como mostram os déficits nominais7. As taxas de juros fortemente elevadas no período mostram o tamanho da contração monetária necessária para atrair capitais externos capazes de financiar o déficit em transações correntes, crescente de 1994 a 1998. Entretanto, esta medida não foi suficiente. A partir de 1997, com a crise internacional da Ásia e o efeito contágio nos mercados emergentes8, a contração da liquidez internacional obrigou o país a elevar seu nível de endividamento externo. Em 1999, este processo se mostrou insustentável e o regime cambial vigente foi substituído. O objetivo deste artigo é analisar o comportamento da economia brasileira neste início de século XXI e explicitar as implicações sobre regime macroeconômico adotado pós-crise cambial de 1999. Para tal, este trabalho será dividido em duas seções. A primeira dedicada à exposição do cenário macroeconômico do Brasil recente e uma análise do comportamento da economia nos últimos anos. A segunda seção evidenciará como os objetivos das políticas econômicas interagiram com as modificações da economia brasileira na última década. 2. O CENÁRIO ECONÔMICO DOS ANOS 2000 Após a crise de balanço de pagamentos de 1998 e a incapacidade do Banco Central brasileiro em manter a taxa de câmbio dentro dos patamares determinados, abandonou-se o regime de metas cambiais e passou-se a adotar um regime de metas de inflação supostamente ancorado pelo superávit primário (ou estabilidade do déficit nominal) e taxa de câmbio flutuante, formando assim o tripé de política macroeconômica. O regime macroeconômico iniciado em 1994 e encerrado em 1998 tinha por objetivo a manutenção da inflação em níveis baixos e estáveis através da ancoragem dos preços domésticos aos preços externos e manutenção da taxa de câmbio estável. Esta ancoragem só foi possível de ser realizada com a abertura comercial e financeira promovida no inicio da década de 1990, que permitiu maiores fluxos de importações e entradas massivas de capitais. Revista de Estudos Sociais \| Ano 2014, N. 32 Vol 16, Pag. 207 Revista de Estudos Sociais \| Ano 2014, N. 32 Vol 16, Pag. 207 Tabela 1: Síntese das Principais Variáveis Macroeconômicas de 1994 a 1999. 1994 - - 49,9 5,33 -1.811,23 7.215,20 44.357,26 1995 0,55 -1,49 40,3 4,42 -18.383,71 12.918,90 38.132,35 1996 0,35 -1,16 24,9 2,15 -23.502,08 8.666,10 31.593,17 1997 -0,25 -2,27 40,8 3,38 -30.452,26 -7.907,16 38.580,30 1998 0,51 -4,61 29,0 0,04 -33.415,90 -7.970,21 57.176,61 1999 2,13 -2,47 19,0 0,25 -25.334,63 -7.822,04 108.768,75 ¹Governo Central. ²Fim de período. ³Em milhões de US dólares. Dívida Externa Líquida do Setor Público³ Ano Resultado Primário por PIB (%)¹ Resultado Nominal por PIB (%)¹ Taxa de Juros (%)² Taxa de Crescimento do PIB (%) Saldo em Transações Correntes³ Saldo em Balanço de Pagamentos³ ²Fim de período. Fonte: Secretaria do Tesouro Nacional e Banco Central do Brasil. O regime macroeconômico que deu início em 1999 tinha por objetivo manter a inflação estável através da restrição fiscal e controle da taxa de juros dos títulos da dívida pública de curto prazo associados à SELIC (Sistema Especial de Liquidação e de Custódia). Segundo o modelo mainstream convencional9, um aumento na taxa de juros reduziria o nível de atividade e elevaria a taxa de desemprego, o que resultaria na queda da inflação. Ano 2014, N. 32, V. 16, Pag. 208 \| Revista de Estudos Sociais Fonte: IBGE, Banco Central do Brasil, Seade. 9 Ver Fisher & Blanchard (1989). ²Índice Nacional de Preços ao Consumidor Amplo (IPCA). ³Região metropolitana de São Paulo. 2. O CENÁRIO ECONÔMICO DOS ANOS 2000 No entanto, como ressalta Farhi (2006), para o caso brasileiro, a taxa de juros teria sido usada para influenciar o nível de inflação via taxa de câmbio: uma elevação da taxa de juros atrairia mais capitais para o país, reduzindo a taxa de câmbio (em reais por dólares), tornando os preços dos bens importados mais baratos, o que reduziria a inflação. 2000 15,84 1,95 5,97 10,00 4,31 2001 19,05 2,32 7,67 11,60 1,31 2002 24,90 3,53 12,53 11,40 2,66 2003 16,33 2,89 9,30 12,00 1,15 2004 17,75 2,65 7,60 10,00 5,71 2005 18,05 2,34 5,69 9,70 3,16 2006 13,19 2,14 3,14 9,00 3,96 2007 11,18 1,77 4,46 9,30 6,09 2008 13,66 2,34 5,90 8,30 5,17 2009 8,65 1,74 4,31 8,50 -0,33 2010 10,67 1,67 5,91 7,40 7,53 2011 10,91 1,88 6,50 6,90 2,73 2012 7,29 2,04 5,84 7,60 1,03 2013 9,90 2,34 5,91 7,50 2,49 ¹Fim de período. ²Índice Nacional de Preços ao Consumidor Amplo (IPCA). ³Região metropolitana de São Paulo. Tabela 2: Principais Variáveis Macroeconômicas da Política Monetária Taxa de crescimento do PIB (%) Ano Taxa de Juros (%)¹ Taxa de Câmbio (R$/US$)¹ Inflação (%)² Taxa de Desemprego (%)³ Tabela 2: Principais Variáveis Macroeconômicas da Política Monetária 9 Ver Fisher & Blanchard (1989). Ano 2014, N. 32, V. 16, Pag. 208 \| Revista de Estudos Sociais Como pode ser visto pela Tabela 2, nos anos em que houve elevação da taxa de câmbio, houve também aumento do nível de inflação, com exceção dos anos de 2007, 2010 e 2012, sendo que os dois primeiros apresentam as mais elevadas taxas de crescimento do PIB, inferindo um efeito inflacionário devido a um aquecimento da demanda doméstica. Já 2012 apresenta um crescimento muito baixo, o que refletiu na redução da inflação mesmo com aumento da taxa de câmbio. Também é possível inferir, pela tabela anterior, que as elevações na taxa de câmbio são acompanhadas por elevações na taxa de juros, com exceção dos anos 2004, 2005, 2010 e 2012. Nos dois primeiro anos mencionados, a inflação ainda se apresentava em níveis elevados devido ao choque cambial de 2002. O ano de 2010 apresentou uma taxa de juros maior em resposta à maior inflação promovida pelo crescimento. Já o ano de 2012 caracterizou-se por menores taxas de juros na tentativa de reerguer o crescimento. Ao longo da década, nota-se que o crescimento do PIB mostrou- se errático. 2. O CENÁRIO ECONÔMICO DOS ANOS 2000 No entanto, a trajetória de queda da taxa de juros acompanha uma trajetória de elevação do produto. É possível observar pela Tabela 2 que de 2002 a 2007 a taxa de juros inicia uma trajetória de queda com índices de inflação estáveis e declinantes. Isto ocorreu devido ao aumento do saldo exportador e surgimento de superávits em transações correntes no período (ver Gráfico 1). Os saldos em transações correntes possibilitaram a entrada de moeda estrangeira no país pela via comercial, dependendo menos das elevadas taxas de juros na tentativa de atrair capitais, valorizar a moeda doméstica e reduzir a inflação. Fonte: Banco Central Fonte: Banco Central Fonte: Banco Central Neste mesmo período a conta capital e financeira mostra uma trajetória declinante em razão da queda da taxa de juros brasileira em contraposição com a elevação da taxa de juros nos mercados de dívidas internacionais, em Revista de Estudos Sociais \| Ano 2014, N. 32 Vol 16, Pag. 209 especial nos mercados norte-americanos10. Este cenário foi possibilitado pela forte expansão das exportações, alavancadas inicialmente com o choque cambial de 2002, como pode ser visto pela Tabela 2, e posteriormente pela elevação dos preços das commodities nos mercados internacionais, como demonstrado pelo Gráfico 2: Fonte: Unctad Fonte: Unctad Desta forma, o boom nos preços internacionais das commodities, em especial os agrícolas, ampliou os saldos comerciais, antes praticamente inexistentes, e possibilitou o surgimento de saldos em transações correntes positivos, acarretando na entrada de moeda externa pela via transacional e reduzindo a dependência da entrada de capitais pela via financeira. Segundo Wray (2008), o movimento de elevação dos preços de commodities não esteve apenas relacionado à elevação da demanda mundial oriunda do elevado crescimento de outros países emergentes como China e Índia. O autor não negligencia a importância destes fatores, mas elenca também a atuação dos investidores institucionais nestes mercados. Esses investidores atuam como captadores de liquidez e tomam decisões de aplicação de capital em prazos mais curtos visando à obtenção de retornos pecuniários associados às oscilações dos preços de ativos líquidos ou de seus preços diferidos nos mercados subjacentes (derivativos financeiros). Portanto, a atuação destes investidores nos mercados de commodities contribuiu para elevar os preços destes ativos, uma vez que o contexto de elevação da demanda por tais bens criava expectativas de preços ascendentes. 10 Ver Souza (2012). 2. O CENÁRIO ECONÔMICO DOS ANOS 2000 Ou seja, os investidores institucionais teriam uma atuação pró- cíclica, contribuindo para elevar de forma drástica os preços destes ativos em momentos de expectativas de preços futuros mais elevados e para baixar de forma igualmente intensa estes preços quando as expectativas se revertiam. A queda nos mercados de commodities também engendraria impactos relevantes: 10 Ver Souza (2012). Ano 2014, N. 32, V. 16, Pag. 210 \| Revista de Estudos Sociais “Falling commodity prices will generate problems; production decisions as well as portfolio allocations have been made on the expectation of rising prices. A lot of leveraged money has gone into commodity markets (including physicals as well as futures), so just as falling real estate prices are devastating for households, for the real estate sector generally, and for financial markets, there will be significant fallout from the slump in commodity prices.” (Wray, 2008, p. 10). O impacto na queda dos mercados de commodities não se traduzirá apenas como receitas menores aos produtores destas, mas na possibilidade de que estes sejam incapazes de honrar as dívidas contratadas para financiar sua produção, gerando impactos produtivos que podem ser amplificados pelas alterações no balanço de pagamentos e na taxa de câmbio para o caso brasileiro. A elevada liquidez nos mercados de dívidas norte-americanos e a busca por retornos pecuniários em um cenário de elevação dos custos de captação (elevação das taxas de juros nos EUA), como ressaltado por Wray (2008), contribuíram para o aumento dos preços das commodities. Vale ressaltar que a importância do aumento da demanda gerada pelo crescimento de mercados emergentes como China e Índia não deve ser negligenciada. O impacto mais imediato deste cenário internacional foi o aumento dos saldos externos da economia brasileira, tornando-a menos depende da necessidade de captação de recursos pela via financeira na tentativa de atrair capitais, valorizar a moeda doméstica, reduzir a taxa de câmbio e, assim, manter estável a inflação. Logo, a bonança da liquidez externa criou espaço para a redução da taxa de juros interna, uma vez que reduziu a dependência de financiamento externo. Revista de Estudos Sociais \| Ano 2014, N. 32 Vol 16, Pag. 211 2. O CENÁRIO ECONÔMICO DOS ANOS 2000 A redução das taxas de juros criou a possibilidade de se alavancar o crescimento da economia brasileira pelo aumento do investimento, de tal forma que a redução das exportações em relação ao PIB (observada pelo Gráfico 1), a partir de 2004, é consequência da forte expansão do PIB e redução da taxa de câmbio e não de uma redução das exportações. Com a crise de 2008 e a inflexão nos mercados de commodities esta situação se reverte, apesar da ligeira recuperação de 2011, iniciando a etapa de estagnação da economia brasileira. Ao observar os dados apresentados pela Tabela 3, pode-se inferir que em praticamente todos os anos nos quais o PIB apresentou aumento na sua trajetória de crescimento, a participação do investimento no produto também aumentou, com a exceção de 2002 e 2008. O ano de 2002 apresentou um aumento no crescimento do PIB em relação ao ano anterior (Tabela 2), mas com redução da participação do investimento, isto em razão do choque cambial que desestimulou importações e ampliou exportações. Neste caso, o saldo exportador contribuiu para o aumento do crescimento, mesmo com a queda da participação do investimento. O ano de 2008 apresentou uma situação contrária: queda do crescimento em relação ao ano anterior com aumento da participação do investimento. Porém, este ano marcou o inicio da crise internacional que teve impactos produtivos globais altamente relevantes. É possível observar, pela análise da Tabela 3, que em 2008, a variação positiva dos estoques foi Revista de Estudos Sociais \| Ano 2014, N. 32 Vol 16, Pag. 211 expressiva, ou seja, os investimentos realizados – produção de máquinas, equipamentos, construção de infraestrutura em geral e outros – não foram vendidos, contribuindo para um nível de crescimento menor neste ano em relação ao anterior. 2000 18,3 7,95 92,1 49,6 42,7 7,67 2001 18,0 5,53 94,5 46,5 45,8 7,68 2002 16,2 -1,17 101,2 47,3 44,3 8,39 2003 15,8 3,13 96,9 44,2 46,8 9,01 2004 17,1 5,96 94,0 43,7 47,9 8,37 2005 16,2 1,65 98,4 42,3 49,9 7,85 2006 16,8 1,94 98,1 40,4 51,6 7,93 2007 18,3 4,84 95,2 38,3 54,1 7,58 2008 20,7 7,64 92,4 36,3 56,7 7,00 2009 17,8 -1,29 101,3 42,3 50,1 7,59 2010 20,2 3,84 96,2 40,6 52,5 6,89 2011 19,7 2,27 97,7 41,4 52,4 6,24 2012 17,5 -3,71 103,8 43,8 49,7 6,52 2013 17,9 -1,62 101,6 41,8 51,8 6,39 Fonte: ipeadata. Ano 2014, N. 32, V. 16, Pag. 212 \| Revista de Estudos Sociais 2. O CENÁRIO ECONÔMICO DOS ANOS 2000 com taxas de juros internas em trajetória declinante e inflação estável, criou-se um ambiente propício para alongamento de prazos com expectativas de expansão prolongadas e uma trajetória de expansão de investimentos e do produto. Neste aspecto, 2008 se apresenta como um turning point de grande relevância. Em um primeiro aspecto, a inversão dos mercados de commodities, redução da participação das exportações no PIB e aumento da participação das importações, engendrando saldos em transações correntes novamente negativos, contribuíram para a interrupção da trajetória de queda da taxa de juros. Os efeitos cambiais foram imediatos, criando uma trajetória de aumento da taxa de câmbio e desvalorização da moeda doméstica, o que acarretou no aumento dos patamares de inflação (Tabela 2). Em segundo lugar, com expectativas de longo prazo corroídas, a lógica de curto prazo se sobrepõe, tornando a administração dos estoques de bens de capital novamente relevante e errática. O baixo crescimento recente vem acompanhado de uma queda da participação do investimento e das vendas de estoques de bens de capital (variação negativa dos estoques na Tabela 3) consolidados nos períodos anteriores (variação positiva dos estoques na Tabela 3). Como ressaltaram Keynes (1930) e Minsky (1975), a decisão de investir precisa ser validada pelo sistema bancário através do fornecimento de crédito. Neste aspecto, como demostraram Dedecca, Trovão & Souza (2014), os períodos de aumento do investimento também foram caracterizados por elevação na relação entre crédito e PIB, em especial pelo aumento da participação do Banco Nacional do Desenvolvimento Econômico e Social (BNDES). O BNDES teve um aspecto crucial no financiamento do investimento ao fornecer crédito a taxas de juros subsidiadas. Stiglitz & Greenwald (2004) demonstram que em mercados com elevadas taxas de juros e demanda por crédito pode-se consolidar uma situação de racionamento de crédito. Nesta situação, infere-se que uma política de fornecimento de crédito subsidiado pode contribuir para a redução do racionamento de financiamento. Revista de Estudos Sociais \| Ano 2014, N. 32 Vol 16, Pag. 213 2. O CENÁRIO ECONÔMICO DOS ANOS 2000 Ano Tabela 3: Participação do Investimento no PIB e Composição do Investimento de 2000 a 2013 Participação do Investimento no PIB (%) Investimento Formação Bruta de Capital Fixo Variação dos Estoques (%) Formação Bruta de Capital Fixo (%) Construção (%) Máquinas e Equipamentos (%) Outros (%) Tabela 3: Participação do Investimento no PIB e Composição do Investimento de 2000 a 2013 Para Keynes (1936) e Kalecki (1977), a variação dos estoques é uma variável relevante para se compreender a dinâmica do investimento no curto prazo, sendo que aumentos expressivos nos estoques em um determinado período tenderiam a comprometer o crescimento do investimento nos períodos subsequentes. Nota-se que, pela Tabela 2 e Tabela 3, de 2000 a 2003 e após 2008, o PIB apresentou taxas de crescimento erráticas acompanhadas de uma trajetória de queda da participação do investimento no PIB e de um movimento de variação dos estoques de bens de capital também errático. De 2004 a 2008, o crescimento do PIB permaneceu em patamares mais elevados, com elevação da participação do investimento no produto e variação dos estoques positivas e estáveis. Os anos nos quais a variação dos estoques é negativa sinalizam que alguns dos investimentos realizados no ano não foram feitos com a produção de novos bens de capital, mas pela venda de bens já produzidos. Este cenário pode sinalizar que o setor produtor de bens de capital não projeta retornos futuros satisfatórios a ponto de ampliar seus investimentos, o que pode incorrer em impactos significativos para o crescimento do produto ao longo dos anos. A crise de balanço de pagamentos brasileira em 1999, acompanhada das crises dos países asiáticos no fim da década de 1990 e da crise bursátil das empresas de tecnologia da informação e comunicação nos EUA em 2000 e a crise da Argentina em 2001, criou, seguindo o raciocínio de Keynes (1936), um estado de expectativas de longo prazo extremamente negativo, o que foi responsável por encurtar os prazos de decisões do setor produtivo, engendrando movimentos erráticos no investimento e no produto. De 2004 a 2008, em um cenário externo positivo de crescimento e expansão da liquidez, com taxas de juros internas em trajetória declinante e inflação estável, criou-se um ambiente propício para alongamento de prazos com expectativas de expansão prolongadas e uma trajetória de expansão de investimentos e do produto. 2. O CENÁRIO ECONÔMICO DOS ANOS 2000 As taxas de juros brasileiras estão entre as mais elevadas do mundo e, ao se comparar os resultados apresentados por Dedecca, Trovão & Souza (2014) para o crédito da economia brasileira com os dados construídos por Dembiermont, Drehmann & Muksakunratana (2013) para o crédito na economia mundial, pode-se observar que a relação entre crédito e PIB do Brasil esteve em valores baixos para os padrões internacionais, aproximando-se de valores apresentados pela Índia ou México, mas muito distantes da China e outros países asiáticos, bem como das economias desenvolvidas. Estes fatores podem atuar como uma sinalização de racionamento de crédito, no qual o setor bancário demanda retornos demasiadamente elevados, estimulados pelas altas taxas da dívida pública, expulsando do mercado projetos de investimento viáveis, mas com taxas de retornos abaixo da taxa de juros dos empréstimos. As questões patrimoniais assumem importância preponderante após a crise de 2008, uma vez que os investimentos foram financiados com expectativas de longo prazo de crescimento elevado, o que acarreta na possibilidade de incapacidade das empresas em honrarem suas dívidas. Segundo Rocca & Santos Júnior (2014), houve uma acentuada na redução nos Revista de Estudos Sociais \| Ano 2014, N. 32 Vol 16, Pag. 213 lucros líquidos das empresas após 2008, apesar da recuperação de 2010, diminuindo a liquidez das empresas, mas ainda a manter sua solvência. A redução dos lucros vem acompanhada da redução da participação do investimento e redução dos estoques de bens de capital (Tabela 3). Uma vez que, como sinalizam Rocca & Santos Júnior (2014), quase 50% do investimento é financiado com lucros retidos, a redução destes também contribuiu para a redução da participação dos investimentos no PIB. Este movimento já havia sido sinalizado por Kalecki (1977) e Steindl (1983): a redução dos investimentos vem acompanhada de queda de lucros, dificuldade em honrar dívidas contratadas nas etapas de execução dos projetos de ampliação da capacidade produtiva e incapacidade de retomar o crescimento econômico. Pela análise da Tabela 3, ao longo da década, os investimentos em máquinas e equipamentos se sobressaem aos investimentos em construção, tornando-se parte mais relevante da composição da formação bruta de capital fixo. Se o ciclo recente encontrou seus limites de expansão, o setor de construção poderia assumir maior preponderância, principalmente através de obras de infraestrutura, como sinaliza IMF (2014). 11 A armadilha da liquidez pode ser caracterizada como a incapacidade de reduções na taxa de juros em elevar o nível de gasto e, consequentemente, a renda agregada. A formulação original de Keynes (1936) envolve uma situação limite na qual os agentes preferem reter a moeda incondicionalmente a gastá-la ou emprestá-la. O atual arcabouço não se trata desta situação limite, mas de um processo de desendividamento. As reduções na taxa de juros não irão elevar o nível de gasto, uma vez que este gasto é financiado pela tomada de empréstimos, e agentes altamente endividados não irão tomar novos empréstimos mesmo a taxas de juros mais baixas devido a suas dificuldades correntes em honrar as dívidas previamente contratadas. Ano 2014, N. 32, V. 16, Pag. 214 \| Revista de Estudos Sociais 2. O CENÁRIO ECONÔMICO DOS ANOS 2000 Este aponta para a necessidade de reutilizar a política fiscal no financiamento de novas obras de infraestrutura na tentativa de recompor a demanda global, uma vez que há evidências de que a política monetária via redução da taxa de juros pode ter se tornado incapaz de recompor a demanda privada pelo aumento de endividamento11. A dinâmica supracitada (saldos em transações correntes que permitiram a redução da inflação via redução da taxa de câmbio de forma mais estável e tornaram factível a redução da taxa de juros e desvencilhamento do investimento privado) impactou, também, nas contas públicas. Como já havia ressaltado Kalecki (1977), saldos externos contribuem para a geração de poupanças internas públicas e privadas. Em outras palavras, os ganhos com as transações correntes se transformaram, em elevação de lucros para o setor privado e em aumento da arrecadação para o setor público. Desta forma, nota-se através da observação da Tabela 4 o aumento do superávit primário de 2002 a 2008, ficando sempre acima dos 2%. Como mencionado anteriormente, em 2002, inicia-se a recuperação da balança comercial devido à forte desvalorização da moeda doméstica (Tabela 2 e Gráfico 1); de 2004 a 2008, apesar da valorização da moeda interna, a expansão do preço das commodities contribui para a geração de saldos externos (Gráficos 1 e 2); após 2008, a crise internacional reverte este quadro e as finanças públicas também ficaram comprometidas. O Gráfico 1 demonstra que em 2004 e 2006 o país obteve os maiores saldos externos; nestes anos o governo também obteve os maiores superávits primários, como pode ser visto pela Tabela 4. Ano 2014, N. 32, V. 16, Pag. 214 \| Revista de Estudos Sociais Tabela 4: Resultados do Governo em relação ao PIB (%) 2000 1,73 -3,85 -2,12 47,7 9,44 38,3 2001 1,69 -3,63 -1,94 52,0 9,61 42,4 2002 2,16 -2,84 -0,68 60,4 15,7 44,7 2003 2,28 -5,94 -3,66 54,8 11,2 43,7 2004 2,70 -4,09 -1,39 50,6 7,89 42,7 2005 2,60 -6,01 -3,41 48,4 3,16 45,3 2006 2,17 -5,31 -3,14 47,3 -1,17 48,4 2007 2,23 -4,47 -2,24 45,5 -7,45 53,0 2008 2,35 -3,17 -0,82 38,5 -11,0 49,5 2009 1,31 -4,62 -3,31 42,1 -9,03 51,1 2010 2,09 -3,30 -1,21 39,1 -9,54 48,7 2011 2,25 -4,36 -2,11 36,4 -13,0 49,4 2012 1,96 -3,35 -1,39 35,3 -14,1 49,4 2013 1,55 -3,84 -2,28 33,6 -14,8 48,3 ¹Governo Central. ²Setor Público. Revista de Estudos Sociais \| Ano 2014, N. 32 Vol 16, Pag. 215 2. O CENÁRIO ECONÔMICO DOS ANOS 2000 Fonte: Secretaria do Tesouro Nacional e Banco Central do Brasil Dívida Externa Líquida² Dívida Interna Líquida² Ano Resultado Primário¹ Juros Nominais¹ Resultado Nominal¹ Dívida Total Líquida² Fonte: Secretaria do Tesouro Nacional e Banco Central do Brasil Segundo Lopreato (2006), a busca pelo superávit primário tem a função de ancorar a política monetária ao criar parâmetros para a estabilidade da dívida e compromisso com o pagamento da mesma, contribuindo para a manutenção da estabilidade das expectativas quanto ao risco de default da dívida pública. Ao criar um ambiente de expectativas favoráveis ao investidor estrangeiro interessado em adquirir títulos da dívida pública, a atração de capitais internacionais seria facilitada mesmo a uma taxa de juros menor. A entrada de tais recursos contribuiria para a queda da taxa de câmbio e redução da inflação. Entretanto, a existência de déficits nominais significativos é responsável pelo aumento do endividamento interno (Tabela 4), o que poderia sugerir ao investidor internacional que risco de default estaria a aumentar. Logo, a suposta ancoragem proposta por Lopreato (2006) dependeria de certa miopia dos investidores domésticos e externos. Na verdade, a combinação de superávit primário e déficit nominal cria uma situação singular, na qual a liquidez criada pelo Estado é captada por agentes privados específicos: os detentores da dívida pública. Como identificou Lerner (1943), a expansão monetária promovida pelo gasto fiscal se transforma em poder de gasto de agentes privados, o qual, uma vez efetivado, contribui para o aumento de preços. Desta forma, do ponto de vista dos fluxos monetários, uma política fiscal contracionista, representada pela meta de superávit primário, contribui para reduzir o poder de gasto de agentes privados, criando uma pressão para a baixa de preços nos mercados de bens. Por outro lado, pela via dos estoques de riqueza, a liquidez pública continua se expandindo através do déficit nominal e aumento da dívida interna, a qual é Revista de Estudos Sociais \| Ano 2014, N. 32 Vol 16, Pag. 215 captada por agentes privados sem passar pelo mercado de bens e inflar seus preços. Em momentos de instabilidade doméstica ou internacional, os agentes nacionais e internacionais realizam a liquidação destes títulos (ou parte deles) e buscam salvaguarda em ativos externos, uma vez que os ativos internos não são vistos como ativos de reserva, mas sim ativos de investimento ou especulação12. 12 Segundo Hicks (1967), os ativos podem ser classificados como de transação, de reserva ou de investimento. Ativos de transação são todos os que podem ser utilizados na compra e venda de bens e outros ativos. A moeda e os depósitos bancários cumpririam esta função. Ativos de reserva são todos os ativos que são mantidos no portfólio de um determinado agente porque estes representam boas reservas de valor. Ativos de investimento são preferidos por determinados agentes porque oferecem um retorno pecuniário ou a possibilidade de valorização de seu preço. As classificações entre ativos de reserva e ativos de investimento não seriam fixas, dependeriam das estratégias de atuação de cada agente e dos mercados nos quais eles podem ou não atuar. 12 Segundo Hicks (1967), os ativos podem ser classificados como de transação, de reserva ou de investimento. Ativos de transação são todos os que podem ser utilizados na compra e venda de bens e outros ativos. A moeda e os depósitos bancários cumpririam esta função. Ativos de reserva são todos os ativos que são mantidos no portfólio de um determinado agente porque estes representam boas reservas de valor. Ativos de investimento são preferidos por determinados agentes porque oferecem um retorno pecuniário ou a possibilidade de valorização de seu preço. As classificações entre ativos de reserva e ativos de investimento não seriam fixas, dependeriam das estratégias de atuação de cada agente e dos mercados nos quais eles podem ou não atuar. 13 Dos Santos & Macedo e Silva (2010) destacam a interação destes balanços para a economia dos EUA, evidenciando como o ciclo de crescimento desta foi alcançado pelos saldos negativos das famílias e empresas norte-americanas, caracterizando um processo de endividamento persistente. No mesmo período, o setor financeiro foi amplamente superavitário, recebendo os ganhos do referido processo de endividamento. Os ganhos financeiros foram ampliados pelo recorrente déficit público. 2. O CENÁRIO ECONÔMICO DOS ANOS 2000 Na tentativa de conter o depauperamento do balanço de pagamentos, a elevação da taxa de câmbio e seu corolário mais imediato, o aceleramento inflacionário, a autoridade monetária é obrigada a elevar a taxa de juros de referência, oferecendo retornos mais atrativos para compensar a percepção negativa acerca da liquidez dos ativos domésticos. Como ressalta Dos Santos & Macedo e Silva (2010), da mesma forma que os saldos externos são responsáveis por ampliar as poupanças pública e privada – movimento observado entre 2003 e 2007 para a economia brasileira – o déficit público é adquirido por agentes privados internos ou externos – movimento perene na economia nacional, como supracitado – e os resultados privados negativos são adquiridos pelo Estado ou pelo setor externo – um possível cenário após a crise de 2008. Qualquer um destes processos (saldos externos, déficit público ou prejuízos privados) terá como corolário o aumento do endividamento dos agentes associados a cada um destes setores13. Desta forma, no período de elevado crescimento, de 2004 a 2008, nota- se a presença de amplos saldos comerciais, sendo que o saldo em transações correntes se manteve superavitário de 2003 a 2007. No mesmo período a dívida pública interna se expandiu significativamente com elevados déficits nominais. Estes fenômenos implicam nos aumentos dos ganhos privados, o que contribuiu para ampliar o investimento, notadamente superior no período. Este mesmo cenário permitiu o desendividamento externo do setor público, como pode ser visto pela Tabela 4. As receitas externas provenientes dos saldos correntes no balanço de pagamentos permitiram o acúmulo de reservas e o desendividamento externo líquido do setor público. Por outro lado, os déficits nominais contribuíram para o aumento da dívida interna líquida, através do processo anteriormente mencionado. Por fim, o processo de acumulação de reservas se mostrou mais intenso do que o de aumento do endividamento interno do setor público, engendrando a redução da dívida líquida total do Estado. 12 Segundo Hicks (1967), os ativos podem ser classificados como de transação, de reserva ou de investimento. Ativos de transação são todos os que podem ser utilizados na compra e venda de bens e outros ativos. A moeda e os depósitos bancários cumpririam esta função. Ativos de reserva são todos os ativos que são mantidos no portfólio de um determinado agente porque estes representam boas reservas de valor. Ano 2014, N. 32, V. 16, Pag. 216 \| Revista de Estudos Sociais 2. O CENÁRIO ECONÔMICO DOS ANOS 2000 Ativos de investimento são preferidos por determinados agentes porque oferecem um retorno pecuniário ou a possibilidade de valorização de seu preço. As classificações entre ativos de reserva e ativos de investimento não seriam fixas, dependeriam das estratégias de atuação de cada agente e dos mercados nos quais eles podem ou não atuar. q p 13 Dos Santos & Macedo e Silva (2010) destacam a interação destes balanços para a economia dos EUA, evidenciando como o ciclo de crescimento desta foi alcançado pelos saldos negativos das famílias e empresas norte-americanas, caracterizando um processo de endividamento persistente. No mesmo período, o setor financeiro foi amplamente superavitário, recebendo os ganhos do referido processo de endividamento. Os ganhos financeiros foram ampliados pelo recorrente déficit público. Ano 2014, N. 32, V. 16, Pag. 216 \| Revista de Estudos Sociais Ano 2014, N. 32, V. 16, Pag. 216 \| Revista de Estudos Sociais Ano 2014, N. 32, V. 16, Pag. 216 \| Revista de Estudos Sociais 3. A POLÍTICA MACROECONÔMICA DOS ANOS 2000 O atual regime de políticas econômicas consiste essencialmente na busca de superávits primário por parte da política fiscal, estabelecimento de metas de inflação a serem atingidas pelo controle da taxa de juros de curto prazo pela via monetária e taxa de câmbio flutuante pela política cambial. Não é possível compreender este atual conjunto de políticas econômicas sem antes delinear alguns aspectos da estabilização inflacionária brasileira. De forma sintética, o plano de estabilização inflacionária consistia em ancorar os preços domésticos aos preços externos através de relativa estabilidade cambial14. Não faz parte do objetivo deste artigo discutir os pormenores da concepção e execução do plano de estabilização. Entretanto, vale ressaltar que, como menciona Franco (1993), a adoção de um plano de estabilização nos moldes do que fora executado dependia dos seguintes fatores: contração fiscal para não ocorrer deterioração das expectativas inflacionárias15; quantidade elevada de reservas em moeda estrangeira na qual ocorresse ancoragem para a sustentação da mesma; e controle sobre a expansão dos agregados monetários para que o aumento do crédito não gerasse excesso de demanda e para que não ocorresse uma liquidação massiva de ativos domésticos e aquisição de ativos externos. O próprio autor sinaliza que as soluções destes problemas potenciais seriam o ajuste fiscal com redução de gastos e (ou) aumento de tributos e a elevação das taxas de juros dos ativos domésticos (essencialmente dívida pública). A contração fiscal impediria a atuação da autoridade pública como um fator de expansão da quantidade de moeda para agentes privados. A elevação das taxas de juros doméstica atuaria na atração de capitais externos e constituição de reservas para a manutenção da âncora cambial e inibiria a expansão do crédito e a fuga de capitais para ativos externos que representavam boas reservas de valor. Nota-se que, com exceção da política cambial, os moldes da política econômica atual foram traçados no plano de estabilização. Como sinalizado anteriormente, ao longo da década de 1990, os elevados déficits em transações correntes corroeram as reservas internacionais e a contração de liquidez internacional, devido às crises em países emergentes no fim da década de 1990, dificultaram a entrada de capitais no país mesmo com taxas de juros elevadas; decretando o fim da ancoragem cambial. Assim, instituiu-se o regime de câmbio flutuante, por parte da política cambial, e o regime de metas de inflação pela via monetária. 14 Era permitida a flutuação da taxa de câmbio em um sistema de bandas assimétricas, determinando um nível máximo para a taxa de câmbio de reais por dólares. Se a taxa de câmbio ameaçasse ultrapassar seu valor de teto, a autoridade monetária atuaria para manter a estabilidade cambial e, por consequência, dos preços. p ç 15 A expansão dos déficits públicos implicaria na expansão dos agregados monetários, frente a isso os agentes poderiam projetar expectativas de inflação mais elevadas, antecipando aumento de preços esperados para períodos correntes. 16 Taylor (1993) sugere uma equação de reação da taxa de juros baseada na diferença entre o produto corrente e o produto potencial. Desta forma, uma economia sobreaquecida, com produto corrente acima do potencial ou natural e vivenciando um processo de elevação de preços, teria sua taxa de juros básica p ç 15 A expansão dos déficits públicos implicaria na expansão dos agregados monetários, frente a isso os agentes poderiam projetar expectativas de inflação mais elevadas, antecipando aumento de preços esperados para períodos correntes. 16 Taylor (1993) sugere uma equação de reação da taxa de juros baseada na diferença entre o produto esperados para períodos correntes. 16 Taylor (1993) sugere uma equação de reação da taxa de juros baseada na diferença entre o produto corrente e o produto potencial. Desta forma, uma economia sobreaquecida, com produto corrente acima do potencial ou natural e vivenciando um processo de elevação de preços, teria sua taxa de juros básica elevada com o intuito de reduzir o produto corrente e, consequentemente, o nível de preços. Seguindo este modelo, os preços domésticos estariam ancorados ao hiato de produto. 17 A equação de reação da taxa de juros utilizada pelo Banco Central do Brasil incorpora tanto o hiato de este modelo, os preços domésticos estariam ancorados ao hiato de produto. 17 A equação de reação da taxa de juros utilizada pelo Banco Central do Brasil incorpora tanto o hiato de produto quanto a variação da taxa de câmbio, mas como o próprio Relatório de Inflação do Banco Central do Brasil (2012) demonstra, os efeitos dos choques cambiais provocam efeitos mais intensos. elevada com o intuito de reduzir o produto corrente e, consequentemente, o nível de preços. Seguindo este modelo, os preços domésticos estariam ancorados ao hiato de produto. 17 A equação de reação da taxa de juros utilizada pelo Banco Central do Brasil incorpora tanto o hiato de produto quanto a variação da taxa de câmbio, mas como o próprio Relatório de Inflação do Banco Central do Brasil (2012) demonstra, os efeitos dos choques cambiais provocam efeitos mais intensos. elevada com o intuito de reduzir o produto corrente e, consequentemente, o nível de preços. Seguindo este modelo, os preços domésticos estariam ancorados ao hiato de produto. 17 A equação de reação da taxa de juros utilizada pelo Banco Central do Brasil incorpora tanto o hiato de produto quanto a variação da taxa de câmbio mas como o próprio Relatório de Inflação do Banco Central Ano 2014, N. 32, V. 16, Pag. 218 \| Revista de Estudos Sociais elevada com o intuito de reduzir o produto corrente e, consequentemente, o nível de preços. Seguindo este modelo, os preços domésticos estariam ancorados ao hiato de produto. 17 A equação de reação da taxa de juros utilizada pelo Banco Central do Brasil incorpora tanto o hiato de produto quanto a variação da taxa de câmbio, mas como o próprio Relatório de Inflação do Banco Central do Brasil (2012) demonstra, os efeitos dos choques cambiais provocam efeitos mais intensos. 3. A POLÍTICA MACROECONÔMICA DOS ANOS 2000 No entanto, a ancoragem dos preços domésticos não estaria associada ao hiato de produto, como sugerido na proposição original de Taylor (1993)16, Revista de Estudos Sociais \| Ano 2014, N. 32 Vol 16, Pag. 217 mas pela taxa de câmbio, mantendo, essencialmente, todo o arcabouço precedente17. A elevada taxa de juros continuaria atuando como fonte de atração de capitais estrangeiros, restrição do mercado de crédito e inibição de liquidação de ativos domésticos e fuga para ativos externos, com o intuito de manter os preços domésticos ancorados nos preços externos pela relativa estabilidade cambial. O regime ganhou um novo nome – “metas de inflação” – mas está constituído sobre a mesma ótica do plano de estabilização. É difícil supor que a simples modificação do nome do regime modifique a maneira como os agentes privados traçam suas expectativas inflacionárias e como constituem seus contratos, principalmente quando os instrumentos são essencialmente os mesmos: taxa de juros elevada e contração fiscal na tentativa de manter estável a taxa de câmbio e o nível de preços. Fonte: IBGE e Banco Central Fonte: IBGE e Banco Central Fonte: IBGE e Banco Central O Gráfico 3 demonstra a reação da inflação (IPCA – Índice Nacional de Preços ao Consumidor Amplo) após variações da taxa de câmbio. Alguns meses após o choque cambial de 2001 e 2002 a inflação se eleva acima de 16%. A tendência de queda da taxa de câmbio, iniciada em 2003, é acompanhada, alguns meses depois, por uma tendência de queda do índice de inflação. Após novo choque cambial em 2008, a inflação se eleva novamente e se estabiliza em torno dos 6% enquanto a taxa de câmbio evidencia uma leve tendência de alta. O Banco Central do Brasil (2012) evidencia que os efeitos de variação da taxa de câmbio, como potencial acelerador de preços, são muito mais intensos do que os efeitos de elevação da taxa de juros e geração de superávits primários como efeitos depressivos de preços. O mesmo trabalho Ano 2014, N. 32, V. 16, Pag. 218 \| Revista de Estudos Sociais Ano 2014, N. 32, V. 16, Pag. 218 \| Revista de Estudos Sociais também ressalta que os efeitos da variação do câmbio são mais intensos sobre os preços dos alimentos e bebidas do que outros bens e serviços. Fonte: IBGE e Banco Central Fonte: IBGE e Banco Central O Gráfico 4 ilustra este movimento. 18 Os autores destacam a dificuldade e possível imprecisão das estimativas das taxas de desemprego e de ocupação da capacidade produtiva não aceleradoras da inflação e do produto potencial. 3. A POLÍTICA MACROECONÔMICA DOS ANOS 2000 O choque cambial de 2001 e 2002 acelerou os preços dos alimentos, alguns meses depois, para além dos 25%. A tendência de queda da taxa de câmbio, iniciada em 2003, também mostrou forte impacto sobre o setor que teve deflação em 2006. No entanto, após esta data, os preços dos alimentos se aceleraram antes mesmo de um impacto cambial, fruto do aumento de preços nos mercados internacionais (Gráfico 2). O comportamento do setor de alimentos é assim caracterizado devido à possibilidade de exportar tais bens para outras economias. As elevações na taxa de câmbio, ou aumento dos preços nos mercados externos, tornam a exportação destes bens altamente vantajosa, restringindo a oferta interna e elevando os preços domésticos. Alves & Correa (2013), na tentativa de adaptar a curva de Phillips novo- keynesiana às especificidades recentes da economia brasileira, dividem a economia em dois setores: bens comercializáveis e bens não comercializáveis. Para os bens comercializáveis, as variações de preços estariam relacionadas ao hiato de capacidade produtiva instalada, diferença entre a capacidade de efetivamente instalada e a taxa de capacidade produtiva instalada não aceleradora da inflação. Para os bens não comercializáveis, adota-se a relação convencional novo-keynesiana de hiato de desemprego, diferença entre o desemprego observado e a taxa de desemprego não aceleradora da inflação. O hiato de produto, diferença entre produto corrente e produto natural, influenciaria os preços de forma geral. Os hiatos de capacidade instalada e produto contribuem positivamente para inflação e o hiato de desemprego contribui negativamente18. Revista de Estudos Sociais \| Ano 2014, N. 32 Vol 16, Pag. 219 A conclusão dos autores indica que o aquecimento do mercado de trabalho ao longo da década marcado pela redução do desemprego teve um impacto significativo nos preços dos bens não comercializáveis. Para os bens comercializáveis, a trajetória mais recente indicaria um movimento contrário devido ao desaquecimento da economia brasileira. Os autores consideraram que o setor de bens comercializáveis é principalmente price-taker, logo a influência interna nos preços destes bens se daria pelo excesso de capacidade de utilização, ou uma inflação de custos. É importante observar que os efeitos da taxa de juros sobre as variáveis não são levados em consideração, logo, não é possível avaliar se o modelo convencional de controle inflacionário pela taxa de juros é factível. Ano 2014, N. 32, V. 16, Pag. 220 \| Revista de Estudos Sociais 3. A POLÍTICA MACROECONÔMICA DOS ANOS 2000 Desta forma, para o período recente, novas elevações na taxa de juros poderiam deprimir ainda mais a atividade industrial e o crescimento sem obter os resultados esperados sobre a taxa de desemprego e o mercado de trabalho. Fonte: IBGE e Seade Fonte: IBGE e Seade O movimento mencionado por Alves & Correa (2013) pode ser observado pelo Gráfico 5. Ao longo da década a redução do nível de desemprego foi acompanhada pelo aumento de preços dos bens comercializáveis. Nota-se que os efeitos cambiais de 2001 e 2002 também são preponderantes sobre a inflação dos referidos bens, uma vez que a indexação de contratos privados, reminiscência da década de 1980, ainda permanece como uma prática comum. Como mencionam Coutinho & Belluzzo (1998), a inflação de bens não comercializáveis já era uma característica comum nos países desenvolvidos ao longo dos ciclos de expansão das décadas de 1990 e 2000. Para Keynes (1936), os processos de expansão de preços são fruto da incapacidade da oferta de determinados bens e serviços reagir aos aumentos de preços precedentes. O setor produtor de alimentos demora a ampliar a oferta de seus produtos, uma vez que cada um deles obedece a um ciclo produtivo natural, mais ou menos extenso. O setor de bens não comercializáveis, notadamente serviços (Gráfico 6), marcado pela sua incapacidade de gerar ganhos de produtividade, só pode ampliar sua oferta Ano 2014, N. 32, V. 16, Pag. 220 \| Revista de Estudos Sociais através do emprego de mais indivíduos; com um mercado de trabalho com baixo desemprego a oferta de serviços reage cada vez mais lentamente ao aumento de preços. Fonte: IBGE e Seade Fonte: IBGE e Seade As características da inflação brasileira demonstraram a dificuldade da atual combinação de políticas macroeconômicas em atingir seu objetivo. A geração de superávits primários teria o intuito de reduzir o poder de gasto de agentes privados, reduzindo, assim, os gastos das famílias brasileiras com os serviços; e as elevadas taxas de juros deveriam atrair capitais, reduzir a taxa de câmbio e controlar a inflação de alimentos pelo desestímulo à oferta externa. No entanto, os efeitos de redução de gastos são abrangentes e não equânimes. Não há justificativa para que a redução de gastos públicos reduza os gastos privados com serviços de forma mais acentuada do que o gasto privado com outros bens. 19 Kaldor (1960) modifica os fatores que compõem as taxas próprias de juros desenvolvidas por Keynes (1936). As taxas próprias de juros delineiam uma teoria de escolha de ativos e são determinadas por quatro fatores: a expectativa de valorização ou desvalorização em termos monetários, os retornos pecuniários associados à posse de um determinado ativo, o custo de manter esse ativo no portfólio do agente e a capacidade deste ativo ser rapidamente transformado em moeda. O risco de iliquidez Revista de Estudos Sociais \| Ano 2014, N. 32 Vol 16, Pag. 221 3. A POLÍTICA MACROECONÔMICA DOS ANOS 2000 No atual cenário, uma redução de gastos privados com bens industriais pode deprimir ainda mais as expectativas de retorno do setor empresarial, reduzindo o investimento corrente. As elevadas taxas de juros têm seu efeito limitado no controle de preços, apesar do impacto severo na determinação do nível de investimento. Em um contexto de moeda inconversível, como salientado por Carneiro (2003 e 2008), a incapacidade da moeda doméstica em ser aceita no plano externo como reserva de valor (tanto para residentes quanto para não residentes) implica em elevada volatilidade cambial, uma vez que a moeda doméstica e os ativos referenciados na mesma sofrem das variações na percepção do risco de iliquidez a eles associados por parte dos agentes globais19. Nas fases de expansão dos ciclos de liquidez global, a percepção do risco de iliquidez dos agentes globais para com os ativos domésticos se reduz, facilitando a entrada de capitais e permitindo a redução da taxa de juros; movimento que pode ser reforçado pela queda das taxas de juros nos mercados de moedas conversíveis. Na reversão destes ciclos, a percepção de risco de iliquidez se eleva, forçando a elevação da taxa de juros como efeito compensatório na tentativa de evitar uma fuga massiva de capitais. No entanto, a reação na taxa de juros não é imediata e novos aumentos desta podem ser seguidos de subsequentes elevações no risco de iliquidez, o que exacerba a volatilidade da taxa de câmbio e, por consequência, dos preços dos bens e serviços internos. Portanto, frente à volatilidade cambial, o Banco Central do Brasil acaba por ser obrigado a intervir no mercado futuro de câmbio na tentativa de diminuir as oscilações cambiais. Mesmo sob um regime de câmbio flutuante, as intervenções se fazem necessárias para diminuir o impacto nos preços e no setor produtivo. Da mesma forma que uma desvalorização drástica da moeda nacional gera aumento dos preços internos, uma valorização acentuada inibe a atividade industrial doméstica, uma vez que este processo desvaloriza artificialmente as importações, tornando-as mais atrativas do que os mesmos produtos fabricados internamente. evidenciaria a dificuldade do ativo em ser transformado em moeda. Desta forma, em equilíbrio, os ganhos oriundos da expectativa de valorização ou dos retornos pecuniários seriam compensados pelo risco de iliquidez. Em momentos de otimismo, as expectativas de valorização ou de retorno pecuniário seriam percebidas como superiores ao risco de iliquidez, contribuindo para a expansão do mercado deste ativo. Em momento de pessimismo, a percepção de risco de iliquidez se sobrepõe às expectativas de ganho e os agentes procuram os ativos com os menores riscos ou a própria moeda, que apresentaria risco nulo. Para o cenário internacional, apenas as moedas conversíveis poderiam apresentar risco de iliquidez nulo ou negligenciável, as demais estariam sujeitas as alterações na percepção do mesmo. 20 McCombie & Thirlwall (1994) afirmam que o processo de industrialização resulta na modificação dos bens exportados e bens importados, o que altera as elasticidades preço e renda da demanda dos mesmos. Ao modificar a base produtiva, uma determinada economia passará a depender menos da importação de bens com elevadas elasticidades renda e preço e passará a exportar mais destes mesmos bens, criando a possibilidade de intensificar ou ampliar o ciclo de crescimento devido aos ganhos oriundos das demandas externas. Economias que não conseguem expandir ou concluir o processo de industrialização estariam fadadas a ciclos curtos de crescimento interrompidos pelas restrições de balanço de pagamentos oriundas da elevação das importações devido ao aumento da renda interna ou interrupção do financiamento externo. 4. CONSIDERAÇÕES FINAIS O ciclo de expansão e retração (ou estagnação) da Economia Brasileira recente está intimamente relacionado ao ciclo de expansão e retração externo. Ao longo da década de 2000, a expansão da economia brasileira foi financiada pela capitação de poupanças externas, representada pelos saldos em transações correntes positivos. Este processo foi possível devido à expansão de liquidez nos mercados internacionais, que elevou os preços das commodities, como ressalta Wray (2008), e a expansão da economia chinesa e outros mercados emergentes demandantes destes produtos. Os saldos em transações correntes reduziram a dependência de elevadas taxas de juros para atrair capitais externos e manter os preços estáveis ou declinantes através da valorização da moeda doméstica; criando espaço para a redução da taxa de juros interna. A queda desta, aliada a política ativa do BNDES, como ressaltado por Dedecca, Trovão & Souza (2014), foi responsável pela ampliação do investimento agregado, o que deu início a fase de expansão vigorosa do produto interno. O aumento da absorção interna, aliado aos saldos externos, elevou a arrecadação pública, o que Ano 2014, N. 32, V. 16, Pag. 222 \| Revista de Estudos Sociais Ano 2014, N. 32, V. 16, Pag. 222 \| Revista de Estudos Sociais Ano 2014, N. 32, V. 16, Pag. 222 \| Revista de Estudos Sociais permitiu a expansão das políticas públicas de distribuição de renda, como sublinham os autores acima mencionados. permitiu a expansão das políticas públicas de distribuição de renda, como sublinham os autores acima mencionados. A expansão da renda interna, somada à ampliação do mercado consumidor, resultado das políticas públicas de distribuição de renda, e a valorização da moeda doméstica, possibilitada pelos saldos externos e atuação branda do Banco Central do Brasil nos mercados de câmbio, acabou por ampliar as importações, reduzindo progressivamente os referidos saldos. Assim, a economia brasileira foi, no decorrer do ciclo, retomando sua vulnerabilidade frente aos eventuais choques de liquidez externos. Em 2008, após o pânico financeiro ser instaurado pela falência do Lehman Brothers, o enxugamento da liquidez global devido à deflação das dívidas hipotecárias securitizadas, como salientou Freitas & Cintra (2008), engendrou em elevadas saídas de capitais do país, desvalorizando significativamente a moeda nacional. 4. CONSIDERAÇÕES FINAIS interrupção do acúmulo de poupanças externas força a autoridade monetária a elevar as taxas de juros domésticas na tentativa de manter os preços estáveis e acaba por condenar o crescimento, inviabilizando a realização de novos investimentos e as possibilidades de surgimento de mecanismos privados de financiamento interno. Portanto, este artigo procurou salientar de maneira sucinta a relação entre o movimento cíclico da economia brasileira recente e seus corolários no atual marco de políticas macroeconômicas. A expansão dos mercados externos permitiu a expansão monetária e creditícia interna, elevando investimento e renda. A vulnerabilidade através de inflexões no balanço de pagamentos permaneceu latente ao longo do período, tornando-se evidente após a contração de liquidez nos mercados externos, o que impediu a continuidade das expansões monetárias e creditícias. 4. CONSIDERAÇÕES FINAIS Os baixos saldos externos, em um cenário de contração da liquidez internacional, aumentaram a percepção de risco de iliquidez dos agentes externos e internos quanto aos ativos nacionais e a moeda na qual estão lastreados, forçando a elevação da taxa de juros interna. Nos anos posteriores, este efeito foi compensado pela significativa redução das taxas de juros nas economias de moeda conversível. Entretanto, os seguidos déficits em transações correntes, que contribuíram para a queda das poupanças públicas e privadas, e a forte redução da demanda internacional, tornaram a economia brasileira novamente dependente da taxa de juros para atrair capitais e manter relativamente estável a taxa de câmbio e o nível de preços. Logo, frente a novos choques de liquidez, e aumento da percepção de risco de iliquidez, o atual arcabouço de políticas macroeconômicas poderá apenas elevar a taxa de juros na tentativa de conter o aumento da taxa de câmbio e dos preços internos. Segundo Carneiro (2008), as vertentes ortodoxas das teorias relacionadas à tese de inconversibilidade monetária afirmam que a redução da dependência das poupanças externas para moedas inconversíveis poderia ser atingida através da obtenção da estabilidade de preços. O autor afirma que isto pode ser contestado, uma vez que a estabilidade de preços foi alcançada em muitos países de moeda inconversível, mas esta condição ainda é presente nestas economias. Na verdade, a estabilidade de preços foi alcançada com a utilização das poupanças externas. Em seus modelos usuais de crescimento com restrição de balanço de pagamentos, McCombie & Thirlwall (1994) sinalizam que economias dependentes de financiamento externo se tornam estagnadas quando este se reduz, em especial, quando o processo de industrialização não resultou em modificação do diferencial de elasticidades de exportação e importação20. A economia brasileira recente desenvolveu um círculo vicioso semelhante: a Revista de Estudos Sociais \| Ano 2014, N. 32 Vol 16, Pag. 223 Revista de Estudos Sociais \| Ano 2014, N. 32 Vol 16, Pag. 223 interrupção do acúmulo de poupanças externas força a autoridade monetária a elevar as taxas de juros domésticas na tentativa de manter os preços estáveis e acaba por condenar o crescimento, inviabilizando a realização de novos investimentos e as possibilidades de surgimento de mecanismos privados de financiamento interno. 5. REFERÊNCIAS BIBLIOGRÁFICAS ALVES, S. & CORREA, A. Um Conto de Três Hiatos: Desemprego, Utilização da Capacidade Instalada da Indústria e Produto. Banco Central do Brasil: Trabalho para Discussão 339, dezembro, 2013. ALVES, S. & CORREA, A. Um Conto de Três Hiatos: Desemprego, Utilização da Capacidade Instalada da Indústria e Produto. Banco Central do Brasil: Trabalho para Discussão 339, dezembro, 2013. BACHA, E. O Fisco e a Inflação: uma Interpretação do caso Brasileiro. Revista de Economia Política, vol. 14, 1994. BACHA, E. O Fisco e a Inflação: uma Interpretação do caso Brasileiro Revista de Economia Política, vol. 14, 1994. BANCO CENTRAL DO BRASIL. Relatório de Inflação, junho, 2012. Ano 2014, N. 32, V. 16, Pag. 224 \| Revista de Estudos Sociais BANCO CENTRAL DO BRASIL. Relatório de Inflação, junho, 2012. CARNEIRO, R. A política macroeconômica da era FHC ao governo Lula: da trindade impossível à autonomia necessária. VELLOSO, João Paulo dos Reis (Coor.) Governo Lula: novas prioridades, 2003. CARNEIRO, R. “Globalização e inconversibilidade monetária”. Revista de Economia Política 28 (4), pp. 539-556, outubro-dezembro 2008. COUTINHO, L. & BELLUZZO, L. G. M. “Financeirização” da riqueza, inflação de ativos e decisões de gasto em economias abertas. Economia e Sociedade, Campinas, (11): 137-50, dez. 1998. DEDECCA, C. TROVÃO, C. & SOUZA, L. Desenvolvimento e Equidade: Desafios do Crescimento Brasileiro. Novos Estudos (98), março, 2014. DEMBIERMONT, C. DREHMANN, M. & MUKSAKUNRATANA, S. How much does the rivate sector really borrow? A new database for total credit to the private non-financial sector. BIS Quarterly Review, Março, 2013. DOS SANTOS, C. % MACEDO E SILVA, A. Revisiting “New Cambridge”: The Three Financial Balances in a General Stock-flow Consistent Applied Modeling Strategy. The Levy Economics Institute of Bard College: Working Paper No. 594, maio, 2010. EICHENGREEN, B. Globalizing capital: a history of the international monetary system. Princeton: Princeton University Press, 1996. FARHI, M. O impacto dos ciclos de liquidez no Brasil: mercados financeiros, taxa de câmbio, preços e política monetária. Em Carneiro, R Ano 2014, N. 32, V. 16, Pag. 224 \| Revista de Estudos Sociais (org) A supremacia dos mercados e a política econômica do Governo Lula. SP: UNESP, 2006. (org) A supremacia dos mercados e a política econômica do Governo Lula. SP: UNESP, 2006. FISHER, S. & BLANCHARD, O. Lectures on Macroeconomics. Cambridge: MIT Press, 1989. FLEMING, J. M. Domestic Financial Policies under Fixed and under Floating Exchange Rates. Staff Papers - IMF, Vol. 9, No. 3. Novembro, 1962. FRANCO, G. Alternativas de Estabilização: gradualismo, dolarização e populismo. Revista de Economia Política, vol. 13, nº 2 (50), abril-junho, 1993. FREITAS, M. & CINTRA. M. Inflação e deflação de ativos a partir do mercado imobiliário americano. Revista de Economia Política, vol. 28, nº 3 (111), pp. 414-433, julho-setembro, 2008. INTERNACIONAL MONETARY FUND. World Economic Outlook, outubro, 2014. GODLEY, W. Seven Unsustainable Processes: Medium-Term Prospects and Policies for the United States and the World. Special Report. Annandale-on-Hudson, N.Y.: Levy Economics Institute of Bard College, 1999. Economics Institute of Bard College, 1999. HICKS, J. Critical Essays in Monetary Theory. Lectures I-III. Oxford, Clarendon Press, 1967. KALDOR, N. Essays on Economic Stability and Growth. Capítulos 1 e 2. Illinois, The Free Press of Glencoe, 1960. KALECKI, M. BANCO CENTRAL DO BRASIL. Relatório de Inflação, junho, 2012. Teoria da Dinâmica Econômica. São Paulo, Editora Nova Cultural, 1977. KALECKI, M. Teoria da Dinâmica Econômica. São Paulo, Editora Nov Cultural, 1977. KEYNES, J. M. (1930). A Treatise on Money: The Pure Theory of Money. The Collected Writings of John Maynard Keynes, v. 5. Cambridge, MacMillan, 1971. KEYNES, J. M. (1936). A Teoria Geral do Emprego, do Juro e da Moeda. São Paulo, Editora Nova Cultural, 1996. LERNER, A. Functional Finance and the Federal Debt. Social Research, 10(1), pp. 38-51, 1943. LOPREATO, F. Política fiscal: mudanças e perspectivas. In: Carneiro, R (org) A supremacia dos mercados e a política econômica do Governo Lula. SP: UNESP, 2006. MCCOMBIE, J. & THIRLWALL, A. Economic Growth and the Balance-of- Payments Constraint. London: Macmillan, 1994. MINSKY, H. (1975). John Maynard Keynes. McGraw-Hill, Nova Iorque, 2008. MODENESI, A. M. Regimes Monetários: teoria e a experiência do Real. Ed. Monole, 2005. MUNDELL, R. A. Capital Mobility and Stabilization Policy under Fixed and Flexible Exchange Rates. The Canadian Journal of Economics and Political Science, Vol. 29, No. 4. Novembro, 1963. Revista de Estudos Sociais \| Ano 2014, N. 32 Vol 16, Pag. 225 OBSTFELD, M. SHAMBAUGH, J. & TAYLOR, A. The Trilemma in History: Tradeoffs among Exchange Rares, Monetary Policies, and Capital Mobility. NBER Working Paper Series: março, 2004. ROCCA, C. & SANTOS JR, L. Redução da Taxa de Poupança e Financiamento dos Investimentos no Brasil – 2010 a 2013. Centro de Estudos de Mercado de Capitais, IBMEC, novembro, 2014. SERRANO, F. Do ouro imóvel ao dólar flexível. Economia e Sociedade, Campinas, v. 11, n. 2 (19), p. 237-253, jul./dez. 2002. SOUZA, L. A Crise Financeira de 2008: Uma Interpretação Teórica Heterodoxa. Dissertação de Mestrado, Campinas, 2012. STEINDL, J. Maturidade e estagnação no capitalismo americano. São Paulo: Nova Cultural, 1983. STIGLITZ, J. & GREENWALD, B. Rumo a um Novo Paradigma em Economia Monetária. São Paulo: Francis, 2004. TAYLOR, J. B. Discretion versus Policy Rules in Practice. Carnegie- Rochester Conference Series on Public Policy, vol. 39, 1993. TRIFFIN, R. Gold and the Dollar Crisis. New Haven, Conn.: Yale University Press, 1960. WRAY, L. R. The Commodities Market Bubble: Money Manager Capitalism and the Financialization of Commodities. Public Policy Brief, The Levy Economics Institute of Bard College, 2008. Ano 2014, N. 32, V. 16, Pag. 226 \| Revista de Estudos Sociais
https://openalex.org/W2125545948	https://www.nature.com/articles/tp201273.pdf	English	null	Serotonin transporter genotype 5HTTLPR as a marker of differential susceptibility? A meta-analysis of child and adolescent gene-by-environment studies	Translational psychiatry	2,012	cc-by	5,521	Serotonin transporter genotype 5HTTLPR as a marker of differential susceptibility? A meta-analysis of child and adolescent gene-by-environment studies MH van IJzendoorn1, J Belsky2 and MJ Bakermans-Kranenburg1 MH van IJzendoorn1, J Belsky2 and MJ Bakermans-K We present results of a meta-analysis of gene-by-environment (G E) studies involving the serotonin transporter genotype 5HTTLPR to evaluate empirical support for two competing conceptual frameworks in developmental psychopathology: diathesis-stress and differential susceptibility. From a diathesis-stress perspective, the cumulative negative effects of the short allele (ss and sl genotypes) and adverse environments on development have been stressed. From a differential-susceptibility perspective, carriers of the s allele are predicted to be more open to adverse as well as positive environments, for better and for worse. Studies with children and adolescents up to 18 years of age (N ¼ 9361) were included. We found 41 effect sizes (N ¼ 5863) for the association between negative environments and developmental outcomes with or without signiﬁcant moderation by 5HTTLPR genotype and 36 effect sizes (N ¼ 3498) for the potentially 5HTTLPR-moderated association between positive environments and developmental outcomes. Five moderators were examined: age, ethnicity, genotyping (biallelic or triallelic) and methods used to assess environment and outcome. In the total set of studies, including studies with mixed ethnicities, we found that ss/sl carriers were signiﬁcantly more vulnerable to negative environments than ll carriers, thus supporting the diathesis-stress model. In the Caucasian samples, however, ss/sl carriers also proﬁted signiﬁcantly more from positive environmental input than ll carriers. Associations between (positive or negative) environment and (positive or negative) developmental outcome were absent for ll carriers. The meta-analytic ﬁndings support the hypothesis that in Caucasian samples 5HTTLPR is a genetic marker of differential susceptibility. G E interactions might be critically dependent on ethnicity. Translational Psychiatry (2012) 2, e147; doi:10.1038/tp.2012.73; published online 7 August 2012 Central to the differential-susceptibility hypothesis is the proposition that individuals vary in their susceptibility to the same environmental inﬂuences. Citation: Transl Psychiatry (2012) 2, e147, doi:10.1038/tp.2012.73 & 2012 Macmillan Publishers Limited All rights reserved 2158-3188/12 Serotonin transporter genotype 5HTTLPR as a marker of differential susceptibility? A meta-analysis of child and adolescent gene-by-environment studies From an evolutionary per- spective, it seems implausible that ‘risk’ alleles would have survived if they did not promote ﬁtness in some circumstances or contexts.8 It is thus a bold but also intuitively plausible conjecture that widespread human characteristics such as a ‘reactive’ temperament or a ‘risky’ genotype are, in fact, markers of susceptibility to positive and negative circum- stances, for better and for worse.2,9,10 Susceptibility markers would not have emerged, survived and spread across a substantial minority of the population if they did not advance adaptation to at least some ecological niches for at least some individuals,10 although random genetic drift cannot be excluded as a possible explanation of their existence. Recent years have witnessed a growing body of correlational evidence consistent with differential-susceptibility thinking.3 The ﬁrst such work documenting genetic differential susceptibility focused on the dopamine D4 receptor gene,11 with a recent meta-analysis of G E studies involving dopamine-related genes and children under 10 years of age indicating that associations between positive environments and positive developmental outcomes were as strong as the negative associations in case of the Received 11 April 2012; revised 11 May 2012; accepted 04 July 2012 1Centre for Child and Family Studies, Leiden University, Leiden, The Netherlands and 2University of California at Davis, King Abdulaziz University, Birkbeck University of London, Jeddah, Saudi Arabia Correspondence: Dr MH van IJzendoorn or Dr MJ Bakermans-Kranenburg, Centre for Child and Family Studies, Leiden University, PO Box 9555, Leiden 2300 RB, Netherlands. E-mail: vanijzen@fsw.leidenuniv.nl or E-mail: bakermans@fsw.leidenuniv.nl Keywords: adolescents; children; diathesis stress; differential susceptibility; 5HTTLPR; meta-analysis; serotonin ild and Family Studies, Leiden University, Leiden, The Netherlands and 2University of California at Davis, King Abdulaziz University, Birkbeck University of Materials and methods Data sources. For our meta-analysis, we systematically searched the PsycLit, Medline and ISI web of knowledge databases, with the key words ‘serotonin’ or ‘5HTT’, and ‘human’ in the title or abstract. The asterisk indicates that the search contained the word or word fragment. In addition, references of three recently published review papers on the same topic were scrutinized.3,4,10 Table 1 Characteristics of studies included in the meta-analyses Table 1 Characteristics of studies included in the meta-analyses Study N Positive/ negativea Age Ethnicityb Genotyping triallelic Assessment environment outcome Bakermans et al. (2012) 37 Negative o10 Caucasian No Observation Observation Benjet et al. (2010) 78 Negative X10 Mixed No Self-report Self-report Brody et al. (2009) 419 Positive X10 Mixed No Observation Self-report Cicchetti et al. (2010) 850 Negative o10 Mixed No Observation Self-report Cicchetti et al. (2011) 92 Positive o10 Mixed No Observation Observation Drury et al. (2012) 100 Positive o10 Caucasian No Observation Other report Eley et al. (2004) 220 Negative X10 Caucasian No Other report Self-report Eley et al. (2011) 344 Positive o10 Caucasian No Observation Observation Fox et al. (2005) 73 Positive o10 Caucasian No Other report Observation Gibb et al. (2011) 74 Negative o10 Caucasian No Observation Observation Gilissen et al. (2008) 87 Positive o10 Caucasian No Observation Observation Hankin et al. (2012) 220 Negative X10 Mixed Yes Other report Self-report Ivorra et al. (2010) 317 Negative o10 Caucasian No Other report Other report Jacobs et al. (2011) 123 Negative X10 Caucasian Yes Other report Observation Kaufman et al. (2004) 101 Negative X10 Mixed No Observation Self-report Kochanska et al. (2011) 88 Positive o10 Caucasian No Observation Other report Luijk et al. (2011) 512 Positive o10 Caucasian No Observation Observation Kumsta et al. (2010) 64 Negative X10 Caucasian No Observation Other report Mueller et al. (2011) 115 Negative o10 Caucasian Yes Self-report Observation Nijmeijer et al. (2010) 194 Negative X10 Caucasian No Other report Other report Nilsson et al. (2005) 196 Positive X10 Caucasian No Self-report Self-report Nobile et al. (2007) 689 Negative X10 Caucasian No Other report Other report Paaver et al. (2008) 435 Positive X10 Caucasian No Self-report Self-report Pauli-Pott et al. (2009) 69 Positive o10 Caucasian No Observation Observation Pluess et al. (2011) 1513 Negative o10 Caucasian No Other report Other report Sadeh et al. (2010) 296 Positive X10 Mixed No Other report self-report Sonuga et al. Hypotheses A series of meta-analyses tested two contrasting hypotheses about the role of 5HTTLPR in G E interactions. The dominant diathesis-stress or cumulative-risk hypothesis stipulates that carriers of the s allele are more vulnerable to the negative effects of adverse environments (double risk). The emerging differential-susceptibility model predicts that carriers of the s-allele are not only more vulnerable to the inﬂuence of negative environments but also proﬁt more from the beneﬁcial inﬂuences of positive environments. From a differential-susceptibility perspective, we expect the interac- tive effects of the 5HTTLPR genotype and positive environ- ments on development to be as large as the interaction effects of the same genotype and risk environments. Data extraction. We identiﬁed 77 pertinent effect sizes on 9361 subjects from 30 reports, providing data for two meta- analyses on the moderating role of 5HTTLPR, when it comes to the impact of the environment on development. Forty-one effect sizes (N ¼ 5863) concerned vulnerability, that is, moderation by ‘risk alleles’ (ss or sl) of the association between adverse environment and negative developmental outcomes. Examples of such studies were the associations between bullying victimization experiences and emotional problems,13 between family risk and depression,14 between prenatal maternal anxiety and infant negative emotionality15 and between early institutional deprivation and emotional problems in adolescence.16 Thirty-six effect sizes (N ¼ 3498)—enabling a focus on the ‘bright side’—pertained Introduction Only refereed reports in the English language were included. Studies reporting the presence or absence of a signiﬁcant G E interaction with 5HTTLPR were included even if the main goal of the research was the presentation of a genetic or environmental main effect. Therefore the literature search was ‘blind’ for the kind of G E outcome and unbiased as to theoretical perspective (see Table 1 for a list of studies). Introduction Most gene-by-environment (G E) studies of the interaction of measured genes by measured environments emphasize the cumulative negative effects of speciﬁc ‘risk’ genes and adverse environments, whereas potentially cumulative positive effects of the same ‘risk’ genes (better called ‘susceptibility’ or ‘plasticity’ genes) interacting with positive environments remain understudied.1–4 From a diathesis- stress perspective, the potentially cumulative negative effects of the 5HTTLPR short allele (ss and sl genotypes) and adverse environments have been stressed in many studies, as in the association between negative childhood experiences and adult depression.5–7 The differential-susceptibility per- spective highlights the need to examine the 5HTTLPR- moderated associations between negative and positive environmental inﬂuences and developmental outcomes from the position that certain individuals are not just more vulnerable to adversity because of their genetic make-up, but disproportionately responsive to positive and negative environmental experiences and exposures. Here we present the ﬁrst meta-analytic evidence addressing the question whether the 5HTTLPR genotype should be considered a marker of vulnerability or susceptibility. Differential susceptibility and 5HTTLPR MH van IJzendoorn et al 2 so-called ‘risk’ polymorphisms.12 Here we test whether the same is the case for the short versus long allele of 5HTTLPR, focusing on children under the age of 18. Study selection. The search was restricted to studies with behavioral, psychiatric or developmental outcomes for children under the age of 18 years, thereby excluding purely medical or physical parameters of child development as outcomes for analysis. We ﬁnished the search on 1 March 2012. Medical treatment, as an environmental factor, and also nonempirical papers or papers with insufﬁcient statistics were excluded. Only refereed reports in the English language were included. Studies reporting the presence or absence of a signiﬁcant G E interaction with 5HTTLPR were included even if the main goal of the research was the presentation of a genetic or environmental main effect. Therefore the literature search was ‘blind’ for the kind of G E outcome and unbiased as to theoretical perspective (see Table 1 for a list of studies). Study selection. The search was restricted to studies with behavioral, psychiatric or developmental outcomes for children under the age of 18 years, thereby excluding purely medical or physical parameters of child development as outcomes for analysis. We ﬁnished the search on 1 March 2012. Medical treatment, as an environmental factor, and also nonempirical papers or papers with insufﬁcient statistics were excluded. Materials and methods These studies concerned, for example, the effect of mothers’ positive emotions expressed about their children with attention deﬁcit/hyperactivity disorder,17 the effects of a family-centered prevention program on adolescents’ risky behaviors,18 the decrease in anxiety in response to cognitive behavior therapy19 or the association of high-quality family functioning (as opposed to neutral or low-quality family functioning) with adolescent alcohol consumption.20 As the 5HTTLPR genotype consists of three variants (ss, sl, and ll), studies sometimes reported on the three variants separately, but in other cases the sl group was combined with either the ss or the ll group. Thus, each of ﬁve possible combinations can be found in the literature (see Table 2). The associations of interest were those between (positive or negative) environment and child outcome within each of the ﬁve genotype groups, consisting of carriers of the short (putative ‘risk’) allele (ss and sl genotypes) and carriers of two long alleles (ll). All effect sizes were computed through consensus of two coders (MHvIJ, MBK). Data synthesis. The Comprehensive Meta-Analysis (CMA) program was used to transform the results of the individual studies (for example, means and s.d. in both genotype groups) into the common metric of correlations and to combine effect sizes.23 Heterogeneity across studies was assessed using the Q-statistic. As most of our data sets were heterogeneous in their effect sizes, and as random effects models are somewhat more conservative than ﬁxed effects parameters in such cases, combined effect sizes and conﬁdence intervals (CIs) from random effects models are presented. We also tested the inﬂuence of the ﬁve moderators on the variation of effect sizes between studies with the Qcontrast statistic in a random effects model.23 The Qcontrast statistic is based on the logic of analysis of variance, with the total variance Qtotal partitioned into Qbetween and Qwithin. Qtotal is the variance with any grouping factors ignored, and Qwithin for each group refers to the variances in the speciﬁc subsets of studies. Qbetween ¼ QtotalQwithin, and is tested for signiﬁcance using the w2 distribution.23 A signiﬁcant Qcontrast value indicates that the difference in effect size between subsets of studies is signiﬁcant. Moderators. We included ﬁve moderators to test whether effect sizes varied signiﬁcantly across moderator categories. First, age was coded in two categories: below 10 years (parallel to a previous meta-analysis on G E with dopamine- related genes12), and 10 years and older. Materials and methods Second, as 5HTTLPR ss genotype has been found associated with higher CSF 5-HIAA levels in African Americans, but with lower levels in Caucasians,21 we categorized the studies into those involving 480% Caucasian participants, and studies involving more than 20% other ethnicities. Third, a single- nucleotide polymorphism (SNP) in the L allele (rs25531) has recently led to differentiation between the high functioning La variant versus the Lg variant that is more functionally equivalent to the S allele (the triallelic approach22). We differentiated between the studies using the traditional biallelic approach and studies applying the triallelic method. Fourth, the method of assessing the environment was categorized into studies using observations, questionnaires or interviews completed by persons other than the children (for example, parents), or self-report questionnaires or interviews. Fifth, the same categorization (observation, other-report, self-report) was used for the assessment of the developmental outcomes. Intercoder reliability was adequate (mean 94%, range 80–100%). Disagreements were discussed to reach consensus. Materials and methods (2009) 681 Positive X10 Caucasian No Observation Other report Spangler et al. (2009) 94 Negative o10 Caucasian No Observation Observation Sugden et al. (2010) 1174 Negative X10 Caucasian No Self-report Other report Sulik et al. (2012) 106 Positive o10 Caucasian No Observation Other report Notes. aPositive ¼ study providing effect size of the association between supportive environment and positive developmental outcomes; negative ¼ study providing effect size of the association between adverse environment and negative developmental outcome. bCaucasian ¼ 480% of the sample Caucasian. Translational Psychiatry Differential susceptibility and 5HTTLPR MH van IJzendoorn et al 3 3 or the associations between positive or negative environmental factors and child outcomes for the various 5HTTLPR genotypes in the Table 2 Combined effect sizes for the associations between positive or negative environmental factors and child outcomes for the various 5HTTLPR genotypes in the total set of studies (k ¼ 77) ( ) Genotype Positive environment Negative environment Contrast K N r 95% CI Q k N r 95% CI Q Q P (1) ss 7 312 0.17* 0.07B0.25 5.12 10 715 0.31 0.24B0.37 6.51 6.17 .01 (2) ss/sl 9 1004 0.27 0.09B0.43 110.81 6 1312 0.09 0.13B0.30 20.16 1.59 .21 (3) sl 4 568 0.17* 0.01B0.33 9.36* 9 1772 0.20 0.08B0.31 30.71 0.07 .80 (4) sl/ll 3 415 0.06* 0.01B0.11 0.33 1 170 0.00 0.15B0.15 n.a. n.a. n.a. (5) ll 13 1199 0.11** 0.04B0.19 24.80* 15 1894 0.06 0.01B0.12 12.34 1.30 .26 (6) all ss/sl (1+2+3) 20 1884 0.21 0.12B0.30 135.48 25 3739 0.22 0.14B0.31 79.86 0.03 .86 Po0.05, Po0.01. Note: direction of effect sizes was labeled according to the a priori hypotheses of this meta-analysis. Po0.05, *Po0.01. Note: direction of effect sizes was labeled according to the a priori hypotheses of this meta-analysis. to the relation between supportive contexts (for example, warm-responsive parenting) and positive developmental outcomes. Results Figure 1 Combined effect sizes for the associations between positive and negative environmental factors and child outcomes for the 5HTTLPR ss/sl and ll genotypes, in the total set of studies (except sl/ll, k ¼ 73) and in studies with 480% Caucasian participants (k ¼ 52). Po0.05, *Po0.01. Note: In the total set of studies, the combined effect sizes for positive environmental factors are signiﬁcant for both ss/sl and ll carriers. The difference between the combined effect sizes for ss/sl and ll is not signiﬁcant, Qcontrast ¼ 0.54, P ¼ 0.46. The combined effect size for negative environmental factors is signiﬁcant for ss/sl carriers, but not for ll carriers. The combined effect sizes for ss/sl and ll differ signiﬁcantly, Qcontrast ¼ 8.35, Po0.01. In the set of studies with mostly Caucasian participants, the combined effect sizes for positive and negative environmental factors are signiﬁcant for the ss/sl carriers, but not for the ll carriers. The differences between the combined effect sizes for ss/sl and ll carriers are signiﬁcant, Qcontrast ¼ 3.92, P ¼ 0.048 for positive environmental factors and Qcontrast ¼ 4.58, P ¼ 0.03 for negative environmental factors. Moderators. Moderator analyses focused on age, use of triallelic genotyping, the type of assessment of the environ- ment and the outcome were not signiﬁcant, and thus did not change our main ﬁndings of ss/sl carriers being signiﬁcantly more vulnerable to negative environments, but not proﬁting signiﬁcantly more from positive environments compared with ll carriers, all ps40.14. However, ethnicity proved to be a signiﬁcant moderator of the association between positive environmental inﬂuences and positive develop- mental outcome for carriers of the ll alleles (Qcontrast ¼ 6.49, P ¼ 0.01). gained more from positive environments than children homozygous for the l allele, consistent with differential suscepti- bility. In summary, then, for children with the ll genotype, the associations between positive or negative environment and positive or negative developmental outcome were absent. Thus, in samples with 480% Caucasian children, ss/sl carriers were more open to environmental inﬂuences than ll carriers, for better and for worse, consistent with differential susceptibility. Figure 1 illustrates these ﬁndings. Caucasian samples. As ethnicity was a signiﬁcant moderator, we recomputed the combined effect sizes for the ss/ll carriers versus the ll carriers using only studies with (mostly) Caucasian participants. Adverse environments. Results p < 0.05, p < 0.01 0 0.05 0.1 0.15 0.2 0.25 ss/sl ll All samples Positive Negative k = 20 k = 13 k = 25 k = 15 N = 1884 N = 1199 N = 3739 N = 1860 Effect size r ** * n.s. > 80% Caucasian Positive Negative k = 14 k = 9 k = 18 k = 11 N = 1471 N = 805 N = 3052 N = 1298 n.s. n.s. ** n.s. * * p < 0.05, p < 0.01 0 0.05 0.1 0.15 0.2 0.25 ss/sl ll All samples Positive Negative k = 20 k = 13 k = 25 k = 15 N = 1884 N = 1199 N = 3739 N = 1860 Effect size r * ** * n.s. > 80% Caucasian Positive Negative k = 14 k = 9 k = 18 k = 11 N = 1471 N = 805 N = 3052 N = 1298 n.s. n.s. ** n.s. * * Figure 1 Combined effect sizes for the associations between positive and negative environmental factors and child outcomes for the 5HTTLPR ss/sl and ll genotypes, in the total set of studies (except sl/ll, k ¼ 73) and in studies with 480% Caucasian participants (k ¼ 52). Po0.05, *Po0.01. Note: In the total set of studies, the combined effect sizes for positive environmental factors are signiﬁcant for both ss/sl and ll carriers. The difference between the combined effect sizes for ss/sl and ll is not signiﬁcant, Qcontrast ¼ 0.54, P ¼ 0.46. The combined effect size for negative environmental factors is signiﬁcant for ss/sl carriers, but not for ll carriers. The combined effect sizes for ss/sl and ll differ signiﬁcantly, Qcontrast ¼ 8.35, Po0.01. In the set of studies with mostly Caucasian participants, the combined effect sizes for positive and negative environmental factors are signiﬁcant for the ss/sl carriers, but not for the ll carriers. The differences between the combined effect sizes for ss/sl and ll carriers are signiﬁcant, Qcontrast ¼ 3.92, P ¼ 0.048 for positive environmental factors and Qcontrast ¼ 4.58, P ¼ 0.03 for negative environmental factors. Results In the Caucasian samples, the combined effect size for developmental problems in the presence of adverse environmental inﬂuences amounted to r ¼ 0.18 (Po0.01, 95% CI 0.11, 0.25) for ss/sl carriers, in a heterogeneous set of effect sizes (Q ¼ 63.06, Po0.01). The combined effect size for the ll carriers was r ¼ 0.04 (NS, 95% CI 0.07, 0.14) in a homogeneous set (Q ¼ 9.98, NS). The difference between the ss/sl versus ll carriers was signi- ﬁcant (Qcontrast ¼ 4.58, P ¼ 0.03). In adverse contexts and consistent with diathesis-stress thinking, ss/sl carriers were more at risk for negative developmental outcomes than ll carriers. Results In Table 2, the combined effect sizes for each of the ﬁve genotype groups are presented. The effect sizes represent the associations between variations in the environmental and developmental outcomes. Here we discuss the contrast between the carriers of one or two s alleles (the combination of the groups with ss, ss/sl or sl carriers) and the carriers of two l alleles (ll carriers), separately for the effect of adverse and supportive environments. Adverse environments. The combined effect size for developmental problems in the presence of negative environmental inﬂuences (that is, ‘the dark side’) amounted to r ¼ 0.22 (Po0.01, 95% CI 0.14, 0.31) for ss/sl carriers, in a heterogeneous set of effect sizes (Q ¼ 79.86, Po0.01). The combined effect size for the ll carriers was r ¼ 0.06 (NS, 95% CI 0.01, 0.12) in a homogeneous set (Q ¼ 12.34, NS). Using a random effects test, the difference was signiﬁcant (Qcontrast ¼ 8.35, Po0.01), supporting the diathesis-stress idea that ss/sl carriers are more vulnerable to environmental adversity than ll carriers. Translational Psychiatry Differential susceptibility and 5HTTLPR MH van IJzendoorn et al 4 Positive environments. Turning to the ‘bright side’—the association between positive environments and better child outcomes—we found a combined effect size of r ¼ 0.21. (Po0.01, 95% CI 0.12, 0.30) for ss/sl carriers in a heterogeneous set of effect sizes (Q ¼ 135.48, Po0.01). The combined effect size for ll carriers was r ¼ 0.11 (P¼ 0.04, 95% CI 0.04, 0.19) in a marginally heterogeneous set (Q ¼ 24.80, Po0.05). Although this differential pattern of associations was consistent with differential susceptibility, the difference in effect sizes across the two genetic groups was nonsigniﬁcant (Qcontrast ¼ 0.54, NS). Thus, although children with s alleles were more negatively affected by adverse contexts than ll carriers with regard to negative outcomes (see above), they did not beneﬁt signiﬁcantly more from positive environments than children homozygous for the l allele. It should be noted that only in the case of the carriers of the short allele in negative environments did the funnel plot reveal potential publication bias, which, when corrected with the Duval and Tweedie trim and ﬁll method,23 changed only marginally the point estimate of the combined effect size. The authors declare no conﬂict of interest. Acknowledgements. MJBK and MHvIJ were supported by awards from the Netherlands Organization for Scientiﬁc Research (MJBK: VICI grant no. 453-09- 003; MHvIJ: SPINOZA prize). Disclaimer. MJBK and MHvIJ had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. 1. Bakermans-Kranenburg MJ, van IJzendoorn MH. Genetic vulnerability or differential susceptibility in child development: the case of attachment. J Child Psychol Psychiatry 2007; 48: 1160–1173. Most G E studies included in the current meta-analyses are correlational in design, making causal inferences impossible. Correlational G E studies are limited in their control of gene–environment correlations that might wrongly be interpreted as G E interactions. Only experimental manipulations of the environment in randomized control trials afford valid conclusions about G E interactions independent of gene–environment correlations.24 Protocolled interven- tions also create standardized and observable changes in the environment; in so doing, they reduce the error component in the measurement of the environment. This should increase the chance of replicable G E ﬁndings,24 as these have been found to be greatly dependent on the quality of the assessment of the environment.7 Furthermore, in most developmental studies, 5HTTPLPR was the only candidate genetic marker of differential susceptibility. However, differ- ential susceptibility will not depend on one genotype only; other genotypes have already been successfully examined as markers of differential susceptibility (perhaps most compelling dopamine-system related genes12). In future studies, genetic pathways related to the serotonin system might be included in the G E equation, and the combined effects of serotonergic and dopaminergic pathways may be examined. This requires studies with large samples that only consortia of researchers are able to conduct. 2. Belsky J, Bakermans-Kranenburg MJ, van IJzendoorn MH. For better and for worse differential susceptibility to environmental inﬂuences. Psychol Sci 2007; 16: 300–305. 2. Belsky J, Bakermans-Kranenburg MJ, van IJzendoorn MH. For better and for worse differential susceptibility to environmental inﬂuences. Psychol Sci 2007; 16: 300–305. 3. Belsky J, Jonassaint C, Pluess M, Stanton M, Brummett B, Williams R. Vulnerability genes or plasticity genes? Mol Psychiatry 2009; 14: 746–754. 4. Homberg JR, Lesch KP. Looking on the bright side of serotonin transporter gene variation. Biol Psychiatry 2011; 69: 513–519. 5. Caspi A, Sugden K, Mofﬁtt TE, Taylor A, Craig IW, Harrington H et al. Discussion The mixed studies with a higher percentage of non-Caucasians may have resulted in population stratiﬁcation or admixture, which may seriously elevate the type I error rate for detecting genes underlying complex traits, and in the end may lead to non-replicable G E ﬁndings. The support for the differential susceptibility theory chronicled here in the case of the 5HTTLPR genotype must be regarded as tentative owing to the fact that the set of studies available for inclusion in our meta-analysis was rather small. Most importantly, the G E studies conducted thus far are quite heterogeneous in terms of environments and outcomes studied. The heterogeneity creates relatively large conﬁdence intervals (CI) around the point estimates of combined effect sizes, which may make the meta-analytic results somewhat unstable. More studies using similar assessments of environments and outcomes will afford stronger tests of the hypothesis that 5HTTLPR is a marker of differential susceptibility, the conclusion drawn here. Furthermore, including assessments of positive environments does not exclude the possibility that the resulting associations with (positive) developmental outcomes are driven by the lower scores on these assessments, thereby representing, perhaps, the higher end of negative environmental input. Evidence that carriers of short versus long allele of 5HTTLPR prove more susceptible to experimental interventions promot- ing development by enhancing the quality of the environment would counter this alternative interpretation. Discussion In this series of meta-analyses on 77 effect sizes with 9361 children and adolescents who were the subject of 30 pertinent research reports, we found some support for 5HTTLPR as a genetic marker of differential susceptibility or plasticity rather than vulnerability. Carriers of the s alleles (ss/sl) were not only at risk for developing poorly in an adverse environment, consistent with the diathesis-stress model, but they were also signiﬁcantly more inﬂuenced by supportive environments enabling them to develop in a more positive way than ll carriers, consistent with differential susceptibility. Indeed, the latter ﬁnding is incompatible with the diathesis-stress frame- work that has informed, implicitly or explicitly, most G E research to date.2,3 Ethnicity was an important moderator in the total set of 5HTTLPR G E studies involving child and adolescent samples. Some evidence for differential suscept- ibility only emerged in studies with (mostly) Caucasian participants. Caucasian participants with the ll genotype were less inﬂuenced by environmental factors, whether supportive Positive environments. The association between positive environments and better child adaptation amounted to a combined effect size of r ¼ 0.17. (Po0.01, 95% CI 0.10, 0.24) for ss/sl carriers, in a heterogeneous set of studies (Q ¼ 29.26, Po0.01), whereas the combined effect size for ll carriers was nonsigniﬁcant, r ¼ 0.05 (95% CI 0.05, 0.14), in a homogeneous set of studies (Q ¼ 5.73, NS). The difference between the ss/sl versus ll carriers was signiﬁcant (Qcontrast ¼ 3.92, P ¼ 0.048). Children with ss/sl genotypes Translational Psychiatry Differential susceptibility and 5HTTLPR MH van IJzendoorn et al 5 environments. At least in the case of children under the age of 18 years, the exclusive ‘dark-side’ view of some genotypes (for example, 5HTT s allele, DRD4-7 repeat allele) as risk factors for psychopathology appears on the basis of this and our previous meta-analysis undeserved and incompatible with emerging empirical evidence. Similar to the diathesis-stress model, differential susceptibility theory acknowledges the negative effects of cumulative genetic and environmental risks, but at the same time draws attention to the bright side of development. The so-called vulnerability genes might make individuals not only vulnerable to negative environmental inﬂuences but also more open to the positive developmental effects of positive changes in the environment. or adverse. 1. Bakermans-Kranenburg MJ, van IJzendoorn MH. Genetic vulnerability or differential susceptibility in child development: the case of attachment. J Child Psychol Psychiatry 2007; 48: 1160–1173. The authors declare no conﬂict of interest. Inﬂuence of life stress on depression: moderation by a polymorphism in the 5-HTT gene. Science 2003; 301: 386–389. 6. Munafo` MR, Durrant C, Lewis G, Flint J. Gene X environment interactions at the serotonin transporter locus. Biol Psychiatry 2009; 65: 211–219. 7. Uher R, McGufﬁn P. The moderation by the serotonin transporter gene of environmental adversity in the etiology of depression: 2009 update. Mol Psychiatry 2010; 15: 18–22. 8. Belsky J. Variation in susceptibility to environmental inﬂuence an evolutionary argument. Psychol Inquiry 1997; 8: 182–186. 9. Belsky J, Pluess M. The nature (and nurture?) of plasticity in early human development. Perspect Psychol Sci 2009; 4: 345–351. 10. Ellis BJ, Boyce WT, Belsky J, Bakermans-Kranenburg MJ, van IJzendoorn MH. Differential susceptibility to the environment: an evolutionary—neurodevelopmental theory. Dev Psychopathol 2011; 23: 7–28. 11. Bakermans-Kranenburg MJ, van IJzendoorn MH. Gene-environment interaction of the Dopamine D4 Receptor (DRD4) and observed maternal insensitivity predicting externalizing behavior in preschoolers. Dev Psychobiol 2006; 48: 406–409. 12. Bakermans-Kranenburg MJ, van IJzendoorn MH. Differential susceptibility to rearing environment depending on dopamine-related genes: new evidence and a meta-analysis. Dev Psychopathol 2011; 23: 39–52. 13. Sugden K, Arseneault L, Harrington H, Mofﬁtt TE, Williams B, Caspi A. Serotonin transporter gene moderates the development of emotional problems among children following bullying victimization. J Am Acad Child Adolesc Psychiatry 2010; 49: 830–840. The diathesis-stress model still dominates the ﬁeld of developmental psychopathology and psychiatric genetics more generally, stressing the negative developmental impact of a vulnerable genetic make-up interacting with adverse environments. The current meta-analyses show, however, just like our earlier one on dopamine-related genes and child development,12 that genetic vulnerability might also imply greater susceptibility to the inﬂuences of positive (changes in) 14. Eley TC, Sugden K, Corsico A, Gregory AM, Sham P, McGufﬁn P. et al. Gene-environment interaction analysis of serotonin system markers with adolescent depression. Mol Psychiatry 2004; 9: 908–915. 15. Pluess M, Velders FP, Belsky J, van IJzendoorn MH, Bakermans-Kranenburg MJ, Jaddoe VW et al. Serotonin transporter polymorphism moderates effects of prenatal maternal anxiety on infant negative emotionality. Biol Psychiatry 2011; 69: 520–525. 16. Kumsta R, Stevens S, Brookes K, Schlotz W, Castle J, Beckett C et al. 5HTT genotype moderates the inﬂuence of early institutional deprivation on emotional problems in adolescence: evidence from the English and Romanian Adoptee (ERA) study. J Child Psychol Psychiatry 2010; 51: 755–762. The authors declare no conﬂict of interest. Translational Psychiatry Differential susceptibility and 5HTTLPR MH van IJzendoorn et al 6 22. Hu X-Z, Lipsky RH, Zhu G, Akhtar LA, Taubman J, Greenberg BD et al. Serotonin transporter promoter gain-of-function genotypes are linked to obsessive-compulsive disorder. Am J Human Genet 2006; 78: 815–826. 17. Sonuga-Barke EJS, Oades RD, Psychogiou L, Chen W, Franke B, Buitelaar J et al. Dopamine and serotonin transporter genotypes moderate sensitivity to maternal expressed emotion: the case of conduct and emotional problems in attention deﬁcit/hyperactivity disorder. J Child Psychol Psychiatry 2009; 50: 1052–1063. 23. Borenstein M, Hedges LV, Higgins JPT. Introduction to meta-analysis. Chicester, NH: Wiley, 2009. 18. Brody GH, Chen Y-F, Beach SRH, Philibert R, Kogan SM. Participation in a family- centered prevention program decreases genetic risk for adolescents’ risky behaviors. Pediatrics 2009; 124: 911–917. 24. van IJzendoorn MH, Bakermans-Kranenburg MJ, Belsky J, Beach S, Brody G, Dodge KA et al. Gene-by-environment experiments: a new approach to ﬁnding the missing heritability. Nat Rev Genet 2011; 12: 881–882. 19. Eley TC, Hudson JL, Creswell C, Tropeano M, Lester KJ, Cooper P et al. Therapy- genetics: the 5HTTLPR and response to psychological therapy. Mol Psychiatry 2012; 17: 236–237. Translational Psychiatry is an open-access journal published by Nature Publishing Group. This work is licensed under the Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/3.0/ 20. Nilsson KW, Sjo¨berg RL, Damberg M, Alm PO, Ohrvik J, Leppert J et al. Role of the serotonin transporter gene and family function in adolescent alcohol consumption. Alcohol Clin Exp Res 2005; 29: 564–570. 21. Williams RB, Marchuk D, Gadde KM, Barefoot JC, Grichnik K, Helms MJ et al. Serotonin- related gene polymorphisms and central nervous system serotonin function. Neuropsychopharmacology 2003; 28: 533–541. Translational Psychiatry
https://openalex.org/W2590378689	https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0172684&type=printable	English	null	Comprehensive growth performance, immune function, plasma biochemistry, gene expressions and cell death morphology responses to a daily corticosterone injection course in broiler chickens	PloS one	2,017	cc-by	14,116	RESEARCH ARTICLE Abstract Citation: Mehaisen GMK, Eshak MG, Elkaiaty AM, Atta A-RMM, Mashaly MM, Abass AO (2017) Comprehensive growth performance, immune function, plasma biochemistry, gene expressions and cell death morphology responses to a daily corticosterone injection course in broiler chickens. PLoS ONE 12(2): e0172684. doi:10.1371/journal. pone.0172684 The massive meat production of broiler chickens make them continuously exposed to poten- tial stressors that stimulate releasing of stress-related hormones like corticosterone (CORT) which is responsible for specific pathways in biological mechanisms and physiological activi- ties. Therefore, this research was conducted to evaluate a wide range of responses related to broiler performance, immune function, plasma biochemistry, related gene expressions and cell death morphology during and after a 7-day course of CORT injection. A total num- ber of 200 one-day-old commercial Cobb broiler chicks were used in this study. From 21 to 28 d of age, broilers were randomly assigned to one of 2 groups with 5 replicates of 20 birds each; the first group received a daily intramuscular injection of 5 mg/kg BW corticosterone dissolved in 0.5 ml ethanol:saline solution (CORT group), while the second group received a daily intramuscular injection of 0.5 ml ethanol:saline only (CONT group). Growth perfor- mance, including body weight (BW), daily weight gain (DG), feed intake (FI) and feed con- version ratio (FC), were calculated at 0, 3 and 7 d after the start of the CORT injections. At the same times, blood samples were collected in each group for hematological (TWBC’s and H/L ratio), T- and B-lymphocytes proliferation and plasma biochemical assays (total pro- tein, TP; free triiodothyronine hormone, fT3; aspartate amino transaminase, AST; and ala- nine amino transaminase, ALT). The liver, thymus, bursa of Fabricius and spleen were dissected and weighed, and the mRNA expression of insulin-like growth factor 1 gene (IGF- 1) in liver and cell-death-program gene (caspase-9) in bursa were analyzed for each group and time; while the apoptotic/necrotic cells were morphologically detected in the spleen. From 28 to 35 d of age, broilers were kept for recovery period without CORT injection and the same sampling and parameters were repeated at the end (at 14 d after initiation of the CORT injection). In general, all parameters of broiler performance were negatively affected by the CORT injection. In addition, CORT treatment decreased the plasma concentration of Comprehensive growth performance, immune function, plasma biochemistry, gene expressions and cell death morphology responses to a daily corticosterone injection course in broiler chickens Gamal M. K. Mehaisen1, Mariam G. Eshak2☯, Ahmed M. Elkaiaty1☯, Abdel-Rahman M. M. Atta1‡, Magdi M. Mashaly1‡, Ahmed O. Abass1 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 1 Department of Animal Production, Faculty of Agriculture, Cairo University, Giza, Egypt, 2 Department of Cell Biology, National Research Centre, Giza, Egypt 1 Department of Animal Production, Faculty of Agriculture, Cairo University, Giza, Egypt, 2 Department of Cell Biology, National Research Centre, Giza, Egypt 1 Department of Animal Production, Faculty of Agriculture, Cairo University, Giza, Egypt, 2 Department of Cell Biology, National Research Centre, Giza, Egypt ☯These authors contributed equally to this work. ‡ These authors also contributed equally to this work. gamoka@cu.edu.eg Introduction Broiler production forms an essential component to meet the increasing demand of animal protein in either the developed or developing countries. The modern large-scale broiler pro- duction provokes a new stressful situation around broilers due to thermal conditions, high- stocking density, immunological challenge, handling, transportation and feed quality. All kinds of such stressors can elicit the biological response of broiler chickens to re-establish homeostatic conditions [1], and directly or indirectly activate the hypothalamic-pituitary- adrenal (HPA) axis [2]. The activated HPA axis results in the release of adrenocorticotropic hormone (ACTH), which stimulates the adrenals to secret and release glucocorticoids, includ- ing corticosterone (CORT), into blood circulation [1]. In chickens, the high levels of CORT in blood circulation is associated with a marked regression in the growth performance, feed effi- ciency and fuel metabolism [3–8]. The weight and function of the lymphoid organs including thymus, bursa of Fabricius and spleen also decreased by the CORT treatment in birds [3,4,9]. In contrast, the effect of CORT on the liver is anabolic and the fat pads are grossly enlarged in liver of broilers treated with CORT [6]. It was found that exposure to stressors elevates the blood concentration of heterophils (the avian counterpart to the mammalian neutrophils) and depresses the blood concentration of lymphocytes [10]. As an important immune parameter, the ratio of circulating heterophils to lymphocytes (H/L ratio) is one of the most recognizable symptoms of stress in chickens [11– 14]. The administration of exogenous corticosterone resulted in an increase in the plasma cor- ticosterone concentration and H/L ratio with a global decrease in the circulating leukocyte number in young chickens [9,13,14]. On the other hand, lymphocytes in birds are specifically separated into T and B cells during the development of the thymus and bursa of Fabricius glands, respectively. Such cells carry receptors for stress hormones produced by the adrenal and pituitary glands, so it could be used as another index to stress [9]. Particularly, high con- centrations of glucocorticoids have immunosuppressive effects by inhibiting, for example, antibody production from B cells, T cell proliferation and phagocytosis [15,16]. In addition, some blood physiological and biochemical changes may occur in the stress- and/or CORT- exposed broiler chickens and may be responsible for their low growth rates [17]. In a recent study, the CORT treatment enhanced the total plasma protein levels in broiler chickens [3]. Corticosterone and comprehensive responses in broiler chickens fT3 and the mRNA expression of hepatic IGF-1. A significant increase in liver weight accom- panied by an increase in plasma TP, AST and ALT was observed with CORT treatment, indicating an incidence of liver malfunction by CORT. Moreover, the relative weights of thy- mus, bursa and spleen decreased by the CORT treatment with low counts of TWBC’s and low stimulation of T & B cells while the H/L ratio increased; indicating immunosuppressive effect for CORT treatment. Furthermore, high expression of caspase-9 gene occurred in the bursa of CORT-treated chickens, however, it was associated with a high necrotic vs. low apoptotic cell death pathway in the spleen. Seven days after termination of the CORT treat- ment in broilers, most of these aspects remained negatively affected by CORT and did not recover to its normal status. The current study provides a comprehensive view of different physiological modulations in broiler chickens by CORT treatment and may set the potential means to mount a successful defense against stress in broilers and other animals as well. fund was awarded to him during the management of the project. The funders have approved the study design, data collection and analysis, decision to publish, and preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist. Editor: Ruud van den Bos, Radboud Universiteit, NETHERLANDS Editor: Ruud van den Bos, Radboud Universiteit, NETHERLANDS Received: September 5, 2016 Accepted: February 8, 2017 Published: February 24, 2017 Copyright: © 2017 Mehaisen et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. Funding: This work was supported by Project of Rapid Climate Change in Poultry Cellular and Molecular Physiology (RCC-PCMP), funded from General Scientific Research Department at Cairo University (GSRD-CU); http://gsrd.cu.edu.eg/. AOA was the principal investigator of the project and the 1 / 25 PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 Introduction It was previously found that corticosterone increased protein degradation as indicated by the PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 2 / 25 Corticosterone and comprehensive responses in broiler chickens reduction in skeletal RNA/protein synthesis [18] and the rising in the muscle proteolysis [19]. The increase in aspartate and alanine amino transaminase (AST and ALT) enzymes levels usu- ally appear in the blood when there is tissue impairment or liver malfunctioning caused by excessive stress [20]. Two other components are mediated by each other and strongly corre- lated with the growth performance in young chickens; triiodothyronine hormone (T3) and insulin-like growth factor 1 (IGF-1) [21,22]. It has been reported that the decreased perfor- mance in stressed broilers is mainly caused by the reduction in T3 and thyroxin (T4) levels [23]. Many studies have also reported the negative effects of environmental and nutritional stress on the expression of IGF-1 gene and its circulating levels in different avian species dur- ing embryonic and post-hatch growth and development [24–26]. It was clearly reported that glucocorticoid administration in chickens has a suppressive effect on the plasma T3 levels [27,28] and the circulating concentrations of hepatic IGF-1 [29]. Rearing stress conditions and physical agents can also activate apoptosis; a mode of cell death that occurs under normal physiological conditions and is triggered by one or more of intracellular factors [30–32]. The apoptotic morphology, including plasma membrane bleb- bing, cytoplasm vacuolization, chromatin condensation and DNA degradation, is essentially the result of the proteolytic action of caspases upon specific cellular substrates [33,34]. Caspase cascades are responsible for both initiating and amplifying early apoptotic signals (e.g. cas- pase-1, -2, -8, -9, -10) as well as executing apoptosis (e.g. caspase-3, -6, -9) [35,36]. It has been well documented that the chronic stress may influence the expression of caspase genes in broiler chickens and other vertebrates [37]. In some cases, stressful stimuli can initiate the necrotic pathway form of cell death programs, a process that has distinct morphological fea- tures; it is accompanied by a rapid collapse of plasma membrane, and it is independent of expression of new genes [38]. In one of few studies that differentiated between apoptosis and necrosis pathways in rat’s Leydig cells [39] and postnatal neuron cells [40] cultured with CORT, a higher necrotic than apoptic cells were detected in groups with higher CORT concentration. Introduction Most researchers have focused on the effect of stressors itself and/or stress-related glucocor- ticoids administration on specific pathways related to stress in birds. However, the administra- tion of corticosterone to broiler chickens simply serves as a practical, controlled and flexible tool in the research of their adaptation to stress. In the present study, the research was to set forth an overall view of modulation induced by corticosterone exposure as simulation for stress condition in broiler chickens. Hence, this work aimed to assess comprehensive responses related to broiler performance, immune function, plasma biochemistry, gene expression and cell death morphology throughout a 7-day course of daily CORT injections. In addition, all measurements were repeated at 7 d after termination of the 7-day course of the CORT treatment to estimate the capability of broiler chickens for recovery to normal physio- logical status after overriding the stress. Corticosterone and comprehensive responses in broiler chickens From 21 to 28 d of age, birds were randomly assigned to one of 2 groups with 5 replicates of 20 birds per replicate. Broilers of the first group received a daily intramuscular injection (at 10:00 AM) of corticosterone (cat# C2505, Sigma, St. Louis, MO 63103, USA) adjusted based on the average body weight of birds at dose of 5 mg/kg; dissolved in 0.5 ml ethanol:saline (1:1, vol/vol) solution (CORT group). The second group received a daily intramuscular injection of 0.5 ml ethanol:saline (1:1, vol/vol) solution and served as positive control (CONT group). Growth performance of broilers was recorded during the experimental period as detailed later. At 21 d of age, before CORT treatment (0 d of injection), blood samples were collected from 5 birds per treatment group (one per replicate) via the wing vein for determination of hemato- logical assay and lymphocyte proliferation. Another 5 birds from each group were sampled and plasma was obtained for biochemical analyses. In addition, 5 birds from each treatment group were sacrificed, by cervical dislocation, and the liver, thymus, bursa of Fabricius and spleen were harvested, weighed and kept on ice until the following analyses. The mRNA expression of insulin-like growth factor 1 gene (IGF-1) was analyzed in liver tissues, while the expression of cell-death-program gene (caspase-9) was analyzed in bursa tissues. Finally, spleen tissues were subjected to a morphological detection of apoptotic/necrotic cells under fluorescent microscope. The same sampling procedures and parameters were repeated during the treatment at 3 and 7 d after the start of CORT injection (at 24 and 28 d of age). After that, birds were kept for recovery period without CORT injection and the same parameters were measured 7 d later (at 14 d after the start of CORT injection = 35 d of age). Each bird in the group was sampled only once and then was removed from the experiment. Birds were monitored closely, twice a day, to detect any signs of stress (breathing difficulty, watery discharge of the peak, decreased appetite, ruffled feathers, or droopy looking) through- out the experimental period. Accordingly, when one or more of these signs appeared, cervical dislocation was used to end the life of these birds. This process was accomplished to minimize suffering of birds and to allow humane endpoints. All experimental protocols were approved by Cairo University Ethics Committee for the Care and Use of Experimental Animals in Edu- cation and Scientific Research (CU-IACUC). Growth performance and organs weight Individual body weights (BW) were recorded at 0, 3 and 7 d after the start of CORT injection course (21, 24 and 28 d of age), and at 7 d after cessation of the 7-day course of the CORT treatment (14 d after initiation of CORT injection; 35 d of age) for each group. The average daily gain (DG), feed intake (FI) and feed conversion ratio (feed/gain; FC) for each replication in the group were obtained at periods 0–3 d and 3–7 d of the CORT treatment course and at 7–14 d of the recovery period (21–24, 24–28 and 28–35 d of age, respectively). After cervical dislocation of broilers, the weights of liver, thymus, bursa of Fabricius and spleen were also cal- culated relatively to the body weights in each group at 0, 3 and 7 d after initiation of the CORT injection, and after 1 week of recovery period. Birds, treatment and ethical statement The current study was conducted in the Poultry Services Center at the Faculty of Agriculture, Cairo University. A total of 200 one-day-old commercial broiler chicks (Cobb500™) were obtained from a commercial hatchery and housed on a deep litter floor brooder. Ambient tem- perature on d 1 was set at 33˚C and then was gradually reduced until 24˚C by d 21. The light regimen was 23L:1D. Chicks were fed commercial broiler diets according to the recommenda- tions of the National Research Council (NRC, 1994). Water and feed were supplied ad libitum during the study. 3 / 25 PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 Plasma biochemical analyses Blood samples of approximately 3 ml were withdrawn from the brachial wing vein in heparin- ized tubes and centrifuged at 1030 xg for 10 min at 4˚C. The plasma was separated and stored at -20˚C until analyzed. The plasma levels of total protein (TP), AST and ALT were analyzed by automatic scanning spectrophotometer (CE1010, Cecil Instruments Limited, Cambridge, United Kingdom) using colorimetric assay kits (TP-2020 for TP, and AT-1034(45) for both AST and ALT; Biodiagnostic Inc, Dokki-Giza, Egypt). The fT3 was measured by the automated ELISA microplate reader using enzyme immunoassay ELISA kits (EIA-10301, Chemux BioSci- ence Inc, San Francisco, CA, USA). The standard curves and calculations, as well as accuracy and sensitivity of the assay were performed following the kits protocol for each assay. Lymphocyte proliferation The T and B lymphocyte proliferation assays were done according to methods described by Zhang and Guo [43] with some modifications. The heparinized blood samples were added to separation medium Histopaque-1077 (cat# 10771, Sigma, USA), and were then centrifuged at 1030 xg for 20 min at 4˚C. Peripheral blood mononuclear cells (PBMCs) were isolated and washed twice with RPMI-1640 (Invitrogen Corp., Grand Island, NY, USA) incomplete cul- ture medium, and then re-suspended in 2 ml of RPMI-1640 complete culture medium. The viable lymphocytes were detected using Trypan Blue dye and plated in triplicate wells (96-well plate) at 1×106 cells per well. Then, 50 μl of either Concanavalin-A (Con-A, 45 μg/ ml, cat# C5275, Sigma, USA) or Lipopolysaccharide (LPS, 10 μg/ml, cat# L4391, Sigma, USA) was added to selected wells to induce the proliferation of T lymphocyte and B lymphocyte, respectively; while control wells received 50 μl of RPMI-1640 medium. Cells were then incu- bated for 48 h at 42˚C with 5% CO2. After incubation, 15 μl of 3-[4,5-dimethylthiazol]- 2,5-diphenyltetrazolium bromide (MTT, 5 mg/ml, cat# M2128, Sigma, USA) was added to each well and the cells were incubated for another 4 h. Subsequently, 100 μl of 10% sodium dodecyl sulfate dissolved in 0.04 M HCl solution was added to each well to lyse the cells and solubilize the MTT crystals. Finally, the absorbance at 570 nm was recorded using an auto- mated ELISA microplate reader (ChroMate Microplate Reader-4300, Awareness Technology Inc., Palm City, FL, USA). Stimulating index (SI) for T and B cells was calculated as follows: SI = OD570(stimulated cells) / OD570(unstimulated cells). Hematological assay Whole blood samples of approximately 3 ml were collected using heparinized syringes and transferred immediately to heparinized tubes. The total white blood cells (TWBC’s) were man- ually determined by mixing 490 μl of brilliant cresyl blue dye with 10 μl of whole blood sample, and then the total leukocytes were counted under a microscope at a magnification of 200x using a hemocytometer slide [41]. The H/L ratio was determined according to Zhang et al. [42] with modification. In brief, a droplet of blood (5 μl) was used to make smears on a clean glass slide (2 slides per each sample). The smears were stained with Hema-3 (cat# 22–122911, 4 / 25 PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 Corticosterone and comprehensive responses in broiler chickens Fisher scientific, USA) after drying and fixing with methyl alcohol. On each slide, heterophils and lymphocytes were counted under a microscope at magnification of 1000x with oil immer- sion until a total of 100 cells were reached. After averaging the cells of 2 slides, the ratios of het- erophils to lymphocytes were calculated. Corticosterone and comprehensive responses in broiler chickens Table 1. Details of primers used for real-time PCR quantitative analysis. Gene symbol GenBank accession no. Primer sequences (5’->3’) Product size (bp) Melting temperature (˚C) IGF-1 FJ977570.1 F: CACCTAAATCTGCACGCT 140 55 R: CTTGTGGATGGCATGATCT Caspase-9 XM_424580.2 F: TCCCGGGCTGTTTCAACTT 208 60 R: CCTCATCTTGCAGCTTGTGC ß-actin NM205518 F: TGCGTGACATCAAGGAGAAG 300 58 R: TGCCAGGGTACATTGTGGTA doi:10 1371/journal pone 0172684 t001 Table 1. Details of primers used for real-time PCR quantitative analysis KCl, 10 mM Tris-HCl, pH 8.3, 10 mM dNTPs, 50 μM oligo-dT primers, 20 U ribonuclease inhibitor and 50 U M-MuLV reverse transcriptase). The RT reaction was carried out at 25˚C for 10 min, followed by 1 h at 42˚C, and the reaction was stopped by heating for 5 min at 99˚C. Afterwards, the reaction tubes containing cDNA were stored at -20˚C until being used for quantitative real time-polymerase chain reaction (qRT-PCR). PCR reactions were set up in 25 μl reaction mixtures containing 12.5 μl of 1× SYBR Premix Ex TaqTM (TaKaRa, Biotech. Co. Ltd., Germany), 0.5 μl sense primers (0.2 μM), 0.5 μl anti- sense primer (0.2 μM), 6.5 μl distilled water, and 5 μl of cDNA template. The reaction program was allocated to 3 steps of thermal cycling parameters. The first step was set to 95.0˚C for 3 min. The second step consisted of 40 cycles in which each cycle was divided to 3 steps: (a) at 95˚C for 15 sec, (b) at 55˚C for 30 sec, and (c) at 72˚C for 30 sec. The last step consisted of 71 cycles which started at 60˚C and then increased by 0.5˚C every 10 sec up to 95˚C. At the end of each qRT-PCR, a melting curve analysis was performed at 95˚C to check the quality of the used primers. Each experiment included a distilled water control. The qRT-PCR of IGF-1and caspase-9 genes were normalized to the main expression of ß- actin and transformed using the comparative cycle threshold (CT) method to quantify expres- sion levels as previously described by Ellestad et al. [44]. Sequence-specific primers (Table 1) for the real-time PCR were designed using the Primer blast web interface (http://www.ncbi. nlm.nih.gov/tools/primer-blast/index.cgi). Quantitative real-time PCR analysis Total RNA was extracted from liver and bursa tissues using the standard TRIzol Reagent extraction method. RNA was dissolved in diethylpyrocarbonate (DEPC)-treated water by pass- ing solution a few times through a pipette tip. Total RNA was treated with 1 U of RQ1 RNase- free DNase (Invitrogen, Germany) to digest DNA residues, then re-suspended in DEPC- treated water. The purity of total RNA was assessed spectrophotometrically at 260/280 nm, and the integrity of extracted RNA was determined by using 1.5% agarose gel electrophoresis. Then total RNA was reverse-transcribed into cDNA by using RevertAidTM First Strand cDNA Synthesis Kit (MBI Fermentas, Germany) according to the manufacturer’s directions. Briefly, an amount of total RNA (5μg) was used with a reaction mixture, termed as master mix (MM). The MM consisted of 50 mM MgCl2 and 5x reverse transcription (RT) buffer (50 mM 5 / 25 PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 Morphological detection of cell death categories Apoptotic changes in the spleen samples were determined morphologically by fluorescent microscope after labeling with acridine orange and ethidium bromide (AO/EB) dyes [45]. Briefly, spleen tissues were washed in PBS (Sigma), chopped finely and centrifuged at 9220 xg for 5 min. The pellet obtained was suspended in trypsin-EDTA solution (0.25%, 53 mM; cat#T4049, Sigma) in PBS for 1 hr at 37˚C and smeared on clean glass slides. Finally, spleen cells smears were air-dried and fixed in a solution of methanol/acetic acid (3:1). The slides were stained with 25 μl of AO/EB dye mixture (4μg/ml AO (cat#A9231, Sigma) and 4μg/ml EB (cat#E7637, Sigma) in PBS, pH 7.4). The slides were examined under fluorescence microscope using B2A filter (450–490 nm excitation, 515 nm emission, at 100x magnification, Nikon Tokyo, Japan), and connected to a COHU 4910 video camera (Cohu, Inc., San Diego, CA, USA) and a personal computer—based image analysis system (Lucia-Comet v.4.51). The cells were divided into three categories as follows: viable cells (green colored), apoptotic cells (yel- low colored), and necrotic cells (red colored). A total of 100 cells were counted, and the % of total apoptotic and necrotic cells were examined for each experimental group (CONT 0, 3, 7 and 14; CORT 0, 3, 7 and 14). Detailed information on the cell death categories and its rates in each experimental group has been demonstrated in (S1 Table). Also, visual examples for apo- ptotic categories in each experimental group are represented in (S1 Fig). PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 6 / 25 Corticosterone and comprehensive responses in broiler chickens Growth performance The effect of the 7 d course of daily corticosterone injections on the BW, DG, FI and FC of broiler chickens during the treatment course and after the post-treatment recovery period is shown in Fig 1. The corticosterone injection, significantly, decreased the BW of broiler chick- ens at 3 and 7 d of the course of the treatment when compared with the control group (516 g vs. 624 g and 584 g vs. 792 g BW for CORT vs. CONT group at 3 and 7 d of the treatment course, respectively), and it was still low after recovery (828 g vs. 1303 g BW of CORT vs. CONT chickens at 14 d of the treatment course, Fig 1-A). A significantly (P<0.05) lower DG was observed for CORT-treated chickens compared to the CONT group during the first 3 days from the start of the CORT injection (52 g vs. 136 g DG during 21–24 d of age). This was also observed during the recovery period (244 g vs. 512 g DG during 28–35 d of age) (Fig 1-B). The FI estimated for CORT-treated chickens was significantly lower than the CONT during 3–7 d from the start of the corticosterone treatment (47.8 vs. 70.2 g FI/chicken/day during 24–28 d of age) and also during the recovery period (47.3 vs. 95.7 g FI/chicken/day during 28–35 d of age) (Fig 1-C). In contrast, the FC was significantly higher in chickens of CORT group than in those of CONT group only within 3 days from the start of the corticosterone injection (3.9 vs. 1.5 FC during 21–24 d of age, Fig 1-D). Statistical analysis All statistical analyses were performed using IBM SPSS 22.0 software package (IBMcorp., NY, USA, 2013). A general linear model (GLM) procedure was performed to analyze all data including CORT treatment (CORT and CONT groups), time (0, 3, 7 and 14 d of the course of the treatment) and their interactions as fixed effects. When interactions between main effects were significant, a multiple pairwise comparison among multiple means within CORT treat- ment group were made by Duncan’s test and the differences between CONT and CORT groups within each time was also performed using independent-samples T-test. Values were considered statistically different at P <0.05 and results were expressed as least square means ± standard error. PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 Organs weight The effect of the 7 d course of daily corticosterone injection on the organs relative weights of broilers during the treatment course and after the post-treatment recovery period is shown in Fig 2. The relative liver weight of broilers in CORT group (4.17%) was significantly (P<0.05) higher than that in CONT group (3.01%) at both 3 and 7 d after the start of treatment (Fig 2-A). As shown in Fig 2-B, the relative weight of thymus glands in treated chickens was sig- nificantly (P<0.05) lower than that in the control chickens during the 7-d course of treat- ment (0.32% vs. 0.54% and 0.21% vs. 0.49% for CORT vs. CONT at 3 and 7 d after the start of injection, respectively). The same trend was also observed at the end of the recovery period (0.35% for CORT vs. 0.47% for CONT at 14 d after the start of injection). The relative weight of bursa was lower in CORT group than in CONT group (Fig 2-C); however, this decrease was significant (P<0.05) only at 7 d after the start of injection (0.08% vs. 0.18% for CORT vs. CONT group, respectively). A significantly lower relative spleen weight was observed in treated chickens (0.07%) compared to the controls (0.10%) at 7 d after the start of the CORT injection (Fig 2-D). 7 / 25 PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 Corticosterone and comprehensive responses in broiler chickens Fig 1. Effect of a 7 d course of daily saline (CONT) or corticosterone at dose of 5 mg/kg BW (CORT injections on the body weight (A), daily gain (B), feed intake (C) and feed conversion (D) of broiler chickens during the treatment course (0, 3 and 7 d after the start of injection; 21, 24 and 28 d of age and one week after cessation of the treatment (14 d after the start of injection; 35 d of age). Bars express the mean ± SEM (n = 36). Bars of CONT and CORT groups, within each time of the treatment cou with different letters (a, b) are significantly different at P<0.05. doi:10.1371/journal.pone.0172684.g001 Fig 1. Organs weight Effect of a 7 d course of daily saline (CONT) or corticosterone at dose of 5 mg/kg BW (CORT) injections on the body weight (A), daily gain (B), feed intake (C) and feed conversion (D) of broiler chickens during the treatment course (0, 3 and 7 d after the start of injection; 21, 24 and 28 d of age) and one week after cessation of the treatment (14 d after the start of injection; 35 d of age). Bars express the mean ± SEM (n = 36). Bars of CONT and CORT groups, within each time of the treatment course, with different letters (a, b) are significantly different at P<0.05. doi:10.1371/journal.pone.0172684.g001 8 / 25 PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 Corticosterone and comprehensive responses in broiler chickens Fig 2. Effect of a 7 d course of daily saline (CONT) or corticosterone at dose of 5 mg/kg BW (CORT injections on the relative weights of liver (A), thymus (B), bursa (C) and spleen (D) to the body we of broiler chickens during the treatment course (0, 3 and 7 d after the start of injection; 21, 24 and of age) and one week after cessation of the treatment (14 d after the start of injection; 35 d of age) Bars express the mean ± SEM (n = 5). Bars of CONT and CORT groups, within each time of the treatmen course, with different letters (a, b) are significantly different at P<0.05. doi:10.1371/journal.pone.0172684.g002 Fig 2. Effect of a 7 d course of daily saline (CONT) or corticosterone at dose of 5 mg/kg BW (CORT) injections on the relative weights of liver (A), thymus (B), bursa (C) and spleen (D) to the body weights of broiler chickens during the treatment course (0, 3 and 7 d after the start of injection; 21, 24 and 28 d of age) and one week after cessation of the treatment (14 d after the start of injection; 35 d of age). Bars express the mean ± SEM (n = 5). Bars of CONT and CORT groups, within each time of the treatment course, with different letters (a, b) are significantly different at P<0.05. doi:10.1371/journal.pone.0172684.g002 9 / 25 PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 Corticosterone and comprehensive responses in broiler chickens Hematological assay The effect of the 7 d course of daily corticosterone injections on the TWBC’s and the H/L ratio of broiler chickens during the treatment course and after the post-treatment recovery period is presented in Fig 3. The corticosterone treatment significantly decreased the TWBC’s at 3 d (37.8 vs. 59.6 x103/μl for CORT vs. CONT group) and 7 d (28.0 vs. 62.1 x103/μl for CORT vs. CONT group) of the course of the treatment (Fig 3-A). Seven days following the termination of CORT injections, the TWBC’s increased but still lower than that observed at the start of the treatment (35.2 vs. 60.8 x103/μl for CORT vs. CONT group). In contrast, the H/L ratio was sig- nificantly (P<0.05) higher due to corticosterone treatment course after 3 d (0.64 vs. 0.30 for Fig 3. Effect of a 7 d course of daily saline (CONT) or corticosterone at dose of 5 mg/kg BW (CORT) injections on the TWBC’s (A) and the heterophil to lymphocyte (H/L) ratio (B) of broiler chickens during the treatment course (0, 3 and 7 d after the start of injection) and one week after cessation of the treatment (14 d after the start of injection). Data express the mean ± SEM (n = 5). Means within a treatment group with different letters (a, b, c) are significantly different at P<0.05.Significant difference between treatment groups within each time of the treatment course (P<0.05). doi:10.1371/journal.pone.0172684.g003 Fig 3. Effect of a 7 d course of daily saline (CONT) or corticosterone at dose of 5 mg/kg BW (CORT) injections on the TWBC’s (A) and the heterophil to lymphocyte (H/L) ratio (B) of broiler chickens during the treatment course (0, 3 and 7 d after the start of injection) and one week after cessation of the treatment (14 d after the start of injection). Data express the mean ± SEM (n = 5). Means within a treatment group with different letters (a, b, c) are significantly different at P<0.05.Significant difference between treatment groups within each time of the treatment course (P<0.05). doi:10.1371/journal.pone.0172684.g003 10 / 25 PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 Corticosterone and comprehensive responses in broiler chickens CORT vs. CONT group) and 7 d (0.91 vs. 0.34 for CORT vs. CONT group) of the CORT injec- tions (Fig 3-B), while it decreased after cessation of the CORT injections (0.65 for the CORT group vs. Hematological assay 0.35 for the CONT group at 14 d after the start of the treatment). Lymphocyte proliferation The effect of a daily injection course of corticosterone over 7 days followed by a 7 days post- treatment recovery period on the lymphocyte proliferation of broiler chickens is illustrated in Fig 4. The stimulation index (SI) in broiler chickens of the CONT group ranged from 4.17 to 5.14 for T cells (Fig 4-A) and from 1.84 to 2.46 for B cells (Fig 4-B). The CORT treatment Fig 4. Effect of a 7 d course of daily saline (CONT) or corticosterone at dose of 5 mg/kg BW (CORT) injections on the stimulation index of T cells (A) and B cells (B) of broiler chickens during the treatment course (0, 3 and 7 d after the start of injection) and one week after cessation of the treatment (14 d after the start of injection). Data express the mean ± SEM (n = 5). Means within a treatment group with different letters (a, b, c) are significantly different at P<0.05.Significant difference between treatment groups within each time of the treatment course (P<0.05). doi:10.1371/journal.pone.0172684.g004 Fig 4. Effect of a 7 d course of daily saline (CONT) or corticosterone at dose of 5 mg/kg BW (CORT) injections on the stimulation index of T cells (A) and B cells (B) of broiler chickens during the treatment course (0, 3 and 7 d after the start of injection) and one week after cessation of the treatment (14 d after the start of injection). Data express the mean ± SEM (n = 5). Means within a treatment group with different letters (a, b, c) are significantly different at P<0.05.Significant difference between treatment groups within each time of the treatment course (P<0.05). doi:10.1371/journal.pone.0172684.g004 11 / 25 PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 Corticosterone and comprehensive responses in broiler chickens significantly (P<0.05) decreased the SI of T cells from 5.01 at the start of injection course to 2.25 and 1.74 at 3 and 7 d after the start of injection course, respectively (Fig 4-A). The SI of B cells was also decreased (P<0.05) by the CORT treatment from 2.56 at the start of injection course to 0.92 and 0.65 at 3 and 7 d after the start of injection course, respectively (Fig 4-B). One week after termination of the CORT injection, the lymphocyte proliferation remained sig- nificantly (P<0.05) lower in the treated chickens than the control chickens (SI of T cells: 2.47 vs. Plasma biochemical analyses The effect of a daily corticosterone injections at dose of 5 mg/kg BW over 7 days followed by one week of recovery period on the plasma concentration of TP, fT3, AST and ALT in broiler chickens is presented in Fig 5. In the CORT group, the TP was significantly higher during the treatment course (4.86 and 4.74 g/dl at 3 and 7 d after the start of injection, respectively) and after the recovery period (4.66 g/dl at 14 d after the start of injection) when compared to the 0 d of injection (2.98 g/dl; Fig 5-A). When compared to the CONT group, the TP was signifi- cantly (P<0.05) higher in CORT group at 3 and 7 d after the start of the CORT injection. In contrast, the CORT treatment significantly decreased the plasma levels of fT3 hormone at 7 d after the start of injection (2.0 pg/ml) when compared to 0 and 3 d after the start of injection (5.5 and 5.0 pg/ml, respectively; P<0.05). Seven days following the termination of the CORT injection (d 14), the fT3 levels returned to similar levels of their controls (5.5 and 5.8 pg/ml at 14 d after the start of injection for CORT and CONT groups, respectively; Fig 5-B). On the other hand, the plasma levels of AST in the CORT group were significantly higher at 3 and 7 d after the start of injection (131.0 and 137.4 U/ml, respectively) and also at 7 d after the interruption of CORT treatment (147.0 U/ml) than those levels at the start of the CORT injection (93.6 U/ml at 0 d of injection); however, this increase was significant only when compared to the CONT group at 7 d after the start of the CORT injection (Fig 5-C). As repre- sented in Fig 5-D, the ALT in the CORT group was significantly higher after 3 and 7 d com- pared to 0 d of starting the injection course (21.2 and 25.0 vs. 14.2 U/ml, respectively). The ALT level decreased after the recovery period of chickens (13.4 U/ml at 14 d after the start of the CORT injection). The differences in ALT between CORT and CONT groups were signifi- cant only at 3 and 7 d after the start of the treatment (P<0.05). Quantitative real-time PCR analysis The relative expression of examined genes IGF-1 and caspase-9 in broiler chickens during the 7 d course of daily corticosterone injections and after 1-week recovery period post-treatment are shown in Fig 6. The current data showed that the relative expression of IGF-1 gene decreased significantly (P<0.05) in the treated broiler chickens by 0.11 and 0.22 fold at 3 and 7 d after the start of the CORT injection (Fig 6-A). This decrease continued to exist even after the termination of the CORT injection course (0.36 fold at the end of recovery period). In con- trast, the expression of caspase-9 gene increased significantly (P<0.05) by 4.0 and 5.0 fold at 3 and 7 d after the start of the CORT injection (Fig 6-B). At the end of recovery period (d 14), the expression of caspase-9 gene decreased during the recovery period, but it still remained sig- nificantly (P<0.05) higher than the controls by 4.2 fold. Lymphocyte proliferation 4.82 (Fig 4-A); SI of B cells: 0.96 vs. 2.46 (Fig 4-B); for CORT group vs. CONT group at 14 d after the start of injection course). Morphological detection of cell death categories The results of splenic cell death category rates in broiler chickens during the 7 d course of daily corticosterone injections and one week after the post-treatment recovery period are illustrated PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 12 / 25 Corticosterone and comprehensive responses in broiler chickens Fig 5. Effect of a 7 d course of daily saline (CONT) or corticosterone at dose of 5 mg/kg injections on the plasma concentration of total protein (A), free T3 hormone (B), AST (C) in broiler chickens during the treatment course (0, 3 and 7 d after the start of injection) a after cessation of the treatment (14 d after the start of injection). Data express the mean ± Means within a treatment group with different letters (a, b, c) are significantly different at P<0.0 difference between treatment groups within each time of the treatment course (P<0.05). doi:10.1371/journal.pone.0172684.g005 Fig 5. Effect of a 7 d course of daily saline (CONT) or corticosterone at dose of 5 mg/kg BW (CORT) injections on the plasma concentration of total protein (A), free T3 hormone (B), AST (C) and ALT (D) in broiler chickens during the treatment course (0, 3 and 7 d after the start of injection) and one week after cessation of the treatment (14 d after the start of injection). Data express the mean ± SEM (n = 5). Means within a treatment group with different letters (a, b, c) are significantly different at P<0.05.Significant difference between treatment groups within each time of the treatment course (P<0.05). 13 / 25 PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 Corticosterone and comprehensive responses in broiler chickens Fig 6. Effect of a 7 d course of daily saline (CONT) or corticosterone at dose of 5 mg/kg BW (CORT) injections on the relative expression of IGF-1 (A) and caspase-9 (B) genes in broiler chickens during the treatment course (0, 3 and 7 d after the start of injection) and one week after cessation of the treatment (14 d after the start of injection). Data express the mean ± SEM (n = 5). Means within a treatment group with different letters (a, b, c) are significantly different at P<0.05.Significant difference between treatment groups within each time of the treatment course (P<0.05). doi:10.1371/journal.pone.0172684.g006 Fig 6. Morphological detection of cell death categories Effect of a 7 d course of daily saline (CONT) or corticosterone at dose of 5 mg/kg BW (CORT) injections on the % of apoptotic (A) and necrotic (B) cells in broiler chickens during the treatment course (0, 3 and 7 d after the start of injection) and one week after cessation of the treatment (14 d after the start of injection). Data express the mean ± SEM (n = 5). Means within a treatment group with different letters (a, b, c) are significantly different at P<0.05.Significant difference between treatment groups within each time of the treatment course (P<0.05). doi:10.1371/journal.pone.0172684.g007 Fig 7. Effect of a 7 d course of daily saline (CONT) or corticosterone at dose of 5 mg/kg BW (CORT) injections on the % of apoptotic (A) and necrotic (B) cells in broiler chickens during the treatment course (0, 3 and 7 d after the start of injection) and one week after cessation of the treatment (14 d after the start of injection). Data express the mean ± SEM (n = 5). Means within a treatment group with different letters (a, b, c) are significantly different at P<0.05.Significant difference between treatment groups within each time of the treatment course (P<0.05). doi:10.1371/journal.pone.0172684.g007 necrotic cells decreased (15.4%) in the CORT group, but it remained significantly different from their controls (10.6% apoptosis and 5.6% necrosis in the CONT group; Fig 7). Morphological detection of cell death categories Effect of a 7 d course of daily saline (CONT) or corticosterone at dose of 5 mg/kg BW (CORT) injections on the relative expression of IGF-1 (A) and caspase-9 (B) genes in broiler chickens during the treatment course (0, 3 and 7 d after the start of injection) and one week after cessation of the treatment (14 d after the start of injection). Data express the mean ± SEM (n = 5). Means within a treatment group with different letters (a, b, c) are significantly different at P<0.05.Significant difference between treatment groups within each time of the treatment course (P<0.05). doi:10.1371/journal.pone.0172684.g006 doi:10.1371/journal.pone.0172684.g006 in Fig 7. The % of total apoptotic and necrotic cells examined in CONT chicken group ranged between 10.6–11.0% and 5.6–6.0%, respectively. The chickens treated with corticosterone showed a significantly (P<0.05) lower % of apoptotic cells during the CORT injection course when compared to its control (9.0% and 6.4% vs. 11.2% at 3 and 7 d vs. 0 d after the start of the CORT injection, respectively; Fig 7-A). In contrast, a significantly (P<0.05) higher % of necrotic cells was observed after CORT injection when compared to its control (14.6% and 21.4% vs. 5.6% at 3 and 7 d vs. 0 d after the start of the CORT injection, respectively; Fig 7-B). At the end of recovery period (d 14), the apoptotic cells increased again (8.2%) while the PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 14 / 25 Corticosterone and comprehensive responses in broiler chickens Fig 7 Effect of a 7 d course of daily saline (CONT) or corticosterone at dose of 5 mg/kg BW (CORT) Fig 7. Effect of a 7 d course of daily saline (CONT) or corticosterone at dose of 5 mg/kg BW (CORT) injections on the % of apoptotic (A) and necrotic (B) cells in broiler chickens during the treatment course (0, 3 and 7 d after the start of injection) and one week after cessation of the treatment (14 d after the start of injection). Data express the mean ± SEM (n = 5). Means within a treatment group with different letters (a, b, c) are significantly different at P<0.05.Significant difference between treatment group within each time of the treatment course (P<0.05). doi:10.1371/journal.pone.0172684.g007 Fig 7. In agreement with previous studies [3–7,29], all growth performance measured in this experiment was negatively affected by the CORT treatment (Fig 1). The final BW, DG, FI and FC of broiler chickens during the week of the CORT injection course significantly decreased when compared with the controls because elevated corticosterone levels inhibits anabolic pro- cesses and suppress appetite [55]. It was also reported that CORT may suppress growth perfor- mance by reducing the absorption of feed through the small intestine [56]. Furthermore, we found that growth performance traits, except FC, of broilers from CORT group remained lower than that of broilers from CONT group after 7 days of cessation of the CORT treatment (Fig 1). In this context, Yang et al. [3] found that feeding CORT-treated broiler chickens a high-energy diet did not compensate for the adverse effects of CORT-induced stress on growth rates. It means that CORT effect may be extended for a long time on stressed broilers even after the termination of CORT treatment or stress condition. The relative weight of liver in broilers from CORT group was higher (P<0.05) than that in broilers from CONT group at 3 and 7 d after initiation of the CORT injection (Fig 2-A). Simi- lar results were obtained by Jiang et al. [6] and Lin et al. [55]. The increase in liver weights after CORT injection is apparently due to the hepatic lipogenesis and concomitant accumula- tions of lipids in liver tissues [57–59]. On the contrary, the CORT treatment significantly decreased the relative weight of thymus, bursa and spleen (Fig 2-B, 2-C and 2-D, respectively). The decrease in relative weights of such immune-organs by CORT administration was also observed in other studies [3,9], and con- firmed the sensitivity of immune-organs to corticosterone [60–62]. It was found that the administration of glucocorticoids resulted in a rapid degeneration of primary lymphoid tis- sues, such as thymus and bursa, due to their apoptotic effects [63,64]. However, the induced apoptosis may vary depending on the cell type and the expression levels of glucocorticoid receptors in these organs [65,66]. A study conducted by Schaumburg and Crone in 1971 [67] demonstrated that bursa lymphocytes in chickens contain higher levels of glucocorticoid receptors than those of the thymus, which makes the bursa more susceptible to the effects of glucocorticoids. Corticosterone and comprehensive responses in broiler chickens of potential stressors [1]. In our previous work [48], we found that endotoxin stress by E. coli infection to laying chickens markedly increased plasma corticosterone concentration by approximately 3 times more than normal chickens only 3 hr after E. coli injection. Such results and previous reports concluded that various forms of stress directly or indirectly activate the hypothalamic-pituitary-adrenal (HPA) axis and the release of corticosterone into blood [2]. Several investigations administrated the corticosterone in chicken diets at a concentration of 20–30 mg/kg [49,50] or in drinking water at 5–20 mg/L [51,52] to induce a stress status. Other investigations injected the corticosterone at doses of 4–6 mg/kg body weight once a day for 7 days to mimic acute stress in broiler chickens [3,53,54]. In this research, we directly injected broiler chickens with corticosterone at a dose of 5 mg/kg BW over 7 consequent days to simulate the stressful situation in these chickens; focusing on a wide range of expected responses at physiological, immunological and molecular levels throughout one-week of daily intramuscular CORT injection and after one-week of the cessation of CORT injection to test the ability of broilers to recover after termination of the stress condition. of potential stressors [1]. In our previous work [48], we found that endotoxin stress by E. coli infection to laying chickens markedly increased plasma corticosterone concentration by approximately 3 times more than normal chickens only 3 hr after E. coli injection. Such results and previous reports concluded that various forms of stress directly or indirectly activate the hypothalamic-pituitary-adrenal (HPA) axis and the release of corticosterone into blood [2]. Several investigations administrated the corticosterone in chicken diets at a concentration of 20–30 mg/kg [49,50] or in drinking water at 5–20 mg/L [51,52] to induce a stress status. Other investigations injected the corticosterone at doses of 4–6 mg/kg body weight once a day for 7 days to mimic acute stress in broiler chickens [3,53,54]. In this research, we directly injected broiler chickens with corticosterone at a dose of 5 mg/kg BW over 7 consequent days to simulate the stressful situation in these chickens; focusing on a wide range of expected responses at physiological, immunological and molecular levels throughout one-week of daily intramuscular CORT injection and after one-week of the cessation of CORT injection to test the ability of broilers to recover after termination of the stress condition. PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 Discussion Any type of stress triggers many biological mechanisms in the body of animals and humans to re-establish the homeostatic conditions and maintain the physiological activity [46]. The chicken is an important animal model well characterized in many biological aspects and brid- ges the evolutionary gap between mammals and other vertebrates [47]. At the same time, the massive meat production of broiler chickens make them continuously exposed to a wide range PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 15 / 25 Corticosterone and comprehensive responses in broiler chickens after a longer time of daily CORT injection (at 7 d after the start of CORT injections, not ear- lier), and these birds were able to recover these organs near to their normal weights as recorded in CONT broilers after cessation of the treatment (Fig 2-C and 2-D). Although we did not determine the expression of corticosterone receptors in thymus, bursa and spleen in this study, it is likely that these receptors might control the variance in the relative weight sup- pression observed among these organs after CORT treatment. It has been emphasized that elevated plasma corticosterone and increased circulating H/L ratio are the two most accepted indicators of the stress condition in birds [11]. In a series of experiments conducted by Shini et al. [9,51,70,71] on young chickens, it was demonstrated that plasma corticosterone concentration and H/L ratio significantly increased post adminis- tration with exogenous corticosterone, and this was associated with a decreased peripheral lymphocyte count and increased peripheral heterophil count, whereas the total circulating leu- kocyte number decreased. In the present study, the TWBC’s was significantly (P<0.05) decreased by the CORT treatment (37.8 vs. 59.6 x103/μl and 28.0 vs. 62.1 x103/μl at 3 and 7 d after the start of the treatment for CORT vs. CONT group, respectively; Fig 3-A). The H/L ratio was significantly (P<0.05) increased at the same times (0.64 vs. 0.30 and 0.91 vs. 0.34 H/L ratio at 3 and 7 d after the start of the treatment for CORT vs. CONT group, respectively; Fig 3-B). In our experiment, the negative effect of CORT on the TWBC’s and H/L ratio in CORT- treated chickens, during the week of the injection course, was followed by a gradual improve- ment that remained low after 1 week of cessation of the CORT injections compared to its val- ues in CONT chickens (the TWBC’s increased again to 35.2 x103/μl vs. 60.8 x103/μl and the H/ L ratio decreased again to 0.65 vs. 0.35 in CORT group vs. CONT group, respectively; Fig 3). These results are in accordance with those obtained by other researchers [14,51,71] who stud- ied the administration of CORT in diets and drinking water of chickens and obtained a high H/L ratio with a high plasma corticosterone values even after interruption of CORT supple- mentation. Conversely, other studies reported that the thymus is one of the body tissues with the highest content of glucocorticoid receptors [68], and that the immature thymocytes show a higher glucocorticoid receptor density than matured thymocytes in human thymus [69]. In the present study, we noted that the decrease in thymus relative weight appeared early at 3 d after the start of the CORT treatment; moreover, the negative effect of CORT treatment on the thymus relative weight continued at 7 d after the start of the CORT treatment, and did not recover even 1 week after cessation of the 7-days CORT treatment (Fig 2-B). In contrast, the negative effect of CORT treatment on the relative weights of bursa and spleen appeared PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 16 / 25 PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 Although we observed a higher relative weight of liver in broilers from the CORT group compared to CONT group (Fig 2-A), the increase observed in plasma ALT and AST levels in CORT group demonstrated the failure in liver function caused by CORT treatment [20]. As previously men- tioned, CORT increased the relative weight of liver mainly due to the increase of lipid deposi- tion, and this in turn induces an impairment of liver tissues and a leakage of enzymes from cell membranes into the blood circulation [79]. injection to 4.86 and 4.74 g/dl at 3 and 7 d after the start of the CORT injection compared to the CONT group (Fig 5-A). Similar results were obtained by a recent study [3], in which the CORT treatment enhanced the total plasma protein levels in broiler chickens. The significant increase in plasma protein of the CORT-treated broilers was accompanied by a significant decrease in the body weight (Fig 1) and the relative weights of immune-organs studied in our experiment (Fig 2). This observation is supported by findings of many authors [18,19,76], There is a strong evidence that the administration of IGF-1 stimulates growth rate and pro- tein synthesis [80] by its role in mediating growth hormone and thyroid hormones [81]. One mechanism by which glucocorticoid hormones depress growth in addition to reduced feed intake in chickens is by depressing circulating concentrations and hepatic expression of IGF-1 [22,29]. In line with these reports, the current data showed that the relative expression of hepatic IGF-1 gene significantly (P<0.05) decreased in the CORT-treated broiler chickens at 3 and 7 d after initiation of the CORT injection by 0.11 and 0.22 fold, respectively. Furthermore, we found that the relative expression of IGF-1 in CORT-treated broiler chickens continued to decrease significantly (P<0.05) after the termination of the CORT injection course (down-reg- ulation by 0.36-fold at 14 d compared to 0 d of the CORT treatment; Fig 6-A). This inhibition of recovery for liver IGF-1 mRNA after termination of the CORT treatment was also observed in fasting-stressed chickens [82] and rats [83] after re-feeding. The data obtained from growth performance also confirmed the incapacity of the CORT-treated broilers to recover to the nor- mal rates of broilers in the CONT group (Fig 1). The other responsible mechanism for growth inhibition by glucocorticoid hormones in chickens is the proteolytic action of caspases on specific cell substrates [33,34]. These observations assumed that corticosterone directly causes a redistribution of leukocyte components in the blood and a proportional change in the H/L ratio to multiply the cells required for the nonspecific immune response such as heterophils [72]. q p p p On the other hand, the CORT treatment negatively affected the lymphocyte proliferation, another important index to stress and correlated to cell mediated response in chickens [9,71]. These findings are confirmed with our results since the SI of T cells was decreased (P<0.05) by the CORT treatment from 5.01 at 0 d of the injection course to 2.25 and 1.74 at 3 and 7 d after the start of the injection course, respectively (Fig 4-A). The SI of B cells was also decreased (P<0.05) from 2.56 at 0 d of the injection course to 0.92 and 0.65 at 3 and 7 d after the start of the injection course, respectively (Fig 4-B). In addition, the SI of T and B lymphocytes were still lower (P<0.05) in the CORT group than in the CONT group after the recovery of chickens and termination of the CORT treatment (Fig 4). In birds, lymphocytes are specifically differen- tiated into T and B cells during development in the thymus and bursa, respectively [73]. There- fore, the low production of T and B cells in CORT-treated broilers in this study may be due to the decrease in the relative weight of these two organs (thymus, as shown in Fig 2-B; and bursa, as shown in Fig 2-C). It may also be attributed to the decrease in the number of lympho- cytes in the CORT-treated chickens (inducing high H/L ratio, as shown in Fig 3-B). Previous studies on captive birds have shown that T-cell mediated acquired immunity is directly and positively linked to food intake and body mass [74], which have been negatively affected by the CORT treatment in our study (as shown in Fig 1) and, consequently, decreased the stimulation of T cells. Moreover, the chicken lymphocyte proliferation may be inhibited by corticosterone due to lipid peroxidation of cellular membranes of these cells [75]. In this study, some plasma biochemical constituents were affected by the corticosterone treatment. The TP significantly (P<0.05) increased from 2.98 g/dl before starting the CORT PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 17 / 25 Corticosterone and comprehensive responses in broiler chickens injection to 4.86 and 4.74 g/dl at 3 and 7 d after the start of the CORT injection compared to the CONT group (Fig 5-A). Similar results were obtained by a recent study [3], in which the CORT treatment enhanced the total plasma protein levels in broiler chickens. The significant increase in plasma protein of the CORT-treated broilers was accompanied by a significant decrease in the body weight (Fig 1) and the relative weights of immune-organs studied in our experiment (Fig 2). This observation is supported by findings of many authors [18,19,76], who reported that glucocorticoids or its synthetic compounds like dexamethasone retard the growth of skeletal muscles by suppressing protein synthesis and increasing protein catabolism in chickens [77]. The apparent elevation of plasma protein in the CORT-treated chickens may also be a result of the concomitant enhancement in protein-derived gluconeogenesis which is activated by glucocorticoids [78]. In addition, insulin-like growth factors and the thyrotrophic [triiodothyronine (T3) and thyroxine (T4)] axis are considered to be prerequisites for the nor- mal growth and development [27]. In our study, we measured the plasma free T3 hormone lev- els and we found a significant decrease after initiation of the CORT treatment at 7 d (2.0 vs. 5.6 pg/ml for CORT vs. CONT group, respectively; P<0.05), and then it returned to similar levels of CONT group after cessation of the treatment (Fig 5-B). Previous studies indicated that exog- enous CORT administration depressed circulating T3 levels by the inhibition of T4 synthesis and peripheral deiodination [27,28]. However, the decrease in fT3 levels in CORT-treated chickens was not observed before the 7th d of the course of the treatment in the present study, indicating a slow response and adjustment of T3 levels after short-term administration of CORT [28]. In addition to the effects on plasma protein and free T3 hormone, CORT treat- ment significantly (P<0.05) increased the plasma levels of AST and ALT enzymes particularly during the week of the CORT injection course (Fig 5-C and 5-D, respectively). This process has been studied in some details [84]; glucocorticoid signaling increases the expression of the pro-apoptotic Bcl-2 family member, which can activate the pro-apoptotic proteins Bax/Bak to disrupt mitochondrial membrane potential, resulting in the release of cytochrome c and other apoptogenic proteins. Consequently, caspase-9 and subsequent effector caspase-3 are PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 18 / 25 Corticosterone and comprehensive responses in broiler chickens activated, finally leading to apoptosis. In the present study, we focused on evaluating the rela- tive expression of caspase-9 gene in the spleen of broiler chickens after corticosterone injection because it is an important member of caspases that is responsible for both initiating and exe- cuting apoptosis [35,36]. The expression of caspase-9 gene significantly (P<0.05) increased by 4.0 and 5.0 fold at 3 and 7 d of the course of the treatment, respectively (Fig 6-B). At the end of the recovery period, the expression of caspase-9 gene started to decrease but remained sig- nificantly (P<0.05) higher, in the treated chickens, than in the control chickens by 4.2 fold. Generally, Collier et al. [85] reported that endogenous glucocorticoids caused an increase in the splenic apoptotic cells in stressed-mice. Our results concluded that CORT inhibited the growth in treated chickens by both down-regulating hepatic IGF-1 gene expression and up- regulating splenic caspase-9 gene expression. Although apoptosis can occur as a normal and beneficial defense mechanism in most organisms, necrosis can also be seen as a part of cellular death during stressful stimulation but it is considered as an unnatural cell death process [86]. The necrotic pathway form of cell death programs is accompanied by a rapid collapse of plasma membrane and it is independent of expression of new genes [38]. In the present study, we targeted at providing further informa- tion about the necrotic versus apoptotic pathways induced by CORT treatment in broilers as an expected result of exposure to stress. As a reliable criterion for distinguishing between apo- ptosis and necrosis, we morphologically detected the apoptotic and necrotic cells in the spleen of treated and untreated broiler chickens with CORT using fluorescent microscope after label- ing with AO/EB dyes. We found that corticosterone significantly (P<0.05) decreased the % of apoptotic cells from the range of 10.6–11.2% in CONT group down to 9.0%, 6.4% and 8.2% in CORT group at 3, 7 and 14 d after the start of the injection course, respectively (Fig 7-A). PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 Corticosterone and comprehensive responses in broiler chickens immune-organs decreased in the CORT-treated chickens with low counts of TWBC’s and lymphocyte proliferation, while the H/L ratio increased; indicating, as expected, immunosup- pressive effect for CORT. Furthermore, high expression of caspase-9 gene was observed in the bursa of CORT-treated chickens, however, it was associated with high necrotic vs. low apopto- tic cell death pathway in the spleen. After termination of the CORT treatment in broilers, most of these aspects remained negatively affected by CORT and did not recover to its normal sta- tus. Such information provides more understanding regarding the pathways of stress in broil- ers and may set the means through which a chicken is able to mount a successful defense against stress. Supporting information S1 Table. Effect of a 7 d course of daily saline (CONT) or corticosterone at dose of 5 mg/kg BW (CORT) injections on apoptotic categories of splenic cells in broiler chickens during the treatment course (0, 3 and 7 d after the start of injection) and one week after cessation of the treatment (14 d after the start of injection). (DOCX) S1 Fig. Visual examples for apoptotic categories determined morphologically under fluo- rescent microscope after labeling with acridine orange and ethidium bromide (AO/EB) dyes. Images represent samples of splenic cells from broiler chickens treated with a 7-d course of daily saline (CONT) or corticosterone at dose of 5 mg/kg BW (CORT) injections during the treatment course (0, 3 and 7 d after the start of injection) and one week after cessation of the treatment (14 d after the start of injection). Viable cells appear green, apoptotic cells appear yellow, and necrotic cells appear orange/red colored. Higher expression of necrosis can be seen in CORT 3, CORT 7 and CORT 14 images. Scale bars: 100 μm. (TIF) Acknowledgments This study was funded by the General Scientific Research Department at Cairo University (GSRD-CU) under activities carried out by the Project of Rapid Climate Change in Poultry Cellular and Molecular Physiology (RCC-PCMP). Dr. Ahmed O. Abass was the principal investigator and research team leader of the project. We would like to express our appreciation to all the personnel from the Poultry Biotechnology Lab and members of Poultry Services Cen- ter at Faculty of Agriculture, Cairo University, for their assistance in sample preparation and monitoring birds throughout the experimental period. Author Contributions Conceptualization: AMMA MMM AOA. On the contrary, the corticosterone treatment significantly (P<0.05) increased the rate of necrotic cells from the range of 5.6–6.0% in CONT group up to 14.6%, 21.4% and 15.4% in CORT group at 3, 7 and 14 d after the start of the injection course, respectively (Fig 7-B). To the best of our knowledge, these results are the first reports concerning the apoptotic and necrotic pathways comparison after CORT treatment in broiler chickens. A study that most closely resembled our examination of apoptosis versus necrosis with CORT treatment was examining Leydig cells treated with CORT in rats [39]; in which signs of necrosis were seen more than apoptosis in treated cells with high doses of CORT compared to untreated cells or cells treated with lower doses of CORT. In another study conducted on rats [40], they reported cells of apoptosis and necrosis as 10.51% and 1.98% at 10−6 mol/L of CORT vs. 4.87% and 26.39% at 10−3 mol/L of CORT, respectively. Whereas the apoptotic pathway can poten- tially be modulated to maintain cell viability under stress condition [87], it seems that CORT treatment blocks this pathway. The high necrosis vs. low apoptosis incidence in splenic cells that we obtained during and after the daily 7-d CORT injection course in broiler chickens indicated that the severe stress and long-term CORT administration increases the possibility of tissue and cell damages, and induces cell death in an irreversible process or incompatible survival recovery [86,88]. This may also explain why the reduction in IGF-1 (Fig 6-A) and the increase in caspase-9 (Fig 6-B) expression in the treated chickens remained high and did not recover to its normal expression in the control chickens after cessation of the CORT treatment in our study. In conclusion, the current study provides overall view of physiological challenges modu- lated in broiler chickens under mimicked stressful condition by CORT treatment. The growth performance was negatively affected by the CORT injection accompanied by low concentra- tion of free T3 hormone in plasma and down-regulation of hepatic IGF-1 gene. Failure in liver functions was seen with CORT treatment by the increase in liver weight with an increase in plasma levels of total protein, and AST and ALT enzymes. Moreover, the relative weight of PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 19 / 25 References 1. Moberg GP, Mench JA, editors. The biology of animal stress: basic principles and implications for ani- mal welfare. Wallingford: CABI; 2000. 1. Moberg GP, Mench JA, editors. The biology of animal stress: basic principles and implications for ani- mal welfare. Wallingford: CABI; 2000. 2. Koolhaas JM, Korte SM, De Boer SF, Van Der Vegt BJ, Van Reenen CG, Hopster H, et al. Coping styles in animals: current status in behavior and stress-physiology. Neurosci Biobehav Rev. 1999; 23: 925– 935. Available: http://www.ncbi.nlm.nih.gov/pubmed/10580307 PMID: 10580307 3. Yang J, Liu L, Sheikhahmadi A, Wang Y, Li C, Jiao H, et al. Effects of corticosterone and dietary energy on immune function of broiler chickens. PLoS One. 2015; 10: e0119750. doi: 10.1371/journal.pone. 0119750 PMID: 25803644 4. Virden WS, Thaxton JP, Corzo A, Dozier WA, Kidd MT. Evaluation of Models Using Corticosterone and Adrenocorticotropin to Induce Conditions Mimicking Physiological Stress in Commercial Broilers. Poult Sci. 2007; 86: 2485–2491. doi: 10.3382/ps.2006-00215 PMID: 18029793 5. Yuan L, Lin H, Jiang KJ, Jiao HC, Song ZG. Corticosterone administration and high-energy feed results in enhanced fat accumulation and insulin resistance in broiler chickens. Br Poult Sci. 2008; 49: 487– 495. doi: 10.1080/00071660802251731 PMID: 18704796 6. Jiang KJ, Jiao HC, Song ZG, Yuan L, Zhao JP, Lin H. Corticosterone administration and dietary glucose supplementation enhance fat accumulation in broiler chickens. Br Poult Sci. 2008; 49: 625–631. doi: 10. 1080/00071660802337241 PMID: 18836910 7. Vahdatpour T, Adl KN, Nezhad YE, Sis NM, Riyazi SR, Vahdatpour S. Effects of Corticosterone Intake as Stress-Alternative Hormone on Broiler Chickens: Performance and Blood Parameters. Asian J Anim Vet Adv. 2009; 4: 16–21. 8. Song Z, Yuan L, Jiao H, Lin H. Effect of corticosterone on hypothalamic corticotropin-releasing hormone expression in broiler chicks (Gallus gallus domesticus) fed a high energy diet. Asian-Australasian J Anim Sci. 2011; 24: 1736–1743. 9. Shini S, Huff GR, Shini A, Kaiser P. Understanding stress-induced immunosuppression: exploration of cytokine and chemokine gene profiles in chicken peripheral leukocytes. Poult Sci. Oxford University Press; 2010; 89: 841–851. 10. Post J, Rebel JMJ, Huurne AAHM, ter Huurne AAHM. Automated Blood Cell Count : A Sensitive and Reliable Method to Study Corticosterone-Related Stress in Broilers. Poult Sci. 2003; 82: 591–595. PMID: 12710478 11. Siegel HS. Stress, strains and resistance. Br Poult Sci. Taylor & Francis Group; 1995; 36: 3–22. 12. Vleck C, Vertalino N, Vleck D, Bucher T. Writing – review & editing: GMKM AOA. Writing – review & editing: GMKM AOA. Writing – review & editing: GMKM AOA. Resources: MGE AME AMMA MMM AOA. Supervision: AMMA MMM AOA. Validation: GMKM AOA. Vi li i GMKM AOA Visualization: GMKM AOA. Writing – original draft: GMKM AOA. Conceptualization: AMMA MMM AOA. Conceptualization: AMMA MMM AOA. Data curation: AOA. Formal analysis: GMKM AOA. Funding acquisition: AOA. Investigation: GMKM MGE AME AOA. Methodology: AMMA MMM AOA. Project administration: AOA. 20 / 25 PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 Corticosterone and comprehensive responses in broiler chickens Resources: MGE AME AMMA MMM AOA. Supervision: AMMA MMM AOA. Validation: GMKM AOA. Visualization: GMKM AOA. Writing – original draft: GMKM AOA. Writing – review & editing: GMKM AOA. PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 References Stress, corticosterone, and heterophil to lymphocyte ratios in free-living Ade´lie Penguins. Condor. 2000; 102: 392–400. 13. Shini S, Shini A, Huff GR. Effects of chronic and repeated corticosterone administration in rearing chick- ens on physiology, the onset of lay and egg production of hens. Physiol Behav. 2009; 98: 73–77. doi: 10.1016/j.physbeh.2009.04.012 PMID: 19393674 14. Babacanoğlu E, Yalc¸in S, Uysal S. Evaluation of a stress model induced by dietary corticosterone sup- plementation in broiler breeders: effects on egg yolk corticosterone concentration and biochemical blood parameters. Br Poult Sci. 2013; 54: 677–685. doi: 10.1080/00071668.2013.847901 PMID: 24397505 15. Grasman KA. Assessing immunological function in toxicological studies of avian wildlife. Integr Comp Biol. 2002; 42: 34–42. doi: 10.1093/icb/42.1.34 PMID: 21708692 16. Engler H, Stefanski V. Social stress and T cell maturation in male rats: transient and persistent alter- ations in thymic function. Psychoneuroendocrinology. 2003; 28: 951–969. Available: http://www.ncbi. nlm.nih.gov/pubmed/14529702 PMID: 14529702 17. Van Hunsel F, Van Gastel A, Neels H, Wauters A, Demedts P, Bruyland K, et al. The influence of psy- chological stress on total serum protein and patterns obtained in serum protein electrophoresis. Psychol Med. 1998; 28: 301–309. Available: http://www.ncbi.nlm.nih.gov/pubmed/9572088 PMID: 9572088 21 / 25 PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 Corticosterone and comprehensive responses in broiler chickens 18. Dong H, Lin H, Jiao HC, Song ZG, Zhao JP, Jiang KJ. Altered development and protein metabolism in skeletal muscles of broiler chickens (Gallus gallus domesticus) by corticosterone. Comp Biochem Phy- siol A Mol Integr Physiol. 2007; 147: 189–195. doi: 10.1016/j.cbpa.2006.12.034 PMID: 17289413 19. Gao J, Lin H, Song ZG, Jiao HC. Corticosterone alters meat quality by changing pre-and postslaughter muscle metabolism. Poult Sci. 2008; 87: 1609–1617. doi: 10.3382/ps.2007-00007 PMID: 18648056 20. Vahdatpour T, Nikpiran H. Effects of Protexin, Fermacto and combination of them on blood enzymes and performance of Japanese quails (Coturnix Japonica). Ann Biol Res. 2011; 2: 283–291. Available: http://scholarsresearchlibrary.com/ABR-vol2-iss3/ABR-2011-2-3-283-291.html 21. Johnson BE, Scanes CG, King DB, Marsh JA. Effect of hypophysectomy and growth hormone on immune development in the domestic fowl. Dev Comp Immunol. 17: 331–339. Available: http://www. ncbi.nlm.nih.gov/pubmed/8375568 PMID: 8375568 22. Tsukada A, Ohkubo T, Sakaguchi K, Tanaka M, Nakashima K, Hayashida Y, et al. Thyroid hormones are involved in insulin-like growth factor-I (IGF-I) production by stimulating hepatic growth hormone receptor (GHR) gene expression in the chicken. Growth Horm IGF Res. 1998; 8: 235–242. Available: http://www.ncbi.nlm.nih.gov/pubmed/10984312 PMID: 10984312 23. Barbour EK, Tayeb I, Shaib H, Ibrahim IM. References Physiological and carcass traits in heat-stressed broilers dif- fering in heat acclimatization, chemical or feed restriction treatments. Agric Biol J North Am. 2010; 1: 65–74. 24. McMurtry JP. Nutritional and developmental roles of insulin-like growth factors in poultry. J Nutr. 1998; 128: 302S–305S. Available: http://www.ncbi.nlm.nih.gov/pubmed/9478012 PMID: 9478012 25. Allan GJ, Flint DJ, Patel K. Insulin-like growth factor axis during embryonic development. Reproduction. 2001; 122: 31–39. Available: http://www.ncbi.nlm.nih.gov/pubmed/11425327 PMID: 11425327 26. Richards MP, Poch SM, McMurtry JP. Expression of insulin-like growth factor system genes in liver and brain tissue during embryonic and post-hatch development of the turkey. Comp Biochem Physiol A Mol Integr Physiol. 2005; 141: 76–86. doi: 10.1016/j.cbpb.2005.04.006 PMID: 15905111 27. Decuypere E, Scanes CG, Ku¨hn ER. Effects of glucocorticoids on circulating concentrations of thyrox- ine (T4) and triiodothyronine (T3) and on peripheral monodeiodination in pre- and post-hatching chick- ens. Horm Metab Res. 1983; 15: 233–236. doi: 10.1055/s-2007-1018680 PMID: 6307846 28. Darras VM, Kotanen SP, Geris KL, Berghman LR, Ku¨hn ER. Plasma thyroid hormone levels and iodothyronine deiodinase activity following an acute glucocorticoid challenge in embryonic compared with posthatch chickens. Gen Comp Endocrinol. 1996; 104: 203–212. doi: 10.1006/gcen.1996.0163 PMID: 8930611 29. Song ZG, Zhang XH, Zhu LX, Jiao HC, Lin H. Dexamethasone alters the expression of genes related to the growth of skeletal muscle in chickens (Gallus gallus domesticus). J Mol Endocrinol. 2011; 46: 217– 225. doi: 10.1530/JME-10-0162 PMID: 21325373 30. Fischer U, Ja¨nicke RU, Schulze-Osthoff K. Many cuts to ruin: a comprehensive update of caspase sub- strates. Cell Death Differ. 2003; 10: 76–100. doi: 10.1038/sj.cdd.4401160 PMID: 12655297 31. Agas D, Marchetti L, Menghi G, Materazzi S, Materazzi G, Capacchietti M, et al. Anti-apoptotic Bcl-2 enhancing requires FGF-2/FGF receptor 1 binding in mouse osteoblasts. J Cell Physiol. 2008; 214: 145–52. doi: 10.1002/jcp.21170 PMID: 17559077 32. Ling Y, Lu N, Gao Y, Chen Y, Wang S, Yang Y, et al. Endostar induces apoptotic effects in HUVECs through activation of caspase-3 and decrease of Bcl-2. Anticancer Res. 2009; 29: 411–417. Available: http://www.ncbi.nlm.nih.gov/pubmed/19331180 PMID: 19331180 33. Enari M, Sakahira H, Yokoyama H, Okawa K, Iwamatsu A, Nagata S. A caspase-activated DNase that degrades DNA during apoptosis, and its inhibitor ICAD. Nature. 1998; 391: 43–50. doi: 10.1038/34112 PMID: 9422506 34. Sahara S, Aoto M, Eguchi Y, Imamoto N, Yoneda Y, Tsujimoto Y. Acinus is a caspase-3-activated pro- tein required for apoptotic chromatin condensation. Nature. 1999; 401: 168–73. doi: 10.1038/43678 PMID: 10490026 35. PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 References Nicholson DW, Thornberry NA. Caspases: killer proteases. Trends Biochem Sci. 1997; 22: 299–306. Available: http://www.ncbi.nlm.nih.gov/pubmed/9270303 PMID: 9270303 36. Kumar S, Colussi PA. Prodomains—adaptors—oligomerization: the pursuit of caspase activation in apoptosis. Trends Biochem Sci. 1999; 24: 1–4. Available: http://www.ncbi.nlm.nih.gov/pubmed/ 10087912 PMID: 10087912 37. Rimoldi S, Lasagna E, Sarti FM, Marelli SP, Cozzi MC, Bernardini G, et al. Expression profile of six stress-related genes and productive performances of fast and slow growing broiler strains reared under heat stress conditions. Meta Gene. 2015; 6: 17–25. doi: 10.1016/j.mgene.2015.08.003 PMID: 26380816 22 / 25 PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 Corticosterone and comprehensive responses in broiler chickens 38. Proskuryakov SY, Konoplyannikov AG, Gabai VL. Necrosis: A specific form of programmed cell death? Exp Cell Res. 2003; 283: 1–16. PMID: 12565815 39. Gao H-B, Tong M-H, Hu Y-Q, Guo Q-S, Ge R, Hardy MP. Glucocorticoid Induces Apoptosis in Rat Ley- dig Cells. Endocrinology. 2002; 143: 130–138. doi: 10.1210/endo.143.1.8604 PMID: 11751601 40. Sun C, Liu N, Li H, Zhang M, Liu S, Liu X, et al. Experimental study of effect of corticosterone on primary cultured hippocampal neurons and their Ca2+/CaMKII expression. J Huazhong Univ Sci Technolog Med Sci. 2004; 24: 543–6. Available: http://www.ncbi.nlm.nih.gov/pubmed/15791835 PMID: 15791835 41. Gehad AE, Mehaisen GM, Abbas AO, Mashaly MM. The role of light program and melatonin on allevia- tion of inflammation induced by lipopolysaccharide injection in broiler chickens. Int J Poult Sci. 2008; 7: 193–201. 42. Zhang L, Yue HY, Zhang HJ, Xu L, Wu SG, Yan HJ, et al. Transport stress in broilers: I. Blood metabo- lism, glycolytic potential, and meat quality. Poult Sci. 2009; 88: 2033–2041. doi: 10.3382/ps.2009- 00128 PMID: 19762854 43. Zhang LB, Guo YM. Effects of liquid DL-2-hydroxy-4-methylthio butanoic acid on growth performance and immune responses in broiler chickens. Poult Sci. 2008; 87: 1370–1376. doi: 10.3382/ps.2007- 00366 PMID: 18577618 44. Ellestad LE, Malkiewicz SA, Guthrie HD, Welch GR, Porter TE. Expression and regulation of glucocorti- coid-induced leucine zipper in the developing anterior pituitary gland. J Mol Endocrinol. 2009; 42: 171– 183. doi: 10.1677/JME-08-0066 PMID: 19060181 45. Thakur P, Sanyal S. Chemopreventive role of preferential COX-2 inhibitor diclofenac in 9, 10-dimethy- benz (a) anthracene induced experimental lung carcinogenesis. Am J Biomed Sci. 2010; 2: 275–288. 46. McEwen BS. Physiology and neurobiology of stress and adaptation: central role of the brain. Physiol Rev. 2007; 87: 873–904. doi: 10.1152/physrev.00041.2006 PMID: 17615391 47. References Hillier LW, Miller W, Birney E, Warren W, Hardison RC, Ponting CP, et al. Sequence and comparative analysis of the chicken genome provide unique perspectives on vertebrate evolution. Nature. 2004; 432: 695–716. doi: 10.1038/nature03154 PMID: 15592404 48. Mehaisen GMK, Eshak MG, El Sabry MI, Abass AO. Expression of Inflammatory and Cell Death Pro- gram Genes and Comet DNA Damage Assay Induced by Escherichia coli in Layer Hens. PLoS One. 2016; 11: e0158314. doi: 10.1371/journal.pone.0158314 PMID: 27347679 49. Hu X, Guo Y. Corticosterone Administration Alters Small Intestinal Morphology and Function of Broiler Chickens. Asian-Australasian J Anim Sci. 2008; 21: 1773–1778. 50. Aslam MA, Groothuis TGG, Smits MA, Woelders H. Effect of Corticosterone and Hen Body Mass on Primary Sex Ratio in Laying Hen (Gallus gallus), Using Unincubated Eggs 1. 2014; 90: 1–9. 51. Shini S, Kaiser P. Effects of stress, mimicked by administration of corticosterone in drinking water, on the expression of chicken cytokine and chemokine genes in lymphocytes. Stress. 2009; 12: 388–399. doi: 10.1080/10253890802526894 PMID: 19006006 52. Fu W, Duan Y, Wang S, Ni Y, Grossmann R, Zhao R. Comparative proteomic analysis of the breast muscle response to chronic corticosterone administration in broiler chickens showing long or short tonic immobility. Poult Sci. 2014; 93: 784–793. doi: 10.3382/ps.2013-03456 PMID: 24706954 53. Zulkifli I, Najafi P, Nurfarahin AJ, Soleimani AF, Kumari S, Aryani AA, et al. Acute phase proteins, inter- leukin 6, and heat shock protein 70 in broiler chickens administered with corticosterone. Poult Sci. 2014; 93: 3112–3118. doi: 10.3382/ps.2014-04099 PMID: 25306460 54. Zhang WH, Gao F, Zhu QF, Li C, Jiang Y, Dai SF, et al. Dietary sodium butyrate alleviates the oxidative stress induced by corticosterone exposure and improves meat quality in broiler chickens. Poult Sci. 2011; 90: 2592–2599. doi: 10.3382/ps.2011-01446 PMID: 22010246 55. Lin H, Sui SJ, Jiao HC, Buyse J, Decuypere E. Impaired development of broiler chickens by stress mim- icked by corticosterone exposure. Comp Biochem Physiol A Mol Integr Physiol. 2006; 143: 400–405. doi: 10.1016/j.cbpa.2005.12.030 PMID: 16458032 56. Hu XF, Guo YM, Huang BY, Zhang LB, Bun S, Liu D, et al. Effect of corticosterone administration on small intestinal weight and expression of small intestinal nutrient transporter mRNA of broiler chickens. Asian-Australasian J Anim Sci. Asian-Australasian Association of Animal Production Societies; 2010; 23: 175–181. 57. Puvadolpirod S, Thaxton JP. Model of physiological stress in chickens 1. Response parameters. Poult Sci. 2000; 79: 363–369. PMID: 10735203 58. PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 References Post J, Rebel JMJ, ter Huurne AAHM. Physiological effects of elevated plasma corticosterone concen- trations in broiler chickens. An alternative means by which to assess the physiological effects of stress. Poult Sci. 2003; 82: 1313–1318. PMID: 12943303 23 / 25 PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 Corticosterone and comprehensive responses in broiler chickens 59. Wang S, Ni Y, Guo F, Fu W, Grossmann R, Zhao R. Effect of corticosterone on growth and welfare of broiler chickens showing long or short tonic immobility. Comp Biochem Physiol—A Mol Integr Physiol. 2013; 164: 537–543. doi: 10.1016/j.cbpa.2012.12.014 PMID: 23266338 60. Khansari DN, Murgo AJ, Faith RE. Effects of stress on the immune system. Immunol Today. 1990; 11: 170–175. PMID: 2186751 61. Dohms JE, Metz A. Stress—mechanisms of immunosuppression. Vet Immunol Immunopathol. 1991; 30: 89–109. Available: http://www.ncbi.nlm.nih.gov/pubmed/1781159 PMID: 1781159 62. Husband AJ. Role of central nervous system and behaviour in the immune response. Vaccine. 1993; 11: 805–816. PMID: 8356843 63. Compton MM, Gibbs PS, Johnson LR. Glucocorticoid activation of deoxyribonucleic acid degradation in bursal lymphocytes. Poult Sci. 1990; 69: 1292–8. Available: http://www.ncbi.nlm.nih.gov/pubmed/ 2235845 PMID: 2235845 64. Compton MM, Johnson LR, Gibbs PS. Activation of thymocyte deoxyribonucleic acid degradation by endogenous glucocorticoids. Poult Sci. 1991; 70: 521–529. Available: http://www.ncbi.nlm.nih.gov/ pubmed/1646445 PMID: 1646445 65. Reichardt HM, Umland T, Bauer A, Kretz O, Schu¨tz G. Mice with an increased glucocorticoid receptor gene dosage show enhanced resistance to stress and endotoxic shock. Mol Cell Biol. 2000; 20: 9009– 9017. Available: http://www.ncbi.nlm.nih.gov/pubmed/11073999 PMID: 11073999 66. Pazirandeh A, Xue Y, Prestegaard T, Jondal M, Okret S. Effects of altered glucocorticoid sensitivity in the T cell lineage on thymocyte and T cell homeostasis. FASEB J. 2002; 16: 727–729. doi: 10.1096/fj. 01-0891fje PMID: 11923224 67. Schaumburg BP, Crone M. Binding of corticosterone by thymus cells, bursa cells and blood lympho- cytes from the chicken. Biochim Biophys Acta. 1971; 237: 494–501. Available: http://www.ncbi.nlm.nih. gov/pubmed/5118633 PMID: 5118633 68. Miller AH, Spencer RL, Stein M, McEwen BS. Adrenal steroid receptor binding in spleen and thymus after stress or dexamethasone. Am J Physiol. 1990; 259: E405–12. Available: http://www.ncbi.nlm.nih. gov/pubmed/2399976 PMID: 2399976 69. Ranelletti FO, Maggiano N, Aiello FB, Carbone A, Larocca LM, Musiani P, et al. Glucocorticoid recep- tors and corticosensitivity of human thymocytes at discrete stages of intrathymic differentiation. J Immu- nol. 1987; 138: 440–445. Available: http://www.ncbi.nlm.nih.gov/pubmed/3098842 PMID: 3098842 70. Shini S, Kaiser P, Shini A, Bryden WL. References Differential alterations in ultrastructural morphology of chicken heterophils and lymphocytes induced by corticosterone and lipopolysaccharide. Vet Immunol Immuno- pathol. 2008; 122: 83–93. doi: 10.1016/j.vetimm.2007.10.009 PMID: 18045696 71. Shini S, Shini A, Kaiser P. Cytokine and chemokine gene expression profiles in heterophils from chick- ens treated with corticosterone. Stress. Taylor & Francis; 2010; 13: 185–194. 72. Cohen JJ. Thymus-derived lymphocytes sequestered in the bone marrow of hydrocortisone-treated mice. J Immunol. 1972; 108: 841–844. Available: http://www.ncbi.nlm.nih.gov/pubmed/4551854 PMID: 4551854 73. Leshchinsky T V, Klasing KC. Divergence of the inflammatory response in two types of chickens. Dev Comp Immunol. 2001; 25: 629–38. Available: http://www.ncbi.nlm.nih.gov/pubmed/11472784 PMID: 11472784 74. Bourgeon S, Kauffmann M, Geiger S, Raclot T, Robin J-P. Relationships between metabolic status, cor- ticosterone secretion and maintenance of innate and adaptive humoral immunities in fasted re-fed mal- lards. J Exp Biol. 2010; 213: 3810–3818. doi: 10.1242/jeb.045484 PMID: 21037060 75. Freeman BA, Crapo JD. Biology of disease: free radicals and tissue injury. Lab Invest. 1982; 47: 412– 426. Available: http://www.ncbi.nlm.nih.gov/pubmed/6290784 PMID: 6290784 76. Wang X, Jia Q, Xiao J, Jiao H, Lin H. Glucocorticoids retard skeletal muscle development and myoblast protein synthesis through a mechanistic target of rapamycin (mTOR)-signaling pathway in broilers (Gal- lus gallus domesticus). Stress. 2015; 18: 686–698. doi: 10.3109/10253890.2015.1083551 PMID: 26371871 77. Siegel HS, Van Kampen M. Energy relationships in growing chickens given daily injections of corticoste- rone. Br Poult Sci. 1984; 25: 477–485. doi: 10.1080/00071668408454889 PMID: 6518401 78. Pardue SL, Thaxton JP, Brake J. Role of ascorbic acid in chicks exposed to high environmental temper- ature. J Appl Physiol. American Physiological Society; 1985; 58: 1511–1516. Available: http://www. ncbi.nlm.nih.gov/pubmed/3997715 79. Mathivanan R, Edwin SC. Hematological and serum biochemical parameters of broilers fed with Andro- graphis paniculata as an alternative to antibiotic growth promoter. J Med Plants Res. 2012; 6: 5647– 5650. 24 / 25 PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 Corticosterone and comprehensive responses in broiler chickens 80. Conlon M, Kita K. Muscle protein synthesis rate is altered in response to a single injection of insulin-like growth factor-I in seven-day-old Leghorn chicks. Poult Sci. 2002; 81: 1543–1547. PMID: 12412921 81. Rosebrough R., McMurtry J. Methimzole and thyroid hormone replacement in broilers. Domest Anim Endocrinol. 2003; 24: 231–242. PMID: 12642163 82. Beccavin C, Chevalier B, Cogburn LA, Simon J, Duclos MJ. Insulin-like growth factors and body growth in chickens divergently selected for high or low growth rate. J Endocrinol. 2001; 168: 297–306. PLOS ONE \| DOI:10.1371/journal.pone.0172684 February 24, 2017 References PMID: 11182767 83. Frystyk J, Delhanty PJ, Skjaerbaek C, Baxter RC, Baxter RC, Baxter RC, et al. Changes in the circulat- ing IGF system during short-term fasting and refeeding in rats. Am J Physiol. 1999; 277: E245–252. PMID: 10444419 84. Smith LK, Cidlowski JA. Chapter 1 –Glucocorticoid-Induced Apoptosis of Healthy and Malignant Lym- phocytes. In: Martini L, editor. Progress in Brain Research. 2010. pp. 1–30. 85. Collier SD, Wu W-J, Pruett SB. Endogenous Glucocorticoids Induced by a Chemical Stressor (Ethanol) Cause Apoptosis in the Spleen in B6C3F1 Female Mice. Toxicol Appl Pharmacol. 1998; 148: 176–182. doi: 10.1006/taap.1997.8324 PMID: 9465277 86. Kanduc D, Mittelman A, Serpico R, Sinigaglia E, Sinha AA, Natale C, et al. Cell death: apoptosis versus necrosis (review). Int J Oncol. 2002; 21: 165–170. PMID: 12063564 87. Renvoize´ C, Biola A, Pallardy M, Bre´ard J. Apoptosis: Identification of dying cells. Cell Biol Toxicol. 1998; 14: 111–120. PMID: 9553722 88. Schmidt S, Rainer J, Ploner C, Presul E, Riml S, Kofler R. Glucocorticoid-induced apoptosis and gluco- corticoid resistance: Molecular mechanisms and clinical relevance. Cell Death Differ. 2004; 11: S45– S55. PMID: 15243581 25 / 25
https://openalex.org/W4283322823	https://rua.ua.es/dspace/bitstream/10045/124520/1/Ferrer-Aracil_etal_2022_EducSci.pdf	English	null	Implementation of Service-Learning as a Strategy to Foster Intercultural Coexistence in the Local Community: A Case Study	Education sciences	2,022	cc-by	11,678	Javier Ferrer-Aracil, Víctor M. Giménez-Bertomeu and Elena M. Cortés-Florín * Department of Social Work and Social Services, University of Alicante, 03690 San Vicente del Raspeig, Alicante, Spain; javier.ferreraracil@ua.es (J.F.-A.); victor.gimenez@ua.es (V.M.G.-B.) * Correspondence: em.cortes@ua.es Abstract: Service-learning (SL) is a participatory teaching–learning methodology through which students learn certain content while also meeting a number of real social needs in their environment. The implementation of SL in different areas and educational stages has been extensively described. However, its potential as a community and intercultural development strategy at the local level has not been widely studied. Through a documentary analysis, the present work sought to understand the characteristics of socio-educational interventions, which, based on the service-learning method- ology, aim at improving coexistence in local communities with a high degree of cultural diversity. A total of 18 projects were included in a community programme implemented in the municipality of Elche (Spain) between 2010 and 2016. They all focused or included the promotion of intercultural coexistence among their objectives. The study design was quantitative, with a descriptive and ex- planatory, univariate and bivariate analysis using the IBM Statistical Package for Social Sciences 26. The results showed that service-learning can contribute to the improvement of intercultural coex- istence. Moreover, a number of SL basic, pedagogical and organisational components are enhanced when integrated into broader community development processes. Citation: Ferrer-Aracil, J.; Giménez-Bertomeu, V.M.; Cortés-Florín, E.M. Implementation of Service-Learning as a Strategy to Foster Intercultural Coexistence in the Local Community: A Case Study. Educ. Sci. 2022, 12, 426. https://doi.org/10.3390/ educsci12070426 Academic Editor: James Albright Received: 30 April 2022 Accepted: 20 June 2022 Published: 22 June 2022 Keywords: service-learning; intercultural coexistence; community education; local development; evaluation Article Javier Ferrer-Aracil, Víctor M. Giménez-Bertomeu and Elena M. Cortés-Florín * 1. Introduction 1.1. Cultural Diversity and Coexistence at School Citation: Ferrer-Aracil, J.; Giménez-Bertomeu, V.M.; Cortés-Florín, E.M. Implementation of Service-Learning as a Strategy to Foster Intercultural Coexistence in the Local Community: A Case Study. Educ. Sci. 2022, 12, 426. https://doi.org/10.3390/ educsci12070426 Academic Editor: James Albright Received: 30 April 2022 Accepted: 20 June 2022 Published: 22 June 2022 1.1. Cultural Diversity and Coexistence at School According to Regueiro and Pérez [1], Spain has always been characterised by its cul- tural diversity, shaped throughout history by the multiple contributions of the peoples who have occupied the land. Today, this diversity is reflected in the different languages, traditions, ways, customs and many other cultural expressions that have been reshaped thanks to the distinct creative capacity of human beings. Moreover, this heterogeneity has expanded further with the incorporation and settlement of immigrant populations of for- eign origin. Publisher’s Note: MDPI stays neu- tral with regard to jurisdictional claims in published maps and institu- tional affiliations. Publisher’s Note: MDPI stays neu- tral with regard to jurisdictional claims in published maps and institu- tional affiliations. g g According to data from the National Institute of Statistics [2], over the last two dec- ades, the number of immigrants in Spain has risen from 923,879 in 2000 to 5,434,153 in 2021. That is, some 11.4% of Spain’s total population is immigrant. This increase in the immigrant population of foreign origin, together with the presence of indigenous cultural minorities such as the Roma people [3], has played a key role in configuring the current policies of diversity management in the education system. The system is facing—and has faced—the challenge of teaching–learning in a context of new multicultural realities [4]. Copyright: © 2022 by the authors. Li- censee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and con- ditions of the Creative Commons At- tribution (CC BY) license (https://cre- ativecommons.org/licenses/by/4.0/). www.mdpi.com/journal/education Educ. Sci. 2022, 12, 426. https://doi.org/10.3390/educsci12070426 Educ. Sci. 2022, 12, 426 2 of 18 2 of 18 Cultural diversity, however, is not the only challenge that the education system, in particular, and Spanish and European society, in general, need to rise up to. So is individ- ualism. Both can lead to social inequality if not adequately addressed. According to the latest editions of the European Values Survey, a notable part of the population approaches their life project, social relations and participation in the public sphere from an individualistic, self-suf- ficient and private profit standpoint [5,6]. As Aristotle remarked, we are relational beings, but a fragmented and polarised economic, social and political context, such as the one we are in today, does not help us to understand how to relate properly, or indeed how to be able to relate. 1.1. Cultural Diversity and Coexistence at School Authors such as Fernández and López call relational illiteracy that “absence of the basic social skills necessary for adequate social interaction” [7] (p. 45). Giménez identified three modes of sociability that can be found in both a community and in an educational community: conviviality, coexistence and hostility [8]. Coexistence is the fact of sharing space and time without maintaining any significant relationships with one another. It represents the major tendency in Spanish areas with high levels of cultural diversity [9,10]. This author divides coexistence into nine dimensions that allow us to measure and achieve it [11]: (1) the relational dimension, that is, the existence and maintenance of relationships between people and/or groups in the community; (2) the attitudinal, i.e., the existence of attitudes of acceptance, inclusion and recognition of di- versity; (3) the normative, i.e., the existence of a normative framework of known and shared citizenship; (4) the axiological, that is, the existence of common values of respect, pluralism and solidarity; (5) the participatory, i.e., when individuals and/or groups are actively involved in the community regardless of their personal traits; (6) communication, referring to respectful messages between people and/or groups; (7) the conflictual, i.e., when mechanisms for the prevention, regulation or peaceful resolution of conflicts exist; (8) identity, referring to feelings of belonging, esteem and identification with the commu- nity; and, lastly (9), the political dimension, i.e., trust in public institutions and the equal enjoyment of rights. 1.2. School in the Community, and Vice Versa, and Service-Learning Methodology Community education is closely related to the concept of intercultural coexistence [12] and the well-known “learning to live together” phrase included in the classic Delors Report [13]. This type of education is the one implemented in the community, for the com- munity and with the community. The community represents the shared physical and re- lational realm in which we can collectively build a positive management of diversity [14– 16], although the purposes and methods may vary depending on the socio-political diver- sity management model [17]. Not only is the community educated in it, but it also actively educates through the organisation of its different agents, among which, school institutions are included [18]. Regarding the latter, and from this perspective, Sales and Moliner refer to schools that are included within territories, overcoming the division between the school and the community [19]. Community education is, above all, an act of socialisation, consisting of the weaving of social and intercultural bonds in order to identify and solve social needs and problems [20]. It encompasses practices such as educating cities, learning communities and service- learning experiences, among others [12]. Based on the results obtained in an action research experience in Queensland (Aus- tralia), Lathouras, Westoby and Shevellar argue that community education contributes to local development, integrating people in community actions related to structural change [21]. Educational processes are linked to the social, economic, cultural and political pro- cesses that take place in every community [22], with the community’s capacity for self- development constituting an educational objective in itself [23]. Service-learning (hereinafter, SL) is a participatory teaching–learning methodology through which students, in their role as active subjects, take part in a project and learn certain contents, competences and values while also improving their environment by meeting certain real social needs [24–26]. Teaching–learning through concrete projects has Educ. Sci. 2022, 12, 426 3 of 18 become popular in recent years for, among other reasons, the positive effects it has on students’ academic performance [27]. become popular in recent years for, among other reasons, the positive effects it has on students’ academic performance [27]. For Uruñuela [28], SL is: (a) a way of understanding citizenship based on its contri- bution to the betterment of society; (b) a way of understanding learning based on social responsibility; and (c) a way of understanding values-education based on experiential learning. 1.2. School in the Community, and Vice Versa, and Service-Learning Methodology This vision coincides with the benefits that community participation brings to personal and social development [29]. SL benefits not only those who receive the service, but also those who offer it, contributing to democratic health by facilitating the active par- ticipation of students—as citizens—in decisions that affect their living conditions [30]. From this perspective, the practice is one of participatory democratic education [31]. p p p p p y The application of SL in different educational settings and stages has already been exten- sively described in recent studies [32,33], together with its impacts on civic awareness [34], creativity [35], social bonding [36], participation [37], critical thinking [38] or sustainability [39], to provide but a few examples. According to Furco, SL improves educational perfor- mance and content learning, as well as participation, commitment and motivation to learn [40], while also helping to link theory and practice, reflection and action [41]. Although the potential of SL as a local/territorial development strategy has been less valued than its implementation in concrete and isolated experiences [42], SL is deemed to have been successfully integrated into a broader context of local/territorial policy [43], and even to have contributed to the implementation of global agendas, such as the Sustainable Development Goals (SDGs). For example, Batlle and Escoda systematised 100 experiences in Spain linked to the SDGs that involved a total of 300 schools and 430 social entities. They found that Goals 4 (quality education), 17 (partnerships to achieve the Goals), 10 (reducing inequalities) and 11 (sustainable communities and cities) were the most wide- spread in the case of students aged between 3 and 18 years [44]. Traver, Moliner and Sales performed a case study framed within an action–partici- patory study and found, among other results, that: (1) SL relates the interests of the par- ticipating actors, as well as those of the school curriculum with those of the territory; (2) SL turns the participants into educators and learners at the same time; (3) SL enables com- munity planning and evaluation (social participation, democratisation of decisions, school–community collaboration and shared knowledge); and (4) places students at the centre of educational action [45]. Despite its importance in the development of any educational project, one of the most problematic aspects of SL is its evaluation. 1.2. School in the Community, and Vice Versa, and Service-Learning Methodology Indeed, there is a certain tendency to overlook the evaluation of the processes, results and impacts of experiences, or to do so superficially [46], partly due to the complexity involved in assessing qualitative processes in quantifi- able terms [47]. Puig et al. [48], among other authors [49], have sought to rectify this by advancing a proposal to conduct SL evaluation regardless of the educational field and stage. This evaluation tool is structured into three dimensions, each based on a series of SL components: basic elements (social needs, service, sense of service and learning), ped- agogical elements (participation, group work, reflection, recognition and evaluation) and organisational elements (partnership, consolidation of schools and social entities). In turn, each component has four possible levels of development. 1.3. Purpose of the Study In accordance with the above, the general objective of this study was to understand the characteristics of socio-educational actions which, in accordance with the SL method- ology, aim at improving local coexistence in highly culturally diverse communities. We took into account the profiles of the institutions and the participants, the issues addressed and the pedagogical components of the experiences, exploring their relationships with variables linked to broader organisational and territorial development processes. Educ. Sci. 2022, 12, 426 4 of 18 2.1. Sample This work was a case study [50] on the use of SL in an intercultural community in- tervention programme implemented in the Carrús neighbourhood of the municipality of Elche (province of Alicante, Spain) between 2010 and 2016. The study was based on the documentary review of 65 documents (18 reports of activities and 47 min of meetings) that were part of 18 SL projects, which were launched with and/or were related to the inter- cultural community intervention programme. This programme was carried out simulta- neously in sixteen other Spanish territories, applying a common community methodology of study–diagnosis–planning–execution–evaluation. The main objective was to foster lo- cal coexistence through the promotion of intercultural relations; a sense of belonging; and the participation of government agents, professionals and the population in the improve- ment of neighbourhoods [51,52]. Within the project, various SL projects were developed, along with other community education initiatives to foster coexistence—such as open schools [53]—based on a territorial vision of education. The review included SL projects that took place in the Carrús neighbourhood in the primary or secondary stages (6–18 years) between 2010 and 2016. The inclusion criteria were that they had to be sufficiently documented and that they focused on or included the promotion of coexistence, or a similar term, among their objectives. Seven experiences that did not meet these criteria were excluded. 2. Development of the SL project: - Elements of the project: The basic, pedagogical and organisational elements evalu- ated were established according to the proposal of Puig et al. [48] and were measured using a 4-point Likert scale, in which 1 indicated the lowest level of development of the element and 4 the highest. Links between the project and the intercultural community programme: We believed it was important to contextualise each SL project within the framework of the wider community programme in which it took place. To this end, the following aspects were considered: o The project’s contribution to coexistence: The main coexistence dimensions eval- uated were established according to the proposal of Giménez [11]. They were measured using a 4-point Likert scale, from 1—Strongly disagree to 4—Strongly agree. g o Socio-educational intervention logic: Project based on social needs collected in the community study–diagnosis (analysis carried out by 197 people, between professionals from different institutions and neighbours of different origins and ages), a project integrated into the neighbourhood’s community programme. - Prior training and qualification: Realisation of an associated training seminar. - Dissemination of the project’s results: Dissemination of the results to the participants, results outreach in the rest of the community, media impact (TV, radio, press). As mentioned above, the data collection process was carried out through the docu- ment analysis [52] to systematize the variables of interest. The qualitative data were con- verted into quantitative data according to the process described. In the case of the ele- ments of the project, we used the available data from a questionnaire on the perception of the promoting institutions on the degree of development of the SL methodology in each project. In the case of the project’s contribution to coexistence, we used the data available from a questionnaire on the promoting institutions’ perceptions of the degree to which each SL project had contributed to establishing or developing the main coexistence di- mensions of interest. All data sources (activity reports and meeting minutes) were dated between 2010 and 2016. 2.2. Instrument and Procedure Quantitative and qualitative data from activity reports and meeting minutes were systematised through document analysis [54], and, in the case of qualitative data, trans- formed into quantitative data. Quantification is the process of assigning numerical values to data conceived as nonnumerical [55]. It was conducted to facilitate the recognition of patterns or, failing that, to extract meaning from qualitative data, allowing us to discern and show regularities or peculiarities in qualitative data that would not be visible other- wise [56–58]. The qualitative data were converted into quantitative data to put qualitative data into a form that could be subjected to statistical analysis together with other data that were already quantitative. The qualitative data were turned into quantitative data over several stages, according to the documentary analysis and quantification process described below: - Reading: The documents were first read in order to become familiar with the quali- tative data. - Categorisation: Secondly, the information was organised into significant analytical categories related to the purpose of the study. - Codification: Finally, analytical categories were transformed into codes and the SL experience data were collected into a data matrix. - Codification: Finally, analytical categories were transformed into codes and the SL experience data were collected into a data matrix. The resulting final variables used to analyse the SL projects are presented below. 1. SL project characteristics: The resulting final variables used to analyse the SL projects are presented below. 1. SL project characteristics: - Type of institution leading the project. - Type of institution participating in the project. - Profile of the people participating in the project (role in the projects, gender, cul- tural diversity, immigrants or members of cultural minorities). - Project theme (according to the SDGs): zero hunger; good health and well-being; quality education; gender equality; clean water and sanitation; affordable and clean energy; decent work and economic growth; industry, innovation and infrastructure; reduced inequalities; sustainable cities and communities; responsible consumption and production; climate action; life below water; life on land; peace, justice and strong institutions; and partnerships for the goals. g - Priority group. 5 of 18 5 of 18 Educ. Sci. 2022, 12, 426 2.3. Data Analysis The data analysis included a univariate and bivariate descriptive and explicative analysis. According to the literature on data analysis [59–63], the descriptive analysis was based on descriptive statistics used to summarise the projects’ characteristics, as well as on measures of central tendency (mean, median) and dispersion (standard deviation, range) used to describe the characteristics of the projects measured using ordinal variables (development of SL elements, contribution to the coexistence dimensions). The explana- tory analysis was based on correlation coefficients and differential group analysis. Spear- man’s rho coefficient was employed to evaluate the relationships between ordinal varia- bles. Given the sample size (n < 30) and the ordinal level of quantitative data, nonpara- metric tests were applied for group differential analyses: the Mann–Whitney U test for independent samples (k = 2) and the Kruskal–Wallis one-way ANOVA test (k > 2) (com- pleted with a Mann–Whitney U test and Bonferroni correction). All statistical analyses were performed using the IBM Statistical Package for Social Sciences 26 software. Educ. Sci. 2022, 12, 426 6 of 18 3.1. SL Project Characteristics 3.1. SL Project Characteristics Of the total number of projects analysed, three quarters (78%) were launched by sec- ondary education, baccalaureate and/or vocational training schools, while the rest (22%) were promoted by pre-primary and primary education schools, as shown in Table 1. Re- garding the total number of participating organisations, secondary education, baccalau- reate and/or vocational training schools (89%), as well as the Elche City Council, stood out. They participated in about 70% of the cases (67%), followed by early childhood and primary education schools (44%) and third sector entities (39%). Table 1. SL project organisations. n % Promoter organisation Early childhood and primary education schools 4 22.2 Secondary, baccalaureate and/or prof. training schools 14 77.8 Participant organisation (multiple choice) Merchants’ association 2 11.1 Rural development partnership 1 5.6 Parent association 5 27.8 Sports association 2 11.1 Neighbourhood association 3 16.7 City council 12 66.7 Early childhood and primary education schools 8 44.4 Secondary, baccalaureate and/or prof. training schools 16 88.9 Health centre 3 16.7 Sports centre 2 11.1 Police force 1 5.6 Third sector entity (social services) 7 38.9 Media 3 16.7 Small business 2 11.1 Animal protection 2 11.1 Residence for the elderly 1 5.6 University 2 11.1 Table 1. SL project organisations. A total of 398 people participated in the SL projects, among which, immigrants or those belonging to indigenous cultural minorities accounted for 15%. They were thus in the minority compared to the total (Table 2). In terms of roles, persons with significant cultural diversity ranged from 0% of public representatives to 16% of citizens. By gender, female citizens (17%) were somewhat more numerous than male citizens (16%). Table 2. SL project participants. n % Men Women Total Men Women Total Participants Politicians/public representatives 9 6 15 60.0 40.0 100.0 Professionals 28 54 82 34.1 65.9 100.0 Citizens 361 299 660 54.7 45.3 100.0 Total 398 359 757 52.6 47.4 100.0 Participants with cultural diversity (among all participants) Total 398 359 757 52.6 47.4 100.0 Participants with cultural diversity (among all participants) Educ. Sci. 2022, 12, 426 7 of 18 Politicians/public representatives 0 0 0 0.0 0.0 0.0 Professionals 4 5 9 14.3 9.3 11.0 Citizens 57 51 108 15.8 17.1 16.4 Total 61 56 117 15.3 15.6 15.5 Table 3 presents the main themes and priority groups of the SL projects. 3.1. SL Project Characteristics The topics were good health and well-being, quality education and/or gender equality, each of which were found in a third of the projects (33%). Around one in four projects focused on reduc- ing inequalities (28%) or sustainable cities and communities (22%). The rest of the topics were present in less than four projects of the total (responsible consumption and produc- tion; life on land; decent work and economic growth; peace, justice and strong institutions; and partnerships for the goals). No projects addressed the following themes: climate ac- tion; life under water; clean water and sanitation; affordable and clean energy; zero hun- ger; and industry, innovation and infrastructure. For their part, the priority groups were highly heterogeneous. Most projects targeted groups with specific needs associated with age—childhood, youth or the elderly (44%). In the same way, the main recipients of one in four projects were groups with specific needs—people with disabilities, migrants or homeless people (28%). Other parts of the projects targeted the general population (17%) or other target audiences (11%). Table 3. SL project characteristics: main theme and priority population group. Characteristic n % Project theme (multiple choice) Partnerships for the goals 1 5.6 Sustainable cities and communities 4 22.2 Quality education 6 33.3 Gender equality 6 33.3 Peace, justice and strong institutions 1 5.6 Responsible consumption and production 3 16.7 Reduced inequalities 5 27.8 Good health and well-being 6 33.3 Decent work and economic growth 2 11.1 Life on land 3 16.7 Priority population group General population 3 16.7 Groups with specific needs by age (childhood, youth, elderly) 8 44.4 Other groups with specific needs (with disabilities, migrants, homelessness) 5 27.8 Other 2 11.1 3 2 Development of SL Projects Table 3. SL project characteristics: main theme and priority population group. Priority population group 3.2. Development of SL Projects The level of development of the basic, pedagogical and organisational elements of the projects varied according to the aspect considered (Table 4): p j g p The aspects with the highest perceived degree of development were social needs, group work and participation. The aspects with the highest perceived degree of development were social needs, group work and participation. The aspects in which a high degree of development was perceived were recognition, sense of service and partnership. Educ. Sci. 2022, 12, 426 8 of 18 8 of 18 The aspects in which an average development was perceived were as follows: reflec- tion, consolidation of schools, service, evaluation and consolidation of entities. Regarding the links to the community programme the projects were part of, the de- gree of their contribution to the coexistence dimensions also varied according to the di- mension considered (Table 4): The dimensions in which a greater degree of contribution was perceived were the relational, attitudinal, participatory and identity dimensions. The dimensions in which a high contribution was perceived were the conflictual, communication, normative and axiological dimensions. The dimension in which an average contribution was perceived was the political di- mension. Table 4. Development of SL projects: development of SL elements and contribution to intercultural coexistence dimensions. Table 4. Development of SL projects: development of SL elements and contribution to intercultural coexistence dimensions. Development of SL elements 3.2. Development of SL Projects n Mean SD Median Range Development of SL elements Social needs 18 3.28 0.958 4.00 2 Service 18 1.83 0.985 1.50 3 Sense of service 18 2.67 0.907 2.50 3 Learning 18 2.78 0.647 3.00 2 Participation 18 3.00 0.594 3.00 2 Group work 18 3.17 0.707 3.00 2 Reflection 18 1.89 0.758 2.00 2 Recognition 18 2.83 1.150 2.50 3 Evaluation 18 1.78 0.647 2.00 2 Partnership 18 2.56 0.984 2.00 3 Consolidation of schools 18 1.89 0.832 2.00 2 Consolidation of entities 18 1.78 0.808 2.00 2 Contribution to intercultural coexistence dimensions Relational 16 3.31 0.479 3.00 1 Attitudinal 16 3.25 1.000 4.00 3 Normative 16 2.88 0.885 3.00 2 Axiological 16 2.75 0.931 2.00 2 Participatory 16 3.13 1.204 4.00 3 Communication 16 2.88 1.408 4.00 3 Conflictual 16 2.94 0.772 3.00 2 Identity 16 3.06 1.289 4.00 3 Political 16 1.94 0.443 2.00 2 T bl 5 h h t j t b d th d d t t d i th i hb Table 5 shows how most projects were based on the needs detected in the neighbour- hood’s community diagnosis (72%), and almost half were integrated into the broader planning of community actions (44%). However, in the latter case, it should be noted that more than one third of the projects (39%) lacked data in this regard in the documents analysed, which was partly due to the fact that they were developed before reaching the community programme phase in the overall programme. Finally, in two out of three experiences, training and training activities were carried out prior to the start of the project (67%). Educ. Sci. 2022, 12, 426 9 of 18 Table 5. Development of SL projects: intervention logic and previous training activities. Table 5. Development of SL projects: intervention logic and previous training activities. n % Intervention logic: based on the needs collected in the community diagnosis Yes 13 72.2 No 5 27.8 Intervention logic: integrated into community planning Yes 8 44.4 No 3 16.7 No data 7 38.9 Training activities prior to the start of the project Yes 12 66.7 No 6 33.3 Regarding the dissemination of the results of the SL projects, in most cases, the results were disseminated to a variable degree according to the recipients (Table 6): g g p Dissemination to project participants (89%). Dissemination to project participants (89%). 3.2. Development of SL Projects Dissemination to the community (67%), although this dissemination did not take place in one in three projects. Media impact (57%). However, the results were not disclosed in these media in a high percentage of projects (44%). Table 6. Dissemination of SL project results. Table 6. Dissemination of SL project results. n % Dissemination of the results to the participants Yes 16 88.9 No 2 11.1 Dissemination of the results to the community Yes 12 66.7 No 6 33.3 Media impact (TV, radio, press) Yes 10 55.6 No 8 44.4 Media impact (TV, radio, press) Media impact (TV, radio, press) Media impact (TV, radio, press) 3.3. Service-Learning and Intercultural Coexistence With Associated Previous Training (n= 12) No Associated Prior Training (n = 6) Mean SD Median Range Mean SD Median Range Development of SL elements Sense of service 3.00 0.853 3.00 2 2.00 0.632 2.00 2 Group work 3.42 0.669 3.50 2 2.67 0.516 3.00 1 Reflection 2.25 0.622 2.00 2 1.17 0.408 1.00 1 Evaluation 2.08 0.515 2.00 2 1.17 0.408 1.00 1 Partnership 3.00 0.853 3.00 2 1.67 0.516 2.00 1 Consolidation of centres 2.25 0.754 2.00 2 1.17 0.408 1.00 1 Intervention logic: Link to community diagnosis. The perceived level of development was significantly different depending on whether or not the projects were linked to the community diagnosis regarding the following SL elements: social needs (U = 12,000; p = 0.019), group work (U = 13,000; p = 0.036), reflection (U = 9500; p = 0.015), evaluation (U = 10,500; p = 0.015), partnership (U = 7500; p = 0.009) and consolidation of centres (U = 11,500; p = 0.028). The differences observed indicated that the perceived degree of development was significantly higher in projects linked to the community diagnosis than in projects without this link (Table 7). Table 7. Group differences in the development of SL elements and contribution to intercultural coexist- ence dimensions by intervention logic (based on the needs collected in the community diagnosis). Link to Community Diagnosis (n= 13) No Link to Community Diagnosis (n = 5) Mean SD Median Range Mean SD Median Range Development of SL elements Social needs 3.62 0.768 4.00 2 2.40 0.894 2.00 2 Group work 3.38 0.650 3.00 2 2.60 0.548 3.00 1 Reflection 2.15 0.689 2.00 2 1.20 0.447 1.00 1 Evaluation 2.00 0.577 2.00 2 1.20 0.447 1.00 1 Partnership 2.92 0.862 4.00 2 1.60 0.548 2.00 1 Consolidation of centres 2.15 0.801 2.00 2 1.20 0.477 1.00 1 Contribution to intercultural coexistence dimensions Communication 3.31 1.182 4.00 3 1.00 0.000 1.00 0 Identity 3.54 0.877 4.00 3 1.00 0.000 1.00 0 Political 2.08 0.277 2.00 1 1.33 0.577 1.00 1 Intervention logic: Integration into the community programme. The perceived level of development of social needs was significantly different depending on whether or not the projects were integrated into the neighbourhood’s community programme (U = 3000; p = 0.034). 3.3. Service-Learning and Intercultural Coexistence 3.3. Service-Learning and Intercultural Coexistence The level of development of the elements of the SL projects and the degree of contri- bution of the projects to the coexistence dimensions varied significantly depending on certain characteristics. First, the level of perceived development of the SL elements showed significant dif- ferences according to the following variables: Priority group: The perceived degree of development of the partnership was signifi- cantly different depending on the project’s priority group (H = 10,697; p = 0.013). Differ- ences were found between projects aimed at the general population and those aimed at groups with specific needs other than age (disability, migrants, homelessness). In the for- mer, the partnership’s perceived degree of development was significantly higher (mean = 3.67; DT = 0.577) (median = 4; range = 1) than in the second groups (mean = 1.60; DT = 0.548) (median = 2; Range = 1). Educ. Sci. 2022, 12, 426 10 of 18 Intervention logic: Link to community diagnosis. The perceived level of development was significantly different depending on whether or not the projects were linked to the community diagnosis regarding the following SL elements: social needs (U = 12,000; p = 0.019), group work (U = 13,000; p = 0.036), reflection (U = 9500; p = 0.015), evaluation (U = 10,500; p = 0.015), partnership (U = 7500; p = 0.009) and consolidation of centres (U = 11,500; p = 0.028). The differences observed indicated that the perceived degree of development was significantly higher in projects linked to the community diagnosis than in projects without this link (Table 7). Table 7. Group differences in the development of SL elements and contribution to intercultural coexist- ence dimensions by intervention logic (based on the needs collected in the community diagnosis). 3.3. Service-Learning and Intercultural Coexistence Link to Community Diagnosis (n= 13) No Link to Community Diagnosis (n = 5) Mean SD Median Range Mean SD Median Range Development of SL elements Social needs 3.62 0.768 4.00 2 2.40 0.894 2.00 2 Group work 3.38 0.650 3.00 2 2.60 0.548 3.00 1 Reflection 2.15 0.689 2.00 2 1.20 0.447 1.00 1 Evaluation 2.00 0.577 2.00 2 1.20 0.447 1.00 1 Partnership 2.92 0.862 4.00 2 1.60 0.548 2.00 1 Consolidation of centres 2.15 0.801 2.00 2 1.20 0.477 1.00 1 Contribution to intercultural coexistence dimensions Communication 3.31 1.182 4.00 3 1.00 0.000 1.00 0 Identity 3.54 0.877 4.00 3 1.00 0.000 1.00 0 Political 2.08 0.277 2.00 1 1.33 0.577 1.00 1 Intervention logic: Integration into the community programme. The perceived level of development of social needs was significantly different depending on whether or not the projects were integrated into the neighbourhood’s community programme (U = 3000; p = 0.034). The data showed that the perceived degree of development for this element was significantly higher in the projects integrated into community programmes (mean = 3.50; DT =.926) (median = 4; range = 2) than in projects where such integration did not occur (mean = 2.00; DT = 0.000) (median = 2; range = 0). Associated previous training: The perceived degree of development was significantly different depending on the existence or nonexistence of previous training associated with the following SL elements: sense of service (U = 14,000; p = 0.028), group work (U =15,000; p = 0.031), reflection (U = 7000; p = 0.004), evaluation (U = 8000; p = 0.003), partnership (U = 8000; p = 0.005) and consolidation of centres (U = 9500; p = 0.008). The results indicated that the perceived degree of development in projects with previous training was signifi- cantly higher than in those that did not execute such training (Table 8). Table 8. Group differences in the development of SL elements and contribution to intercultural co- existence dimensions by training activities prior to the start of the project. 3.3. Service-Learning and Intercultural Coexistence The empir- ical data showed that, in the cases of projects whose results were disseminated to partici- pants, the perceived level of development of this element was higher than in projects where this dissemination was not carried out (Table 9). However, this result should be taken with caution since there were only two projects in which the dissemination men- tioned above did not take place. Table 9. Group differences in development of SL elements and contribution to intercultural coexist- ence dimensions, by participant dissemination. Participant Dissemination (n = 16) No Participant Dissemination (n = 2) Mean SD Median Range Mean SD Median Range Development of SL elements Participation 3.13 .500 3.00 2 2.00 0.00 2.00 0 Contribution to intercultural coexistence dimensions Communication 3.14 1.292 4.00 3 1.00 0.000 1.00 0 Identity 3.36 1.082 4.00 3 1.00 0.000 1.00 0 Community information: Dissemination to the community. The perceived degree of development was significantly different depending on whether or not the project had dis- seminated the results to the community in relation to the following SL elements: service (U = 15,500; p = 0.038), sense of service (U = 14,000; p = 0.028), reflection (U = 14,000; p = 0.027), evaluation (U = 16,000; p = 0.035), and consolidation of entities (U = 13,000; p = 0.021). The differences observed indicated that the perceived degree of development with respect to these elements was significantly higher in cases of projects whose results were disseminated to the community than in those in which they were not (Table 10). Table 10. Group differences in the development of SL elements and contribution to intercultural coexistence dimensions by community dissemination. Community Dissemination to the (n= 12) No Community Dissemination (n = 6) Mean SD Median Range Mean SD Median Range Development of SL elements Service 2.17 1.030 2.00 3 1.17 0.408 1.00 1 Sense of service 3.00 0.853 3.00 2 2.00 0.632 2.00 2 Reflection 2.17 0.718 2.00 2 1.33 0.516 1.00 1 Evaluation 2.00 0.603 2.00 2 1.33 0.516 1.00 1 Consolidation of entities 2.08 0.793 2.00 2 1.17 0.408 1.00 1 Contribution to intercultural coexistence dimensions Communication 3.45 0.036 4.00 3 1.60 1.342 1.00 3 Identity 3.64 0.924 4.00 3 1.80 1.095 1.00 2 Community information: Media impact. 3.3. Service-Learning and Intercultural Coexistence The data showed that the perceived degree of development for this element was significantly higher in the projects integrated into community programmes (mean = 3.50; DT =.926) (median = 4; range = 2) than in projects where such integration did not occur (mean = 2.00; DT = 0.000) (median = 2; range = 0). Associated previous training: The perceived degree of development was significantly different depending on the existence or nonexistence of previous training associated with the following SL elements: sense of service (U = 14,000; p = 0.028), group work (U =15,000; p = 0.031), reflection (U = 7000; p = 0.004), evaluation (U = 8000; p = 0.003), partnership (U = 8000; p = 0.005) and consolidation of centres (U = 9500; p = 0.008). The results indicated that the perceived degree of development in projects with previous training was signifi- cantly higher than in those that did not execute such training (Table 8). Table 8. Group differences in the development of SL elements and contribution to intercultural co- existence dimensions by training activities prior to the start of the project. Table 8. Group differences in the development of SL elements and contribution to intercultural co- existence dimensions by training activities prior to the start of the project. With Associated Previous Training (n= 12) No Associated Prior Training (n = 6) Mean SD Median Range Mean SD Median Range Development of SL elements Sense of service 3.00 0.853 3.00 2 2.00 0.632 2.00 2 Group work 3.42 0.669 3.50 2 2.67 0.516 3.00 1 Reflection 2.25 0.622 2.00 2 1.17 0.408 1.00 1 Evaluation 2.08 0.515 2.00 2 1.17 0.408 1.00 1 Partnership 3.00 0.853 3.00 2 1.67 0.516 2.00 1 Consolidation of centres 2.25 0.754 2.00 2 1.17 0.408 1.00 1 With Associated Previous Training (n= 12) No Associated Prior Training (n = 6) Mean SD Median Range Mean SD Median Range Educ. Sci. 2022, 12, 426 11 of 18 Contribution to intercultural coexistence dimensions Participatory 3.42 1.165 4.00 3 2.25 0.957 2.50 2 Identity 3.50 0.905 4.00 3 1.75 1.500 1.00 3 Political 2.08 0.289 2.00 1 1.50 0.577 1.50 1 Community information: Participant dissemination. The perceived level of develop- ment of the Participation was significantly different depending on whether or not the pro- ject had disseminated the results among the participants (U = 1000; p = 0.012). Community information: Media impact. Finally, the perceived level of development of the recognition element was significantly different depending on whether the project 3.3. Service-Learning and Intercultural Coexistence Finally, the perceived level of development of the recognition element was significantly different depending on whether the project Contribution to intercultural coexistence dimensions Participatory 3.42 1.165 4.00 3 2.25 0.957 2.50 2 Identity 3.50 0.905 4.00 3 1.75 1.500 1.00 3 Political 2.08 0.289 2.00 1 1.50 0.577 1.50 1 Community information: Participant dissemination. The perceived level of develop- ment of the Participation was significantly different depending on whether or not the pro- ject had disseminated the results among the participants (U = 1000; p = 0.012). The empir- ical data showed that, in the cases of projects whose results were disseminated to partici- pants, the perceived level of development of this element was higher than in projects where this dissemination was not carried out (Table 9). However, this result should be taken with caution since there were only two projects in which the dissemination men- tioned above did not take place. Table 9. Group differences in development of SL elements and contribution to intercultural coexist- ence dimensions, by participant dissemination. Table 9. Group differences in development of SL elements and contribution to intercultural coexist- ence dimensions, by participant dissemination. Participant Dissemination (n = 16) No Participant Dissemination (n = 2) Mean SD Median Range Mean SD Median Range Development of SL elements Participation 3.13 .500 3.00 2 2.00 0.00 2.00 0 Contribution to intercultural coexistence dimensions Communication 3.14 1.292 4.00 3 1.00 0.000 1.00 0 Identity 3.36 1.082 4.00 3 1.00 0.000 1.00 0 Community information: Dissemination to the community. The perceived degree of development was significantly different depending on whether or not the project had dis- seminated the results to the community in relation to the following SL elements: service (U = 15,500; p = 0.038), sense of service (U = 14,000; p = 0.028), reflection (U = 14,000; p = 0.027), evaluation (U = 16,000; p = 0.035), and consolidation of entities (U = 13,000; p = 0.021). The differences observed indicated that the perceived degree of development with respect to these elements was significantly higher in cases of projects whose results were disseminated to the community than in those in which they were not (Table 10). Table 10. Group differences in the development of SL elements and contribution to intercultural coexistence dimensions by community dissemination. coexistence dimensions by community dissemination. 3.3. Service-Learning and Intercultural Coexistence Community Dissemination to the (n= 12) No Community Dissemination (n = 6) Mean SD Median Range Mean SD Median Range Development of SL elements Service 2.17 1.030 2.00 3 1.17 0.408 1.00 1 Sense of service 3.00 0.853 3.00 2 2.00 0.632 2.00 2 Reflection 2.17 0.718 2.00 2 1.33 0.516 1.00 1 Evaluation 2.00 0.603 2.00 2 1.33 0.516 1.00 1 Consolidation of entities 2.08 0.793 2.00 2 1.17 0.408 1.00 1 Contribution to intercultural coexistence dimensions Communication 3.45 0.036 4.00 3 1.60 1.342 1.00 3 Identity 3.64 0.924 4.00 3 1.80 1.095 1.00 2 Community information: Media impact. Finally, the perceived level of development of the recognition element was significantly different depending on whether the project Community information: Media impact. Finally, the perceived level of development of the recognition element was significantly different depending on whether the project Educ. Sci. 2022, 12, 426 12 of 18 12 of 18 had had any media impact or not (U = 3000; p = 0.000). The results showed that, in projects with media impact, the degree of perceived development of the recognition element was significantly higher (mean = 3.70; DT =.675) (median = 4; range = 2) than that of projects that did not have these impacts (mean = 1.75; DT = 0.463) (median = 2; range = 1). No significant differences were observed in the level of development perceived de- pending on the institution promoting the project. Second, the degree of contribution of the SL projects to the development of the coex- istence dimensions showed significant differences in some cases according to the follow- ing variables: Intervention logic: Link to community diagnosis. The degree of the perceived contri- bution of the projects was significantly different depending on whether or not the projects were linked to the community diagnosis in the communication (U= 3000; p = 0.013), iden- tity (U = 1500; p = 0.007) and political (U = 6000; p = 0.008) dimensions. In the projects linked to this diagnosis, their degree of perceived contribution to the dimensions mentioned above was significantly higher than in the projects without this link (Table 7). 3.3. Service-Learning and Intercultural Coexistence Previous associated training: A project’s degree of perceived contribution to the de- velopment of coexistence also varied significantly depending on whether or not the pro- ject was associated with previous training, in particular in the participation (U = 8000; p = 0.031), identity (U = 9500; p = 0.050) and political (U = 11,000; p = 0.021) dimensions. In all the dimensions mentioned, the degree of contribution received was higher if the project had previous associated training (Table 8). Community information: Participant dissemination. In the same way, the level of contribution of the projects to the communication (U = 3000; p = 0.050) and identity (U = 2000; p = 0.033) dimensions was significantly different depending on whether the project disseminated its results to the participants or not. In both dimensions, a greater degree of contribution was perceived in projects that conducted this dissemination compared to those that did not (Table 9). However, this result should be taken with caution since there were only two projects in which the dissemination did not take place. Community information: Dissemination to the community. Finally, there was also evidence of a significantly different degree of contribution to the development of the com- munication (U = 9000; p = 0.019) and identity dimensions depending on whether or not the project results were disseminated to the community. As for the previous variable, a greater degree of contribution was perceived in both dimensions in the case of the projects that performed this merger compared to those that did not (Table 10). No significant differences were observed in the degree of perceived contribution to the coexistence dimensions according to the promoting institution, the priority group, in- tegration into a community programme and media impact. 4. Discussion and Conclusions An essential objective when working with local communities with a high degree of cultural diversity is to establish encounters and exchanges between the people, groups and institutions that compose them and live in them [52]. In short, the aim is to establish an inclusive model of management for cultural diversity [51]. In this article, we described a case study of the use of service-learning, a participatory and educational methodology, to achieve this objective. j The projects were evaluated from a dual or, one could also say, triple perspective. On the one hand, we considered the structural characteristics of SL projects: the type of insti- tutions that promote them and participate in them, participant profiles, the topics ad- dressed and the targeted priority groups. On the other, we assessed them according to their dynamic components [48], that is, the basic, pedagogical and organisational elements that shape them. Finally, we analysed their contribution to the community’s intercultural coexistence [11]. Table 11 summarises group differences in the development of basic, ped- agogical and organisational SL components and the contributions to intercultural coexist- ence dimensions by each project characteristic; in other words, the SL elements and the Educ. Sci. 2022, 12, 426 13 of 18 13 of 18 intercultural coexistence dimensions in which significant statistical differences were ob- served according to different characteristics of the projects. intercultural coexistence dimensions in which significant statistical differences were ob- served according to different characteristics of the projects. Table 11. Summary of group differences in development of SL elements and contribution to inter- cultural coexistence dimensions by project characteristics. cultural coexistence dimensions by project characteristics. 4. Discussion and Conclusions It is worth noting that the projects analysed contemplated and promoted the partici- pation of three types of actors: (a) politicians/public representatives, (b) professionals and (c) citizens, especially when the local/territorial development process was understood to have resulted from—positive or negative—interactions among them [16], each according to their own roles and without confusing them [52]. Generally, the results of the projects under study indicated a medium-to-high per- ceived development of the set of basic, pedagogical and organisational elements proposed by Puig et al. to evaluate SL experiences [48]. Notable among the elements that obtained the highest scores was the identification of social needs by the participating students. The objective of this identification was to understand not only social situations that could be of concern to the people living and/or working in the community, but also the potential for educational action in terms of re- sources. SL projects must be based on the existing social reality and on the recognition of its weaknesses and strengths [24,42]. In this case, the students had access to the identifi- cation of social needs as well as to the elaboration and execution of the corresponding proposal of services to the community. This element was enhanced in cases where SL was incorporated into a broader intercultural community programme, through which local governments, professionals and citizens participated both in the analysis of realities and in the consequent improvement of actions. This result supports that of other works [42,45], which stressed the value of participants’ mobilisation and the shared construction of knowledge with respect to their own reality. Other SL elements with a significant impact were group work and participation. SL fostered the groups’ capacity of development and organisation in the face of the commu- nity’s social needs. Within the collective work process, participants coordinated their con- tributions in order to modify some aspect of the realities linked to health and well-being, quality education and gender equality, among other issues. These results were compatible with those of previous studies [24,28,30,32,37,40,45], in which participation was considered to somehow take shape based on group work in the different phases of SL development. p g p p p The perception of the development of group work differed significantly when it was associated with the intercultural community programme variables, specifically, the com- munity diagnosis. Group work perception increased when the SL was connected to the local community through this participatory research process. 4. Discussion and Conclusions Variables with Statistically Significant Differences According to Project Characteristics Project Characteristics Development of SL elements Social needs - Link to community diagnosis - Integration into community pro- gramme - Community dissemination Service - Community dissemination Sense of service - Associated pretraining - Community dissemination Learning --- Participation - Participant dissemination Group work - Link to community diagnosis - Associated pretraining Reflection - Link to community diagnosis - Associated pretraining - Community dissemination Recognition - Media impacts Evaluation - Link to community diagnosis - Associated pretraining - Community dissemination Partnership - Priority group - Link to community diagnosis - Associated pretraining Consolidation of centres - Link to community diagnosis - Associated pretraining Consolidation of entities - Community dissemination Contribution to intercultural coexistence dimensions Relational --- Attitudinal --- Normative --- Axiological --- Participatory - Associated pretraining Communication - Link to community diagnosis - Participant dissemination - Community dissemination Conflictual --- identity - Link to community diagnosis - Associated pretraining - Participant dissemination - Community dissemination Political - Link to community diagnosis - Integration into community pro- gramme - Associated pretraining Educ. Sci. 2022, 12, 426 14 of 18 14 of 18 The results showed that SL projects represented a meeting point for multiple and different institutions from various domains—not just educational ones. This outcome partly resembled what was observed in the practices systematised by Batlle and Escoda [44]. One implication is the need to assume that a community’s social, cultural, sports or health institutions adopt an educational role; at the same time, school institutions need to recognise the educational role of other institutions [42]. The involvement of formal schools is necessary but not sufficient to develop a genuine community education [12,20]. y p g y The involvement of the city councils in two out of three cases was a factor that re- flected the local/territorial orientation of the projects. The literature on community work in Spain widely recognises the central role of the local administration as a catalyst for the partic- ipation of the population and of all public and private technical resources in matters of general interest [7,16]. Other recognised factors include experiences such as that of Barcelona (Spain), in which the Diputació de Barcelona helps local authorities to implement SL initiatives through training, communication and technical support actions, among others [43]. 4. Discussion and Conclusions A similar situation applied to other elements: reflection, evaluation, partnership and the consolidation of schools. All of them improved their performance to a greater or lesser extent. This could be explained in part because true community participation–organisation—i.e., that which creates social networks of action and incorporates other local agents beyond the initial group—begins with community diagnosis—not when the diagnosis has already been elaborated [52]. Community diagnosis goes beyond a mere description of problematic social situations, as it highlights improvements to these situations, actively involving leaders, professionals Educ. Sci. 2022, 12, 426 15 of 18 15 of 18 and citizens in decision-making [52]. It represents a method of research and social dy- namization aimed at facilitating the collective processes of reflection, planning and action within communities [7,16,52]. The existence of prior SL methodology training led to significantly more favourable perceptions regarding the development of several basic, pedagogical and organisational project elements. Specifically, these elements were related to service, group work, reflec- tion, evaluation, partnerships and the consolidation of schools. Prior training is a neces- sary and important action, but it must be pursued through other training proposals that respond to specific demands and each community’s particular identity [42]. The latter also applies to project dissemination. In cases where the projects were dis- seminated, we observed a greater degree of perceived development with respect to the following elements: social needs, service, sense of service, participation, reflection, recog- nition, evaluation and consolidation of entities. Bär, Campo and Rubio have also defended the importance of making the work visible to the community [42]. As Marchioni et al. [52] argue, a local/territorial development process cannot take place without extensive and continuous information on the actions that sustain it since participation and sustainability are not possible without information. Finally, the perception of the contribution of SL projects to the development of some intercultural coexistence dimensions (identity, the political dimension, communication and participation) was significantly higher depending on some of the characteristics of these projects (link to the community diagnosis, integration into community program- ming, prior associated training, dissemination to participants and dissemination to the community). Thus, we can affirm that the SL projects analysed were conducive to pro- moting coexistence in communities with diversity. However, unlike the case described by Ochoa and Pérez [37], the coexistence here developed beyond the school boundaries. Author Contributions: Conceptualization, J.F.-A., V.M.G.-B. and E.M.C.-F.; data curation, J.F.-A., V.M.G.-B. and E.M.C.-F.; formal analysis, J.F.-A., V.M.G.-B. and E.M.C.-F.; funding acquisition, J.F.- A., V.M.G.-B. and E.M.C.-F.; investigation, J.F.-A., V.M.G.-B. and E.M.C.-F.; methodology, J.F.-A., V.M.G.-B. and E.M.C.-F.; project administration, J.F.-A., V.M.G.-B. and E.M.C.-F.; resources, J.F.-A., V.M.G.-B. and E.M.C.-F.; software, J.F.-A., V.M.G.-B. and E.M.C.-F.; supervision, J.F.-A., V.M.G.-B. and E.M.C.-F.; validation, J.F.-A., V.M.G.-B. and E.M.C.-F.; visualization, J.F.-A., V.M.G.-B. and E.M.C.-F.; writing—original draft, J.F.-A., V.M.G.-B. and E.M.C.-F.; writing—review & editing, J.F.-A., V.M.G.-B. and E.M.C.-F. All authors have read and agreed to the published version of the manuscript. 4. Discussion and Conclusions From the perspective of the promoting institutions, the projects fostered relationships among individuals, groups and institutions, enhancing a sense of identification with the diverse community. These data are significant because they indicate that SL had an impact on how community members related to each other. These findings support Essomba and Leiva [12], who maintain, in the same line as Giménez [8,11], that placing people from different backgrounds and/or cultural belongings within the same space–time dimension does not necessarily lead to intercultural coexistence. Other anthropological conditions are required: adequate relationships, attitudes of respect, shared norms and values, active participation, communication, conflict management, awareness of belonging and political action. Based on the results obtained, SL can help to facilitate these conditions. To finish, the study presented a number of limitations. First, the number of SL pro- jects analysed was small (n = 18). Second, all of the SL projects studied took place within the same geographical area (the Carrús neighbourhood in the municipality of Elche, Spain). The results cannot, therefore, be generalised. Moreover, it was based entirely on the analysis of secondary data. To obtain further information on how SL can locally contribute to intercul- tural coexistence, it would be necessary to broaden the sample of experiences, diversify the geographical locations and incorporate primary sources of information. Author Contributions: Conceptualization, J.F.-A., V.M.G.-B. and E.M.C.-F.; data curation, J.F.-A., V.M.G.-B. and E.M.C.-F.; formal analysis, J.F.-A., V.M.G.-B. and E.M.C.-F.; funding acquisition, J.F.- A., V.M.G.-B. and E.M.C.-F.; investigation, J.F.-A., V.M.G.-B. and E.M.C.-F.; methodology, J.F.-A., V.M.G.-B. and E.M.C.-F.; project administration, J.F.-A., V.M.G.-B. and E.M.C.-F.; resources, J.F.-A., V.M.G.-B. and E.M.C.-F.; software, J.F.-A., V.M.G.-B. and E.M.C.-F.; supervision, J.F.-A., V.M.G.-B. and E.M.C.-F.; validation, J.F.-A., V.M.G.-B. and E.M.C.-F.; visualization, J.F.-A., V.M.G.-B. and E.M.C.-F.; writing—original draft, J.F.-A., V.M.G.-B. and E.M.C.-F.; writing—review & editing, J.F.-A., V.M.G.-B. and E.M.C.-F. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Funding: This research received no external funding. Institutional Review Board Statement: Not applicable. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Informed Consent Statement: Not applicable. Informed Consent Statement: Not applicable. Educ. Sci. 2022, 12, 426 16 of 18 Data Availability Statement: The data presented in this study are available on request from the corresponding author. The data are not publicly available due to they are an ad hoc quantitative data matrix based on document analysis. 4. Discussion and Conclusions Acknowledgments: We wish to thank the people and institutions participating in the SL projects analysed, as well as the “la Caixa” Foundation, the main promoter of the intercultural community programme they were part of, for the support it provided to realise this work. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest. References Puntos De Vista 2005, 1, 7–31. 9. Giménez, C.; Lobera, J. Convivencia Social E Intercultural en Territorios de Alta Diversidad. Encuesta 2012 Sobre Convivencia Intercul- tural en el Ámbito Local; Obra Social “la Caixa”: Barcelona, Spain, 2014. p 10. Giménez, C.; Lobera, J.; Mora, T.; Roche, D. Convivencia Social E Intercultural en Territorios de Alta D Convivencia Intercultural en el Ámbito Local; Obra Social “la Caixa”: Barcelona, Spain, 2015. 11. Giménez, C. Promoviendo la convivencia ciudadana intercultural en barrios de alta diversidad. Ideas y experiencias para la praxis comunitaria. In Hagamos de Nuestro Barrio un Lugar Habitable. Manual de Intervención Comunitaria en Barrios; Buades, J., Giménez, C., Eds.; Tirant lo Blanch: Valencia, Spain, 2013; pp. 36–57, ISBN 978-84-15731-85-6. p pp 12. Essomba, M.A.; Leiva, L. Joining Forces to Live Together Keys to the Intercultural Community Interventi Social “la Caixa”: Barcelona, Spain, 2015. p 13. Delors, J. La Educación Encierra un Tesoro; UNESCO: Madrid, Spain, 1996; ISBN 9788402944974. 14. Pérez-Sindín, X. Definiendo y teorizando el significado de comunidad de lugar en la era de la globalización. RIPS Revista de Investigaciones Políticas y Sociológicas 2020, 19, 107–122. https://doi.org/10.15304/rips.19.2.6938. 15. Zúñiga, M. La comunidad del siglo XXI. Un marco interpretativo desde la perspectiva del Trabajo So Social 2020, 33, 197–207. https://doi.org/10.5209/cuts.64416. p g 16. Pastor-Seller, E. Trabajo Social con Comunidades: Teoría, Metodología y Prácticas; Universitas: Madrid, Spain, 2021; ISBN 9788479915476. 17. Raya-Diez, E.; Ezquerro, M.; Serrano-Martínez, C. Gestión de la Diversidad Cultural: Recursos y herramientas del Trabajo So- cial. Comunitania 2019, 18, 65–84. https://doi.org/10.5944/comunitania.18.4. p g 18. Ferrer-Aracil, J. Trabajo social con comunidades en el ámbito de la educación. In La escuela comunitaria: Bases Teóricas y Prácticas; Universidad de Alicante: Alicante, Spain, 2020; ISBN 978-84-9717-725-2. 19. Sales, A.; Moliner, O. (Eds.) La escuela incluida en el territorio. In La Transformación Educativa Desde la Participación Ciudadana; Octaedro: Barcelona, Spain, 2020; ISBN 9788418083549. p 20. Essomba, M.A. Educación comunitaria: Crear condiciones para la transformación educativa. Rizoma Freireano 2019, 27, 1–14. Available online: https://www.rizoma-freireano.org/articles-2727/educacion-comunitaria (accessed on 11 April 2022). 20. Essomba, M.A. Educación comunitaria: Crear condiciones para la transformación educativa. Rizoma Freireano 2019, 27, 1 14 Available online: https://www.rizoma-freireano.org/articles-2727/educacion-comunitaria (accessed on 11 April 2022). p g 21. Lathouras, A.; Westoby, P.; Shevellar, L. Reimagining and radicalizing community development pract land through popular education action research. Community Dev. J. 2021, 56, 283–299. https://doi.org/1 22. Smith, M.K. References 1. Regueiro, M.T.; Pérez, S. Gestión de la Diversidad Cultural en las Sociedades Contemporáneas; Tirant lo Blanch: Valencia, Spain, 2014; ISBN 978-84-9004-885-6. 2. Instituto Nacional de Estadística. Available Online: https://www.ine.es/ (accessed on 23 April 2022). 2. Instituto Nacional de Estadística. Available Online: https://www.ine.es/ (accessed on 23 April 2022). p p 3. Haz-Gómez, F.E.; Andreu, I.; Hernández, M.; Guerrero, C.; Romero, E.; Abellán, M.D. La investigación sobre comunidades gitanas en España en el contexto europeo: Un análisis documental. Revista de Estudios Socioeducativos 2019, 7, 117–136. https://doi.org/10.25267/rev_estud_socioeducativos.2019.i7.08. 4. Rubio, M.; Martínez, R.; Olmos, A. Formación universitaria, migraciones e interculturalidad en España: Una revisión de la oferta educativa de los estudios de Grado de Educación Infantil, Educación Primaria, Pedagogía y Educación Social. Revista de Socio- logía de la Educación 2019, 12, 337–350. https://doi.org/10.7203/RASE.12.2.14655. g p educativa de los estudios de Grado de Educación Infantil, Educación Primaria, Pedagogía y Educación Social. Revista de Socio- logía de la Educación 2019, 12, 337–350. https://doi.org/10.7203/RASE.12.2.14655. g p g 5. Elzo, J.; Silvestre, M. Un individualismo placentero y protegido. In Cuarta Encuesta Europea de Valores en su Aplicación a España; Universidad de Deusto: Bilbao, Spain, 2010; ISBN 978-84-9830-246-2. g p g 5. Elzo, J.; Silvestre, M. Un individualismo placentero y protegido. In Cuarta Encuesta Europea de Valores en su Aplicación a España; Universidad de Deusto: Bilbao, Spain, 2010; ISBN 978-84-9830-246-2. p 6. Silvestre, M. Valores en la era de la incertidumbre: Individualismos y solidaridades. In Quinta Encuesta Europea de Valores en su Aplicación a España; Catarata: Madrid, Spain, 2020; ISBN 978-84-1352-016-2. 6. Silvestre, M. Valores en la era de la incertidumbre: Individualismos y solidaridades. In Quinta Encuesta Europea de Valores en su Aplicación a España; Catarata: Madrid, Spain, 2020; ISBN 978-84-1352-016-2. p p p 7. Fernández, T.; López, A. Trabajo Social Comunitario: Afrontando Juntos los Desafíos del Siglo XXI; Alianza Editorial: Madrid, Spain, 2008. 8. Giménez, C. Convivencia. Conceptualización y sugerencias para la praxis. Puntos De Vista 2005, 1, 7–31. 7. Fernández, T.; López, A. Trabajo Social Comunitario: Afrontando Juntos los Desafíos del Siglo XXI; Alianza Editorial: Madrid, Spain, 2008. 8. Giménez, C. Convivencia. Conceptualización y sugerencias para la praxis. Puntos De Vista 2005, 1, 7–31. 7. Fernández, T.; López, A. Trabajo Social Comunitario: Afrontando Juntos los Desafíos del Siglo XXI; Alianza E G é C C C l ó l p j f f g p 8. Giménez, C. Convivencia. Conceptualización y sugerencias para la praxis. References Local Education: Community, Conversation, Praxis; Open University Press: Philadelphia, PA, USA, 1994; ISBN 0-335- 19274-2. 23. Ander-Egg, E. Metodología y Práctica del Desarrollo de la Comunidad. 1 ¿Qué es el Desarrollo de la Comunidad?; Lumen: Buenos Aires, Argentina, 2003; ISBN 950-724-830-7. g 24. Puig, J.M.; Batlle, R.; Bosch, C.; Palos, J. Aprendizaje Servicio: Educar para la Ciudadanía; Octaedro: Barcelona, Spain, 2007; ISBN 978 8480639019. 25. Bringle, R.G.; Hatcher, J.A.; Jones, S.G. International Service Learning: Conceptual Frameworks and Research; Stylus Publishing: Ster- ling, VA, USA, 2010; ISBN 978-1579223397. 17 of 18 Educ. Sci. 2022, 12, 426 17 of 18 26. Hatcher, J.A.; Bringle, R.G. Understanding Service-Learning and Community Engagement: Crossing Boundaries through Research; In- formation Age Pub.: Charlotte, NC, USA, 2011; ISBN 978 1617356575. g 27. Guo, P.; Saab, N.; Post, L.S.; Admiraal, W. A review of project-based learning in higher education: Student outcomes and measures. Int. J. Educ. Res. 2020, 102, 101586. https://doi.org/10.1016/j.ijer.2020.101586. p g j j 28. Uruñuela, P.M. Aprendizagem-trabalho: Uma metodologia para o desenvolvimento da convivência. AMAzônica 2011, 6, 35–61. 29. Miller, L.; McCauley, J.; Sanjay, M. Measuring the Impact of Community Engagement: Development of the BACE Scale. Int. J. Res. Serv. Learn. Community Engagem. 2018, 6, 1–14. https://doi.org/10.37333/001c.6982. 28. Uruñuela, P.M. Aprendizagem-trabalho: Uma metodologia para o desenvolvimento da convivência. A p g g p 29. Miller, L.; McCauley, J.; Sanjay, M. Measuring the Impact of Community Engagement: Development of the BACE Scale. Int. J. Res. Serv. Learn. Community Engagem. 2018, 6, 1–14. https://doi.org/10.37333/001c.6982. y g g p g 30. Lebovits, H.; Bharath, D.M. Service-learning as a tool to cultivate democratically minded students: A conceptual framework. J. Public Nonprofit Aff. 2019, 5, 277–292. https://doi.org/10.20899/jpna.5.3.277-292. 30. Lebovits, H.; Bharath, D.M. Service-learning as a tool to cultivate democratically minded st Public Nonprofit Aff. 2019, 5, 277–292. https://doi.org/10.20899/jpna.5.3.277-292. 31. Sant, E. Democratic Education: A Theoretical Review (2006–2017). Rev. Educ. Res. 2019, 89, 655–696. https://doi.org/10.3102/0034654319862493. 32. Redondo-Cordobado, P.; Sources, J.L. Research on Service-Learning in Spanish Scientific Production: A Systematic Review. Complut. J. Educ. 2020, 31, 69–83. https://doi.org/10.5209/rced.61836. p p g 33. Sotelino-Losada, A.; Arbués-Radigales, E.; García-Docampo, L.; González-Geraldo, J.L. Service-learnin and understanding from academic publication. Front. Educ. 2021, 6, 604825. https://doi.org/10.3389/fed p g Arbués-Radigales, E.; García-Docampo, L.; González-Geraldo, J.L. Service-learning in Europe. Dimensions g p p g 34. Billig, S.H. Service and Service-Learning in International Baccalaureate High Schools: An International Comparison of Out- comes and Moderators. Int. J. Res. Serv. Learn. References Community Engagem. 2017, 5, 57–83. https://doi.org/10.37333/001c.2975. 35. Desmet, O.A.; Roberts, A.M. Teaching for Positive and Transformational Creativity through Service Learning. Educ. Sci. 2022, 12, 234. https://doi.org/10.3390/educsci12040234. p g 36. Young, S.; Karme, T. Service learning in an Indigenous not-for-profit organization. Educ. Train. 2015, 57, 774–790. https://doi.org/10.1108/ET-04-2014-0041. p g 37. Ochoa, A.; Pérez, L.M. El aprendizaje servicio, una estrategia para impulsar la participación y mejorar la convivencia escolar. Psicoperspectivas 2019, 18, 1–13. https://doi.org/10.5027/psicoperspectivas-vol18-issue1-fulltext-1478. 38. Campbell, C.G.; Oswald, B.R. Promoting Critical Thinking Through Service Learning: A Home-Visiting Case Study. Teach. Psy- chol. 2018, 45, 193–199. https://doi.org/10.1177/0098628318762933. p g 39. Aboytes, J.G.R.; Barth, M. Transformative learning in the field of sustainability: A systematic litera J. Sustain. High. Educ. 2020, 21, 993–1013. https://doi.org/10.1108/IJSHE-05-2019-0168. 40. Furco, A. A Research Agenda for K-12 School-based Service-Learning: Academic Achievement and School Success. Int. J. Res. Serv. Learn. Community Engagem. 2013, 1, 11–22. Available online: https://journals.sfu.ca/iarslce/index.php/journal/arti- cle/view/51/16 (accessed on 12 April 2022). p 41. Rubio, L. ApS: Aterrizaje entre teoría y práctica. Aula de Innovación Educative 2011, 203–204, 3 http://hdl.handle.net/11162/87401 (accessed on 12 April 2022). 42. Bär, B.; Campo, L.; Rubio, L. Líneas de acción y principios para la incorporación del aprendizaje-servicio en el ámbito local. Trabajo en red en el territorio. Teoría de Educación Revista Interuniversitaria 2021, 33, 243–263. https://doi.org/10.14201/teri.23660. 42. Bär, B.; Campo, L.; Rubio, L. Líneas de acción y principios para la incorporación del aprendizaje-servicio en el ámbito local. Trabajo en red en el territorio. Teoría de Educación Revista Interuniversitaria 2021, 33, 243–263. https://doi.org/10.14201/teri.23660. 43. Maroto, S. La educación se mueve en el territorio. Estrategias locales de aprendizaje-servicio. Revista Iberoamericana de Aprendi- zaje-Servicio 2018, 6, 19–32. https://doi.org/10.1344/RIDAS2018.6.4. j p g 43. Maroto, S. La educación se mueve en el territorio. Estrategias locales de aprendizaje-servicio. Revista Iberoamericana de Aprendi- zaje-Servicio 2018, 6, 19–32. https://doi.org/10.1344/RIDAS2018.6.4. j p g 44. Batlle, R.; Escoda, E. 100 buenas prácticas de aprendizaje-servicio. In Inventario de Experiencias Educativas con Finalidad Social; Batlle, R., Escoda, E., Jesús Cuñado, M., García Laso, A., Martín, D.A., Prats, D., Eds.; Santillana: Madrid, Spain, 2019. 45. Traver, J.A.; Moliner, O.; Sales, A. Negociating curriculum: Learning-Service in included school. Alteridad 2019, 14, 195–206. https://doi.org/10.17163/alt.v14n2.2019.04. 46. León-Carrascosa, V.; Sánchez-Serrano, S.; Belando-Montoro, M.R. Diseño y validación de un cuestionario para evaluar la me- todología Aprendizaje-Servicio. Estudios sobre Educación 2020, 39, 247–266. https://doi.org/10.15581/004.39.247-266. 47. Puig, J.M.; Martín, X.; Rubio, L. ¿Cómo evaluar proyectos de aprendizaje servicio? g 27. Guo, P.; Saab, N.; Post, L.S.; Admiraal, W. A review of project-based learning in higher education measures. Int. J. Educ. Res. 2020, 102, 101586. https://doi.org/10.1016/j.ijer.2020.101586. 26. Hatcher, J.A.; Bringle, R.G. Understanding Service-Learning and Community Engagement: Crossing Boundaries through Research; In- formation Age Pub.: Charlotte, NC, USA, 2011; ISBN 978 1617356575. 57. Ryan, G.W.; Bernard, H.R. Data management and analysis methods. In Handbook of Qualitative Research; Denzin, N.K., Lincoln, Y.S., Eds.; Sage: Thousand Oaks, CA, USA, 2000; pp. 769–802. g pp 58. Sandelowski, M. Real qualitative researchers do not count: The use of numbers in qualitative researc 24, 230–240. https://doi.org/10.1002/nur.1025. y 60. Grinnell, R.M.; Unrau, Y.A. I Social Work Research and Evaluation: Quantitative and Qualitative Approaches; Oxford University Press: New York, NY, USA, 2005; ISBN 0-195179498. , , , ; 61. Rubin, A.; Babbie, E.R. Research Methods for Social Work, 8th ed.; Brooks/Cole: Belmont, CA, USA, 2014; y 63. Thyer, B. The Handbook of Social Work Research Methods, 2nd ed.; Sage: Thousand Oaks, CA, USA, 2010; ISBN 978-14-12958-40-0. p g 59. Fortune, A.E.; Reid, W.J. Research in Social Work, 3rd ed.; Columbia University Press: New York, NY, U 61. Rubin, A.; Babbie, E.R. Research Methods for Social Work, 8th ed.; Brooks/Cole: Belmont, CA, USA, 2014; ISBN 978-12-85173-46-7. 62. Siebert, C.F.; Siebert, D.C. Data Analysis with Small Samples and Non-Normal Data: Nonparametrics and Other Strategies; Oxford U i it P N Y k NY USA 2018 ISBN 978 0199391493 f 62. Siebert, C.F.; Siebert, D.C. Data Analysis with Small Samples and Non-Normal Data: Nonparametrics University Press: New York, NY, USA, 2018; ISBN 978-0199391493. y 63. Thyer, B. The Handbook of Social Work Research Methods, 2nd ed.; Sage: Thousand Oaks, CA, USA, 2010; 61. Rubin, A.; Babbie, E.R. Research Methods for Social Work, 8th ed.; Brooks/Cole: Belmont, CA, USA, 2014; ISBN 978-12-85173-46-7. 62 Siebert C F ; Siebert D C Data Analysis with Small Samples and Non Normal Data: Nonparametrics and Other Strategies; Oxford g pp , M. Real qualitative researchers do not count: The use of numbers in qualitative research. Res. Nurs. Heal https://doi.org/10.1002/nur.1025. References Voces de la Educació able online: https://hal.archives-ouvertes.fr/hal-02534364 (accessed on 20 April 2022). p ( p ) 48. Puig, J.M.; Martín, X.; Rubio, L.; Palos, J.; Gijón, M.; de la Cerda, M.; Graell, M. Rúbrica para la Autoevaluación y la Mejora de los Proyectos de ApS; Fundació Jaume Bofill: Barcelona, Spain, 2014. 49. Ruiz-Corbella, M.; García-Gutiérrez, J. (Eds.) Aprendizaje-Servicio: Los Retos de la Evaluación; Narcea Ediciones: Madrid, Spain, 2019; ISBN 978-84-277-2531-7. 50. Goode, W.J.; Hatt, P.K. Métodos de Investigación Social; Trillas: Madrid, Spain, 2004; ISBN 968-24-3836-5 51. Giménez, C.; Álamo, J.M.; Pérez, F. Joining Forces to Live Together Keys to the Intercultural Community Intervention Project (1: Living Together and Social Cohesion); Obra Social “la Caixa”: Barcelona, Spain, 2015. g p 52. Marchioni, M.; Morín, L.M.; Giménez, C.; Rubio, J.A. Joining Forces to Live Together Keys to the Intercultural Community Intervention Project (2: Methodology); Obra Social “la Caixa”: Barcelona, Spain, 2015. 53. Ferrer-Aracil, J.; Giménez-Bertomeu, V.M.; Cortés-Florín, E.M. La Escuela Abierta: Educación comunitaria para la convivencia ciudadana intercultural. In Construyendo Juntos Una Escuela para la Vida; Jiménez, A.S., Cáceres, J., Vergara, E., Pereira, M.R., Eds.; Dykinson: Madrid, Spain, 2021; pp. 205–212, ISBN 978-84-1377-352-0. y p pp 54. Bowen, G. Document Analysis as a Qualitative Research Method. Qual. Res. J. 2009, 9, 27–40. https://doi.org/10.3316/QRJ0902027. ual. Res. J. 2009, 9, 27–40. https://doi.org/10.3316/QRJ0902 54. Bowen, G. Document Analysis as a Qualitative Research Method. Qual. Res. J. 2009, 9, 27–40. https://doi.org 55. Sandelowski, M.; Voils, C.I.; Knafl, G. On Quantitizing. J. Mix. Methods Res. 2009, 3, 208–222. https://doi.org/1 55. Sandelowski, M.; Voils, C.I.; Knafl, G. On Quantitizing. J. Mix. Methods Res. 2009, 3, 208–222. https://doi.org/10.1177/1558689809334210. 56. Miles, M.B.; Huberman, A.M. Qualitative Data Analysis: An Expanded Sourcebook, 2nd ed.; Sage: Thousand Oaks, CA, USA, 1994; ISBN 0-8039-4653-8. 56. Miles, M.B.; Huberman, A.M. Qualitative Data Analysis: An Expanded Sourcebook, 2nd ed.; Sage: Thousand Oaks, CA, USA, 1994; ISBN 0-8039-4653-8. 18 of 18 Educ. Sci. 2022, 12, 426 18 of 18 g pp 58. Sandelowski, M. Real qualitative researchers do not count: The use of numbers in qualitative research. Res. Nurs. Health 2001, 24, 230–240. https://doi.org/10.1002/nur.1025. y 60. Grinnell, R.M.; Unrau, Y.A. I Social Work Research and Evaluation: Quantitative and Qualitative Approaches; Oxford University Press: New York, NY, USA, 2005; ISBN 0-195179498.
https://openalex.org/W2992969074	https://europepmc.org/articles/pmc6889359?pdf=render	English	null	Maternal and perinatal outcomes of hypertensive disorders of pregnancy in Ethiopia: systematic review and meta-analysis	BMC pregnancy and childbirth	2,019	cc-by	7,978	Mersha et al. BMC Pregnancy and Childbirth (2019) 19:458 https://doi.org/10.1186/s12884-019-2617-8 Mersha et al. BMC Pregnancy and Childbirth (2019) 19:458 https://doi.org/10.1186/s12884-019-2617-8 Maternal and perinatal outcomes of hypertensive disorders of pregnancy in Ethiopia: systematic review and meta- analysis Amanual Getnet Mersha1, Tadesse Melaku Abegaz2 and Mohammed Assen Seid2 Amanual Getnet Mersha1, Tadesse Melaku Abegaz2 and Mohammed Assen Seid2 © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Abstract Background: Hypertensive disorders of pregnancy complicate around 6% of pregnancies and accounts for 19% of maternal death in Ethiopia. The current review aimed to assess maternal and perinatal outcomes of pregnancies complicated by hypertension in Ethiopia. Methods: A systematic review and meta-analysis was done on the outcome of hypertensive disorder among pregnant women in Ethiopia. Literature search was made in five databases and Statistical analyses were carried out by using Stata 14 software. The pooled prevalence of maternal death, HELLP syndrome, perinatal death, and low birth weight was calculated using a random-effects model. Egger’s test and funnel plot were used to evaluate publication bias. The Cochran Q test and I2 test statistics were used to test the heterogeneity of studies. Result: Thirteen studies included in the review, with an overall sample size of 5894 women diagnosed to have hypertensive disorder of pregnancy. The pooled prevalence of maternal death was estimated to be 4% (95% CI: 2, 6%). The pooled prevalence of HELLP syndrome was 13% (95% CI: 10, 16%). Other complications such as pulmonary edema, kidney injury, hepatic injury, placental abruption, and aspiration pneumonia were also reported. Perinatal death was observed in one-fourth of women with HDP 25% (95% CI: 18, 32%). The pooled prevalence of low birth weight neonate in a woman with HDP is 37% (95% CI, 27, 48%). Conclusions: In Ethiopia, the prevalence of perinatal and maternal mortality among pregnant women with one of the hypertensive disorders were found to be higher than rates reported from high income as well as most of the low and middle income countries. For instance, one in four of pregnancies complicated by hypertensive disorder end up in perinatal death in Ethiopia. HELLP syndrome, placental abruption, pulmonary edema, renal damage, prematurity, perinatal asphyxia, and low birth weight were also commonly reported. To improve the health outcomes of hypertensive disorders of pregnancy, it is recommended to improve utilization of maternal health service; early detection and early referral of pregnant women with hypertensive disorder; advocating policies and strategies that improves the quality of health care that a pregnant woman and her newborn receive. Keywords: Ethiopia, Hypertensive disorders of pregnancy, Meta-analysis, Outcomes, Systematic review * Correspondence: amanuelget16@gmail.com p Full list of author information is available at the end of the article p g g 1Department of Gynecology and Obstetrics, School of Medicine, College of Medicine and Health Sciences, University of Gondar, P.O. Box: 196, Gondar, Ethiopia Maternal death The death of a woman while pregnant or within forty- two completed days of termination of pregnancy irre- spective of duration and site of pregnancy, from any cause related to or aggravated by the pregnancy or by its management but not due to accidental or incidental causes. Although hypertensive disorders of pregnancy are among the leading causes of maternal and perinatal deaths in Ethiopia, there is no pooled national evidence that dem- onstrates the feto-maternal outcomes of pregnancies com- plicated by the disorder. Therefore, the current review aimed to assess the maternal as well as perinatal outcomes of pregnancies complicated by hypertension in Ethiopia. Operational definitions of outcomes y g p g yp According to the 2016 Demographic and Health Survey (EDHS), the perinatal mortality rate in Ethiopia was 33 per 1000 pregnancies by the end of 2016 [9]. Perinatal mortality is three to five folds higher in women with pre- eclampsia/eclampsia syndrome as compared to those without the disorders [10, 11]. Different studies docu- mented high perinatal mortality rate among women with hypertensive disorders of pregnancy. For instance, the rate of perinatal death was found to be 317/1000 births in Ethiopia [29]; 230/1000 births in Pakistan [12]; and, 144/ 1000 births in Turkey [13]. A Southern Ethiopia study re- ported a considerable association of perinatal death with maternal death, antepartum occurrence of the disease, low birth weight, hepatic injury, earlier gestational age at diag- nosis, having eclampsia, and multipara [22]. The primary outcomes of interest included in the Meta- analysis are maternal death, HELLP syndrome, perinatal death, and low birth weight. The secondary outcomes included in the review are preterm delivery, perinatal as- phyxia, acute kidney injury, aspiration pneumonia, pul- monary edema, ANC service utilization, and placental abruption. Page 2 of 12 Mersha et al. BMC Pregnancy and Childbirth (2019) 19:458 Page 2 of 12 HELLP syndrome A syndrome consists of haemolysis (H), elevated liver enzymes (EL), and Low platelet count (LP). Eligibility criteria All b i l All observational studies that reported at least one of the maternal and/or fetal outcomes of hypertensive disor- ders of pregnancy and conducted in Ethiopia were in- cluded in the current review. Perinatal death Study design and search strategy Background disorders of pregnancy in Ethiopia. The studies were re- trieved through internet search from the databases of MEDLINE, Scopus, PubMed, ScienceDirect, and Google Scholar. A combination of keywords and phrases like: preeclampsia (Mesh), preeclamp(all fields), eclampsia (Mesh),eclamp(all fields), hypertensive disorders of pregnancy (Mesh), hypertensive disorders of pregnancy (all fields), fetal outcome (Mesh), fetal outcome (all fields), maternal outcome (Mesh), maternal outcome (all fields), gestational hypertension (Mesh),gestational hypertension (all fields), pregnancy induced hyperten- sion (all fields), and Ethiopia (all fields), were used to search articles in the databases. The reference lists of identified studies were also screened to recover other ar- ticles and one unpublished study was retrieved from Addis Ababa University electronic library. All published articles up to 21 September 2018 were included in the review. Globally, hypertensive disorders of pregnancy complicate 3–10% of all pregnancies and it is a major cause of mater- nal and perinatal complications [1]. A recent review re- ported that hypertensive disorders of pregnancy complicate around 6% of all pregnancies in Ethiopia [2]. Hypertensive disorders of pregnancy (HDP) accounts for 18% of maternal deaths worldwide, with an estimated number of about 62, 000–77, 000 deaths occur each year [3]. Due to the existing low level of health service utilization and poor quality of maternal and neonatal care, the maternal and perinatal morbidities are much higher in low and middle -income countries (LMICs) [4, 5]. For instance, 19% of all maternal deaths in Ethiopia are attributed to hypertensive disorders of pregnancy [6]. Hypertensive disorders of pregnancy were also reported to account for 30% of maternal mortality in Ghana [5]. Maternal complications of hypertensive disorders of preg- nancy include placental abruption, pulmonary edema, thrombocytopenia, hemolytic anemia, stroke, recurrent seizure, renal damage, hepatic injury and others [7]. HELLP syndrome comprises of the following: haemolysis, elevated liver enzymes, and low platelets. HELLP syndrome occurs in about 0.5 to 0.9% of all pregnancies and complicates 10 to 20% of women with severe preeclampsia. HELLP syn- drome is one of the common cause of maternal and fetal mortality among pregnant women with hypertension [8]. Quality assessment Q y All reviewers (AGM, MAS, TMA) independently assessed the quality of studies using strengthening the reporting of observational studies in epidemiology (STROBE) scale checklist quality assessment tool [17]. All of the included studies were assessed to have a qual- ity of > 70% and, there were no studies excluded based on quality assessment. Study design and search strategy The death of a fetus/neonate in the perinatal period (from age of viability or twenty-eight weeks of gestation in Ethiopian context to first six days after birth). A systemic review and meta-analysis was conducted to assess maternal and fetal outcomes of hypertensive Page 3 of 12 Mersha et al. BMC Pregnancy and Childbirth (2019) 19:458 Page 3 of 12 Studies identified A total of 127 articles were identified from five elec- tronic databases (MEDLINE, Scopus, PubMed, Science- Direct, and Google Scholar). Out of these identified articles: 38 articles found duplicated and were removed; 49 articles excluded after reviewing their title; 16 articles excluded after reviewing their abstracts; and 11 articles excluded after a full text review (did not report the out- come variables). Finally, thirteen studies were included in the systematic review and meta-analysis (Fig. 1). Perinatal asphyxia Perinatal asphyxia A neonatal condition defied by five minute APGAR score of less than seven. Data extraction All of the research articles that were identified from searches of the electronic databases were imported into the ENDNOTE software version X5 (Tomson Reuters, USA) and duplicates were removed. Two authors (AGM and TMA) screened the titles and abstracts of identified articles by applying the inclusion criteria. Two authors (AGM and MAS) independently reviewed the full text. Final inclusion of the studies was determined by agree- ment of both reviewers and when there is disagreement, a third author (TMA) was involved. All the authors were involved in the discussion and agreed on the final inclu- sion. Before data extraction had begun, full-length arti- cles of the selected studies were read to confirm for fulfilling the inclusion criteria. Then, data extraction was performed by two reviewers (AGM and TMA) inde- pendently. The selected studies were reviewed to extract data like; year of publication; author(s); study design; sample size; maternal outcomes; fetal outcomes; type of HDP; period of occurrence of the HDP; gestational age at the time of diagnosis; and antenatal care visit. When there was a disagreement in data extraction between the reviewers, it was resolved through discussion and mutual agreement between the investigators. Socio-demographic, ANC service utilization, and clinical characteristics of study participants Overall, thirteen studies with a total sample size of 5894 pregnant women were found to have one of the HDPs and included in the review [20–32]. A maximum of 52.5% in a study conducted in Jimma specialized hospital [20] to a minimum of 3.5% among women admitted at teaching hospitals in Addis Ababa [24] were not having any antenatal care visit. Furthermore, 52% of women from another study [26]; 22.2% from a study in Amhara region [28]; and, 7.5% in Somalia regional state of Ethiopia [21] had no antenatal care visit. For other components of socio-demographic and clinical Description of the studies Out of thirteen studies as demonstrated in table one; eleven studies were cross-sectional studies [20, 21, 23– 26, 28–32] while two studies followed a retrospective co- hort study design [22, 27]. Four studies conducted in Southern Nations Nationalities and People (SNNPR) [22, 23, 27, 30]; four studies were conducted in Addis Ababa (capital city of Ethiopia) [24, 25, 31, 32]; three studies were from Oromia regional state [20, 26, 29]; one study in Amhara region [28]; and one study conducted in So- mali regional state of Ethiopia [21]. The sample size of studies ranges from a minimum of 93 women [21] to a maximum of 1809 women [25]. All of the included stud- ies defined hypertensive disorders of pregnancy as a Sys- tolic blood pressure (SBP) of 140 mmHg or higher and/ or diastolic blood pressure (DBP) of 90 mmHg or higher on two or more consecutive occasions during pregnancy. Low birth weight Low birth weight Birth weight of less than 2500 g. Birth weight of less than 2500 g. Preterm delivery probably a few studies are contributing to the final re- sult. Random-effects model for estimating pooled effects was employed due to the high level of observed hetero- geneity and was measured as proportions of outcomes with 95% confidence intervals (CIs). Egger’s regression asymmetry test with p-value < 0.05 used as a cutoff to declare presence of statistically significant publication bias. The detail description of the original studies was presented in a table and forest plot. Birth of baby after age of viability or twenty-eight weeks of gestation in Ethiopian context but before thirty-seven completed weeks of gestation. Statistical analysis and heterogeneity Seven studies re- ported the prevalence of HELLP syndrome [20, 21, 24– 26, 29, 31]. As demonstrated in the forest plotthe preva- lence of HELLP syndrome was estimated to be 13% [0.13 (0.10–0.16), I2 = 72.2%, p < 0.01] (Fig. 3). Publica- tion bias was checked by using the Egger’s test that showed non-significant publication bias, p-value > 0.428 (Fig. 6). Acute kidney injury was reported by six of the included studies [20, 21, 23, 24, 26, 31]: the maximum rate of acute kidney injury was reported from a study conducted in Jimma specialized hospital (24.2%) [20]; and, the minimum rate was reported from a study conducted in Addis Ababa (6.5%) [24]. Placental abruption complicates as high as 15.3% of women with HDP in Addis Ababa [25]; and, as low as 1.3% in Southern Nations and Nationalities of Ethiopia [30]. The highest rate of pulmonary edema was reported from one study conducted in Jimma specialized hospital (20.1%) [20]; followed by a study from Addis Ababa (17.5%) [25]. Aspiration pneumonia complicates 17.5% of women with the diagnosis of HDP in one study [25]. Fig. 1 Flow diagram of the studies included in the review Fig. 1 Flow diagram of the studies included in the review Fig. 1 Flow diagram of the studies included in the review Maternal outcomes of HDP characteristics of women included in the review there were no adequate data to generate results. Among thirteen studies included in the review, nine studies reported rate of maternal death in women diag- nosed to have HDP [20–25, 27, 28, 32]. As illustrated in the forest plot, the overall all rate of maternal death in Ethiopian women with HDP was estimated to be 4% [0.04 (0.02–0.06), I2 = 94.64%, P < 0.01] (Fig. 2). The test of publication bias using the Egger’s test was non- significant, p-value > 0.092 (Fig. 6). Seven studies re- ported the prevalence of HELLP syndrome [20, 21, 24– 26, 29, 31]. As demonstrated in the forest plotthe preva- lence of HELLP syndrome was estimated to be 13% [0.13 (0.10–0.16), I2 = 72.2%, p < 0.01] (Fig. 3). Publica- tion bias was checked by using the Egger’s test that showed non-significant publication bias, p-value > 0.428 (Fig. 6). Statistical analysis and heterogeneity Acute kidney injury was reported by six of the included studies [20, 21, 23, 24, 26, 31]: the maximum rate of acute kidney injury was reported from a study conducted in Jimma specialized hospital (24.2%) [20]; and, the minimum rate was reported from a study conducted in Addis Ababa (6.5%) [24]. Placental abruption complicates as high as 15.3% of women with HDP in Addis Ababa [25]; and, as low as 1.3% in Southern Nations and Nationalities of Ethiopia [30]. The highest rate of pulmonary edema was reported from one study conducted in Jimma specialized hospital (20.1%) [20]; followed by a study from Addis Ababa (17.5%) [25]. Aspiration pneumonia complicates 17.5% of women with the diagnosis of HDP in one study [25]. Statistical analysis and heterogeneity Statistical analyses were carried out by using Stata 14 (Stata Corp LP, College Station, TX) software to esti- mate the pooled prevalence of selected maternal and perinatal outcomes [18]. Statistical heterogeneity be- tween studies was evaluated using the Cochran’s Q test and I2 statistic [19]. I2 value of greater than 75 demon- strates that heterogeneity among the studies is high and Mersha et al. BMC Pregnancy and Childbirth (2019) 19:458 Page 4 of 12 characteristics of women included in the review there were no adequate data to generate results. Hypertensive disorder of pregnancy (HDP) Among the five groups of HDP, preeclampsia – eclamp- sia syndrome was the most commonly reported type of HDP. For instance, preeclampsia accounts for 66.2% of mothers among women admitted with HDP in Jimma specialized hospital [20]; 72.9% among mother with HDP in a Southern Ethiopian study [23]; and, 82.7% in a study done in Addis Ababa [25]. Eclampsia was reported to account for 27.8% of women admitted for HDP in a study conducted at Amhara regional state [28]; 34.1% in Southern study [27]; and, 24.2% in a study from Addis Ababa study. A study conducted in Somalia regional state of Ethiopia reported that 61% of the women with HDP were diagnosed during the antepartum period; 28%were diagnosed during the intrapartum period; and the remaining 10.7% were diagnosed in the postpartum period [21]. Most convulsions started during the ante- partum period as compared to onsets during the intra- partum and postpartum periods. For instance, 38% of all convulsions in a study conducted at Oromia regional state started in the antepartum period; 18% occur during the intrapartum period; and the rest 44% convulsions started during the postpartum period [20]. Additionally, 86% of all the convulsions started during the antepartum period in a study conducted at teaching hospitals located at Addis Ababa [24] (Table 1). Maternal outcomes of HDP Among thirteen studies included in the review, nine studies reported rate of maternal death in women diag- nosed to have HDP [20–25, 27, 28, 32]. As illustrated in the forest plot, the overall all rate of maternal death in Ethiopian women with HDP was estimated to be 4% [0.04 (0.02–0.06), I2 = 94.64%, P < 0.01] (Fig. 2). The test of publication bias using the Egger’s test was non- significant, p-value > 0.092 (Fig. 6). Hypertensive disorder of pregnancy (HDP) yp p g y Among the five groups of HDP, preeclampsia – eclamp- sia syndrome was the most commonly reported type of HDP. For instance, preeclampsia accounts for 66.2% of mothers among women admitted with HDP in Jimma specialized hospital [20]; 72.9% among mother with HDP in a Southern Ethiopian study [23]; and, 82.7% in a study done in Addis Ababa [25]. Eclampsia was reported to account for 27.8% of women admitted for HDP in a study conducted at Amhara regional state [28]; 34.1% in Southern study [27]; and, 24.2% in a study from Addis Ababa study. A study conducted in Somalia regional state of Ethiopia reported that 61% of the women with HDP were diagnosed during the antepartum period; 28%were diagnosed during the intrapartum period; and the remaining 10.7% were diagnosed in the postpartum period [21]. Most convulsions started during the ante- partum period as compared to onsets during the intra- partum and postpartum periods. For instance, 38% of all convulsions in a study conducted at Oromia regional state started in the antepartum period; 18% occur during the intrapartum period; and the rest 44% convulsions started during the postpartum period [20]. Additionally, 86% of all the convulsions started during the antepartum period in a study conducted at teaching hospitals located at Addis Ababa [24] (Table 1). Page 5 of 12 Mersha et al. BMC Pregnancy and Childbirth (2019) 19:458 Table 1 Overview of studies included in the systematic review and Meta-analysis Sources Location Study design Quality (%) Sample size No ANC PND SB END LBW PTB PNA MD Renal injury Hepatic injury HELLP Syndrome Pulmonary edema Placental Abruption Preeclampsia Eclampsia Others Wondimu et al. (2018) Karamara hospital Cross sectional 77.3% 93 59 26 19 7 34 47 26 10 22 3 5 9 7 93 Obsa and Wolka (2018) Wolita Cross sectional 81.8% 225 34 4 16 13 164 58 3 Seid et al. (2017) Gandhi Memorial hospital, Addis Abeba Cross sectional 72.7% 200 7 1 13 28 200 Maereg Wagnew et al. (2016) Tikur Anbesa, Zewditu, St. Paul’s hospitals, Adis Abeba Cross sectional 83.4% 1809 387 513 363 150 532 395 102 6 257 114 100 1496 313 Nega et al. (2015) Jimma university hospital Cross sectional 82.1% 314 165 23 76 57 82 208 106 Seyom et al. Hypertensive disorder of pregnancy (HDP) (2016) 0.04 (0.02, 0.06) 0.00 (0.00, 0.03) 0.11 (0.08, 0.15) 0.11 (0.06, 0.19) 0.05 (0.04, 0.07) 0.03 (0.01, 0.05) 0.05 (0.04, 0.07) ES (95% CI) 0.02 (0.01, 0.04) 0.07 (0.05, 0.11) 0.00 (0.00, 0.01) 100.00 13.01 8.70 5.39 12.61 11.86 % 12.61 Weight 12.11 10.26 13.44 0.04 (0.02, 0.06) 0.00 (0.00, 0.03) 0.11 (0.08, 0.15) 0.11 (0.06, 0.19) 0.05 (0.04, 0.07) 0.03 (0.01, 0.05) 0.05 (0.04, 0.07) ES (95% CI) 0.02 (0.01, 0.04) 0.07 (0.05, 0.11) 0.00 (0.00, 0.01) 100.00 13.01 8.70 5.39 12.61 11.86 % 12.61 Weight 12.11 10.26 13.44 .25 .5 .75 1 Fig. 2 Forest plot displaying rate of Maternal death Overall (I^2 = 94.64%, p = 0.00) Seid I et al. (2017) Abate Met al. (2006) Wondimu et al. (2018) Yifruet al. (2015) Wubanchi et al. (2015) Endesha et al. (2014) Obsa and Wolka (2018) Nega et al. (2015) Maereg Wagnew et al. (2016) 0.04 (0.02, 0.06) 0.00 (0.00, 0.03) 0.11 (0.08, 0.15) 0.11 (0.06, 0.19) 0.05 (0.04, 0.07) 0.03 (0.01, 0.05) 0.05 (0.04, 0.07) 0.02 (0.01, 0.04) 0.07 (0.05, 0.11) 0.00 (0.00, 0.01) 100.00 13.01 8.70 5.39 12.61 11.86 12.61 12.11 10.26 13.44 0.04 (0.02, 0.06) 0.00 (0.00, 0.03) 0.11 (0.08, 0.15) 0.11 (0.06, 0.19) 0.05 (0.04, 0.07) 0.03 (0.01, 0.05) 0.05 (0.04, 0.07) 0.02 (0.01, 0.04) 0.07 (0.05, 0.11) 0.00 (0.00, 0.01) 100.00 13.01 8.70 5.39 12.61 11.86 12.61 12.11 10.26 13.44 .25 .5 .75 1 Fig. 2 Forest plot displaying rate of Maternal death Fig. 2 Forest plot displaying rate of Maternal death Perinatal outcomes of HDP six studies [21, 26, 28–31]: the highest rate being 52.8% in Addis Ababa [31] and the lowest rate being 13.4% in Southern Ethiopia [30]. Preterm birth complicates as high as 65.3% of women with HDP in Somalia regional state of Ethiopia [21] to as low as 31% in a study con- ducted at Oromia regional state [26]. Among thirteen studies included in the review, nine studies reported rate of perinatal death in mothers diag- nosed with HDP [21, 22, 25, 26, 28–32]. As illustrated in the forest plot, perinatal death was observed in one- fourth of women [0.25 (0.18–0.32), I2 = 95.94%, p < 0.01] (Fig. 4). Egger’s test was conducted and illustrated that the publication bias is not statistically significant, p- value > 0.576 (Fig. 6). The rate of stillbirths were most common than the rate of early neonatal deaths. For in- stance, in one study the rate of stillbirth was almost four folds higher than early neonatal deaths (81% vs. 19%) [22]. The frequency of giving birth to a low birth weight neonate was assessed using eight studies [21, 22, 25, 26, 28–31]. The overall rate of having a low birth weight newborn was found to be 37% [0.37 (0.27–0.48%), I2 = 97.40%, p < 0.01] (Fig. 5). Egger’s test was employed to see for publication bias and found non-significant, p- value > 0.859 (Fig. 6). Perinatal asphyxia was reported by Hypertensive disorder of pregnancy (HDP) (2015) Mettu Kari hospital Cross sectional 72.9% 121 62 13 11 2 46 34 20 8 14 15 2 76 23 22 Yifruet al. (2015) Southren Ethiopia Retrospective cohort 71.2% 1015 382 51 521 326 612 346 57 Wubanchi et al. (2015) Debre Berhan hospital Cross sectional 81.2% 270 49 81 89 56 88 7 182 75 13 Vata et al. (2015) Dilla University hospital Cross sectional 75% 172 16 16 23 27 2 143 29 Selamawit et al. (2015) Zewditu Memorial hospital Cross sectional 76.5% 250 14 29 19 10 126 91 132 20 38 15 121 17 112 Endesha et al. (2014) Hawassa,Hosanna, Yirgalem Hospital Retrospective cohort 77.9% 1015 382 322 261 61 545 51 521 326 612 346 57 Zenebe et al. (2011) Jimma university hospital Cross sectional 74.8% 153 40 52 42 10 70 33 14 108 37 8 Abate et al. (2006) St. Paul’s and Tikur anbesa hospitals Cross sectional 73.5% 257 113 69 44 25 28 PND perinatal death, SB still birth, END early neonatal death, LBW ow birth weight, PNA perinatal asphyxia, PTB preterm birth, MD maternal death, ANC antenatal care Mersha et al. BMC Pregnancy and Childbirth (2019) 19:458 Page 6 of 12 Overall (I^2 = 94.64%, p = 0.00) Seid I et al. (2017) Abate Met al. (2006) Wondimu et al. (2018) Yifruet al. (2015) Wubanchi et al. (2015) Endesha et al. (2014) Study Obsa and Wolka (2018) Nega et al. (2015) Maereg Wagnew et al. (2016) 0.04 (0.02, 0.06) 0.00 (0.00, 0.03) 0.11 (0.08, 0.15) 0.11 (0.06, 0.19) 0.05 (0.04, 0.07) 0.03 (0.01, 0.05) 0.05 (0.04, 0.07) ES (95% CI) 0.02 (0.01, 0.04) 0.07 (0.05, 0.11) 0.00 (0.00, 0.01) 100.00 13.01 8.70 5.39 12.61 11.86 % 12.61 Weight 12.11 10.26 13.44 0.04 (0.02, 0.06) 0.00 (0.00, 0.03) 0.11 (0.08, 0.15) 0.11 (0.06, 0.19) 0.05 (0.04, 0.07) 0.03 (0.01, 0.05) 0.05 (0.04, 0.07) ES (95% CI) 0.02 (0.01, 0.04) 0.07 (0.05, 0.11) 0.00 (0.00, 0.01) 100.00 13.01 8.70 5.39 12.61 11.86 % 12.61 Weight 12.11 10.26 13.44 .25 .5 .75 1 Fig. 2 Forest plot displaying rate of Maternal death Overall (I^2 = 94.64%, p = 0.00) Seid I et al. (2017) Abate Met al. (2006) Wondimu et al. (2018) Yifruet al. (2015) Wubanchi et al. (2015) Endesha et al. (2014) Study Obsa and Wolka (2018) Nega et al. (2015) Maereg Wagnew et al. Discussion In the current review, 4% of women in Ethiopia diag- nosed with hypertensive disorders of pregnancy ended up in maternal death. This finding is similar with a study conducted in Pakistan that reported maternal death rate of 6.23% among women with hypertension [33]. While the case fatality rate for eclampsia ranged from 0 to 1.8% in high-income countries. Such a wide disparity is due to both differences in incidence and quality of obstetric care for hypertensive disease in pregnancy [41]. The rate of maternal death from this review is much higher than a study conducted in Saudi Arabia which reported a Mersha et al. BMC Pregnancy and Childbirth (2019) 19:458 Page 7 of 12 Overall (I^2 = 72.22%, p = 0.00) Nega et al. (2015) Selamawit D et al. (2015) Wondimu et al. (2018) Zenebe W et al. (2011) Study Seyom et al. (2015) Seid I et al. (2017) Maereg Wagnew et al. (2016) 0.13 (0.10, 0.16) 0.18 (0.14, 0.23) 0.15 (0.11, 0.20) 0.05 (0.02, 0.12) 0.09 (0.06, 0.15) ES (95% CI) 0.12 (0.08, 0.19) 0.14 (0.10, 0.19) 0.14 (0.13, 0.16) 100.00 14.37 13.97 13.68 13.71 Weight 11.13 13.20 19.94 % 0.13 (0.10, 0.16) 0.18 (0.14, 0.23) 0.15 (0.11, 0.20) 0.05 (0.02, 0.12) 0.09 (0.06, 0.15) ES (95% CI) 0.12 (0.08, 0.19) 0.14 (0.10, 0.19) 0.14 (0.13, 0.16) 100.00 14.37 13.97 13.68 13.71 Weight 11.13 13.20 19.94 % .25 .5 .75 1 Fig. 3 Forest plot displaying rate of HELLP syndrome Overall (I^2 = 72.22%, p = 0.00) Nega et al. (2015) Selamawit D et al. (2015) Wondimu et al. (2018) Zenebe W et al. (2011) Study Seyom et al. (2015) Seid I et al. (2017) Maereg Wagnew et al. (2016) 0.13 (0.10, 0.16) 0.18 (0.14, 0.23) 0.15 (0.11, 0.20) 0.05 (0.02, 0.12) 0.09 (0.06, 0.15) ES (95% CI) 0.12 (0.08, 0.19) 0.14 (0.10, 0.19) 0.14 (0.13, 0.16) 100.00 14.37 13.97 13.68 13.71 Weight 11.13 13.20 19.94 % 0.13 (0.10, 0.16) 0.18 (0.14, 0.23) 0.15 (0.11, 0.20) 0.05 (0.02, 0.12) 0.09 (0.06, 0.15) ES (95% CI) 0.12 (0.08, 0.19) 0.14 (0.10, 0.19) 0.14 (0.13, 0.16) 100.00 14.37 13.97 13.68 13.71 Weight 11.13 13.20 19.94 % .25 .5 .75 1 Fig. 3 Forest plot displaying rate of HELLP syndrome Overall (I^2 = 72.22%, p = 0.00) Nega et al. (2015) Selamawit D et al. (2015) Wondimu et al. (2018) Zenebe W et al. (2011) Seyom et al. Discussion (2015) Seid I et al. (2017) Maereg Wagnew et al. (2016) 0.13 (0.10, 0.16) 0.18 (0.14, 0.23) 0.15 (0.11, 0.20) 0.05 (0.02, 0.12) 0.09 (0.06, 0.15) 0.12 (0.08, 0.19) 0.14 (0.10, 0.19) 0.14 (0.13, 0.16) 100.00 14.37 13.97 13.68 13.71 11.13 13.20 19.94 0.13 (0.10, 0.16) 0.18 (0.14, 0.23) 0.15 (0.11, 0.20) 0.05 (0.02, 0.12) 0.09 (0.06, 0.15) 0.12 (0.08, 0.19) 0.14 (0.10, 0.19) 0.14 (0.13, 0.16) 100.00 14.37 13.97 13.68 13.71 11.13 13.20 19.94 .25 .5 .75 1 Fig. 3 Forest plot displaying rate of HELLP syndrome Fig. 3 Forest plot displaying rate of HELLP syndrome example, defective placental vascular remodeling around mid second trimester of pregnancy with the second round of trophoblastic invasion into the deciduas results in inadequate placental perfusion. The hypoxic placenta then releases various placental factors such as soluble vascular endothelial growth factor receptor-1 (sVEGFR- 1), which then binds vascular endothelial growth factor (VEGF) and placental growth factor (PGF), causing endothelial cell and placental dysfunction by preventing them from binding endothelial cell receptors. This re- sults in increased platelet activation and aggregation leading to low platelet count, haemolysis and hepatic in- jury [36, 37]. maternal death of 1.3%. The Saudi Arabian study was conducted in a tertiary center that may explain the lower reported rate of maternal death as compared to the current finding [34]. This could also be justified by the high number of participants with no antenatal care (ANC) follow up in some of the studies included in the current review. Furthermore, differences in the quality of maternal and neonatal health care service may have been caused this disparity. p y In Ethiopia, HELLP syndrome complicates 13% of women with hypertensive disorders of pregnancy. This finding is line with a study conducted by Karumanchi et al. which reported a 10 to 20% rate of HELLP syn- drome among women admitted for the diagnosis of pre- eclampsia [35]. The much higher rate of HELLP syndrome as compared to women without hypertensive disorders (0.5–0.9%) could be explained by the patho- logical course of the disease. In preeclampsia for Pulmonary edema, acute kidney injury, hepatic injury, placental abruption, aspiration pneumonia, and other life treating complications were also reported by included studies. The above mentioned complications were re- ported by the 2014 world health organization (WHO) Mersha et al. Discussion This finding is in line with another study conducted in Pakistan which re- ported a perinatal mortality of 17.5% [33]. However, this is in contrary to a Norway study conducted among preg- nant women that reported a perinatal mortality of only 9.2%. This disparity may be explained by differences in the quality of follow up a pregnant woman receives [38]. In this review, low birth weight complicates more than one-third of women with HDP in Ethiopia (37%). This rate is much higher than the rates reported in a study conducted in China (6.8%); and, a review conducted among women with chronic hypertension (16.9%). As per the 2016 Ethiopian Demographic and Health Statis- tics (EDHS) report, the rate of low birth weight in the general population is 13% that could explain the existing higher number of low birth weight in Ethiopia [9]. In this review, low birth weight complicates more than one-third of women with HDP in Ethiopia (37%). This rate is much higher than the rates reported in a study conducted in China (6.8%); and, a review conducted among women with chronic hypertension (16.9%). As per the 2016 Ethiopian Demographic and Health Statis- tics (EDHS) report, the rate of low birth weight in the general population is 13% that could explain the existing higher number of low birth weight in Ethiopia [9]. Discussion BMC Pregnancy and Childbirth (2019) 19:458 Page 8 of 12 Overall (I^2 = 95.94%, p = 0.00) Selamawit D et al. (2015) Zenebe W et al. (2011) Endesha et al. (2014) Seyom et al. (2015) Wondimu et al. (2018) Vata et al. (2015) Maereg Wagnew et al. (2016) Study Wubanchi et al. (2015) Abate Met al. (2006) 0.25 (0.18, 0.32) 0.12 (0.08, 0.16) 0.34 (0.27, 0.42) 0.32 (0.29, 0.35) 0.11 (0.06, 0.18) 0.28 (0.20, 0.38) 0.09 (0.06, 0.15) 0.28 (0.26, 0.30) ES (95% CI) 0.30 (0.25, 0.36) 0.44 (0.38, 0.50) 100.00 11.49 10.49 11.70 11.11 9.92 11.41 11.81 % Weight 11.12 10.95 0.25 (0.18, 0.32) 0.12 (0.08, 0.16) 0.34 (0.27, 0.42) 0.32 (0.29, 0.35) 0.11 (0.06, 0.18) 0.28 (0.20, 0.38) 0.09 (0.06, 0.15) 0.28 (0.26, 0.30) ES (95% CI) 0.30 (0.25, 0.36) 0.44 (0.38, 0.50) 100.00 11.49 10.49 11.70 11.11 9.92 11.41 11.81 % Weight 11.12 10.95 .25 .5 .75 1 Fig. 4 Forest plot demonstrating frequency of Perinatal death Fig. 4 Forest plot demonstrating frequency of Perinatal death multinational analysis using 29 countries from Africa, Asia, Latin America and Middle East [3, 4]. the review. The above mentioned complications have been reported by other original articles and systematic reviews [39, 40]. Although the exact mechanisms for the above mentioned perinatal complications are not yet well known, the most acceptable theory for the develop- ment of preeclampsia is defective remodeling of spiral arteries. Defective placentation affects utero-placental blood flow and leads to complications such as preterm birth, low birth weight, perinatal asphyxia, and fetal growth restriction [14, 15]. An Indian study reported that the most common neonatal complication was pre- maturity (23.65%), low birth weight (7.52%) and intra- uterine growth restriction (9.67%) [16]. In this review, the prevalence of perinatal mortality among women with hypertensive disorders of pregnancy in Ethiopia was found to be 25%. This finding is in line with another study conducted in Pakistan which re- ported a perinatal mortality of 17.5% [33]. However, this is in contrary to a Norway study conducted among preg- nant women that reported a perinatal mortality of only 9.2%. This disparity may be explained by differences in the quality of follow up a pregnant woman receives [38]. In this review, the prevalence of perinatal mortality among women with hypertensive disorders of pregnancy in Ethiopia was found to be 25%. Strength and limitations of the study g y This review is the first to review the maternal and fetal outcomes among women with all types of HDPs in Ethiopia. The limited numbers of studies evaluating out- comes made it impossible to generate strong conclu- sions. The included studies were also conducted primarily in tertiary health centers, and the data may not be representative of outcomes in low-level facilities or in Perinatal asphyxia, preterm birth and other complica- tions have been also reported in the studies included in Mersha et al. BMC Pregnancy and Childbirth (2019) 19:458 Page 9 of 12 Overall (I^2 = 97.40%, p = 0.00) Study Zenebe W et al. (2011) Vata et al. (2015) Maereg Wagnew et al. (2016) Endesha et al. (2014) Selamawit D et al. (2015) Seyom et al. (2015) Wondimu et al. (2018) Wubanchi et al. (2015) 0.37 (0.27, 0.48) ES (95% CI) 0.46 (0.38, 0.54) 0.13 (0.09, 0.19) 0.29 (0.27, 0.32) 0.54 (0.51, 0.57) 0.50 (0.44, 0.57) 0.38 (0.30, 0.47) 0.37 (0.27, 0.47) 0.33 (0.28, 0.39) 100.00 Weight 12.20 12.74 13.08 % 13.00 12.55 12.03 11.75 12.65 0.37 (0.27, 0.48) ES (95% CI) 0.46 (0.38, 0.54) 0.13 (0.09, 0.19) 0.29 (0.27, 0.32) 0.54 (0.51, 0.57) 0.50 (0.44, 0.57) 0.38 (0.30, 0.47) 0.37 (0.27, 0.47) 0.33 (0.28, 0.39) 100.00 Weight 12.20 12.74 13.08 % 13.00 12.55 12.03 11.75 12.65 .25 .5 .75 1 Fig. 5 Forest plot showing frequency of Low birth weight Overall (I^2 = 97.40%, p = 0.00) Study Zenebe W et al. (2011) Vata et al. (2015) Maereg Wagnew et al. (2016) Endesha et al. (2014) Selamawit D et al. (2015) Seyom et al. (2015) Wondimu et al. (2018) Wubanchi et al. (2015) 0.37 (0.27, 0.48) ES (95% CI) 0.46 (0.38, 0.54) 0.13 (0.09, 0.19) 0.29 (0.27, 0.32) 0.54 (0.51, 0.57) 0.50 (0.44, 0.57) 0.38 (0.30, 0.47) 0.37 (0.27, 0.47) 0.33 (0.28, 0.39) 100.00 Weight 12.20 12.74 13.08 % 13.00 12.55 12.03 11.75 12.65 0.37 (0.27, 0.48) ES (95% CI) 0.46 (0.38, 0.54) 0.13 (0.09, 0.19) 0.29 (0.27, 0.32) 0.54 (0.51, 0.57) 0.50 (0.44, 0.57) 0.38 (0.30, 0.47) 0.37 (0.27, 0.47) 0.33 (0.28, 0.39) 100.00 Weight 12.20 12.74 13.08 % 13.00 12.55 12.03 11.75 12.65 .25 .5 .75 1 Fig. 5 Forest plot showing frequency of Low birth weight Fig. 5 Forest plot showing frequency of Low birth weight pregnant woman and her baby. Strength and limitations of the study Giving the above mentioned fact of poor antenatal care (ANC) service utilization, community based health education can be one possible level of intervention to improve the health outcomes of women with HDP by improving maternal health service utilization. The second pos- sible level of intervention could be early detection and early referral of pregnant women with hyperten- sive disorder. Policies and strategies that may en- hance the health professionals’ capacity should be advocated. The other possible level of intervention could be, optimizing the quality of care that a preg- nant woman with hypertensive disorders receives. Additionally, it is recommended to have easily avail- able and affordable laboratory testing; frequent ma- ternal and fetal monitoring; increase accessibility of antihypertensive and anticonvulsive drugs; increase preparation for neonatal resuscitation; and, improv- ing quality of neonatal intensive care units. It is also recommended to develop a uniform systematic regis- tration system of maternal and perinatal outcomes of HDP in the country. the community. Furthermore, there is a high level of heterogeneity among the included studies that should be taken in consideration while using the results of the re- view. Despite these possible limitations, the review pro- vides useful information that may contribute to both the filling of the gaps in the national maternal morbidity re- search agenda and guiding practice and policy about the most frequent complications of HDP. References 24. Seid I, et al. Maternal outcomes of preeclampsia in an Ethiopian Gynecologic Hospital. Ann Med Health Sci Res. 2017;7:16–21. 1. Oberts JM, Pearson G, Cutler J, et al. Summary of the NHLBI working group on research on hypertension during pregnancy. Hypertension. 2003;41:437– 45. 1. Oberts JM, Pearson G, Cutler J, et al. Summary of the NHLBI working group on research on hypertension during pregnancy. Hypertension. 2003;41:437– 45. 25. Wagnew M, Dessalegn M, Worku A, Nyagero J. Trends of preeclampsia/ eclampsia and maternal and neonatal outcomes among women delivering in Addis Ababa selected government hospitals, Ethiopia: a retrospective cross sectional study. Pan Afr Med J. 2016;25(Suppl 2):12. 2. Berhe AK, Kassa GM, Fekadu GA, Muche AA. Prevalence of hypertensive disorders of pregnancy in Ethiopia: a systemic review and meta-analysis. BMC Pregnancy Childbirth. 2018;18(1):34. 2. Berhe AK, Kassa GM, Fekadu GA, Muche AA. Prevalence of hypertensive disorders of pregnancy in Ethiopia: a systemic review and meta-analysis. BMC Pregnancy Childbirth. 2018;18(1):34. 26. Seyom E, Abera M, Tesfaye M, Fentahun N. Maternal and fetal outcome of pregnancy related hypertension in Mettu Karl Referral Hospital, Ethiopia. J Ovarian Res. 2015;8:10. 3. Khan KS, Wojdyla D, Say L, Gulmezoglu AM, Van Look PF. WHO analysis of causes of maternal death: a systematic review. Lancet. 2006;367:1066–74. 3. Khan KS, Wojdyla D, Say L, Gulmezoglu AM, Van Look PF. WHO analysis of causes of maternal death: a systematic review. Lancet. 2006;367:1066–74. 4. Firoz T, Sanghvi H, Merialdi M, von Dadelszen P. Pre-eclampsia in low and middle income countries. Best Pract Res Clin Obstet Gynaecol. 2011;25:537– 48. 27. Berhan Y, Endeshaw G. Maternal mortality predictors in women with hypertensive disorders of pregnancy: a retrospective cohort study. Ethiop J Health Sci. 2015;25(1):89–98. 5. Adu-Bonsaffoh K, Obed SA, Seffah JD. Maternal outcomes of hypertensive disorders in pregnancy at Korle Bu Teaching Hospital, Ghana. Int J Gynaecol Obstet. 2014;127(3):238–42. 28. Terefe W, Getachew Y, Hiruye A, Derbew M, Mariam DH, Mammo D, et al. Patterns of hypertensive disorders of pregnancy and associated factors at DebreBerhan referral hospital, north Shoa, Amhara region. Ethiop Med J. 2015;(Suppl 2):57–65. 6. Berhan Y, Berhan A. Causes of maternal mortality in Ethiopia: a significant decline in abortion related death. Ethiop J Health Sci. 2014;24:15–28. 29. Wolde Z, Segni H, Woldie M. Hypertensive disorders of pregnancy in Jimma university specialized hospital. Ethiop J Health Sci. 2011;21(3):147–54. 7. Received: 26 November 2018 Accepted: 21 November 2019 23. Obsa MS, Wolka E. Maternal outcome of pregnant mothers with hypertensive disorder of pregnancy at hospitals in Wolaita Zone, Southern Ethiopia. J Preg Child Health. 2018;5:375. https://doi.org/10.4172/2376-127X.1000375. No funding source. No funding source. 16. Aabidha PM, Cherian AG, Paul E, Helan J. Maternal and fetal outcome in pre-eclampsia in a secondary care hospital in South India. J Family Med Prim Care. 2015;4(2):257–60. Author details 1 Author details 1Department of Gynecology and Obstetrics, School of Medicine, College of Medicine and Health Sciences, University of Gondar, P.O. Box: 196, Gondar, Ethiopia. 2Department of Clinical Pharmacy, School of Pharmacy, College of Medicine and Health Sciences, University of Gondar, P.O. Box: 196, Gondar, Ethiopia. 21. Gudu W, Bekele D. A prospective review of eclampsia at a regional hospital, Eastern Ethiopia: incidence, clinical correlates, management and pregnancy outcome. Ethiop Med J. 2018;56:125-132. 22. Endeshaw G, Berhan Y. Perinatal outcome in women with hypertensive disorders of pregnancy: a retrospective cohort study. Int Schol Res Not. 2015;2015:208043. Conclusions and recommendations This review demonstrated the high prevalence of peri- natal and maternal mortality among pregnant women with one of the HDP in Ethiopia. Moreover, other severe maternal and perinatal complications such as HELLP syndrome, placental abruption, pulmonary edema, hemolytic anemia, renal damage, prematurity, perinatal asphyxia, as well as low birth weight were also com- monly reported. Based on the above mentioned results, it is recom- mended to intervene at three levels so as to improve the maternal and fetal outcomes of hypertensive dis- orders in Ethiopia and to ensure the safety of a Mersha et al. BMC Pregnancy and Childbirth (2019) 19:458 Page 10 of 12 Fig. 6 Publication bias of maternal and fetal outcomes Fig. 6 Publication bias of maternal and fetal outcomes Fig. 6 Publication bias of maternal and fetal outcomes Page 11 of 12 Page 11 of 12 Mersha et al. BMC Pregnancy and Childbirth (2019) 19:458 Abbreviations 11. Roberts JM, Pearson G, Cutler J, Lindheimer M. Summary of the NHLBI Working Group on research on hypertension during pregnancy. Hypertension. 2003;41(3 I):437–45. ANC: Antenatal care; CI: Confidence interval; HDP: Hypertensive disorders of pregnancy; WHO: World Health Organization 12. Nusrat N, Ahson M, Munir A. Hypertensive disorders of pregnancy; frequency, maternal and perinatal out comes. J Pakistan Army Med Corps. 2010;26(1):119–23. Acknowledgements We acknowledge the authors of the included articles. 13. Yücesoy G, Ozkan S, Bodur H, et al. Maternal and perinatal outcome in pregnancies complicated with hypertensive disorder of pregnancy: a seven year experience of a tertiary care center. Arch Gynecol Obstet. 2005;273(1): 43–9. Availability of data and materials All relevant materials and data supporting the findings of this review are included within the manuscript. 17. vonElm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61(4):344–9. Competing interests The authors declare that they have no competing interests. Competing interests The authors declare that they have no competing interests. 20. Desalegn N, Haile M. Causes of Admission and out Comes Among Preeclampsia and Eclampsia Mothers Admitted to Jimma University Specialized Hospital Intensive Care Unit. Clin Med Res. 2015;4(5):154–9. https://doi.org/10.11648/j.cmr.20150405.16. Authors’ contribution AGM conceived the study and wrote the final manuscript. AGM, TMA, MAS involved in the acquisition of data, performed review, analyzed the data, and wrote the initial draft of the manuscript. All authors read and approved the final version of the manuscript. 14. Palatnik A, Grobman WA, Miller ES. Is a history of preeclampsia associated with an increased risk of a small for gestational age infant in a future pregnancy? Am J ObstetGynecol. 2016;215(3):355.e1. 15. Bramham K, Briley AL, Seed P, et al. Adverse maternal and perinatal outcomes in women with previous preeclampsia: a prospective study. Am J ObstetGynecol. 2011;204(6):512.e1. Consent for publication Not applicable. 19. Huedo-Medina TB, Sánchez-Meca J, Marín-Martínez F, Botella J. Assessing heterogeneity in meta-analysis: Q statistic or I2 index? Psychol Methods. 2006;11(2):193. Competing interests Competing interests The authors declare that they have no competing interests. Ethics approval Our study is an investigation of the literature and does not need ethical approval for retrieving the already available public content. 18. StataCorp L. Stata Statistical Software: Release 14. [computer program]. StataCorp LP. 2015. Consent for publication Not applicable. Mersha et al. BMC Pregnancy and Childbirth (2019) 19:458 34. Subki AH, Algethami MR, Baabdullah WM, Alnefaie MN, Alzanbagi MA, Alsolami RM, et al. Prevalence, risk factors, and fetal and maternal outcomes of hypertensive disorders of pregnancy: a retrospective study in Western Saudi Arabia. Oman Med J. 2018;33(5):409–15. Saudi Arabia. Oman Med J. 2018;33(5):409–15. 35. Karumanchi SA, Maynard SE, Stillman IE, Epstein FH, Sukhatme VP. Preeclampsia: a renal perspective. Kidney Int. 2005;67(6):2101–13. 36. Steinborn A, Rebmann V, Scharf A, Sohn C, Grosse-Wilde H. Soluble HLA-DR levels in the maternal circulation of normal and pathologic pregnancy. Am J Obstet Gynecol. 2003;188:473–9. 37. Agatisa PK, Ness RB, Roberts JM, Costantino JP, Kuller LH, McLaughlin MK. Impairment of endothelial function in women with a history of preeclampsia: an indicator of cardiovascular risk. Am J Physiol Heart Circ Physiol. 2004;286:H1389–93. 37. Agatisa PK, Ness RB, Roberts JM, Costantino JP, Kuller LH, McLaughlin MK. Impairment of endothelial function in women with a history of preeclampsia: an indicator of cardiovascular risk. Am J Physiol Heart Circ Physiol. 2004;286:H1389–93. 38. Ahmad A, Samuelsen S. Hypertensive disorders in pregnancy and fetal 371° 121° death at different gestational lengths: a population study of 2 pregnancies. BJOG. 2012;119:1521–8. 39. Browne JL, Vissers KM, Antwi E, Srofenyoh EK, Van der Linden EL, Agyepong IA, et al. Perinatal outcomes after hypertensive disorders in pregnancy in a low resource setting. Trop Med Int Health. 2015;20(12):1778–86. 40. Ye C, Ruan Y, Zou L, Li G, Li C, et al. The 2011 survey on hypertensive disorders of pregnancy (HDP) in China: prevalence, risk factors, complications, pregnancy and perinatal outcomes. PLoS One. 2014;9(6): e100180. https://doi.org/10.1371/journal.pone.0100180. 40. Ye C, Ruan Y, Zou L, Li G, Li C, et al. The 2011 survey on hypertensive disorders of pregnancy (HDP) in China: prevalence, risk factors, complications, pregnancy and perinatal outcomes. PLoS One. 2014;9(6): e100180. https://doi.org/10.1371/journal.pone.0100180. 41. Ronsmans C, Campbell O. Quantifying the fall in mortality associated with interventions related to hypertensive diseases of pregnancy. BMC Public Health. 2011;11(3):S8. 41. Ronsmans C, Campbell O. Quantifying the fall in mortality associated with interventions related to hypertensive diseases of pregnancy. BMC Public Health. 2011;11(3):S8. Mersha et al. BMC Pregnancy and Childbirth (2019) 19:458 References Ukah UV, De Silva DA, Payne B, Magee LA, Hutcheon JA, Brown H, et al. Prediction of adverse maternal outcomes from pre-eclampsia and other hypertensive disorders of pregnancy: a systematic review. Pregnancy Hypertens. 2018;11:115–23. 30. Vata PK, Chauhan NM, Nallathambi A, Hussein F. Assessment of prevalence of preeclampsia from Dilla region of Ethiopia. BMC Res Notes. 2015;8:816. 31. Selamawit D, Sisay T. Maternal and perinatal outcomes of pregenanciescomplecated by preeclampsia/eclampsia at zewditu memorial hospital. Addis Ababa Aniversity School of Graduate Studies Faculty of Medicine 2015. 8. Steegers EA, von Dadelszen P, Duvekot JJ, Pijnenborg R. Pre-eclampsia. Lancet. 2010;376:631–44. eegers EA, von Dadelszen P, Duvekot JJ, Pijnenborg R. Pre-eclampsia. 9. Central Statistical Agency (CSA) [Ethiopia] and ICF. Ethiopia Demographic and Health Survey 2016. Addis Ababa and Rockville: CSA and ICF; 2016. 32. Abate M, Lakew Z. Eclampsia a 5 years retrospective review of 216 cases managed in two teaching hospitals in Addis Ababa. Ethiop Med J. 2006; 44(1):27–31. 10. Abalos E, Cuesta C, Carroli G, Qureshi Z, Widmer M, Vogel JP, Souza JP, on behalf of the WHO Multicountry Survey on Maternal and Newborn Health Research Network. Pre-eclampsia, eclampsia and adverse maternal and perinatal outcomes: a secondary analysis of the World Health Organization Multicountry Survey on Maternal and Newborn Health. BJOG. 2014; 121(Suppl. 1):14–24. 33. Hossain N, Shah N, Khan N, Lata S, Khan NH. Maternal and perinatal outcome of hypertensive disorders of pregnancy at a tertiary care hospital. J Dow Univ Health Sci Karachi. 2011;5(1):12–6. Page 12 of 12 Mersha et al. BMC Pregnancy and Childbirth (2019) 19:458 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
https://openalex.org/W1990026351	https://discovery.ucl.ac.uk/1469956/1/Jameel%20et%20al.%20Great%20Expectations.pdf	English	null	Great Expectations: The Role of Rules in Guiding Pro-social Behaviour in Groups with High Versus Low Autistic Traits	Journal of autism and developmental disorders	2,015	cc-by	9,886	Great Expectations: The Role of Rules in Guiding Pro-social Behaviour in Groups with High Versus Low Autistic Traits Leila Jameel • Karishma Vyas • Giulia Bellesi • Diana Cassell • Shelley Channon The Author(s) 2015. This article is published with open access at Springerlink.com disorder (ASD), very little work has examined how these translate into everyday social functioning and impact on speciﬁc aspects of social interaction, in particular pro- social behaviour such as helping and sharing with others, volunteering time or donating money. Two studies exam- ining charitable giving found that adults with ASD donated less than matched controls (Lin et al. 2012; Izuma et al. 2011). Parental reports indicated that children with ASD tend to display signiﬁcantly less pro-social behaviour compared to typically developing children, e.g. being kind and considerate to others, sharing or offering practical help (Meyer et al. 2006; Allik et al. 2006). Several small-scale interventions using social stories have attempted to pro- mote pro-social behaviour in children with ASD (e.g. see Crozier and Tincani 2007; Leaf et al. 2009). Recent work has examined the use of oxytocin, which is thought to play a role in regulating and promoting social behaviour, in individuals with ASD, and this seems to show some pro- mise (e.g. see Andari et al. 2010). Abstract Measuring autistic traits in the general population has proven sensitive for examining cognition. The present study extended this to pro-social behaviour, investigating the inﬂuence of expectations to help others. A novel task describing characters in need of help was administered to students scoring high versus low on the Autism-Spectrum Quotient. Scenarios had two variants, describing either a ‘clear-cut’ or ‘ambiguous’ social rule. Participants with high versus low autistic traits were less pro-social and sympathetic overall towards the characters. The groups’ ratings of char- acters’ expectations were comparable, but those with high autistic traits provided more rule-based rationales in the clear- cut condition. This pattern of relatively intact knowledge in the context of reduced pro-social behaviour has implications for social skill training programmes. Keywords Autistic traits Pro-social behaviour Empathy Mentalising Social rules Social knowledge g In the light of the established social and emotional deﬁcits associated with ASD, it is important to understand what predicates pro-social behaviour, and how factors in the social environment might facilitate or inhibit this. Pro- social behaviour is thought to be driven by empathy (Eisenberg 2007; Minio-Paluello et al. 2009), via ‘self’- or ‘other’-oriented processes (Schaller and Cialdini 1988). J Autism Dev Disord DOI 10.1007/s10803-015-2393-x J Autism Dev Disord DOI 10.1007/s10803-015-2393-x ORIGINAL PAPER Great Expectations: The Role of Rules in Guiding Pro-social Behaviour in Groups with High Versus Low Autistic Traits Vicariously invoked feelings of distress and increased physiological arousal when witnessing someone in need may result in a desire to alleviate the pain shared by the onlooker, thus stimulating a response to help. The resulting action is thus self-oriented and motivated by a need to reduce the vicarious empathic arousal experienced. On the other hand, pro-social behaviour may be driven by an intuitive understanding of the other’s thoughts, feeling and needs (Jameel et al. 2014). L. Jameel (&) K. Vyas G. Bellesi S. Channon Department of Experimental Psychology, University College London (UCL), Bedford Way Building, Gower Street, London WC1E 6BT, UK e-mail: l.jameel@ucl.ac.uk D. Cassell Child and Adolescent Mental Health Services, South West London and St George’s Mental Health Trust, London, UK Introduction Successful social functioning entails processing and responding sensitively to subtle and complex information provided by the social world. Whilst there is a wealth of literature exploring cognitive accounts of autism spectrum L. Jameel (&) K. Vyas G. Bellesi S. Channon Department of Experimental Psychology, University College London (UCL), Bedford Way Building, Gower Street, London WC1E 6BT, UK e-mail: l.jameel@ucl.ac.uk D. Cassell Child and Adolescent Mental Health Services, South West London and St George’s Mental Health Trust, London, UK 12 3 3 J Autism Dev Disord The self- versus other-orientated routes for motivating pro- social behaviour can be viewed as analogous to the separable ‘emotional’ and ‘cognitive’ components that contribute to- wards the experience of empathy. ‘Emotional’ empathy refers to the ability to resonate with and share another’s emotional state. By contrast, ‘cognitive’ empathy, or mentalising (also synonymous with theory of mind or perspective-taking), de- scribes the ability to infer others’ mental states and thus what they might be thinking or feeling in any given situation (for a discussion of cognitive vs emotional ‘ﬁne cuts’ see: Blair 2008). Mentalising is thought to be impaired in individuals with ASD, with difﬁculties in interpreting others’ intentions, thoughts and feelings, whereas most evidence suggests that ability to empathise with others’ emotional states is intact. However, upon close inspection the literature reveals a slightly messier picture. Some studies examining emotional empathy have found impairment in those with ASD (Hum- phreys et al. 2007) and others have not (Adolphs et al. 2001). Where there is evidence for intact emotional empathy, this is thought to reﬂect somecapacity of those with ASD to resonate emotionally with others, but via a self-stance (Frith and de Vignemont 2005; Minio-Paluello et al. 2009). However, whilst the cognitive and emotional components of empathy are thought to be dissociable in principle, they are likely to be used in concert in everyday life. knowledge about social roles, norms, and scripts, it also requires ﬂexible application of these resources to navigate novel scenarios accurately (Riggio and Reichard 2008). Evidence from real-life-type tasks suggests that high- functioning individuals with ASD may show proﬁciency in simple structured social tasks, but difﬁculties with more advanced and naturalistic tasks. For instance, when asked to make judgments about subtle social behaviours (e.g. Introduction faux-pas or deception), individuals with high-functioning ASD may provide correct answers, but the reasoning behind their judgments is often inaccurate or inappropriate (Bowler 1992; Moran et al. 2011). Moreover, in real-life-type problem sol- ving tasks, high-functioning participants with ASD have been found to display difﬁculty in generating problem solutions, but not in selecting from alternatives presented to them (Channon et al. 2001, 2014). Participants with ASD were thus able to recognise the best solution from a selection of options (i.e. low task demand), but not to produce high quality solu- tions spontaneously (i.e. high task demand). Studies of individuals with high autistic trait scores drawn from the general population has suggested a pattern of social, cognitive and emotional features similar to, but less severe than that typically observed in ASD. For example, Go¨kc¸en et al. (2014) found that higher numbers of autistic traits were associated with lower self-rated levels of emotional perception and expression, poorer performance on a classic task involving predicting char- acters’ emotional states from pictures of their eyes, and impairment on a task examining cognitive ﬂexibility. It has been suggested that individuals with high-function- ing ASD sometimes rely on compensatory strategies, such as the application of learned social rules to alleviate mentalising deﬁcits (Hill and Frith 2003). This is supported by neuroi- maging evidence demonstrating that during online mentalis- ing, high-functioning individuals with ASD show reduced activation in brain regions typically associated with this type of activity, even when correct mental state attributions are made (Happe et al. 1996). Conversely, individuals with high- functioning ASD tend to show greater activation in areas typically associated with more general problem-solving abilities (Happe et al. 1996). Reliance on compensatory mechanisms such as drawing on previously acquired social knowledge may disguise social difﬁculties to an extent, but clumsy and inﬂexible patterns of social behaviour may occur in more complex, unpredictable social circumstances. A recent study (Jameel et al. 2014) examined everyday pro- social performance in students scoring high versus low on a measure of autistic traits (Autism-Spectrum Quotient (AQ); Baron-Cohen et al. 2001). Participants were presented with everyday situations involving a character in need of their help. Each scenario thus involved a difﬁcult social judgment with respect to balancing the needs of the character against the participant’sowninterests.Forinstance,ifsomeoneseesaman fall over heavily in front of them, do they stop to help, even if it means being late to work? Introduction Those scoring high on the AQ were found to behave less pro-socially than their low AQ counter- parts,both when asked togenerate pro-social coursesofaction, and to select them from alternatives. Participants were also asked to give self- versus other-satisfaction ratings corre- sponding to three different courses of helping actions, ranging from low to high pro-social value (i.e. carry on walking, stop brieﬂy to help the man up, or stop to help the man up and see what additional assistance he may require). Contrary to pre- dictions, those with high AQ scores gave similar estimates of the characters’ sense of satisfaction, but reported less personal satisfaction than those with low AQ scores for going ‘above and beyond’ to help others at their own expense. Clinical accounts tend to support the notion that high- functioning individuals with ASD draw upon compensa- tory mechanisms to navigate the social world. Mu¨ller et al. (2008) individually interviewed adults with ASD to discuss their social and communication challenges. Their descrip- tions included difﬁculties with ‘chit-chat’ conversations that tend not to follow predictable sets of rules, and pro- blems in following unstructured dialogue that require improvised responses. References were also made to actively learning how to behave socially via observation of others’ social interactions. Whilst appropriate social behaviour is thought to involve the acquisition of Lower levels of pro-social behaviour might be linked to limited perception of societal expectations, a reduced sense of 123 123 J Autism Dev Disord pressure to comply with such expectations, and lower rewards for helping others. The present study sought to explore this by examining the role of societal expectations and sympathy for others’ needs in guiding pro-social decision making. Par- ticipants with high or low AQ scores were presented with a new set of scenarios featuring a character in need. Each sce- nario had two variant endings: a clear-cut versus ambiguous social rule. In the ‘clear-cut’ condition there was a strong social rule guiding the participants to behave pro-socially (i.e. offer to give up your seat to an elderly woman walking with a stick).Inthe ‘ambiguous’conditionthesocialrulewasweaker (i.e. offer to give up your seat to a young woman carrying a heavy parcel). which measures personality traits associated with the autistic spectrum in adults of typical intelligence. It uses a four point Likert scale and yields total AQ trait scores ranging from 0 to 50, with higher scores indicating higher numbers of autistic traits. Participants and Procedure Ethical approval was obtained from the UCL Research Ethics Committee. An opportunistic sample of 645 full- time university students (41.39 % female) who were ﬂuent in English and aged 18 or over (mean age 20 years) was recruited for the screening phase of the study. All partici- pants provided informed consent before completing the AQ (Baron-Cohen et al. 2001). Participants were entered into a prize draw and informed that they might be invited to take part in the second phase of the study, for which they would be paid. Total AQ scores were calculated for the whole sample. AQ traits are more common in males than in females (Baron-Cohen et al. 2001), thus in order to form gender balanced high AQ and low AQ groups for the experimental phase of the study participants within the highest-scoring and lowest-scoring 10 % of males, and highest-scoring and lowest-scoring 10 % of females, were selected. Participants were contacted via email or tele- phone and invited to take part in the second stage of the study. Participants were asked to reason about why someone might act pro-socially in the situation, and were also asked to rate the characters’ expectations of help, their own likelihood to help, and their sympathy for the characters in need. It was predicted that the highAQ groupwould generate rationales for pro-social behaviour that relied upon social rules, rather than engaging with the individual perspectives of the characters, at least in the clear-cut condition where there was a readily available social rule. However, on the lower-demand measure ofratingcharacters’expectationsofhelp,itwaspostulatedthat the high AQ group might not differ from the low AQ group, at leastfortheclear-cutcondition.Itwaspredictedthatthosewith highAQtraitswouldbeslowertoproduceresponses,atleastin the ambiguous condition when less salient cues guiding pro- social behaviour were provided. Although little work has ex- amined this, it seemed probable that sympathy ratings and likelihood of helping would be reduced in the high AQ group, especially in the ambiguous condition where social rules were less clear. However, it was also considered possible that the high AQ group might differentiate more than the low AQ group between the clear-cut and ambiguous conditions in their sympathy ratings and likelihood of helping, showing more ‘black and white’ thinking consistent with a rigid reliance on social rules. In support of this prediction, some previous work with individuals with ASD reported that they showed height- ened sensitivity to ‘good’ versus ‘poor’ justiﬁcations for wrongdoing (Channon et al. Participants and Procedure 20 (10 female, 10 male) individuals from the lower range and 21 (11 male, 10 female) individuals from the upper range took part in the experimental phase of the study, forming two groups of low and high AQ participants. AQ scores ranged from 2 to 9 in the low AQ group (2–9 for male and 4–8 for female participants), and 26–46 in the high AQ group (28–37 for male and 26–46 for female participants). A t test conﬁrmed that AQ scores differed signiﬁcantly between groups, t(1,39) = 23.23, p \ .001; mean AQ scores were 30.52 (SD = 4.40) and 6.15 (SD = 1.66) for the high and low AQ groups respectively. The groups did not differ signiﬁcantly in age, t(1,39) = .064, p = .950; mean age was 21.11 (SD = 2.62) and 21.06 (SD = 2.57) for the high and low groups respectively. 20 (10 female, 10 male) individuals from the lower range and 21 (11 male, 10 female) individuals from the upper range took part in the experimental phase of the study, forming two groups of low and high AQ participants. AQ scores ranged from 2 to 9 in the low AQ group (2–9 for male and 4–8 for female participants), and 26–46 in the high AQ group (28–37 for male and 26–46 for female participants). A t test conﬁrmed that AQ scores differed signiﬁcantly between groups, t(1,39) = 23.23, p \ .001; mean AQ scores were 30.52 (SD = 4.40) and 6.15 (SD = 1.66) for the high and low AQ groups respectively. The groups did not differ signiﬁcantly in age, t(1,39) = .064, p = .950; mean age was 21.11 (SD = 2.62) and 21.06 (SD = 2.57) for the high and low groups respectively. Participants and Procedure 2010), and greater differentiation between intentional and unintentional actions when assigning blame (Channon et al. 2011). Experimental Phase Introduction This measure was not designed as a diagnostic tool, but rather for the identiﬁcation of traits and behaviours related to the autistic proﬁle, containing 50 statements covering ﬁve different aspects of autistic symptomatology: social skill, attention switching, attention to detail, communication and imagination. Methods Screening Phase Screening Phase Screening Phase The Autism-Spectrum Quotient The Autism-Spectrum Quotient (AQ; Baron-Cohen et al. 2001) is a brief, self-report questionnaire with good internal consistency, construct validity and test–retest reliability, All participants were tested individually, and provided written informed consent before completing the ‘Social 12 3 J Autism Dev Disord Expectations’ task. Participants were also asked to com- plete a brief health-screening questionnaire that asked about any serious accidents or illnesses, psychological or emotional difﬁculties; in practice no exclusions were required. Participants were paid for their time and efforts. were presented in separate booklets such that relevant scenarios remained on display throughout task perfor- mance, in order to minimise any memory demands. Each scenario was followed by four questions. Participants were ﬁrst asked to rate how likely they would be to help the characters, and then to rate how sympathetic they felt towards the characters. Participants were then asked to indicate the strength of the characters’ expectations for help, and ﬁnally to provide a rationale explaining why they might offer to help the character. Response times for ra- tionale production were recorded using a stopwatch. Total response time was calculated by summing the speed of response time across all 10 scenarios. The ‘‘Social Expectations’’ Task The ‘‘Social Expectations’’ Task The ‘‘Social Expectations Task’’ was designed to examine pro-social behaviour in relation to some of the unwritten social rules that govern everyday interactions, comparing scenarios based on both clear-cut and ambiguous rules. A range of scenarios were devised and piloted, in order to reﬁne the items and develop the scoring systems. The ﬁnal task consisted of 10 hypothetical scenarios involving the participant and an unfamiliar character. The scenarios consisted of everyday social situations where the par- ticipant had the opportunity to engage in pro-social be- haviour aiding the character, in line with a social rule. Scenarios were designed such that it was clear the par- ticipant was the only individual who could aid the char- acter, and that engaging in the pro-social behaviour would be inconvenient to the participant (e.g. offering to give up your seat and stand for someone). The character was male in half the scenarios, and female in the other half, and the social context varied across scenarios to reﬂect a range of natural situations. To control for order effects, two differ- ent scenario orders were created and counterbalanced within each group. 123 Example Scenario You are sitting in a crowded waiting room with a small bag, waiting for a delayed train. All the other seats are taken by passengers with lots of luggage. A woman enters the waiting room looking for a seat. Clear-cut Ending: She is elderly and walking with a stick Ambiguous Ending: She is a young adult and is carrying a large parcel Questions for Each Scenario Likelihood of Pro-social Behaviour Ratings: How likely is it that you would offer her your seat? On a scale of 1–10: 1 = not at all likely, 10 = very likely Sympathy for Character Ratings: How sympathetic do you feel towards her? On a scale of 1–10: 1 = not at all sympathetic, 10 = very sympathetic Strength of Character Expectation Ratings: How much do you think she expects you to offer her your seat? On a scale of 1–10: 1 = not at all, 10 = very much Verbal Rationales of Societal Expectation Understanding: Why might you offer her your seat? Scoring Likelihood of Pro-social Behaviour Ratings Each scenario stem had two endings, manipulating the strength of the social rule guiding pro-social behaviour in the situation. ‘Clear-cut’ endings implied a strong social rule (e.g. ‘‘she is elderly and walking with a stick’’), cueing an appropriate response (i.e. you should offer to give up your seat). ‘Ambiguous’ endings still referred to a char- acter that would beneﬁt from help, but did not rely on such strongly endorsed social rules (e.g. ‘‘she is a young adult and is carrying a large parcel’’). For each scenario ending participants were asked to give three ratings, indicating how likely they would be to behave pro-socially, how sympathetic they would feel towards the character, and how strongly the character expected their help. Participants were also asked to generate verbal responses explaining why they might choose to help the character in the given situation. All participants ﬁrst read a sheet of instructions about the task. This explained that they would see short scenarios about everyday situations and would respond verbally to questions, supplying either ratings or free responses. Par- ticipants were requested to answer as quickly and truthfully as possible. The scenarios were presented on paper, and participants were taken through an example before com- pleting the 10 experimental items. Scenarios and questions Sympathy for Character Ratings Sympathy for Character Ratings For each scenario, participants rated the degree of sym- pathy they experienced for the characters on a scale of 1–10, where higher scores indicated greater sympathy. Ratings were then summed across all 10 scenarios to create a total score for each condition (range 10–100), creating 2 scores: (1) clear-cut rule sympathy rating, (2) ambiguous rule sympathy rating. Social Expectations Task Likelihood of Pro-social Behaviour Ratings A repeated measures 2 9 2 ANOVA was conducted to examine ratings for likelihood of behaving pro-socially for all scenarios. There was one between-participant factor (AQ group: high vs low AQ) and one within-participant factor (ambiguity of social rule: clear-cut vs ambiguous). There were signiﬁcant main effects of condition, F(1,39) = 491.87, p = .0001, and of group, F(1,39) = 6.79, p = .013. The condition by group interaction was not sig- niﬁcant F(1,39) = .274, p = .604. Scoring For each scenario, participants rated the likelihood of offering to help the characters on a scale of 1–10, where 123 J Autism Dev Disord higher scores indicated greater pro-social behaviour. Ratings were then summed across all 10 scenarios to create a total score for each condition (range 10–100), creating 2 scores: (1) clear-cut rule pro-social behaviour rating, (2) ambiguous rule pro-social behaviour rating. Strength of Character Expectation Ratings For each scenario, participants rated how much they thought the characters’ expected their help on a scale of 1–10, where higher scores indicated a greater expectation to help. Ratings were then summed across all 10 scenarios to create a total score for each condition (range 10–100), creating 2 scores: (1) clear-cut rule character expectation for help rating, (2) ambiguous rule character expectation for help rating. From inspection of the mean scores presented in Table 2, it is clear that all participants were more likely to behave pro-socially when the social rule was clear-cut versus ambiguous. This is in line with the prediction that a clear-cut rule would enhance the characters’ expectation for help, and thus also the likelihood of complying with it. The high AQ group was less pro-social overall; the lack of condition of social rule by group interaction suggests that the groups were not however differentially affected by the strength of the social rule. Verbal Rationales of Societal Expectation Understanding Verbal responses were categorised according to two dimensions: rule-based or person-based rationales. The person-based rationales reﬂected responses that referred to the characters’ needs, and/or conveyed a sense of self- sacriﬁce on the part of the participants, in order to meet the characters’ needs. Rule-based rationales reﬂected re- sponses that made explicit reference to a social rule guiding an expectation to help, or implied a social rule by simply referring to the facts of the scenario. Scoring of the example scenario is shown below in Table 1. Responses could only score for one of the two dimen- sions; if both dimensions were met then the best answer was taken, and thus participants scored for person-based rationales. Sympathy for Character Ratings A repeated measures 2 9 2 ANOVA was also conducted to examine ratings for sympathy for all scenarios. There were signiﬁcant main effects of condition, F(1,39) = 356.63, p = .0001, and group, F(1,39) = 11.4, p = .002. How- ever, the condition by group interaction was not signiﬁcant F(1,39) = .486, p = .490. A repeated measures 2 9 2 ANOVA was also conducted to examine ratings for sympathy for all scenarios. There were signiﬁcant main effects of condition, F(1,39) = 356.63, p = .0001, and group, F(1,39) = 11.4, p = .002. How- ever, the condition by group interaction was not signiﬁcant F(1,39) = .486, p = .490. The mean scores suggested that all participants were more sympathetic when the rule was clear-cut versus ambiguous. The high AQ participants were less sympa- thetic towards characters overall; the lack of condition of social rule by group interaction suggests that the groups were not however differentially affected by the strength of the social rule. The responses were classiﬁed by one blind independent rater, and one rater who was not blind to group member- ship. There was an inter-rater agreement rate of 90.73 %; all disagreements were resolved between the raters via discussion. Once all responses had been classiﬁed and disagreements resolved, participants’ scores were summed across all 10 scenarios, and the percentage of person-based versus rule-based responses was calculated for both the clear-cut and ambiguous conditions. Data Analysis Means and standard deviations (SD) for each of the mea- sures are presented below in Table 2. A signiﬁcance level of .05 was adopted, with adjustment for post hoc t tests (.05/2) to control for multiple comparisons. Verbal Rationales of Societal Expectation Understanding Finally the high and low AQ groups were compared for their verbal responses outlining why one might choose to help the character in each scenario. A repeated measures 2 9 2 ANOVA was conducted to examine the percentage of rationales classiﬁed as rule-based versus person-based for the two conditions. The main effect of condition was not signiﬁcant F(1,39) = .114, p = .738, nor was the main effect of group F(1,39) = 2.161, p = .150. However, there was a signiﬁcant condition by group interaction, F(1,39) = 5.57, p = .023. Post-hoc t tests, using a strict signiﬁcance level, showed that the high AQ group used signiﬁcantly more rule-based versus person-based ratio- nales than the low AQ group in the clear-cut condition, Rule-based rationales t (1,39) = 2.327, p = .025; there was no signiﬁcant group difference in the ambiguous condition, t(1,39) = .269, p = .790. expectation for help across all scenarios. There was a signiﬁcant main effect of condition, F(1,39) = 175.41, p = .0001. The main effect of group, F(1,39) = 1.24, p = .272, and the condition by group interaction, were not signiﬁcant F(1,39) = .30, p = .864. Total response time for rationale production was also examined by summing the speed of response time across all 10 scenarios and comparing the group averages. A re- peated measure 2 9 2 ANOVA was conducted to examine the speed of response time in the clear-cut versus am- biguous condition. The main effect of condition was not signiﬁcant F(1,39) = .635, p = .431, neither was the main effect of group F(1,39) = 1.672, p = .204. As with the rationale classiﬁcation measure, there was a signiﬁcant condition by group interaction, F(1,39) = 7.632, p = .009. Post-hoc t tests, using a strict signiﬁcance level, did not reach signiﬁcance. However, inspection of the mean scores indicated a tendency for the high AQ group to be selec- tively slower in the ambiguous, t (1,39) = 1.967, p = .060, vs clear-cut condition, t(1,39) = .387, p = .701. This conﬁrms that the social rule manipulation operated as intended; all participants identiﬁed a stronger character expectation to help when the rule was clear-cut versus ambiguous. The high AQ group rated the scenarios similarly to the low AQ group identifying a stronger character expectation for help in the clear-cut versus am- biguous condition. Strength of Character Expectation Ratings A repeated measures 2 9 2 ANOVA was also conducted to examine ratings for the strength of the characters’ 12 3 J Autism Dev Disord Table 1 Scoring of example scenario from the ‘Social Expectations’ task Description of criteria Person-based rationales A response that either referred to the characters’ needs, and/or conveyed a sense of self-sacriﬁce on the participants’ part in order to meet the characters’ needs Rule-based rationales A response that made explicit reference to a social rule guiding an expectation to help, or implied a social rule by simply referring to the facts of the scenario Example responses: Clear-cut ending: ‘‘She is elderly and walking with a stick’’ Person-based rationale examples e.g. ‘‘I would feel sorry for her and it would be difﬁcult for her to stand in a crowded waiting room’’ e.g. ‘‘She needs it more than I do’’ Rule-based rationale examples e.g. ‘‘You should always offer your seat to women, elderly and the disabled’’ e.g. ‘‘She is walking with a stick’’ Ambiguous ending: ‘‘She is a young adult and is carrying a large parcel’’ Person-based rationale examples e.g. ‘‘She must be feeling very tired’’ e.g. ‘‘I think she needs it more than I do because she is carrying a large parcel’’ Rule-based rationale examples e.g. ‘‘To be polite’’ e.g. ‘‘It is common courtesy to offer your seat’’ Table 1 Scoring of example scenario from the ‘Social Expectations’ task Description of criteria Person-based rationales A response that either referred to the characters’ needs, and/or conveyed a sense of self-sacriﬁce on the participants’ part in order to meet the characters’ needs Rule-based rationales A response that made explicit reference to a social rule guiding an expectation to help, or implied a social rule by simply referring to the facts of the scenario Example responses: Clear-cut ending: ‘‘She is elderly and walking with a stick’’ Person-based rationale examples e.g. ‘‘I would feel sorry for her and it would be difﬁcult for her to stand in a crowded waiting room’’ e.g. ‘‘She needs it more than I do’’ Rule-based rationale examples e.g. ‘‘You should always offer your seat to women, elderly and the disabled’’ e.g. ‘‘She is walking with a stick’’ Ambiguous ending: ‘‘She is a young adult and is carrying a large parcel’’ Person-based rationale examples e.g. ‘‘She must be feeling very tired’’ e.g. Strength of Character Expectation Ratings ‘‘I think she needs it more than I do because she is carrying a large parcel’’ Rule-based rationale examples e.g. ‘‘To be polite’’ e.g. ‘‘It is common courtesy to offer your seat’’ Discussion The present study examined the role of social rules in guiding pro-social behaviour, and how this might be in- ﬂuenced by autistic traits. A scenario-based task was de- veloped describing everyday situations in which a character required help. Each scenario had two conditions: it ended with either a clear-cut or an ambiguous social rule 123 123 J Autism Dev Disord Table 2 Mean percentage scores and standard deviations for all measures for the ‘Social Expectations’ task Low AQ group (N = 20) M (SD) High AQ group (N = 21) M (SD) Signiﬁcance Effect Size Likelihood (%) Condition ** Gp * Gp 9 condition NS Clear-cut 86.95 (9.26) 78.42 (9.68) – 0.90 Ambiguous 55.10 (10.16) 48.05 (12.73) – 0.61 Sympathy (%) Condition Gp Gp 9 condition NS Clear-cut 80.25 (9.25) 68.81 (13.09) – 1.01 Ambiguous 43.85 (9.55) 35.00 (12.62) – 0.79 Strength of expectation (%) Condition Gp NS Gp 9 condition NS Clear-cut 76.05 (9.01) 73.24 (2.17) – 0.43 Ambiguous 54.15 (8.77) 50.76 (11.09) – 0.34 Verbal rationale classiﬁcation (%) Condition NS Gp NS Gp 9 condition Clear-cut Rule 33.00 (18.38) 49.05 (25.08) 025 0.73 Person 67.00 (18.38) 50.95 (25.08) 0.73 0.73 Ambiguous Rule 40.5 (20.64) 39.05 (13.38) – 0.08 Person 59.5 (20.64) 60.95 (13.38) – 0.08 Verbal rationale response time (s) Condition NS Gp NS Gp 9 condition Clear-cut 87.35 (24.09) 90.76 (31.61) – 0.12 Ambiguous 81.30 (20.75) 101.71 (42.53) – 0.61 Signiﬁcant at p = .01; Signiﬁcant at p = .025 Table 2 Mean percentage scores and standard deviations for all measures for the ‘Social Expectations’ task Signiﬁcant at p = .01; Signiﬁcant at p = .025 manipulation. Thus, all participants were more compliant with the expectation to help characters in the clear-cut versus ambiguous condition. For instance, in the example scenario, participants were much more likely to help when the character was elderly and walking with a stick, but far less likely to help when she was young and carrying a heavy parcel; the high AQ group gave lower likelihood of helping ratings than the low AQ group across conditions. With respect to the sympathy ratings, all participants were more sympathetic in the clear-cut versus ambiguous con- dition. The high AQ group was thus less likely to feel sympathy for the characters, regardless of the clarity of the social rule. guiding pro-social behaviour. Discussion Participants’ likelihood of complying with these societal expectations and their sym- pathy for the character was assessed using rating scales. Their ‘understanding’ of the social expectation to help was also assessed by two measures. Firstly, they indicated the strength of the characters’ expectations of help using a rating scale. Secondly, their justiﬁcation for why one would help was assessed via the generation of verbal rationales. The key ﬁnding with respect to pro-social behaviour was that, as expected, the high AQ participants were less likely to behave pro-socially overall, but the groups were not differentially affected by the strength of social rule 12 3 3 J Autism Dev Disord capabilities of individuals with ASD and/or a variety of task demands (Blair 2008). Did those with high AQ traits understand the societal expectations inherent in the scenarios? One way of esti- mating these societal expectations was by asking par- ticipants to rate the extent to which the characters expected help. Interestingly, the groups did not differ in this; the high AQ group was equally able to identify the stronger expectation to behave pro-socially in the clear-cut versus ambiguous condition. However, when asked to complete the more demanding task of providing verbal rationales outlining why one might behave pro-socially, the picture was more complex. In the clear-cut condition the low AQ group used more person-based rationales (e.g. she needs the seat more than me) than rule-based rationales reﬂecting societal expectations (e.g. you should always give up your seat for the elderly), whereas the high AQ group used these equally. In the ambiguous condition, both groups used slightly more person-based than rule-based rationales. In- spection of the rationale production response times re- vealed a condition by group interaction suggesting that the high AQ group was selectively slower for the ambiguous versus clear-cut condition when producing rationales, although this did not hold as signiﬁcant when post hoc t-tests were conducted. It has also been suggested that those with ASD may possess limited capacity to resonate emotionally with oth- ers via a self-focused stance, mediated by a heightened sense of egocentricity, at the expense of the ability to take an allocentric (other-focused) stance (Frith and de Vigne- mont 2005; Minio-Paluello et al. 2009). Discussion With respect to the present AQ study, it is possible that an imbalance in the priorities placed on self versus other needs resulted in less motivation to behave pro-socially, regardless of whether the high AQ group could correctly identify and understand the characters’ needs. The high AQ participants may therefore have focused on themselves and prioritised their own interests at the expense of helping others, and expe- riencing reduced resonance with the characters’ points of view may have compounded this. However, impairment in emotional empathy could not account for the greater tendency of the high AQ group to use rule-based rationales in the clear-cut condition when reasoning about why one should behave pro-socially, nor the differences observed in the speed of rationale produc- tion. Impairment in emotional empathy in the context of intact mentalising abilities would not be expected to affect the high AQ group’s capacity to understand how they should behave and why; rather, it should selectively in- ﬂuence their actual behaviour. Nonetheless, a possible contribution of emotional empathy cannot be dismissed, since it is difﬁcult to disentangle the relative contributions of this versus mentalising in relation to measures designed to explore everyday social behaviour. The ﬁndings of the present study corroborate those of Jameel et al. (2014), providing further evidence of reduced pro-social behaviour in individuals with high numbers of autistic traits. The two studies taken together also reveal some hint of preserved social knowledge in the high AQ group, but differences between groups in the socio-emo- tional processes thought to motivate pro-social behaviour. Various theoretical accounts associated with ASD may be relevant for explaining the pattern of ﬁndings reported in those with high AQ traits, in particular the role of emo- tional empathy, mentalising, and the potential inﬂuence of social knowledge and learning. It is also possible that non- social models such as impaired executive functioning (Hill 2004) or weak central coherence (Frith 1989, 2003) make a contribution, although they are not discussed in detail here. Impaired Mentalising? A mentalising explanation would contend that other-ori- ented pro-social behaviour is dependent upon appreciating others’ perspectives, and acting accordingly; failure to do so may have reduced the participants’ incentive to help the characters, resulting in reduced compliance with the soci- etal expectation to behave pro-socially. This explanation is consistent with some evidence in studies of those with high autistic traits suggesting differences on tasks tapping mentalising including social attention (Freeth et al. 2013) and poorer performance on false belief tasks (Best et al. 2008). Impaired Emotional Empathy? Since both emotional and cognitive components of empa- thy are thought to play a role in motivating pro-social behaviour (Eisenberg 2007), this theoretical account could explain the decreased pro-social behaviour displayed by the high AQ group. With regard to ASD, although it has been posited that this is associated with impaired menta- lising but intact emotional empathy (Blair 2008), there is some conﬂict in the literature exploring this. Some studies examining the recognition of, and response to, affective expressions have reported impairment in ASD (Humphreys et al. 2007) and others have not (Adolphs et al. 2001). This inconsistent picture may reﬂect the differing intellectual The sympathy ratings may provide the most direct measure of mentalising. Sympathy refers to feelings of concern about the welfare of others, and is thought to play a motivating role in pro-social behaviour via other-oriented processes (Decety and Michalska 2010). Whilst sympathy and empathy are often conﬂated, they are in fact distinct concepts; the experience of sympathy is said to be depen- dent upon the mentalistic ability to apprehend another’s 123 123 J Autism Dev Disord mental state, but it does not necessarily require a vicarious emotional experience (Decety and Chaminade 2003). On this basis, sympathising with characters in the social expectations task requires participants to put themselves in others’ shoes and imagine how they would feel in such a situation, and is potentially mediated by mentalising abil- ities. Lower sympathy ratings by the high AQ group are therefore consistent with the evidence of impaired men- talising in those with ASD. produce person-based rationales as often as the low AQ group. This may indicate a stylistic preference for rule- based reasoning over reliance on mentalistic processes, which may be more effortful for them. This interpretation is also supported by the ﬁnding that those with high AQ traits tended to be slower to produce rationales only in the ambiguous condition. Impaired Mentalising? A lack of salient social rules un- derpinning the scenarios may have forced them to exert more effort and employ person-based rationale, making heavier demands on mentalising ability. Some previous work has reported ‘black and white’ sympathy ratings in those with ASD, with heightened sensitivity to ‘good’ versus ‘poor’ justiﬁcations for wrongdoing (Channon et al. 2010), and greater differ- entiation between intentional and unintentional actions when assigning blame (Channon et al. 2011). No evidence of such ‘black and white’ thinking was found in the present AQ study, since sympathy ratings were lower overall in the high versus low AQ group, and the groups were not dif- ferentially affected by the strength of the social rule. However, the nature of the present study is different to those previously used, since in the Channon et al. tasks, lack of sympathy in the group with ASD related to char- acters acting for reasons that generally contravene societal, and indeed legal, expectations (e.g. drunk driving after a party, or intentionally giving a spouse an overdose of their medication). By contrast, in the present AQ study, the characters were all deserving of help and thus sympathy, regardless of the condition. This is also consistent with previous literature indicating that individuals with ASD might provide correct answers, but that the reasoning behind their judgments is often more limited (Moran et al. 2011). For instance, children with ASD have been found to show accurate moral judgements (Grant et al. 2005; Leslie et al. 2006), and intact under- standing of transgressions involving breaking social versus moral rules (Blair 1996), but to display difﬁculties in jus- tifying their choices. Zalla et al. (2009) similarly found that adults with ASD were able to use compensatory strategies to carry out social judgments regarding faux-pax, but failed to justify their responses adequately. Interpreting the ﬁndings of the present AQ study in the light of previous relevant work in ASD, it appears that individuals with high numbers of autistic traits may show some preservation of social judgement and assimilate the characters’ expectations (a mentalistic demand), by seem- ingly exploiting their knowledge of rules regarding societal norms. Further evidence supporting the use of compen- satory mechanisms within the autistic spectrum can be derived from a recent neuroimaging study that asked why some children with ASD pass classic mentalising tasks, and others do not (White et al. 2014). Impaired Mentalising? They subdivided ASD participants into groups who either failed or passed such tasks, and compared their brain activation patterns with those of typically developing children during an online mentalising task. Regardless of whether they belonged to the ‘fail’ or ‘pass’ groups, participants with ASD showed differences in brain regions associated with mentalising in comparison to typically developing children. This suggests that even individuals with ASD who pass such tasks may be doing so via an alternative route. The Role of Social Knowledge and Learning One potential caveat for an interpretation consistent with a mentalising deﬁcit is that those with high AQ traits did not differ from the low trait group on some measures assessing participants’ understanding of characters’ expectations for help. At ﬁrst glance it might appear that a mentalistic ap- praisal is required to grasp the characters’ expectations. However, routes other than mentalising may also lead to similar judgments of the characters’ expectations, such as reliance on knowledge of societal norms. More broadly, it has been suggested that individuals with ASD may not rely on intuitive socio-emotional processes for solving social and moral dilemmas (Greene and Haidt 2002). Rather, it may be a more laborious process in which individuals with ASD learn about and apply social rules, especially when these are readily available. In the present AQ study, reliance on compensatory mechanisms such as social knowledge may have cir- cumvented the need to employ mentalistic processes when estimating the characters’ expectations. However, this may be at the expense of resonating with someone emotionally, and hence this may explain why the high AQ group reported less sympathy with the characters, and were less likely to help them. Both mentalisitic abilities and emo- tional empathy play a key role in facilitating socially sensitive behaviour, and it is likely that healthy individuals use both social knowledge stores and socio-emotional Thus, in the clear-cut task condition of the current study where the social rules were more salient, the high AQ group appeared to draw upon these rules by providing more rule-based and fewer person-based justiﬁcations for the reasons surrounding why one should act pro-socially. On the other hand, when salient rules were not available in the ambiguous condition, the high AQ group was able to 123 12 3 J Autism Dev Disord symptoms may interact with the environment and disrupt everyday functioning. At present, this level of under- standing is very limited, with most work focusing on either abstract laboratory based tasks or clinical reports. Tasks such as the current one provide a rich source of material for understanding of the nature of everyday difﬁculties that those with high AQ traits, or possibly a diagnosis of ASD, may face. It could potentially be used in clinical settings to identify factors that may facilitate or impinge upon suc- cessful everyday social functioning, and to better inform interventions (Channon et al. 2014). Implications and Applications for Everyday Social Functioning The present study examined individuals drawn from the general population with high AQ traits and did not require a clinical diagnosis of ASD. One might expect those with high AQ traits to show a muted, albeit similar, pattern of social performance to those with a clinical diagnosis of ASD. Although there is no evidence of a bimodal distribution separating out clinical and non-clinical levels of impair- ments (Skuse et al. 2005), a clear dose–response relationship for degree of autistic traits has yet to be established. Whilst the pattern of ﬁndings from the present study is broadly consistent with literature examining social functioning in individuals diagnosed with ASD versus neurotypical con- trols, it would nevertheless be important to extend these measures to a clinical sample before clinical applications could be developed from these ﬁndings. This could also help to establish the validity of the assumption that there is a continuum of social ability relating to the number and severity of autistic traits, regardless of a clinical threshold. A potential limitation of the study concerns the validity and sensitivity of measures such as the Social Expectations task that attempt to tap everyday social behaviour. The strength of tasks such as this is that they use real-life-type materials to investigate performance. Real-life-type mate- rials have advantages over more traditional abstract labora- tory tasks, since they are more likely to draw upon both social and practical knowledge acquired both directly and indi- rectly through life experience, in addition to intellectual skills. Despite these advantages, however, any task that is carried out under experimental conditions with structured cues that constrain and prompt performance in a manner that can be easily measured and interpreted, has limitations with respect to real world validity. Such cues are often lacking in everyday life and their presence in experimental tests may thus lead to a lack of correspondence with real world per- formance and an underestimation of social difﬁculties outside of the laboratory (Channon et al. 2001). This is particularly pertinent when assessing individuals who share characteristics of ASD, where high-functioning individuals have been found to pass traditional mentalising tasks but to Social difﬁculties such as problems in forming or maintaining interpersonal relationships or engaging in inappropriate behaviour are often central to the everyday struggles that individuals with ASD face (Troisi 2008; Crespi and Badcock 2008). The Role of Social Knowledge and Learning The ﬁndings of the present study with respect to reduced pro-social behaviour in those with high AQ traits, highlight the need for social skills training to embrace both expanding individuals’ knowledge of social rules (e.g. you should give up your seat to someone elderly), and to target the ﬂexible appli- cation of social rules to novel situations. In individuals diagnosed with ASD an overreliance on social rules, without a deeper understanding of the social implications or caveats, may lead to awkward and inﬂexible patterns of social behaviour. Compensatory cognitive strategies can be useful in the face of reduced or absent capacity to access appropriate socio-emotional processing. However, they must be combined with enhanced social motivation; otherwise, individuals may be aware of the rules, but fail to act on them, as appears to have been the case in the present study. Social motivation is an area of research that has been neglected in ASD, and which requires further exploration. For instance, it would be interesting to study how rewards are processed by individuals with high AQ traits or ASD, and which kind of rewards might facilitate appropriate behaviour. processes to assess social scenarios and respond appro- priately. In order to make good use of social rules to deal with complex social stimuli in a range of contexts, ﬂex- ibility is required to learn about the contingencies for applying rules appropriately to different conditions (Nelson and Guyer 2011; Bunge 2004). This is likely to draw upon executive skills, which are thought to be impaired in individuals diagnosed with ASD (see Hill 2004 for a review). From a developmental perspective, as children become adolescents, and in turn adults, the complexity of the social environments they navigate increases dramati- cally. In typically developing children, this should be supported by gradually enhanced social knowledge, as a result of exposure to more challenging and ambiguous social environments. However, in children and adolescents with ASD, who tend to avoid social engagement (Richer 1976), and/or experience the social world in an ‘atypical’ fashion (Hughes and Leekam 2004), such learning may be deﬁcient. Thus, the acquisition of social rule knowledge and its use may be further protracted or stunted in those with ASD. With respect to the current study, this could account for the high AQ group’s failure to put their rela- tively preserved knowledge of social rules into practice by choosing to help the characters. Conclusion The present study found that those with high AQ traits were overall less pro-social and sympathetic towards characters in need of their help, regardless of the strength of the social rule underpinning this. However, individuals with high AQ traits were equally able to estimate char- acters’ expectations for their help. This apparent conﬂict may reﬂect a reliance on social knowledge to estimate characters’ expectations on the basis of learned societal expectations, amongst other possible contributing factors. In line with this, those with high AQ traits tended to pro- duce more rule-bound rationales to reason about why they were expected to help characters in the clear-cut condition, although not in the ambiguous condition where the social rules were less salient. This may again reﬂect a level of understanding that is more reliant on social knowledge, at the expense of utilising socio-emotional processes to en- gage directly with the characters’ needs. This reinforces and extends the ﬁndings of a previous study (Jameel et al. 2014), which indicated that those with high AQ traits demonstrated a relatively intact ability to assess characters’ perspectives in the context of reduced pro-social behaviour, and reported less personal reward for helping them. Allik, H., Larsson, J. O., & Smedje, H. (2006). Health-related quality of life in parents of school-age children with Asperger syndrome or high-functioning autism. Health and Quality of Life Out- comes, 4(1), 1–8. Andari, E., Duhamel, J. R., Zalla, T., Herbrecht, E., Leboyer, M., & Sirigu, A. (2010). Promoting social behavior with oxytocin in high-functioning autism spectrum disorders. Proceedings of the National Academy of Sciences, 107(9), 4389–4394. Baron-Cohen, S., Wheelwright, S., Skinner, R., Martin, J., & Clubley, E. (2001). The autism spectrum quotient (AQ): Evidence from Asperger syndrome/high-functioning autism, males and females, scientists and mathematicians. Journal of Autism and Develop- mental Disorders, 31(1), 5–17. Best, C. S., Moffat, V. J., Power, M. J., Owens, D. G., & Johnstone, E. C. (2008). The boundaries of the cognitive phenotype of autism: Theory of mind, central coherence and ambiguous ﬁgure perception in young people with autistic traits. Journal of Autism and Developmental Disorders, 38(5), 840–847. Blair, R. J. R. (1996). Brief report: Morality in the autistic child. Journal of Autism and Developmental Disorders, 26(5), 571–579. Blair, R. J. R. (2008). Fine cuts of empathy and the amygdala: Dissociable deﬁcits in psychopathy and autism. The Quarterly Journal of Experimental Psychology, 61(1), 157–170. Bowler, D. M. (1992). References Adolphs, R., Sears, L., & Piven, J. (2001). Abnormal processing of social information from faces in autism. Journal of Cognitive Neuroscience, 13(2), 232–240. Implications and Applications for Everyday Social Functioning In order to design effective interventions, such as social skill training, it is important to have a detailed understanding of how individuals’ 123 123 J Autism Dev Disord show difﬁculties with more open-ended and naturalistic measures (Heavey et al. 2000). It is therefore important to try to replicate these ﬁndings in more naturalistic environments, which can often be much more stimulating and inter- personally demanding. One way forward may be through the use of virtual reality paradigms. Open Access This article is distributed under the terms of the Creative Commons Attribution License which permits any use, dis- tribution, and reproduction in any medium, provided the original author(s) and the source are credited. show difﬁculties with more open-ended and naturalistic measures (Heavey et al. 2000). It is therefore important to try to replicate these ﬁndings in more naturalistic environments, which can often be much more stimulating and inter- personally demanding. One way forward may be through the use of virtual reality paradigms. Acknowledgments This work was supported by the Economic and Social Research Council [Grant Number ES/J500185/1]. Conclusion 12 3 J Autism Dev Disord skills and pro-social behavior for three children diagnosed with autism through the use of a teaching package. Research in Autism Spectrum Disorders, 3(1), 275–289. Decety, J., & Michalska, K. J. (2010). Neurodevelopmental changes in the circuits underlying empathy and sympathy from childhood to adulthood. Developmental Science, 13(6), 886–899. Leslie, A. M., Mallon, R., & DiCorcia, J. A. (2006). Transgressors, victims, and cry babies: Is basic moral judgment spared in autism? Social Neuroscience, 1(3–4), 270–283. Eisenberg, N. (2007). Empathy-related responding and prosocial behaviour. Novartis Foundation Symposium, 278, 71–80. Freeth, M., Foulsham, T., & Kingstone, A. (2013). What affects social attention? Social presence, eye contact and autistic traits. PLoS ONE, 8(1), e53286. Lin, A., Tsai, K., Rangel, A., & Adolphs, R. (2012). Reduced social preferences in autism: Evidence from charitable donations. Journal of Neurodevelopmental Disorders, 4, 8. Frith, U. (1989). Autism and ‘‘theory of mind’’. In C. Gillberg (Ed.), Diagnosis and treatment of autism (pp. 33–52). New York: Plenum Press. Meyer, J. A., Mundy, P. C., Van Hecke, A. V., & Durocher, J. S. (2006). Social attribution processes and comorbid psychiatric symptoms in children with Asperger syndrome. Autism, 10(4), 383–402. Frith, U. (2003). Autism, explaining the enigma (2nd ed.). Oxford: Blackwell. Minio-Paluello, I., Baron-Cohen, S., Avenanti, A., Walsh, V., & Aglioti, S. M. (2009). Absence of embodied empathy during pain observation in Asperger syndrome. Biological Psychiatry, 65(1), 55–62. Frith, U., & de Vignemont, F. (2005). Egocentrism, allocentrism, and Asperger syndrome. Consciousness and Cognition, 14(1), 719–738. Go¨kc¸en, E., Petrides, K. V., Hudry, K., Frederickson, N., & Smillie, L. D. (2014). Sub-threshold autism traits: The role of trait emotional intelligence and cognitive ﬂexibility. British Journal of Psychology, 105(2), 187–199. Moran, J. M., Young, L., Saxe, R., Lee, S. M., O’Young, D., Mavros, P., & Gabrieli, J. (2011). Impaired theory of mind for moral judgment in high-functioning autism. Proceedings of the National Academy of Sciences, 108(7), 2688–2692. Grant, C. M., Boucher, J., Riggs, K. J., & Grayson, A. (2005). Moral understanding in children with autism. Autism, 9(3), 317–331. Mu¨ller, E., Schuler, A., & Yates, G. B. (2008). Social challenges and supports from the perspective of individuals with Asperger syndrome and other autism spectrum disabilities. Autism, 12(2), 173–190. Greene, J., & Haidt, J. (2002). How (and where) does moral judgment work? Trends in Cognitive Sciences, 6(12), 517–523. Conclusion Happe, F., Ehlers, S., Fletcher, P., Frith, U., Johansson, M., Gillberg, C., & Frith, C. (1996). ‘Theory of mind’ in the brain. Evidence from a PET scan study of Asperger syndrome. NeuroReport, 8(1), 197–201. Nelson, E. E., & Guyer, A. E. (2011). The development of the ventral prefrontal cortex and social ﬂexibility. Developmental Cognitive Neuroscience, 1(3), 233–245. Heavey, L., Phillips, W., Baron-Cohen, S., & Rutter, M. (2000). The Awkward Moments Test: A naturalistic measure of social understanding in autism. Journal of Autism and Developmental Disorders, 30(3), 225–236. Richer, J. (1976). The social avoidance of autistic children. Animal Behavior, 24(4), 898–906. Riggio, R. E., & Reichard, R. J. (2008). The emotional and social intelligences of effective leadership: An emotional and social skill approach. Journal of Managerial Psychology, 23(2), 169–185. Hill, E. L. (2004). Evaluating the theory of executive dysfunction in autism. Developmental Review, 24(2), 189–233. Schaller, M., & Cialdini, R. B. (1988). The economics of empathic helping: Support for a mood management motive. Journal of Experimental Social Psychology, 24(2), 163–181. Hill, E. L., & Frith, U. (2003). Understanding autism: insights from mind and brain. Philosophical Transactions of the Royal Society of Biological Sciences, 358(1430), 281–289. Hughes, C., & Leekam, S. (2004). What are the links between theory of mind and social relations? Review, reﬂections and new directions for studies of typical and atypical development. Social Development, 13(4), 590–619. Skuse, D. H., Mandy, W. P., & Scourﬁeld, J. (2005). Measuring autistic traits: Heritability, reliability and validity of the social and communication disorders checklist. The British Journal of Psychiatry, 187(6), 568–572. Troisi, A. (2008). Psychiatric disorders and the social brain: Distinguishing mentalizing and empathizing. Behavioral and Brain Sciences, 31(03), 279–280. Humphreys, K., Minshew, N., Leonard, G. L., & Behrmann, M. (2007). A ﬁne-grained analysis of facial expression processing in high-functioning adults with autism. Neuropsychologia, 45(4), 685–695. White, S. J., Frith, U., Rellecke, J., Al-Noor, Z., & Gilbert, S. J. (2014). Autistic adolescents show atypical activation of the brain’s mentalizing system even without a prior history of mentalizing problems. Neuropsychologia, 56, 17–25. Izuma, K., Matsumoto, K., Camerer, C., & Adolphs, R. (2011). Insensitivity to social reputation in autism. Proceedings of the National Academy of Sciences, 108(42), 17302–17307. Zalla, T., Sav, A. M., Stopin, A., Ahade, S., & Leboyer, M. (2009). Faux pas detection and intentional action in Asperger syndrome. A replication on a French sample. Journal of Autism and Developmental Disorders, 39(2), 373–382. Conclusion ‘‘Theory of mind’’ in Asperger’s syndrome. Journal of Child Psychology and Psychiatry, 33(5), 877–893. Bunge, S. A. (2004). How we use rules to select actions: A review of evidence from cognitive neuroscience. Cognitive, Affective, & Behavioral Neuroscience, 4(4), 564–579. Overall, the current study suggests that certain aspects of social judgement may be intact in those with high AQ traits, perhaps as a result of reliance on knowledge previously ac- quired through everyday experience in the social world. However, whilst reliance upon social knowledge may be useful for navigating social situations, it may be insufﬁcient for motivating identiﬁcation with others’ needs and subse- quent pro-social behaviour. Although there is a wealth of lit- erature exploring cognitive accounts of ASD, there is very little work either in those with high AQ traits or with ASD examining how these differences translate into everyday so- cial functioning and impact on broader aspects of social in- teraction. Tasks such as that used in the present AQ study could be instrumental in providing a basis for a sensitive methodology to improve understanding of the nature of ev- eryday difﬁculties associated with autistic characteristics, and could potentially be applied to clinical populations to improve interventions such as social-skill training. Channon, S., Charman, T., Heap, J., Crawford, S., & Rios, P. (2001). Real life type problem solving in Asperger’s Syndrome. Journal of Autism and Developmental Disorders, 31(5), 461–469. Channon, S., Crawford, S., Orlowska, D., Parikh, N., & Thoma, P. (2014). Mentalising and social problem solving in adults with Asperger’s syndrome. Cognitive Neuropsychiatry, 19(2), 149–163. Channon, S., Fitzpatrick, S., Drury, H., Taylor, I., & Lagnado, D. (2010). Punishment and sympathy judgements: Is the quality of mercy strained in Asperger’s Syndrome? Journal of Autism and Developmental Disorders, 40(10), 1219–1226. Channon, S., Lagnado, D., Fitzpatrick, S., Drury, H., & Taylor, I. (2011). Judgments of cause and blame: Sensitivity to intention- ality in Asperger’s Syndrome. Journal of Autism and Develop- mental Disorders, 41(11), 1534–1542. Crespi, B., & Badcock, C. (2008). Psychosis and autism as diametrical disorders of the social brain. Behavioral and Brain Sciences, 31(3), 241–260. Crozier, S., & Tincani, M. (2007). Effects of social stories on prosocial behavior of preschool children with autism spectrum disorders. Journal of Autism and Developmental Disorders, 37(9), 1803–1814. Decety, J., & Chaminade, T. (2003). Neural correlates of feeling sympathy. Neuropsychologia, 41(2), 127–138. Conclusion Jameel, L., Vyas, K., Bellesi, G., Roberts, V., & Channon, S. (2014). Going ‘Above and Beyond’: Are those high in autistic traits less pro-social? Journal of Autism and Developmental Disorders, 44(8), 1–13. Leaf, J. B., Taubman, M., Bloomﬁeld, S., Palos-Rafuse, L., Leaf, R., McEachin, J., & Oppenheim, M. L. (2009). Increasing social 123 123
https://openalex.org/W2790763837	https://hal.archives-ouvertes.fr/hal-01705503/file/Chapter-20%20%28Aftab%20Mahesr-Majeed%20Shar-Khalil%20Memon-Hafeez%20Memon-AQ%20Tunio%29.pdf	English	null	Comparison of Klinkenberg-Corrected Gas and Liquid Permeability in Kirthar Fold Belt Tight Gas Sands	Mehran University research journal of engineering and technology	2,017	cc-by	5,852	Comparison of Klinkenberg-Corrected Gas and Liquid Permeability in Kirthar Fold Belt Tight Gas Sands Aftab Ahmed Mahesar, Khalil Rehman Memon, Hafeez-Ur-Rahman Memon, Abdul Haque Tunio Comparison of Klinkenberg-Corrected Gas and Liquid Permeability in Kirthar Fold Belt Tight Gas Sands Aftab Ahmed Mahesar, Khalil Rehman Memon, Hafeez-Ur-Rahman Memon, Abdul Haque Tunio To cite this version: Aftab Ahmed Mahesar, Khalil Rehman Memon, Hafeez-Ur-Rahman Memon, Abdul Haque Tunio. Comparison of Klinkenberg-Corrected Gas and Liquid Permeability in Kirthar Fold Belt Tight Gas Sands. Mehran University Research Journal of Engineering and Technology, 2017, 36 (4), pp.957-964. hal-01705503 Distributed under a Creative Commons Attribution 4.0 International License HAL Id: hal-01705503 https://hal.science/hal-01705503v1 Submitted on 9 Feb 2018 L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. Distributed under a Creative Commons Attribution 4.0 International License p g ( g g ) * Institute of Petroleum & Natural Gas Engineering, Mehran Universities of Engineering & Technology, Jamshor Comparison of Klinkenberg-Corrected Gas and Liquid Permeability in Kirthar Fold Belt Tight Gas Sands AFTAB AHMED MAHESAR, KHALIL REHMAN MEMON, AND HAFEEZ-UR-RAHMAN MEMON, AND ABDUL HAQ TUNIO RECEIVED ON 17.11.2016 ACCEPTED ON 21.02.2017 ABSTRACT ABSTRACT A detailed laboratory study was carried out to investigate the controls and differences between Klinkenberg- corrected gas and liquid permeability. For this reason outcrop core samples were collected from Kirthar fold belt of lower Indus basin, the plugs were horizontally taken, in cylindrical shape having dimension of 3.5-5.32 cm length and with maximum of 3.2 cm diameter. The sample porosity measurements were performed using calculations from grain volume and bulk volume method. For the purpose of comparison, slippage free gas permeability tests using nitrogen gas was measured and liquid permeability of samples was measured using brine (NaCl) of different compositions. The data obtained showed that liquid permeability was lower by an order of magnitude than the permeability of samples measured with gas. However, the gas permeability corrected for Klinkenberg effects showed difference of half an order of magnitude when compared with liquid permeability. Hence the differences in liquid permeability and gas permeability could be described by other mechanism of particles mobilization and dissolution and pore blocking phenomena. Moreover, the obtained data of gas and liquid permeability was then used to develop permeability estimation correlations. The results suggests that there is scatter in the measured values and predicted values of gas and liquid permeability data, which means that such correlations should not be used where accurate liquid permeability values of tight sandstones are needed. Permeability predicted using the existing correlations developed based on gas permeability data lead to an overestimation of permeability also the flow rates might be over predicted within such low permeability reservoirs. Key Words: Tight Gas Sands, Brine, Permeability, Stress Sensitivity Corresponding Author (E-Mail: engr.aftabmahesar@gmail.com) Corresponding Author (E-Mail: engr.aftabmahesar@gmail.com) Key Words: Tight Gas Sands, Brine, Permeability, Stress Sensitivity Mehran University Research Journal of Engineering & Technology, Volume 36, No. 4, October, 2017 [p-ISSN: 0254-7821, e-ISSN: 2413-7219] 957 esearch Journal of Engineering & Technology, Volume 36, No. 4, October, 2017 [p-ISSN: 0254-7821, e-ISSN: 2413-7219 Mehran University Research Journal of Engineering & Technology, Volume 36, No. 4, October, 2017 [p-ISSN: 02 * Institute of Petroleum & Natural Gas Engineering, Mehran Universities of Engineering & Technology, Jamshoro. 1. INTRODUCTION reservoirs; as it is difficult to obtain because it takes a lot of time to reach stabilized flow rates and pressure differentials [1-2]. Hence; if liquid permeability is needed for input into reservoir modeling, the reservoir engineers generally obtain this by making equivalent to slippage T he permeability is the main property of any reservoir rock, which controls the fluid flow and is an essential parameter for modeling reservoir production behavior. Permeability to liquid either brine or distilled water are generally not present for most of the T 957 Comparison of Klinkenberg-Corrected Gas and Liquid Permeability in Kirthar Fold Belt Tight Gas Sands corrected gas permeability values of routine core analysis [3-4]. However, gas permeability measured in laboratory always show higher value than absolute liquid permeability for the same sample [3-5]. Many of the previous studies have reported correlations between gas permeability and liquid permeability which were performed on sandstones samples [5-6]. The main difference in the pore system of samples between sandstones and other reservoirs could be the pores itself and pore; throat internal structure [2-5,7-8]. For sandstone, high to medium porosity has spherical internal pore structure with high to medium aspect ratio (pore body to throat size ratio). It is also reported that the samples with less clay minerals, clay contents have higher permeability, in which pores with low aspect ratio if filled with sand grains, and are not completely spherical, also lead to low permeability [7-9]. production from tight gas reservoirs ranges from 2 MMscfd to few thousand cubic feet [10,13]. Hence it is essential to hydraulically fracture these reservoirs and it is the most critical technique in tight-reservoirs development [14-15]. The many of the low permeability reservoirs productivity is being raised to the level of their commercial production rate by hydraulic fracturing. “A gas reservoir estimated value of average permeability if it is less than 0.1md [16-17] with initial wellhead absolute flow potential (flow at atmospheric pressure at the well head) of less than or equal to 5 MMscfd” [15-18]. Due to sufficient availability of unconventional gas resources, no resource assessment for tight gas and other non-conventional resource has been carried out in Pakistan. However, the situation is changing due to growing energy deficit [9-10,16-19]. rch Journal of Engineering & Technology, Volume 36, No. 4, October, 2017 [p-ISSN: 0254-7821, e-ISSN: 2413-7219] Mehran University Research Journal of Engineering & Technology, Volume 36, No. 4, October, 2017 [p-ISSN: 0254- 1. INTRODUCTION Gas 958 Comparison of Klinkenberg-Corrected Gas and Liquid Permeability in Kirthar Fold Belt Tight Gas Sands The purpose of this study was to assess the extent of variations in between gas and liquid permeability, that might exists either due to the gas slippage effects or could be due to other reason such as physiochemical reactions with water as a pore fluid. Moreover, the other aim was to develop a correlation by which a liquid permeability could be estimated from gas permeability values. To attain these objectives the permeability of samples collected from Kirthar fold belt outcrop were obtained, initially the permeability was measured using dry gas (nitrogen gas) and then the samples were flooded with brines of different concentrations, finally the permeability measured using demineralized water. As the nitrogen gas is non-reactive does not reacts with the samples grain surfaces and is chemically non-reactive compared with demineralized water (distilled water). The measured liquid and gas permeability corrected for slippage effects was then compared each other [10-21]. The purpose of this study was to assess the extent of variations in between gas and liquid permeability, that might exists either due to the gas slippage effects or could be due to other reason such as physiochemical reactions with water as a pore fluid. Moreover, the other aim was to develop a correlation by which a liquid permeability could be estimated from gas permeability values. To attain these objectives the permeability of samples collected from Kirthar fold belt outcrop were obtained, initially the permeability was measured using dry gas (nitrogen gas) and then the samples were flooded with brines of different concentrations, finally the permeability measured using demineralized water. As the nitrogen gas is non-reactive does not reacts with the samples grain surfaces and is chemically non-reactive compared with demineralized water (distilled water). The measured liquid and gas permeability corrected for slippage effects was then compared each other [10-21]. well rounded to sub rounded. Marl is very thin-bedded light grey and is laminated finely. Mudstone is bio-turbated and is brownish reported two depositional systems. One is shallow marine and other one is fluviodeltaic to deep marine turbidites. There were three facies were recognized as transitional proximal to distal settings, as the deep shelf facies and shore face associations in the North depositional system [20,23-24]. 1. INTRODUCTION Gas is the dominant component of conventional energy supply of the country, which is projected to decline from 4 BCFD in 2008 to 1.6 BCFD in 2022 giving rise to deficit of 7.2 BCFD, which is almost twice the current production level. This demand is not expected to be met by the conventional gas plays, and has necessitated the need to look for non- conventional indigenous gas resources such as tight gas. This study aims to investigate permeability of tight gas sands so that the marginally profitable reservoirs productivity can accurately be assessed, so that the uncertainties can be decreased in commercial extraction of these readily available gas resources [15,19-20]. The objective of this study was to develop correlation in between gas and liquid permeability values based on the data obtained in laboratory for the tight rocks of Kirthar fold belt area Sindh Pakistan. The core samples in fact are the ground reality in understanding and exploitation of hydrocarbon resources [9-11]. The petro physical properties and geological attributes of rocks encompass the operator to realize the production potential of a reservoir and map out the best scenarios for optimal field development. Moreover; the authors [11-12] reported that the influence of confining stress related to permeability of few tight sand samples was also studied.The permeability stress sensitivity results for these samples will be useful for providing a sound basis of the Kirthar fold tight gas potential development [13-14]. This study aims to investigate permeability of tight gas sands so that the marginally profitable reservoirs productivity can accurately be assessed, so that the uncertainties can be decreased in commercial extraction of these readily available gas resources [15,19-20]. Gas in excess of 40TCF (Trillion Cubic Feet) might be lying un-tapped in the TGR (Tight Gas Reservoirs) of Pakistan based on exploration activities carried out in the mature basins of Pakistan and is expected that there might be substantial rise in gas resources . Based on the drilling and preliminary assessment of the industry, there are substantial tight gas reservoirs in the existing plays of Pakistan, summarized in Table 1 [16,20-21]. The tight gas potential of Pakistan tight gas reservoirs is very challenging technically and commercially [10]. Hence, for this reason it is essential to perform detailed petro physical analysis and to produce gas from such low permeability reservoirs at commercial rates. 3. (1) The plugs were taken horizontally to the bedding section of samples obtained, near Kirthar fold belt outcrop in district Dadu Sindh, Pakistan. About 3.5-5.90 cm length and with maximum diameter of 2.8 cm cylindrical shape plug samples were taken. (1) The plugs were taken horizontally to the bedding section of samples obtained, near Kirthar fold belt outcrop in district Dadu Sindh, Pakistan. About 3.5-5.90 cm length and with maximum diameter of 2.8 cm cylindrical shape plug samples were taken. (2) After plugging the samples were cleaned in a Soxhlet extractor for removal of any of the contaminated fluids and residues. The samples were then placed into oven for about 48 hours for drying purposes. 2. The samples collection location is specified with Fig. 1 of Kirthar Fold Belt and is located in (West Pakistan Fold Belt) [10]. This imparts a good formation (outcrop) and is realized to have good potential of hydrocarbon reserves as well as having good trapping mechanism [10,21-22]. Kirthar Fold Belt, the Pab Formation is part of sedimentary succession. The thick marine siliciclastic pab formations of Kirthar fold ranges from 50-450m. These mostly consist of sandstone of inter bedded with marl and mudstones. The sandstone is yellow, grey light brown and greenish, which is medium to coarse gained moderate well sorted, The samples collection location is specified with Fig. 1 of Kirthar Fold Belt and is located in (West Pakistan Fold Belt) [10]. This imparts a good formation (outcrop) and is realized to have good potential of hydrocarbon reserves as well as having good trapping mechanism [10,21-22]. FIG. 1. THE STUDY AREA REPRESENTED IN THIS MAP [19] TABLE 1. EXISTING TIGHT GAS POTENTIAL WITHIN THE PAKISTAN [19] esearch Journal of Engineering & Technology, Volume 36, No. 4, October, 2017 [p-ISSN: 0254-7821, e-ISSN: 2413-7219 TABLE 1. EXISTING TIGHT GAS POTENTIAL WITHIN THE PAKISTAN [19] Mehran University Research Journal of Engineering & Technology, nis a B k c ol B stc e p s o r P rio v re s e R )fc b ( P II G s u d n I eld di M 8 1 1 7 0 0 0 5 1 tle b dlo F n a m ialu S 4 7 - 0 0 0 9 1 tle b dlo F ra htri K 7 8 2 0 0 3 7 lato T 9 1 6 2 9 0 0 3 1 4 TABLE 1. EXISTING TIGHT GAS POTENTIAL WITHIN THE PAKISTAN [19] nis a B k c ol B stc e p s o r P rio v re s e R )fc b ( P II G s u d n I eld di M 8 1 1 7 0 0 0 5 1 tle b dlo F n a m ialu S 4 7 - 0 0 0 9 1 tle b dlo F ra htri K 7 8 2 0 0 3 7 lato T 9 1 6 2 9 0 0 3 1 4 FIG. 1. THE STUDY AREA REPRESENTED IN THIS MAP [19] ng & Technology, Volume 36, No. 4, October, 2017 [p-ISSN: 0254-7821, e-ISSN: 2413-7219] 959 Comparison of Klinkenberg-Corrected Gas and Liquid Permeability in Kirthar Fold Belt Tight Gas Sands (3) The measurements were performed using helium gas as a pore fluid for gas permeability measurement and (NaCl) brine of 5%. (3) The measurements were performed using helium gas as a pore fluid for gas permeability measurement and (NaCl) brine of 5%. (3) The measurements were performed using helium gas as a pore fluid for gas permeability measurement and (NaCl) brine of 5%. experimental setup is shown in Fig. 2; few samples of very low permeability were measured with pulse decay experimental setup of Core lab. The pulse decay experimental set-up outline is also presented in Fig. 3. The permeability was calculated using following Equation (1): (4) The porosity measurements were performed by calculations, from GV (Grain Volume) and BV (Bulk Volume). The BV, was determined by measuring the sample diameter and length with caliper as well as weighing the water saturated sample while immersing into water using Archimedes principle. GVs of sample plugs were obtained using a helium porosimeter. Mehran University Research Journal of Engineering & Technology, Volume 36, No. 4, October, 2017 [p-ISSN: 0254- TABLE 1. EXISTING TIGHT GAS POTENTIAL WITHIN THE PAKISTAN [19] However; the BV calculations were performed using the digital Vernier caliper by determining the dimensions of the samples. Around the three measurements of lengths and diameters were taken for accuracy and minimization of error in calculated values. (4) ( ) 2 2 2 1 2 P P A P Q L k s a a s a − = μ (2) (2) Where μ is the viscosity of the fluid flows across sample; L is the length of the samples in centimeters, Q is the flow rate in cc/sec, P is the pressure in atm and A is the cross- sectional area of sample in cm2. Where μ is the viscosity of the fluid flows across sample; L is the length of the samples in centimeters, Q is the flow rate in cc/sec, P is the pressure in atm and A is the cross- sectional area of sample in cm2. The liquid permeability was measured using brine of NaCl, experiments were performed by flowing brine of 5% NaCl FIG. 2. DIAGRAM SHOWS THE STEADY STATE GAS PERMEABILITY SETUP FIG. 3. PULSE-DECAY APPARATUS FOR UNSTEADY-STATE PERMEABILITY MEASUREMENT TO GAS [31] FIG. 2. DIAGRAM SHOWS THE STEADY STATE GAS PERMEABILITY SETUP FIG. 2. DIAGRAM SHOWS THE STEADY STATE GAS PERMEABILITY SETUP The porosity was determined using following Equation (1): BV GV BV − = φ (1) (1) Where φ is porosity in percentage or in fractions, BVand GV. FIG. 2. DIAGRAM SHOWS THE STEADY STATE GAS PERMEABILITY SETUP (1) For brine permeability measurements (NaCl) brine solutions of 5%was prepared in the laboratory. These brines were prepared based on reported formation waters. The brine solutions were filtered by using 0.40 mm filter paper and the brine were degassed in a vacuum apparatus for removing the air prior to use for permeability measurement. Viscosity of the brine samples was measured separately using the Rolling ball viscometer. For each core samples the steady state permeability was obtained using four flow rates, calculations were followed by the application of Darcy’s equation. The steady state gas permeability (1) FIG. 3. PULSE-DECAY APPARATUS FOR UNSTEADY-STATE PERMEABILITY MEASUREMENT TO GAS [31] FIG. 3. PULSE-DECAY APPARATUS FOR UNSTEADY-STATE PERMEABILITY MEASUREMENT TO GAS [31] 960 Comparison of Klinkenberg-Corrected Gas and Liquid Permeability in Kirthar Fold Belt Tight Gas Sands across the samples. The pressure drop at each flow rate was recorded. TABLE 2. BASIC PETROPHYSICAL PROPERTIES OF KIRTHAR FOLD BELT SAMPLES elp m a S ) D m ( g K b K b K /g K 1 T K 2 0.1 9 8.0 6 3.0 2 T K 3 4.5 6 5.2 7 2.2 3 T K 6 6.3 0 6.1 9 2.2 4 T K 9 8.6 4 0.5 7 3.1 TABLE 2. BASIC PETROPHYSICAL PROPERTIES OF KIRTHAR FOLD BELT SAMPLES FIG. 5. PERMEABILITY VERSUS POROSITY RELATION FOR ALL SAMPLES STUDIED ( ) 2 1 P P A L Q K − = μ (3) ( ) 2 1 P P A L Q K − = μ (3) elp m a S ) D m ( g K b K b K /g K 1 T K 2 0.1 9 8.0 6 3.0 2 T K 3 4.5 6 5.2 7 2.2 3 T K 6 6.3 0 6.1 9 2.2 4 T K 9 8.6 4 0.5 7 3.1 elp m a S ) D m ( g K b K b K /g K 1 T K 2 0.1 9 8.0 6 3.0 2 T K 3 4.5 6 5.2 7 2.2 3 T K 6 6.3 0 6.1 9 2.2 4 T K 9 8.6 4 0.5 7 3.1 (3) Where Q is the flow in cc/sec, A is the cross-sectional area of the sample in cm2, μ is the viscosity of sample and P1 and P2 are the upstream and downstream pressure in atm. TABLE 1. EXISTING TIGHT GAS POTENTIAL WITHIN THE PAKISTAN [19] A graph between pressure drop across samples versus flow rate was developed for each core plug to make sure that intersect at zero (Fig. 4).The permeability measurements were performed at ambient temperature, and viscosity was calculated at the measured temperature. Following Equation (3) was used for liquid permeability calculations: Fig. 6 provides the results from permeability measured at different net confining stresses. The experiments initially performed at lower confining stress of 500 psi and gradually increased to high confining stress upto 2500 psi. The geometric mean of the permeability to gas is 0.209mDarcy and for brine permeability is 0.080mDarcy (Table 2). 4. RESULTS AND DISCUSSION The low permeability samples would be more affected by layer of clay bound water, as clay swelling occur, which results in lessening of size of the pore throat resulting in reduced area for flow of fluid, consequently reduction in permeability [25,27-30]. The relation of brine and gas permeability of tight gas sands developed is shown in Fig. 4. Utilizing the permeability data from both gas and liquid a general equation was derived (Equation (4)), which shows the relation between two measurements i.e. the gas permeability and brine permeability (Fig. 6). Moreover; slippage-corrected gas permeability and liquid permeability data were plotted on a log–log plot and a power regression was employed to establish a correlation (Fig. 6).The results showed a difference of half an order magnitude and the brine permeability was 5-10 times less than gas when measured on the same core plug. The plotted data revealed a direct proportionality relationship (Fig. 6). Slippage corrected gas permeability plotted against liquid had shown that the gas permeability is high compared to liquid permeability. The samples with lower permeability values have shown more reduction when measured with liquid permeability compared to gas permeability. The other possible reason could be the sample dimensions. If length of the core sample is less than the diameter of samples then it will disobey Darcy’s assumption [23-26]. Such differences in permeability values provide the relative control of fluid flow within tight sand reservoirs. The dissimilarity observed in permeability of sandstones using different pore fluids could be the physiochemical reactions between water and clay mineralogy .It was assumed that the presence of clay mineralogy could be the reason of reduced liquid permeability, as it can result in formation of layer of clay bound water on the grain surfaces of samples. The low permeability samples would be more affected by layer of clay bound water, as clay swelling occur, which results in lessening of size of the pore throat resulting in reduced area for flow of fluid, consequently reduction in permeability [25,27-30]. 4.2 The measured permeability of tight gas samples from all plugs was plotted against net confining stress and is shown in Fig. 7. In this plot, ‘y’ axes correspond to the permeability measured at 2500 psi confining pressure and the ‘x’ axes is permeability which is measured at net confining pressure of 500psi. It can be seen from the data plotted in Fig. 7 that there is a greater degree of permeability reduction with low permeability cores than samples having high permeability. In core samples with initial absolute gas permeability less than <0.1mDarcy, the permeability in those samples significantly decreased at high net confining pressure of 2500psi. This behavior could be compared with other tight gas formations, which exhibit permeability, lower than 0.1 mDarcy. The reduction in permeability observed in present study is also similar to that of the [26-30]. The trend line black colored on Fig. 8 is the power law fit of the data points on the log-log plot, which also shows that the permeability with less than 1mD is more sensitive to confining stress. However; there is a moderate impact on brine permeability reduction was observed compared to the gas permeability of the samples studied. Reduction in permeability due to stress increase is reported in Fig. 7. The reduction in permeability in tight gas samples from Kirthar fold belt cannot be generalized for all types of low permeability reservoirs, as every reservoir will let fluid flow differently based on their composition and mineral contents as well depositional environment. By increased confining stress, the permeability of these samples showed reduction particularly for samples having very low absolute initially permeability. More work is required to provide reasons for reduction in permeability due to stress increase. Based on the gas and permeability data a correlation was established Equation (4) to predict liquid permeability. The fitted results are also shown in Fig. 6. Kliquid = 0.479K1.14 (R2 = 0.92) (4) (4) 4. RESULTS AND DISCUSSION 4. In this study permeability measurements were made on samples supplied from Kirthar fold belt tight gas sands, with results presented in Table 2. Porosity-permeability results as a cross-plot are shown in Fig. 5. Presenting a type I and II rock facies. In addition to one of the sample permeability was circled as outlier. This sample could be fractured however, it is recommended to perform micro scanning using electronic microscope by preparing thin sections. FIG. 5. PERMEABILITY VERSUS POROSITY RELATION FOR ALL SAMPLES STUDIED Mehran University Research Journal of Engineering & Technology, Volume 36, No. 4, October, 2017 [p-ISSN: 0254-7821, e-ISSN: 2413-7219] 961 FIG.. 4. A TYPICAL PRESSURE DROP VERSUS FLOW RATE FOR ONE OF THE SAMPLE FIG. 6. GAS AND LIQUID PERMEABILITY DATA OF KIRTHAR FOLD BELT SINDH PAKISTAN FIG. 6. GAS AND LIQUID PERMEABILITY DATA OF KIRTHAR FOLD BELT SINDH PAKISTAN Mehran University Research Journal of Engineering & Technology, Volume 36, No. 4, October, 2017 [p-ISSN: 0254-7821, e-ISSN: 2413-7219] 961 Mehran University Research Journal of Engineering & Technology, Volume 36, No. 4, October, 2017 [p-ISSN: 0254-7821, e-ISSN: 2413-7219] 961 961 Comparison of Klinkenberg-Corrected Gas and Liquid Permeability in Kirthar Fold Belt Tight Gas Sands Moreover; slippage-corrected gas permeability and liquid permeability data were plotted on a log–log plot and a power regression was employed to establish a correlation (Fig. 6).The results showed a difference of half an order magnitude and the brine permeability was 5-10 times less than gas when measured on the same core plug. The plotted data revealed a direct proportionality relationship (Fig. 6). Slippage corrected gas permeability plotted against liquid had shown that the gas permeability is high compared to liquid permeability. The samples with lower permeability values have shown more reduction when measured with liquid permeability compared to gas permeability. The other possible reason could be the sample dimensions. If length of the core sample is less than the diameter of samples then it will disobey Darcy’s assumption [23-26]. Such differences in permeability values provide the relative control of fluid flow within tight sand reservoirs. The dissimilarity observed in permeability of sandstones using different pore fluids could be the physiochemical reactions between water and clay mineralogy .It was assumed that the presence of clay mineralogy could be the reason of reduced liquid permeability, as it can result in formation of layer of clay bound water on the grain surfaces of samples. rch Journal of Engineering & Technology, Volume 36, No. 4, October, 2017 [p-ISSN: 0254-7821, e-ISSN: 2413-7219] Mehran University Research Journal of Engineering & Technology, Volume 36, No. 4, October, 2017 [p-ISSN: 0254-7821, e-ISSN: 2413-7219] 962 4.1 The correlations were validated based on the data of core samples measured in the laboratory. Various samples from the Lower Indus basin of Kirthar fold belt formation were used for the development of the correlation. 962 Comparison of Klinkenberg-Corrected Gas and Liquid Permeability in Kirthar Fold Belt Tight Gas Sands FIG. 7. PLOT IS THE GASPERMEABILITY OFFIVE SAMPLES MEASURED UNDER STRESS CONDITIONS FIG. 8. PERMEABILITY RESULTS PLOTTED AT AMBIENT STRESS OF 500 VERSUS IN-SITU STRESS OF 2500PSI OVERBURDEN STRESS FIG. 7. PLOT IS THE GASPERMEABILITY OFFIVE SAMPLES MEASURED UNDER STRESS CONDITIONS it was found that there was 0.50% error which seems to be negligible. it was found that there was 0.50% error which seems to be negligible. (iv) (iv) Permeability of samples measured showed sensitivity to stress; the less permeable samples were more sensitive to stress particularly those samples which were having less than 0.1 mDarcy absolute permeability. For further stress sensitivity analysis it is recommended to run more permeability experiments on several different samples at different stress conditions. (iv) Permeability of samples measured showed sensitivity to stress; the less permeable samples were more sensitive to stress particularly those samples which were having less than 0.1 mDarcy absolute permeability. For further stress sensitivity analysis it is recommended to run more permeability experiments on several different samples at different stress conditions. FIG. 7. PLOT IS THE GASPERMEABILITY OFFIVE SAMPLES MEASURED UNDER STRESS CONDITIONS FIG. 7. PLOT IS THE GASPERMEABILITY OFFIVE SAMPLES MEASURED UNDER STRESS CONDITIONS ACKNOWLEDGEMENTS FIG. 8. PERMEABILITY RESULTS PLOTTED AT AMBIENT STRESS OF 500 VERSUS IN-SITU STRESS OF 2500PSI OVERBURDEN STRESS FIG. 8. PERMEABILITY RESULTS PLOTTED AT AMBIENT STRESS OF 500 VERSUS IN-SITU STRESS OF 2500PSI OVERBURDEN STRESS The authors acknowledge support from number of people who made this manuscript possible including: parents, teachers, and friends. Authors are also indebted to their parental institutes, Mehran University of Engineering & Technology, Jamshoro, Pakistan, who provided the opportunity to produce this paper. REFERENCES [1] Al-Muthane, A.S., “Best Practices in Conventional Core Analysis a Laboratory Investigation”, Society of Petroleum Engineers, Section Technical Symposium, Alkhobar, Saudi Arabia, 10-12 May, 2008. [1] Al-Muthane, A.S., “Best Practices in Conventional Core Analysis a Laboratory Investigation”, Society of Petroleum Engineers, Section Technical Symposium, Alkhobar, Saudi Arabia, 10-12 May, 2008. [2] [2] Al-Jabri, R.A., Al-Maamari, R.S., and Wilson, O.B., “Klinkenberg-Corrected Gas Permeability Correlation for Shuaiba Carbonate Formation”, Journal of Petroleum Science & Engineering, Volume 131, pp. 172-176, 2015. [3] Amthor, J., Kerans, C., and Gautheir, P., “Reservoir Characterization of a Shu’aiba Carbonate Ramp-Margin Field, Northern Oman”, Barremian-Aptian Stratigraphy and Hydrocarbon Habitat of the Eastern Arabian Plate, 2010. [2] Al-Jabri, R.A., Al-Maamari, R.S., and Wilson, O.B., “Klinkenberg-Corrected Gas Permeability Correlation for Shuaiba Carbonate Formation”, Journal of Petroleum Science & Engineering, Volume 131, pp. 172-176, 2015. Mehran University Research Journal of Engineering & Technology, Volume 36, No. 4, October, 2017 CONCLUSIONS [23] [9] [9] Raza, H.A., Ahmed, R., Ali, S., Sheikh, A., and Shafique, N., “Exploration Performance in Sedimentary Zones of Pakistan”, Pakistan Journal Hydrocarbon Research, Volume 1, pp. 1-7, 1989. [23] Loznyuk, O., Surtaev, V., Sakhan, A., Murtazin, R., Latkin, K., Sitdikov, S., Pestrikov, A., Gusakov, V., Politov, M., and Yudin, A., “A Multistage Stimulation Operation to Unlock the Gas Potential of the Turonian Siltstone Formation in Western Siberia”, Proceedings Conference on Society of Petroleum Engineers Russian Petroleum Technology. [10] Kadri, I., “Petroleum Geology of Pakistan: Pakistan Petroleum Limited”, Karachi, Pakistan, 1995. [11] Brace, W.F., Walsh, J., and Frangos, W., “Permeability of Granite Under High Pressure”, Journal of Geophysical Research, Volume 73, No. 6, pp. 2225-2236, 1968 [24] Civan, F., Rai, C.S., and Sondergeld, C.H., “Shale-Gas Permeability and Diffusivity Inferred by Improved Formulation of Relevant Retention and Transport Mechanisms”, Transport in Porous Media, Volume 86, No. 3, pp. 925-944, 2011. [12] Coward, M.P., and Johnson, H., “Structural Geology in Reservoir Characterization”, 1998 [13] Zhang, M., Takeda, M., Esaki, T., Takahashi, M., and Endo, H., “Effects of Confining Pressure on Gas and Water Permeabilities of Rocks”, Proceedings Materials Research Society Symposium, Cambridge University Press, pp. 851-860, 2002 [25] [25] Lushev, M., Markin, M., Dubnitskiy, I., and Vorobyev, V., “Determining Methods of Static Mechanical Properties of Poorly Consolidated Sand-Rocks (by the Example of the Yuzhno-Russkoye Field)”, Proceedings of Confreence on Society of Petroleum Engineers Russian Petroleum Technology, Society of Petroleum Journal, 2015. [14] Shar, A.M., and Mahesar, A.A., “Pakistan’s Kirthar Fold Belt Show Development Potential”, Oil and Gas Journal, Volume 114, No. 12, 2016. [26] [26] Mulkamanov, A., Dorofeev, A., Vorobyev, V., Bolshakova, A., Chupeev, A., Sauve, R., Tonkin, T., and Vernus, J., “Integrating Asset Modelling for Strategic Gas Field Development Planning and Short-Term Optimization”, Proceedings of Conference on Society of Petroleum EngineersRussian Petroleum Technology, Society of Petroleum Journal, 2015. [15] [15] Unconventional Resources in Pakistan, Internal Documents of Ministry of Petroleum and Natural Resources, 2012. [16] Jones, F.O., and Owens, W., “A Laboratory Study of Low-Permeability Gas Sands”, Journal of Petroleum Technology, Volume 32, No. 9, pp. 1,631-631, 980. [27] Sampath, K., and Keighin, C.W., “Factors Affecting Gas Slippage in Tight Sandstones of Cretaceous Age in the Uinta Basin”, Journal of Petroleum Technology, Volume 34, No. 11, pp. 2,715-712,720,1982. CONCLUSIONS The gas and liquid permeability of samples collected from Kirthar fold belt conducted using different pore fluids (e.g. NaCl brines of different composition and distilled water). The results were compared and the following conclusions were drawn: [3] Amthor, J., Kerans, C., and Gautheir, P., “Reservoir Characterization of a Shu’aiba Carbonate Ramp-Margin Field, Northern Oman”, Barremian-Aptian Stratigraphy and Hydrocarbon Habitat of the Eastern Arabian Plate, 2010. (i) Gas permeability is higher than brine permeability’s gas permeability values are around 5-10 times those values measured by distilled water. [4] Bloomfield, J., and Williams, A., “An Empirical Liquid Permeability-Gas Permeability Correlation for Use in Aquifer Properties Studies”, Quarterly Journal of Engineering Geology and Hydrogeology, Volume 28 (Supplement-2), pp. S143-S150, 1995. (ii) (ii) There was around half an order of magnitude difference in gas and brine permeability of the samples studied from Kirthar fold belt sands. The permeability measured with brine was lower, particularly for those samples, which have less than 0.1mDarcy permeability. [5] Macary, S.M., “Conversion of Air Permeability to Liquid Permeabilities Extracts Huge Source of Information for Reservoir Studies”, Proceedings of Conference on Petroleum Science and Engineering, Middle East Oil Show, pp. 65-71, 1999. [6] Ambrose, R.J., Hartman, R.C., Diaz Campos, M., Akkutlu, I.Y., and Sondergeld, C., “New Pore-Scale Considerations for Shale Gas in Place Calculations”, Proceedings of Conference on Society of Petroleum Engineers Unconventional Gas, 2012. (iii) The equation derived for liquid permeability estimation, and the validity was tested utilizing the existing data sets of the measured permeability that were available in our database, (iii) The equation derived for liquid permeability estimation, and the validity was tested utilizing the existing data sets of the measured permeability that were available in our database, (iii) 963 Comparison of Klinkenberg-Corrected Gas and Liquid Permeability in Kirthar Fold Belt Tight Gas Sands [7] [7] Pugh, V.J., “Correlations of Liquid anc Air Permeabilities for Use in Reservoir Engineering Studies”, Log Analyst, 1991. [21] Viqar-un-nisa, Q.S., “Hydrocarbon Prospects of Southern Indus Basin, Pakistan”, AAPG Bulletin, Volume 70, No 6, pp. 730-747, 1986. [21] [22] [22] Kawata, Y., and Fujita, K., “Some Predictions of Possible Unconventional Hydrocarbons Availability Until 2100”, Proceedings of Society of Petroleum Engineers Asia Pacific Oil and Gas Conference and Exhibition, 2003 [8] Tanikawa, W., and Shimamoto, T., “Klinkenberg Effect for Gas Permeability and its Comparison to Water Permeability for Porous Sedimentary Rocks”, Hydro Earth System, Science, pp. 1315-1338, 2006. Mehran University Research Journal of Engineering & Technology, Volume 36, No. 4, October, 201 rch Journal of Engineering & Technology, Volume 36, No. 4, October, 2017 [p-ISSN: 0254-7821, e-ISSN: 2413-7219] CONCLUSIONS [17] [17] McDougall, J.W., and Hussain, A., “Fold and Thrust Propagation in the Western Himalaya Based on a Balanced Cross Section of the Surghar Range and Kohat Plateau, Pakistan (1)”, AAPG Bulletin, Volume 75, No. 3, pp. 463-478, 1991. [28] [28] Faulkner, D.R., and Rutter, E.H., “Comparisons of Water and Argon Permeability in Natural Clay-Bearing Fault Gouge under High Pressure at 20 Degrees C”, Journal of Geophysical. Research, Volume 105, No. 16, pp. 415-16 426, 2000. [18] [18] Heid, J., McMahon, J., Nielsen, R., and Yuster, S., “Study of the Permeability of Rocks to Homogeneous Fluids”, Proceedings of Drilling and Production Practice, American Petroleum Institute,1950. [29] [29] Jones, S.C., “A Rapid Accurate Unsteady – State Klinkenberg Parameter”, Society of Petroleum Engineers Journal, pp. 383-397, 1972. [19] [19] DGPC (Directorate General of Petroleum Concessions) Pakistan, Petroleum, Exploration and Production Policy, Ministry of Petroleum and Natural Resources, Islamabad, Pakistan, (Accessed, 27 April, 2011). [30] Klinkenberg, L.J., “The Permeability of Porous Media to Liquids and Gases”, American Petroleum Institute, Drilling and Productions Practices, pp. 200-213, 1941. [20] [20] Khan, A., Kelling, G., Umar, M., and Kassi, A., “Depositional Environments and Reservoir Assessment of Late Cretaceous Sandstones in the South Central Kirthar Foldbelt, Pakistan”, Journal of Petroleum Geology, Volume 25, No. 4, pp. 373-406, 2002. [31] [31] James, S.C., “A Technique for Faster Pulse-Dacay Permeability Measurements in Tight Rocks”, Society of Petroleum Journal, pp. 19-26, 1997. 964
https://openalex.org/W2552335180	https://publications.anl.gov/anlpubs/2016/08/128763.pdf	English	null	Investigation of Aerodynamic Interference between Twin Deck Bridges	null	2,016	cc-by	18,750	ANL/ESD-16/15 DOCUMENT AVAILABILITY Online Access: U.S. Department of Energy (DOE) reports produced after 1991 and a growing number of pre-1991 documents are available free via DOE’s SciTech Connect (http://www.osti.gov/scitech/) Reports not in digital format may be purchased by the public from the National Technical Information Service (NTIS): U.S. Department of Commerce National Technical Information Service 5301 Shawnee Rd Alexandra, VA 22312 www.ntis.gov Phone: (800) 553-NTIS (6847) or (703) 605-6000 Fax: (703) 605-6900 Email: orders@ntis.gov Reports not in digital format may be purchased by the public from the National Technical Information Service (NTIS): U.S. Department of Commerce National Technical Information Service 5301 Shawnee Rd Alexandra, VA 22312 www.ntis.gov Phone: (800) 553-NTIS (6847) or (703) 605-6000 Fax: (703) 605-6900 Email: orders@ntis.gov Reports not in digital format are available to DOE and DOE contractors from the Office of Scientific and Technical Information (OSTI): U.S. Department of Energy Office of Scientific and Technical Information P.O. Box 62 Oak Ridge, TN 37831-0062 www.osti.gov Phone: (865) 576-8401 Fax: (865) 576-5728 Email: reports@osti.gov Reports not in digital format are available to DOE and DOE contractors from the Office of Scientific and Technical Information (OSTI): U.S. Department of Energy Office of Scientific and Technical Information P.O. Box 62 Oak Ridge, TN 37831-0062 www.osti.gov Phone: (865) 576-8401 Fax: (865) 576-5728 Email: reports@osti.gov Office of Scientific and Technical Information U.S. Department of Energy Office of Scientific and Technical Information P.O. Box 62 Oak Ridge, TN 37831-0062 www.osti.gov Phone: (865) 576-8401 Fax: (865) 576-5728 Email: reports@osti.gov Office of Scientific and Technical Information U.S. Department of Energy Office of Scientific and Technical Information P.O. Box 62 Oak Ridge, TN 37831-0062 www.osti.gov Phone: (865) 576-8401 Fax: (865) 576-5728 Email: reports@osti.gov Disclaimer This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor UChicago Argonne, LLC, nor any of their employees or officers, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. Investigation of Aerodynamic Interference between Twin Deck Bridges. May 2016 May 2016 About Argonne National Laboratory Argonne is a U.S. Department of Energy laboratory managed by UChicago Argonne, LLC under contract DE-AC02-06CH11357. The Laboratory’s main facility is outside Chicago, at 9700 South Cass Avenue, Argonne, Illinois 60439. For information about Argonne and its pioneering science and technology programs, see www.anl.gov. About Argonne National Laboratory DOCUMENT AVAILABILITY The views and opinions of document authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof, Argonne National Laboratory, or UChicago Argonne, LLC. Disclaimer Thi ANL/ESD-16/15 by M.A. Sitek, C. Bojanowski, and S.A. Lottes Transportation Research and Analysis Computing Center (TRACC) Energy Systems Division, Argonne National Laboratory Investigation of Aerodynamic Interference between Twin Deck Bridges Page I May 2016 Table of Contents 1. Introduction ............................................................................................................................. 1 1.1. Background .......................................................................................................................... 1 1.2. Proposed analysis methods ................................................................................................ 3 2. CFD model ............................................................................................................................... 4 2.1. General information ........................................................................................................... 4 2.2. Mesh study .......................................................................................................................... 5 3. Analysis of the flow around a single deck bridge ...................................................................10 4. Analysis of the flow around a twin deck bridge ...................................................................... 17 4.1. Pressure acting on a twin deck bridge ............................................................................... 17 4.2. Influence of gap-to-width ratio ......................................................................................... 27 4.3. Influence of varying angle of attack .................................................................................. 32 5. Dynamic response ................................................................................................................. 38 5.1. Configuration 1: both decks suspended on springs ......................................................... 39 5.2. Configuration 2: the upstream deck is suspended on springs, the downstream deck is constrained ....................................................................................................................... 48 5.3. Configuration 3: the upstream deck is constrained, the downstream deck is suspended on springs .......................................................................................................................... 50 5.4. Comparison of results for configurations 1, 2, and 3 ......................................................... 51 6. Validation of the CFD modeling approach ............................................................................ 54 7. Conclusions ........................................................................................................................... 60 8. Acknowledgements ................................................................................................................ 62 9. References ............................................................................................................................. 62 Page I Page I List of Figures List of Figures List of Figures Investigation of Aerodynamic Interference between Twin Deck Bridges Page II Figure 1-1: Thaddeus Kosciusko Bridge, NY, by Njhepler at English Wikipedia, CC BY 2.5, https://commons.wikimedia.org/w/index.php?curid=12452015 .................................................. 1 Figure 1-2: Fred Hartman Bridge, TX, by United States Coast Guard, PA2 James Dillard - U.S. Coast Guard Visual Information GalleryU.S. Coast Guard Visual Information Gallery Home, Public Domain, https://commons.wikimedia.org/w/index.php?curid=3499375 ......................... 2 Figure 2-1: The computational domain with twin deck dimensions in meters and locations of local CSYS origins ........................................................................................................................... 4 Figure 2-2: An example mesh around single and twin deck cross sections ................................... 5 Figure 2-3: Force component values vs. wall Y+ (P – polyhedral mesh, H – hexahedral mesh, ratio – stretch factor for the cell thickness in the prism layer) .......................................................7 Figure 2-4: Pitch moment component values vs. Y+ (P – polyhedral mesh, H – hexahedral mesh, ratio – stretch factor for the cell thickness in the prism layer) ............................................ 8 Figure 2-5: Mesh influence on force and moment components values acting on the upstream deck ................................................................................................................................................. 9 Figure 2-6: Mesh influence on force and moment components values acting on the downstream deck ................................................................................................................................................10 Figure 3-1: Single deck. Pressure plot on the mid section of the domain (a) zoomed out view, (b) the area around the deck ............................................................................................................... 11 Figure 3-2: Single deck. Horizontal velocity plot on the mid section of the domain (a) general view, (b) the area around the deck................................................................................................. 12 Figure 4-1: Pressure contour plot around the decks, (a) URANS, (b) LES at an instant in time, polyhedral mesh, Y+=1 .................................................................................................................. 18 Figure 4-2: Pressure on the symmetry line of the top surface of upstream deck, (a) location of measured points, (b) pressure values ............................................................................................ 19 Figure 4-3: Pressure on the symmetry line of the bottom surface of the upstream deck, (a) location of measured points, (b) pressure values ......................................................................... 20 Figure 4-4: Pressure on the symmetry line of the top surface of downstream deck, (a) location of measured points, (b) pressure values ............................................................................................ 21 Figure 4-5: Pressure on the symmetry line of the bottom surface of downstream deck, (a) location of measured points, (b) pressure values ......................................................................... 22 Figure 4-6: Pressure on the symmetry line of the upstream surface of the upstream deck, (a) location of measured points, (b) pressure values ......................................................................... List of Figures 23 Figure 4-7: Pressure on the symmetry line of the downstream surface of upstream deck, (a) location of measured points, (b) pressure values ......................................................................... 24 Figure 4-8: Pressure on the symmetry line of the upstream surface of the downstream deck , (a) location of measured points, (b) pressure values ......................................................................... 25 Figure 4-9: Pressure on the symmetry line of the downstream surface of the downstream deck, (a) location of measured points, (b) pressure values .................................................................... 26 Figure 4-10: Gap-to width ratio influence on the drag force in parallel flow. Deck 1 – upstream deck, deck 2 – downstream deck. .................................................................................................. 31 Figure 4-11: Gap-to width ratio influence on the lift force in parallel flow. Deck 1 – upstream deck, deck 2 – downstream deck. .................................................................................................. 31 Figure 4-12: Gap-to width ratio influence on the pitch moment in parallel flow. Deck 1 – upstream deck, deck 2 – downstream deck. .................................................................................. 31 Figure 4-13: Influence of angle of attack on drag force, L/B=0.13. Deck 1 – upstream deck, deck 2 – downstream deck. ................................................................................................................... 37 Figure 4-14: Influence of angle of attack on lift force, L/B=0.13. Deck 1 – upstream deck, deck 2 – downstream deck. ...................................................................................................................... 37 Figure 4-15: Influence of angle of attack on pitch moment, L/B=0.13. Deck 1 – upstream deck, deck 2 – downstream deck. .......................................................................................................... 37 Figure 5-1: Twin deck model on elastic springs ............................................................................ 39 Figure 5-2: Vertical translations of the decks in time, (a) during the entire simulation, (b) during last 5 seconds of the simulation time. ........................................................................................... 40 Figure 5-3: Fast Fourier Transform of the vertical vibrations of the decks .................................. 41 Figure 5-4: Rotations of the decks in time, (a) during the entire simulation, (b) during last 5 seconds of the simulation time ..................................................................................................... 42 Figure 5-5: Fast Fourier Transform of the rotational vibrations of the decks ............................. 42 Figure 5-6: Forces acting on the decks in configuration 1, (a) drag, (b) lift and (c) moment ...... 43 Figure 5-7: Forces and moment acting on the decks in configuration 2, (a) drag force, (b) lift force, (c) moment. ......................................................................................................................... 49 Figure 5-8: Forces and moment acting on the decks in configuration 3, (a) drag force, (b) lift force, (c) moment ........................................................................................................................... 51 Figure 5-9: Comparison of force values acting on deck 1 at different configurations .................. 52 Figure 5-10: Comparison of force values acting on deck 2 at different configurations ................ List of Figures 53 Figure 2-1: Stonecutters bridge cross-section. All dimensions are given in meters. .................... 54 Figure 2-2: Sign convention ......................................................................................................... 54 Figure 2-3: Sensitivity of force coefficients to Reynolds number ................................................. 55 Figure 2-4: The drag coefficient at different angles of attack ....................................................... 56 Figure 2-5: The lift coefficient at different angles of attack ......................................................... 56 Figure 2-6: The pitch moment coefficient at different angles of attack ........................................ 57 Figure 2-7: Velocity fields on a vertical plane at (a) Re=1.2e4, (b) 9.1e4 ..................................... 58 Figure 2-8: Vorticity fields on a vertical plane at (a) Re=1.2e4, (b) 9.1e4 .................................... 59 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page IV Page IV Page IV Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges List of Tables List of Tables Table 2-1. Types of meshes used in the simulations ....................................................................... 6 Table 3-1: Pressure field contour plots for a single deck bridge at varying angle of attack .......... 13 Table 3-2: Velocity field contour plots for a single deck bridge at varying angle of attack ........... 15 Table 3-3: Extreme values of pressure and velocity on a single deck ............................................ 17 Table 4-1: Extreme values of pressure and velocity for a twin deck with varying gap ................. 27 Table 4-2: Pressure and velocity fields around twin deck with varying gap ................................ 27 Table 4-3: Velocity field contour plots for a twin deck bridge at varying gap .............................. 29 Table 4-4: Pressure field contour plots for a twin deck bridge with 0.1 m gap (L/B=0.26) at varying angle of attack .................................................................................................................. 32 Table 4-5: Velocity field contour plots for a twin deck bridge with 0.1 m gap (L/B=0.26) at varying angle of attack .................................................................................................................. 34 Table 4-6: Extreme values of pressure and velocity for a twin deck ............................................ 36 Table 5-1. Comparison of drag force, lift force and pitch moment for a stationary twin deck and a twin deck suspended on elastic springs ........................................................................................ 44 Table 5-2: Velocity vector field on the middle plane in consecutive time steps ........................... 44 Table 5-3: Velocity streamlines in consecutive time steps ........................................................... 46 Table 2-1: Force coefficients and Strouhal number obtained from CFD simulations .................. 59 Investigation of Aerodynamic Interference between Twin Deck Bridges Page V Page V 1.1. Background Construction of a twin bridge can be a cost effective and minimally disruptive way to increase capacity when an existing bridge is not near the end of its service life. With ever growing vehicular traffic, the demand approaches the capacity of many existing roads and bridges. Remodeling a structure with an insufficient number of lanes can be a good solution in case of smaller and less busy bridges. Closing down or reducing traffic on crossings of greater importance for the construction period, however, can result in major delays and revenue loss for commerce and transportation as well as increasing the traffic load on alternate routes. Multiple-deck bridges are considered to be the answer to this issue. A parallel deck can be built next to the existing one without reducing the flow. Moreover, it can be decided on the design stage that a new bridge has a twin or multi-deck structure. Several such structures have already been built throughout the United States, among them: the new NY Bridge Project - the Tappan Zee Hudson River Crossing, SR-182 Columbia River Bridge, the Thaddeus Kosciusko Bridge (I-87) (see Figure 1-1), the Allegheny River Bridge, PA, which carries I-76, and Fred Hartman Bridge, TX (see Figure 1-2). The structures located outside of the US include: Stonecutters Bridge and Ting Kau Bridge in Hong Kong, Zhejiang Xihoumen Bridge, Haihe Bridge in China, and Yi Sun-sin Bridge in Korea. Figure 1-1: Thaddeus Kosciusko Bridge, NY, by Njhepler at English Wikipedia, CC BY 2.5, https://commons.wikimedia.org/w/index.php?curid=12452015 Figure 1-1: Thaddeus Kosciusko Bridge, NY, by Njhepler at English Wikipedia, CC BY 2.5, https://commons.wikimedia.org/w/index.php?curid=12452015 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 1 Page \| 1 Page \| 1 Figure 1-2: Fred Hartman Bridge, TX, by United States Coast Guard, PA2 James Dillard - U.S. Coast Guard Visual Information GalleryU.S. Coast Guard Visual Information Gallery Home, Public Domain, https://commons.wikimedia.org/w/index.php?curid=3499375 Figure 1-2: Fred Hartman Bridge, TX, by United States Coast Guard, PA2 James Dillard - U.S. Coast Guard Visual Information GalleryU.S. Coast Guard Visual Information Gallery Home, Public Domain, https://commons.wikimedia.org/w/index.php?curid=3499375 The experimental studies mostly cover the topic of aerodynamic interference effect on aerostatic coefficients. A set of tests at different Reynolds number is usually performed and the coefficients obtained for a twin deck are compared to those of a single deck. 1.1. Background The main design parameter considered is the central gap width (see: [1], [2], [3], [4]). The results of static tests show that the upstream box is not significantly affected and behaves similarly to a single deck, as shown in [1] and [2]. In contrast, the pressure acting on the downstream box changes due to the characteristics of the slot. The other design parameter is the geometry of the bridge girders (see: [1], [3], [5]). The researchers analyze sets of chosen cross-sections to find the one with the most favorable aerodynamic characteristics and try to optimize an existing shape. The flow pattern, the vortex- shedding frequency, and Strouhal number are also of interest, see [3], [4], [6], [7], [8], [9], [10]. The measurements reveal that as the gap ratio increases, the turbulent kinetic energy in the gap region and the fluctuating pressures on the downstream deck increase. Results show that, at certain flow velocities, the vortex induced vibrations of the upstream deck tend to be amplified due to the interference effect between parallel decks, see [8]. The literature review delivers a discussion on, other than gap width, modifications of the cross- section geometry, which would improve the aerodynamic properties. The proposed control measures include: bracings, wind fairings, stabilizing plates, and spoiler gratings fixed on the railing, among others, as presented in papers [5], [8], [11], [12], and [13]. Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 2 Page \| 2 CFD modeling compliments the wind tunnel testing and is used in parametric studies, where multiple cases need to be considered, like in [4]. 2D and 3D CFD URANS computations are performed to obtain aerostatic coefficients and to study the vortex shedding effect. CFD also helps to find an optimized cross-section shape and add modifications to the original design, e.g. adding wind barriers and sharpening the wind fairing noses of the two decks, (see [14], [15], [16], [17], [18], [19], [20]). In this research, one of the existing twin-deck bridges, the Stonecutters Bridge in Hong Kong, is studied and used as a reference in the validation of the computational fluid dynamics analysis approach. Stonecutters Bridge is a cable-stayed bridge with a twin deck. Extensive studies, both experimental and computational, were performed on this bridge design in the past. Researchers concentrated on the investigation of the effects of gap-width and the angle of wind incidence on the aerodynamic characteristics of the bridge. Experimental measurements were presented in [1], [14] and [20]-[24], and the results of computational simulations are shown in [25] and [26], among others. Experimental tests were performed to assess the bridge model response to smooth and turbulent flow. Pressure distributions were studied to investigate the influence of the gap width in [1] and [3]. The analysis showed that the gap width significantly affects the pressure distribution on the decks and therefore their aerodynamic performance. Vortex shedding mechanisms of the bridge deck were also studied. They revealed that the bridge is susceptible to vortex shedding and that increasing the gap width causes the Strouhal number to increase. Dynamic tests of the twin-deck bridge model were conducted [21] to analyze the influence of the central gap. A comparative analysis showed that increasing the gap causes a higher torsional damping ratio, which increases the aerodynamic stability of the bridge. A detailed CFD study of flutter derivatives was performed in [26] at the in-service stage of the Stonecutters Bridge as well as during construction and the results indicate stability of the bridge. The adopted approach is validated by an analysis of aerodynamic behavior of the Stonecutters Bridge. The computational results are compared to experimental measurements found in literature. 1.2. Proposed analysis methods This study was performed on a generic cross-section segment of a twin bridge deck model in an unsteady air flow. The analysis methods cover preliminary simulations, such as selection between Reynolds Averaged Navier-Stokes (RANS) equations with different turbulence models versus Large Eddy Simulations (LES), and a mesh density study. The main purpose of the research is to: characterize static and dynamic responses of the model twin decks by computing steady state aerodynamic forces and pitching moment in RANS simulations as well as computing the time history of aerodynamic forces and pitching moment obtained from LES simulations. The research also seeks to characterize the influence of the upstream deck wake on the downstream deck. Static simulations include a parametric study with various angles of wind incidence, different wind speed values, and a study of gap-to-width ratios. 2. CFD model 2. Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 3 2.1. General information The CFD model is set up to represent wind tunnel experiments. The domain geometry consists of three parts: the wind tunnel nozzle, trapezoidal transition part, and the main cube modeling the room. The dimensions are such that the boundary conditions do not have a significant influence on the flow around the bridge model. The inlet is far enough from the bridge model for the flow to be developed in front of it and also the domain is long enough that there is no influence of reverse flow that may occur near the pressure outlet. All outer surfaces of the model, excluding the inlet and outlet surfaces, have a no-slip wall boundary condition applied. The CFD model with a twin deck test section inside is presented in Figure 2-1. The deck dimensions are as follows: deck width (B) - 0.381 m, deck thickness - 0.022 m, girder height - 0.031 m, length – 1.524 m. In the study of twin deck models, four gap widths (L) are considered: 0.05 m, 0.1 m, 0.2 m, 0.3m, which correspond to the gap to single deck width (L/B) ratio of 0.13, 0.26, 0.52, and 0.79. An example mesh around the decks is presented in Figure 2-2. The mesh is denser around the decks and the wake area to capture the character of the flow and the interference between the decks. Although the research is focused on the flow around twin-deck bridges, a set of runs is performed with a single deck as well, for comparison purposes. Figure 2-1: The computational domain with twin deck dimensions in meters and locations of local CSYS origins Figure 2-1: The computational domain with twin deck dimensions in meters and locations of local CSYS origins Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 4 Figure 2-2: An example mesh around single and twin deck cross sections Single deck Twin-deck Single d k Twin-deck Figure 2-2: An example mesh around single and twin deck cross sections 2.2. Mesh study The influence of mesh density on the results is investigated in the first step of this study. A model of a twin deck bridge with a 0.1 m gap (gap ratio=0.26) was used. In all cases a prism layer with constant thickness is created around the decks to provide layers of prism cells next to the wall. The ‘all Y+ treatment’ option is used, which uses a blended wall function to estimate shear stress if 5<Y+<30 and the standard wall function for Y+>30. Two prism layer stretching values are considered. The first stretching value is 1.0, with 1, 2, 4, 8 and 16 layers (which gives Y+=40, 21, 9.4, 3.7 and 2.1 accordingly). The second stretching value is 1.5, with 8 layers (which gives Y+=1.3). The number of volume cells varies from approximately 3 to 4 million cells. Two types of meshing techniques were used. The polyhedral mesher is suited to complex, multi- region geometries. These cells usually have 12 to 14 faces. The trimmer mesher uses predominantly hexahedral mesh cells with trimmed cells next to surfaces. Table 2-1 shows which meshing techniques were used in the simulations, along with a close-up view of the resulting prism layer around a corner of the deck. Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 5 Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 5 Table 2-1. Types of meshes used in the simulations Mesher Prism layer stretching Resulting mesh around the deck Polyhedral 1 1.5 Hexahedral 1.5 An unsteady RANS (URANS) solver was used in the computations as well as the Large Eddy Simulation model. The simulations were kept running until the flow reached a steady state and force and moment components on the stationary decks converged to a constant value. Computations using the URANS solver with the k-ε turbulence model and polyhedral mesh were performed for different thicknesses of the first prism layer. Moreover, for the densest mesh (and thinnest first layer) LES computations were done using both polyhedral and trimmer meshers. Plots of the computed drag and lift force are presented in Figure 2-3 and Figure 2-4. The parametric mesh study shows that the biggest differences in drag and lift forces for both decks occur between the lowest (Y+=1) and highest values of Y+ (Y+=40). Two highest values of Y+ give very close values for drag and lift force components, but not for pitch moment on the upstream deck. For the instance of the lowest Y+, most simulations give very similar results except for LES with hexahedral mesh Table 2-1. Types of meshes used in the simulations Mesher Prism layer stretching Resulting mesh around the deck Polyhedral 1 1.5 Hexahedral 1.5 Mesher Prism layer stretching Resulting mesh around the deck Polyhedral 1 1.5 Hexahedral 1.5 Resulting mesh around the deck An unsteady RANS (URANS) solver was used in the computations as well as the Large Eddy Simulation model. The simulations were kept running until the flow reached a steady state and force and moment components on the stationary decks converged to a constant value. Computations using the URANS solver with the k-ε turbulence model and polyhedral mesh were performed for different thicknesses of the first prism layer. Moreover, for the densest mesh (and thinnest first layer) LES computations were done using both polyhedral and trimmer meshers. Plots of the computed drag and lift force are presented in Figure 2-3 and Figure 2-4. The parametric mesh study shows that the biggest differences in drag and lift forces for both decks occur between the lowest (Y+=1) and highest values of Y+ (Y+=40). Two highest values of Y+ give very close values for drag and lift force components, but not for pitch moment on the upstream deck. Investigation of Aerodynamic Interference between Twin Deck Bridges For the instance of the lowest Y+, most simulations give very similar results except for LES with hexahedral mesh. Computations using the URANS solver with the k-ε turbulence model and polyhedral mesh were performed for different thicknesses of the first prism layer. Moreover, for the densest mesh (and thinnest first layer) LES computations were done using both polyhedral and trimmer meshers. Plots of the computed drag and lift force are presented in Figure 2-3 and Figure 2-4. The parametric mesh study shows that the biggest differences in drag and lift forces for both decks occur between the lowest (Y+=1) and highest values of Y+ (Y+=40). Two highest values of Y+ give very close values for drag and lift force components, but not for pitch moment on the upstream deck. For the instance of the lowest Y+, most simulations give very similar results except for LES with hexahedral mesh. Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 6 Figure 2-3: Force component values vs. Investigation of Aerodynamic Interference between Twin Deck Bridges wall Y+ (P – polyhedral mesh, H – hexahedral mesh, ratio – stretch factor for the cell thickness in the prism layer) -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 0 5 10 15 20 25 30 35 40 Force [N] Wall Y+ Drag force - deck 1 Lift force- deck 1 Drag force- deck 2 Lift force - deck 2 Drag force - deck 1 (LES P) Lift force - deck 1 (LES P) Drag force - deck 2 (LES P) Lift force - deck 2 (LES P) Drag force - deck 1 (ratio=1.5) Lift force - deck 1 (ratio=1.5) Drag force - deck 2 (ratio=1.5) Lift force - deck 2 (ratio=1.5) Drag force - deck 1 (LES H) Lift force - deck 1 (LES H) Drag force - deck 2 (LES H) Lift force - deck 2 (LES H) -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 0 5 10 15 20 25 30 35 40 Force [N] Wall Y+ Drag force - deck 1 Lift force- deck 1 Drag force- deck 2 Lift force - deck 2 Drag force - deck 1 (LES P) Lift force - deck 1 (LES P) Drag force - deck 2 (LES P) Lift force - deck 2 (LES P) Drag force - deck 1 (ratio=1.5) Lift force - deck 1 (ratio=1.5) Drag force - deck 2 (ratio=1.5) Lift force - deck 2 (ratio=1.5) Drag force - deck 1 (LES H) Lift force - deck 1 (LES H) Drag force - deck 2 (LES H) Lift force - deck 2 (LES H) Figure 2-3: Force component values vs. wall Y+ (P – polyhedral mesh, H – hexahedral mesh, ratio – stretch factor for the cell thickness in the prism layer) Figure 2-3: Force component values vs. wall Y+ (P – polyhedral mesh, H – hexahedral mesh, ratio – stretch factor for the cell thickness in the prism layer) Figure 2-3: Force component values vs. wall Y+ (P – polyhedral mesh, H – hexahedral mesh, ratio – stretch factor for the cell thickness in the prism layer) Page \| 7 Page \| 7 Page \| 7 Investigation of Aerodynamic Interference between Twin Deck Bridges Figure 2-4: Pitch moment component values vs. Y+ (P – polyhedral mesh, H – hexahedral mesh, ratio – stretch factor for the cell thickness in the prism layer) -0.06 -0.04 -0.02 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0 5 10 15 20 25 30 35 40 Moment [Nm] Wall Y+ Moment - deck 1 Moment - deck 2 Moment - deck 1 (LES P) Moment - deck 2 (LES P) Moment - deck 1 (ratio=1.5) Moment - deck 2 (ratio=1.5) Moment - deck 1 (LES H) Moment - deck 2 (LES H) -0.06 -0.04 -0.02 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0 5 10 15 20 25 30 35 40 Moment [Nm] Wall Y+ 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0 5 10 15 20 25 30 35 40 Moment [Nm] Moment - deck 1 Moment - deck 2 Moment - deck 1 (LES P) Moment - deck 2 (LES P) Moment - deck 1 (ratio=1.5) Moment - deck 2 (ratio=1.5) Moment - deck 1 (LES H) Moment - deck 2 (LES H) Figure 2-4: Pitch moment component values vs. Y+ (P – polyhedral mesh, H – hexahedral mesh, ratio – stretch factor for the cell thickness in the prism layer) The influences of mesh type and mesh density on the force components are also shown in Figure 2-5 and Figure 2-6 in the form of histograms. The first two columns refer to polyhedral and hexahedral mesh with Y+=20, the next two columns refer to polyhedral and hexahedral mesh with Y+=1, and the last column refers to LES simulations with polyhedral mesh. Page \| 8 Page \| 8 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges (a) (b) (c) Figure 2-5: Mesh influence on force and moment components values acting on the upstre deck 0.000 0.200 0.400 0.600 Drag force on Deck 1 [N] polyhedral mesh, Y+=20 hexahedral mesh, Y+=20 polyhedral mesh, Y+=1.3 hexahedral mesh, Y+=1.1 LES, Y+=1.5 -0.100 0.100 0.300 0.500 0.700 Lift force on Deck 1 [N] polyhedral mesh, Y+=20 hexahedral mesh, Y+=20 polyhedral mesh, Y+=1.3 hexahedral mesh, Y+=1.1 LES, Y+=1.5 -0.040 0.010 0.060 0.110 Moment on Deck 1 [Nm] polyhedral mesh, Y+=20 hexahedral mesh, Y+=20 polyhedral mesh, Y+=1.3 hexahedral mesh, Y+=1.1 LES, Y+=1.5 (a) 0.000 0.200 0.400 0.600 Drag force on Deck 1 [N] polyhedral mesh, Y+=20 hexahedral mesh, Y+=20 polyhedral mesh, Y+=1.3 hexahedral mesh, Y+=1.1 LES, Y+=1.5 (a) Dra polyhedral mesh, Y+=20 hexahedral mesh, Y+=20 polyhedral mesh, Y+=1.3 hexahedral mesh, Y+=1.1 LES, Y+=1.5 (b) -0.100 0.100 0.300 0.500 0.700 Lift force on Deck 1 [N] polyhedral mesh, Y+=20 hexahedral mesh, Y+=20 polyhedral mesh, Y+=1.3 hexahedral mesh, Y+=1.1 LES, Y+=1.5 (b) polyhedral mesh, Y+=20 hexahedral mesh, Y+=20 polyhedral mesh, Y+=1.3 hexahedral mesh, Y+=1.1 LES, Y+=1.5 (c) -0.040 0.010 0.060 0.110 Moment on Deck 1 [Nm] polyhedral mesh, Y+=20 hexahedral mesh, Y+=20 polyhedral mesh, Y+=1.3 hexahedral mesh, Y+=1.1 LES, Y+=1.5 (c) M polyhedral mesh, Y+=20 hexahedral mesh, Y+=20 polyhedral mesh, Y+=1.3 hexahedral mesh, Y+=1.1 LES, Y+=1.5 Figure 2-5: Mesh influence on force and moment components values acting on the upstream deck Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 9 Figure 2-5: Mesh influence on force and moment components values acting on the upstream deck (a) 0.000 0.200 0.400 0.600 Drag force on Deck 2 [N] polyhedral mesh, Y+=20 hexahedral mesh, Y+=20 polyhedral mesh, Y+=1.3 hexahedral mesh, Y+=1.1 LES, Y+=1.5 (a) 0.000 0.200 0.400 0.600 Drag force on Deck 2 [N] polyhedral mesh, Y+=20 hexahedral mesh, Y+=20 polyhedral mesh, Y+=1.3 hexahedral mesh, Y+=1.1 LES, Y+=1.5 (a) Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 9 Investigation of Aerodynamic Interference between Twin Deck Bridges (b) (c) -0.100 0.100 0.300 0.500 0.700 Lift force on Deck 2 [N] polyhedral mesh, Y+=20 hexahedral mesh, Y+=20 polyhedral mesh, Y+=1.3 hexahedral mesh, Y+=1.1 LES, Y+=1.5 -0.040 0.010 0.060 0.110 Moment on Deck 2 [Nm] polyhedral mesh, Y+=20 hexahedral mesh, Y+=20 polyhedral mesh, Y+=1.3 hexahedral mesh, Y+=1.1 LES- Y+1.5 (b) -0.100 0.100 0.300 0.500 0.700 Lift force on Deck 2 [N] polyhedral mesh, Y+=20 hexahedral mesh, Y+=20 polyhedral mesh, Y+=1.3 hexahedral mesh, Y+=1.1 LES, Y+=1.5 (b) polyhedral mesh, Y+=20 hexahedral mesh, Y+=20 polyhedral mesh, Y+=1.3 hexahedral mesh, Y+=1.1 LES, Y+=1.5 (c) -0.040 0.010 0.060 0.110 Moment on Deck 2 [Nm] polyhedral mesh, Y+=20 hexahedral mesh, Y+=20 polyhedral mesh, Y+=1.3 hexahedral mesh, Y+=1.1 LES- Y+1.5 (c) Figure 2-6: Mesh influence on force and moment components values acting on the downstream deck Figure 2-6: Mesh influence on force and moment components values acting on the downstream deck Figure 2-6: Mesh influence on force and moment components values acting on the downstream deck 3. Analysis of the flow around a single deck bridge A set of runs was performed to assess the character of the flow around a single deck. The results presented in this chapter are used as a reference for the twin deck study. The pressure and velocity contour plots for a single deck in a wind tunnel are shown in Figure 3-1 and Figure 3-2. The zoomed out view shows the pressure distribution relative to the atmospheric reference pressure and the velocity magnitude in a cross section of the domain. The pressure is nearly uniform away from the deck, and the flow separation and wake are visible in the velocity magnitude color plot. The close-up view of the flow around the deck makes it possible to see the distributions around the deck in greater detail. Pressure reaches the highest values on the upstream side – vertical and bottom surfaces where the approach flow stagnates. The lowest (negative in value) pressure is located on the top surface, right at the front. Moreover, multiple separation points can be seen in the velocity contour plot as the deck cross-section is not streamlined. Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 10 Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 10 (a) (b) Figure 3-1: Single deck. Pressure plot on the mid section of the domain (a) zoomed out view, (b) the area around the deck (a) (b) Figure 3-1: Single deck. Pressure plot on the mid section of the domain (a) zoomed out view, (b) the area around the deck (b) Figure 3-1: Single deck. Pressure plot on the mid section of the domain (a) zoomed out view, (b) the area around the deck Page \| 11 Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 12 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges (a) (b) Figure 3-2: Single deck. Horizontal velocity plot on the mid section of the domain (a) general view, (b) the area around the deck (a) (b) (b) Figure 3-2: Single deck. Horizontal velocity plot on the mid section of the domain (a) general view, (b) the area around the deck Figure 3-2: Single deck. Horizontal velocity plot on the mid section of the domain (a) general view, (b) the area around the deck The pressure distribution around the deck was checked for different flow directions. Angles of attack in the range of -10 deg to 10 deg with a 5 deg interval were selected. The resultant pressure fields on the center cross section of the deck are collected in Table 3-1. Table 3-2 summarizes velocity field contour plots. The extreme values of pressure and velocity for these cases are listed in Table 3-3. The minimum pressure on the deck, equal to -8.37 Pa, is located at the top upstream corner of the deck and occurs when the angle of attack equals 5 deg. The highest positive pressure value, 5.7 Pa, is found on the upstream surface of the deck. The velocity field on the central plane changes with the angle of attack. It is smooth in the -5 deg to 5 deg range. For the angles -10 deg and 10 deg a more turbulent flow forms behind the deck. Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 12 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 12 age \| 12 Pressure field contour plots for a single deck bridge at varying angle of attack Table 3 1: Pressure field contour plots for a single deck bridge at varying angle of attack Angle [deg] Pressure contour plot -10 -5 Angle [deg] Pressure contour plot -10 Page \| 13 Investigation of Aerodynamic Interference between Twin Deck Bridges 0 5 10 0 stigation of Aerodynamic Interference between Twin Deck Bridges Page 0 5 10 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 14 : Velocity field contour plots for a single deck bridge at varying angle of attack Angle [deg] Velocity contour plot -10 -5 Angle [deg] Velocity contour plot -10 Velocity contour plot -5 Page \| 15 Investigation of Aerodynamic Interference between Twin Deck Bridges f d f b k d 0 5 10 10 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 16 Table 3-3: Extreme values of pressure and velocity on a single deck Table 3-3: Extreme values of pressure and velocity on a single deck Angle [deg] Pressure [Pa] Velocity [m/s] min max max -10 -6.76 5.37 3.54 -5 -7.51 5.48 3.58 0 -6.98 5.70 3.85 5 -8.37 5.34 3.86 10 -6.23 5.38 3.98 4.1. Pressure acting on a twin deck bridge Figure 4-1 illustrates pressure contour plots around twin deck cross-sections with a 0.1 m gap (gap ratio=0.26). The figures show a comparison of pressure fields for two models: URANS solver and LES with a low Y+. A polyhedral mesh was used in all cases. The value ranges are kept constant for an easier comparison. The pressure distribution around the upstream deck is very similar to the one observed around the single deck, described in Chapter 3. The downstream deck is located in its wake and the pressure acting on it is much lower. Figure 4-2 to Figure 4-9 provide detailed information about the pressure values at a set of points chosen on each surface of the model. The results presented for LES computations are time-averaged. The results obtained with the k-ε model are very similar, regardless the mesh size at most of the points, except for the ones located in the corners. As the pressure gradients are high in these regions it is important to use a denser mesh, so that no information is lost. A comparison of two models with the same Y+ shows that they give a similar pressure distribution in the areas of the deck where the flow is less turbulent and does not have active vortex shedding. There are larger differences between the URANS and LES model results along the top of the upstream deck shown in Figure 4.2 (b) where there are passing vortices shed from the leading edge of the upstream deck. These differences show that the time averaged LES pressure values do not closely match the URANS results in a region of active vortex transport. Page \| 17 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges (a) (b) Figure 4-1: Pressure contour plot around the decks, (a) URANS, (b) LES at an instant in time, polyhedral mesh, Y+=1 (a) (b) Figure 4-1: Pressure contour plot around the decks, (a) URANS, (b) LES at an instant in time, polyhedral mesh, Y+=1 Page \| 18 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges (a) (b) Figure 4-2: Pressure on the symmetry line of the top surface of upstream deck, (a) location of measured points, (b) pressure values -8 -6 -4 -2 0 2 4 6 8 1 2 3 4 5 6 7 8 9 10 11 Pressure [m] k-epsilon, polyhedral mesh, Y+=1.1 LES, polyhedral mesh, Y+=1.1 k-epsilon, polyhedral mesh, Y+=20.6 (a) (b) Figure 4-2: Pressure on the symmetry line of the top surface of upstream deck, (a) locatio measured points, (b) pressure values -8 -6 -4 -2 0 2 4 6 8 1 2 3 4 5 6 7 8 9 10 11 Pressure [m] k-epsilon, polyhedral mesh, Y+=1.1 LES, polyhedral mesh, Y+=1.1 k-epsilon, polyhedral mesh, Y+=20.6 (a) (b) k-epsilon, polyhedral mesh, Y+=1.1 LES, polyhedral mesh, Y+=1.1 k-epsilon, polyhedral mesh, Y+=20.6 Figure 4-2: Pressure on the symmetry line of the top surface of upstream deck, (a) location of measured points, (b) pressure values Figure 4-2: Pressure on the symmetry line of the top surface of upstream deck, (a) location of measured points, (b) pressure values Page \| 19 Investigation of Aerodynamic Interference between Twin Deck Bridges (a) (b) Figure 4-3: Pressure on the symmetry line of the bottom surface of the upstream deck, (a) location of measured points, (b) pressure values -8 -6 -4 -2 0 2 4 6 8 1 2 3 4 5 6 7 8 9 1011121314151617181920212223242526272829 Pressure [m] k-epsilon, polyhedral mesh, Y+=1.1 LES, polyhedral mesh, Y+=1.1 k-epsilon, polyhedral mesh, Y+=20.6 (a) (b) 8 Figure 4-3: Pressure on the symmetry line of the bottom surface of the upstream deck, (a) location of measured points, (b) pressure values Figure 4-3: Pressure on the symmetry line of the bottom surface of the upstream deck, (a) location of measured points, (b) pressure values Page \| 20 Page \| 20 Investigation of Aerodynamic Interference between Twin Deck Bridges (a) (b) Figure 4-4: Pressure on the symmetry line of the top surface of downstream deck, (a) location of measured points, (b) pressure values -8 -6 -4 -2 0 2 4 6 8 1 2 3 4 5 6 7 8 9 10 11 Pressure [m] k-epsilon, polyhedral mesh, Y+=1.1 LES, polyhedral mesh, Y+=1.1 k-epsilon, polyhedral mesh, Y+=20.6 (a) (b) -8 -6 -4 -2 0 2 4 6 8 1 2 3 4 5 6 7 8 9 10 11 Pressure [m] k epsilon polyhedral mesh Y+=1 1 LES polyhedral mesh Y+=1 1 (a) (b) k-epsilon, polyhedral mesh, Y+=1.1 LES, polyhedral mesh, Y+=1.1 k-epsilon, polyhedral mesh, Y+=20.6 Figure 4-4: Pressure on the symmetry line of the top surface of downstream deck, (a) location of measured points, (b) pressure values Figure 4-4: Pressure on the symmetry line of the top surface of downstream deck, (a) location of measured points, (b) pressure values Page \| 21 Investigation of Aerodynamic Interference between Twin Deck Bridges (a) (b) Figure 4-5: Pressure on the symmetry line of the bottom surface of downstream deck, (a) location of measured points, (b) pressure values -8 -6 -4 -2 0 2 4 6 8 1 2 3 4 5 6 7 8 9 1011121314151617181920212223242526272829 Pressure [m] k-epsilon, polyhedral mesh, Y+=1.1 LES, polyhedral mesh, Y+=1.1 k-epsilon, polyhedral mesh, Y+=20.6 (a) (a) (b) (b) -8 -6 -4 -2 0 2 4 6 8 1 2 3 4 5 6 7 8 9 1011121314151617181920212223242526272829 Pressure [m] k-epsilon, polyhedral mesh, Y+=1.1 LES, polyhedral mesh, Y+=1.1 k-epsilon, polyhedral mesh, Y+=20.6 Figure 4-5: Pressure on the symmetry line of the bottom surface of downstream deck, (a) location of measured points, (b) pressure values Figure 4-5: Pressure on the symmetry line of the bottom surface of downstream deck, (a) location of measured points, (b) pressure values Page \| 22 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges (a) (b) Figure 4-6: Pressure on the symmetry line of the upstream surface of the upstream deck, (a) location of measured points, (b) pressure values -8 -6 -4 -2 0 2 4 6 8 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Pressure [m] k-epsilon, polyhedral mesh, Y+=1.1 LES, polyhedral mesh, Y+=1.1 k-epsilon, polyhedral mesh, Y+=20.6 (a) (b) Figure 4-6: Pressure on the symmetry line of the upstream surface of the upstream deck, (a location of measured points, (b) pressure values -8 -6 -4 -2 0 2 4 6 8 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Pressure [m] k-epsilon, polyhedral mesh, Y+=1.1 LES, polyhedral mesh, Y+=1.1 k-epsilon, polyhedral mesh, Y+=20.6 (a) (b) k-epsilon, polyhedral mesh, Y+=1.1 LES, polyhedral mesh, Y+=1.1 k-epsilon, polyhedral mesh, Y+=20.6 Figure 4-6: Pressure on the symmetry line of the upstream surface of the upstream deck, (a) location of measured points, (b) pressure values Figure 4-6: Pressure on the symmetry line of the upstream surface of the upstream deck, (a) location of measured points, (b) pressure values Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 23 Investigation of Aerodynamic Interference between Twin Deck Bridges (a) (b) igure 4-7: Pressure on the symmetry line of the downstream surface of upstream deck, ( location of measured points, (b) pressure values -8 -6 -4 -2 0 2 4 6 8 1 2 3 4 5 6 7 8 9 10 11 12 13 Pressure [m] k-epsilon, polyhedral mesh, Y+=1.1 LES, polyhedral mesh, Y+=1.1 k-epsilon, polyhedral mesh, Y+=20.6 (a) (a) (b) -8 -6 -4 -2 0 2 4 6 8 1 2 3 4 5 6 7 8 9 10 11 12 13 Pressure [m] (b) k-epsilon, polyhedral mesh, Y+=1.1 LES, polyhedral mesh, Y+=1.1 k-epsilon, polyhedral mesh, Y+=20.6 Figure 4-7: Pressure on the symmetry line of the downstream surface of upstream deck, (a) location of measured points, (b) pressure values Page \| 24 Page \| 24 Investigation of Aerodynamic Interference between Twin Deck Bridges (a) (b) gure 4-8: Pressure on the symmetry line of the upstream surface of the downstream deck , location of measured points, (b) pressure values -8 -6 -4 -2 0 2 4 6 8 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Pressure [m] k-epsilon, polyhedral mesh, Y+=1.1 LES, polyhedral mesh, Y+=1.1 k-epsilon, polyhedral mesh, Y+=20.6 (a) (a) (b) -8 -6 -4 -2 0 2 4 6 8 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Pressure [m] k il l h d l h Y 1 1 LES l h d l h Y 1 1 (b) Figure 4-8: Pressure on the symmetry line of the upstream surface of the downstream deck , ( location of measured points, (b) pressure values -8 -6 -4 -2 0 2 4 6 8 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Pressure [m] k-epsilon, polyhedral mesh, Y+=1.1 LES, polyhedral mesh, Y+=1.1 k-epsilon, polyhedral mesh, Y+=20.6 (b) k-epsilon, polyhedral mesh, Y+=1.1 LES, polyhedral mesh, Y+=1.1 k-epsilon, polyhedral mesh, Y+=20.6 k-epsilon, polyhedral mesh, Y+=1.1 LES, polyhedral mesh, Y+=1.1 k-epsilon, polyhedral mesh, Y+=20.6 Figure 4-8: Pressure on the symmetry line of the upstream surface of the downstream deck , (a) location of measured points, (b) pressure values Figure 4-8: Pressure on the symmetry line of the upstream surface of the downstream deck , (a) location of measured points, (b) pressure values Page \| 25 Investigation of Aerodynamic Interference between Twin Deck Bridges (a) (b) gure 4-9: Pressure on the symmetry line of the downstream surface of the downstream d (a) location of measured points, (b) pressure values -8 -6 -4 -2 0 2 4 6 8 1 2 3 4 5 6 7 8 9 10 11 12 13 Pressure [m] k-epsilon, polyhedral mesh, Y+=1.1 LES, polyhedral mesh, Y+=1.1 k-epsilon, polyhedral mesh, Y+=20.6 (a) (a) (b) (b) -8 -6 -4 -2 0 2 4 6 8 1 2 3 4 5 6 7 8 9 10 11 12 13 Pressure [m] k-epsilon, polyhedral mesh, Y+=1.1 LES, polyhedral mesh, Y+=1.1 k-epsilon, polyhedral mesh, Y+=20.6 Figure 4-9: Pressure on the symmetry line of the downstream surface of the downstream deck, (a) location of measured points, (b) pressure values Figure 4-9: Pressure on the symmetry line of the downstream surface of the downstream deck, (a) location of measured points, (b) pressure values Page \| 26 Page \| 26 4.2. Influence of gap-to-width ratio The influence of the horizontal distance between two decks on pressure and velocity fields around them is investigated in this section. The gap-to-width ratios in the model cases are: 0.13, 0.26, 0.52, and 0.79. Table 4-1 summarizes the maximum and the minimum pressure and velocity values acting on the upstream deck with different spacing between decks. The differences in values are very small, less than 4%. Table 4-2 shows field plots of pressure and velocity at various deck spacings. able 4-1: Extreme values of pressure and velocity for a twin deck with varying gap Table 4-1: Extreme values of pressure and velocity for a twin deck with varying gap Pressure [Pa] Velocity [m/s] L/B min max max 0.13 -8.37 5.38 3.81 0.26 -8.38 5.40 3.80 0.52 -8.36 5.4 3.8 0.79 -8.39 5.40 3.81 Table 4-2: Pressure and velocity fields around twin deck with varying gap Table 4-2: Pressure and velocity fields around twin deck with varying gap L/B Pressure contour plots 0.13 Table 4-2: Pressure and velocity fields around twin deck with varying gap L/B Pressure contour plots 0.13 L/B Page \| 27 Investigation of Aerodynamic Interference between Twin Deck Bridges 0.26 0.52 0.79 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 28 Page \| 28 Table 4-3: Velocity field contour plots for a twin deck bridge at varying gap L/B Velocity contour plots 0.13 0.26 Table 4-3: Velocity field contour plots for a twin deck bridge at varying gap /B Velocity contour plots Table 4-3: Velocity field contour plots for a twin deck bridge at varying gap /B V l i l Velocity contour plots L/B Page \| 29 Investigation of Aerodynamic Interference between Twin Deck Bridges 0.52 0.79 0.52 0.79 0.52 0.79 The combined results for drag, lift, and pitch moment acting on a single deck and a twin deck are shown in Figure 4-10, Figure 4-11, and Figure 4-12. A comparison of force components on deck 1 (the upstream deck in the twin deck bridge system) and the single deck model show that there is only a slight difference in values. The deck spacing does not have a big influence on the drag forces acting on deck 1, which is almost constant and is equal to 0.41 N. The drag is lower on the downstream deck, because it is located in the upstream deck wake. This effect diminishes as the gap increases. A change of spacing from 0.05 m to 0.3 m (L/B=0.13 and 0.79) causes the drag force on deck 2 (downstream deck in the twin deck bridge system) to increase from 0.15 N to 0.25 N, and the pitch moment changes from -0.016 Nm to -0.032 Nm. In contrast, lift force does not vary significantly. The values of lift force for the upstream deck are 15-20% higher than for a single deck and they are almost constant (varying between 0.45 N and 0.48 N) regardless the gap. The lift force on the downstream deck increases as the gap gets bigger, from 0.11 N for the smallest gap to 0.135 N for 0.2 m gap (L/B=0.52). The combined results for drag, lift, and pitch moment acting on a single deck and a twin deck are shown in Figure 4-10, Figure 4-11, and Figure 4-12. A comparison of force components on deck 1 (the upstream deck in the twin deck bridge system) and the single deck model show that there is only a slight difference in values. The deck spacing does not have a big influence on the drag forces acting on deck 1, which is almost constant and is equal to 0.41 N. The drag is lower on the downstream deck, because it is located in the upstream deck wake. This effect diminishes as the gap increases. A change of spacing from 0.05 m to 0.3 m (L/B=0.13 and 0.79) causes the drag force on deck 2 (downstream deck in the twin deck bridge system) to increase from 0.15 N to 0.25 N, and the pitch moment changes from -0.016 Nm to -0.032 Nm. In contrast, lift force does not vary significantly. Investigation of Aerodynamic Interference between Twin Deck Bridges The values of lift force for the upstream deck are 15-20% higher than for a single deck and they are almost constant (varying between 0.45 N and 0.48 N) regardless the gap. The lift force on the downstream deck increases as the gap gets bigger, from 0.11 N for the smallest gap to 0.135 N for 0.2 m gap (L/B=0.52). Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 30 Page \| 30 Page \| 30 Figure 4-10: Gap-to width ratio influence on the drag force in parallel flow. Deck 1 – upstream deck, deck 2 – downstream deck. 0.00 0.10 0.20 0.30 0.40 0.50 0.000 0.200 0.400 0.600 0.800 1.000 Drag force [N] Gap-to-width ratio [-] single deck twin deck - deck 1 twin deck - deck2 Figure 4 10: Gap to width ratio influence on the drag force in parallel flow Deck 1 upstream 0.00 0.10 0.20 0.30 0.40 0.50 0.000 0.200 0.400 0.600 0.800 1.000 Drag force [N] Gap-to-width ratio [-] single deck twin deck - deck 1 twin deck - deck2 Figure 4-10: Gap-to width ratio influence on the drag force in parallel flow. Deck 1 – upstream deck, deck 2 – downstream deck. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.000 0.200 0.400 0.600 0.800 1.000 Lift force [N] Gap-to-width ratio [-] single deck twin deck - deck 1 twin deck - deck 2 Figure 4-11: Gap-to width ratio influence on the lift force in parallel flow. Deck 1 – upstream deck, deck 2 – downstream deck. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.000 0.200 0.400 0.600 0.800 1.000 Lift force [N] Gap-to-width ratio [-] single deck twin deck - deck 1 twin deck - deck 2 Figure 4-11: Gap-to width ratio influence on the lift force in parallel flow. Deck 1 – upstream deck, deck 2 – downstream deck. Figure 4-12: Gap-to width ratio influence on the pitch moment in parallel flow. Deck 1 – upstream deck, deck 2 – downstream deck. Investigation of Aerodynamic Interference between Twin Deck Bridges -0.06 -0.01 0.04 0.09 0.000 0.200 0.400 0.600 0.800 1.000 Moment [Nm] Gap-to-width ratio [-] single deck twin deck - deck 1 twin deck - deck 2 -0.06 -0.01 0.04 0.09 0.000 0.200 0.400 0.600 0.800 1.000 Moment [Nm] Gap-to-width ratio [-] single deck twin deck - deck 1 twin deck - deck 2 Figure 4-12: Gap-to width ratio influence on the pitch moment in parallel flow. Deck 1 – upstream deck, deck 2 – downstream deck. Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 31 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 31 4.3. Influence of varying angle of attack A set of five angles of attack were tested, which ranged from -10 degrees to 10 degrees in 5 degree increments. Pressure contour plots for the twin deck model with a 0.1 m gap (L/B=0.26) are collected in Table 4-4 and velocity contour plots for the same setup are presented in Table 4-5. Table 4-4: Pressure field contour plots for a twin deck bridge with 0.1 m gap (L/B=0.26) at varying angle of attack Angle [deg] Pressure field contour plot -10 -5 Table 4-4: Pressure field contour plots for a twin deck bridge with 0.1 m gap (L/B=0.26) at varying angle of attack Angle [deg] Pressure field contour plot -10 Table 4-4: Pressure field contour plots for a twin deck bridge with 0.1 m gap (L/B=0.26) at varying angle of attack Table 4-4: Pressure field contour plots for a twin deck bridge with 0.1 m gap (L/B=0.26) at varying angle of attack Pressure field contour plot -5 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 32 0 5 0 Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 33 0 5 10 10 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 33 Table 4-5: Velocity field contour plots for a twin deck bridge with 0.1 m gap (L/B=0.26) at varying angle of attack Angle [deg] Velocity field contour plot -10 -5 Table 4-5: Velocity field contour plots for a twin deck bridge with 0.1 m gap (L/B=0.26) at varying angle of attack Velocity field contour plot Velocity field contour plot Page \| 34 Investigation of Aerodynamic Interference between Twin Deck Bridges 0 5 Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 35 0 5 10 10 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 35 Maximum and minimum values of pressure acting on deck surfaces, with respect to changing flow angle of attack, are presented in Table 4-6. Differences in maximum values are not significant with the maximum equal 5.52 Pa at -5 deg angle. The minimum value for a twin deck at parallel flow is close to the single deck case and it increases with the increase of the absolute value of angle of attack. The maximum velocity in the vicinity of the decks does not experience big changes and reaches a maximum of 3.8 m/s in parallel flow. Table 4-6: Extreme values of pressure and velocity for a twin deck Table 4-6: Extreme values of pressure and velocity for a twin deck Angle [deg] Pressure [Pa] Velocity [m/s] min max max -10 -6.31 5.48 3.50 -5 -7.11 5.52 3.53 0 -8.38 5.40 3.80 5 -7.45 5.37 3.74 10 -5.21 5.46 3.43 The variation of the force and angle of attack is illustrated in Figure 4-13 to Figure 4-15 for a twin deck with a 0.1 m distance between decks (gap ratio 0.26). Figure 4-13 shows plots of the drag forces acting on the decks. The force stays positive and increases with the increase of the absolute value of the angle, and reaches the maximum of approximately 0.6 N at 10 degrees. Drag on the upstream deck has values that are close to the single deck model for the selected flow directions. For upstream/downstream bridge deck, the biggest difference can be seen for -10 deg angle, where the force equals 0.35 N (as compared to 0.6 N), whereas for 10 deg the values are the closest to each other (and equal approximately 0.6 N). The lift force, displayed in Figure 4-14, changes sign with the angle of attack. This tendency is true for the single deck model as well as for the twin deck model. Investigation of Aerodynamic Interference between Twin Deck Bridges For negative values of the angle of attack the lift force assumes negative values (the lowest: -0.61 N for the downstream deck in a twin deck and -0.736 N for upstream deck in a twin deck), goes through zero for angles in the range from -5 deg to 0 deg, and reaches positive values for non-negative angles. The highest value of the lift force was obtained for the downstream deck in the twin deck bridge at 10 deg angle. It is almost two times higher for the downstream deck in the twin deck bridge than the upstream deck (it is equal to 1.5 N and 0.84 N respectively). The lift force acting on the single deck at 10 deg angle falls in between these values and is equal 1.1 N. The pitch moment at various flow directions is shown in Figure 4-15. The pitch moment for the upstream deck in a twin deck bridge reaches a maximum of 0.09 Nm for parallel flow and decreases as the absolute value of the angle of attack increases. The values of the pitch moment for the downstream deck in a twin deck are negative for most of the angles of attack and equal between -0.02 Nm to -0.03 Nm. At the angle of attack equal to 5 deg, the pitch moment is smaller and equals 0.005 Nm. Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 36 Figure 4-13: Influence of angle of attack on drag force, L/B=0.13. Deck 1 – upstream deck, deck 2 – downstream deck. -1.000 -0.500 0.000 0.500 1.000 -10 -5 0 5 10 Drag force [N] Angle [deg] single deck twin deck - deck 1 twin deck - deck 2 Figure 4-13: Influence of angle of attack on drag force L/B=0 13 Deck 1 – upstream deck deck -1.000 -0.500 0.000 0.500 1.000 -10 -5 0 5 10 Drag force [N] Angle [deg] single deck twin deck - deck 1 twin deck - deck 2 Figure 4-13: Influence of angle of attack on drag force, L/B=0.13. Deck 1 – upstream deck, deck 2 – downstream deck. -1.000 -0.500 0.000 0.500 1.000 1.500 2.000 -10 -5 0 5 10 Lift force [N] Angle [deg] single deck twin deck - deck 1 twin deck - deck 2 Figure 4-14: Influence of angle of attack on lift force, L/B=0.13. Deck 1 – upstream deck, deck 2 – downstream deck. 5. Dynamic response The aerodynamic response of the decks can be investigated with the use of the Dynamic Fluid Body Interaction (DFBI) solver available in STAR-CCM+. The solver accounts for six degrees of freedom of body motions, uses mesh morphing to maintain a high quality mesh when bodies move in the domain, and handles solving for fluid transport through the deforming mesh. The URANS solver with the k-ε turbulence model is used to solve for the fluid flow. The following model configurations are taken into account and compared:  decks are stationary  DFBI configuration 1: both decks are suspended on springs,  DFBI configuration 2: deck 1 is suspended on springs and deck 2 is constraine  DFBI configuration 3: deck 1 is constrained and deck 2 is suspended on spring The decks are modeled as rigid bodies and each of them is suspended on 8 springs of assumed length (0.5 m) and stiffness (1000 N/m). Mass, center of mass, and all components of moment of inertia of the deck model have to be provided. These values are obtained using the geometry property computation capabilities of the LS-DYNA, LS-PrePost preprocessor [27] with an assumed material density of 1300 kg/m3. The total mass of the deck equals 20 kg. The center of mass is located at Z=-0.015 m, with Z=0 m on the top surface of the deck. The moments of inertia are: 𝐼𝑥𝑥= 11.6 kg m2, 𝐼𝑦𝑦= 0.69 kg m2, and 𝐼𝑧𝑧= 12.3 kg m2, about x, y, and z axis respectively. The release time is specified to be 30 sec. The period before the release time allows an initial flow solution to be computed including fluid forces on the deck. After steady state flow on stationary decks is achieved, the body suspended on springs is released. The body forces, including fluid forces, can now move the deck. The initial time and the ramp up time (together with the damping forces) should be long enough to reduce sudden non-physical application of the gravitational force to a reasonable level. This time depends on several parameters, like mass of the deck, stiffness of the springs, size of the domain, and cell sizes. It should be selected for each simulation. If the full gravitational force is applied at once, the mesh deformation can be big enough to cause zero- volume cells to appear and the run ends with an error. Investigation of Aerodynamic Interference between Twin Deck Bridges -1.000 -0.500 0.000 0.500 1.000 1.500 2.000 -10 -5 0 5 10 Lift force [N] Angle [deg] single deck twin deck - deck 1 twin deck - deck 2 Figure 4-14: Influence of angle of attack on lift force, L/B=0.13. Deck 1 – upstream deck, deck 2 – downstream deck. Figure 4-15: Influence of angle of attack on pitch moment, L/B=0.13. Deck 1 – upstream deck, deck 2 – downstream deck. -1.000 -0.500 0.000 0.500 1.000 -10 -5 0 5 10 Moment [Nm] Angle [deg] single deck twin deck - deck 1 twin deck - deck 2 Figure 4-15: Influence of angle of attack on pitch moment, L/B=0.13. Deck 1 – upstream deck, -1.000 -0.500 0.000 0.500 1.000 -10 -5 0 5 10 Moment [Nm] Angle [deg] single deck twin deck - deck 1 twin deck - deck 2 10 Figure 4-15: Influence of angle of attack on pitch moment, L/B=0.13. Deck 1 – upstream deck, deck 2 – downstream deck. Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 37 Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 37 Page \| 37 5. Dynamic response A ramp up of the force avoids this problem. The deck suspended on springs is allowed to move only vertically and rotationally along the centerline of the deck. All other rigid body motions are constrained. The stationary deck is positioned 2.46 cm below the initial location of the moveable deck to take into account the elongation of top springs due to the weight of the deck. When the body is released, i.e. when the springs are activated in the model, it still moves downwards, due to gravity. The displacement oscillates around a constant value with a decreasing amplitude. No structural damping is involved in the simulation; therefore, this change is a result of aerodynamic damping. Force components were recorded and compared for both decks. The forces acting on stationary decks are established during the first 30 sec of the simulation time. The drag force acting on the upstream deck converges to a value of 0.413 N and the force acting on the downstream deck converges to 0.178 N. The lift forces for the upstream and downstream decks are 0.455 N and 0.108 N, accordingly, and pitch moments are equal to 0.086 Nm and -0.02 Nm for the upstream and downstream decks, respectively. Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 38 Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 38 Page \| 38 5.1. Configuration 1: both decks suspended on springs 5.1. Configuration 1: both decks suspended on springs Figure 5-1 illustrates DFBI configuration 1, where both decks are suspended on elastic springs. Their vertical motion and rotation are illustrated in Figure 5-2 for the time after the springs are activated. High initial amplitudes, reaching 4.7 cm, decrease in time due to aerodynamic damping. After 100 sec of simulation time they are equal 1.1 mm. Both decks oscillate around the same value of -0.0246 m, which is a result of the elongation of the upper springs and compression of the lower springs due to the weight of the model. The amplitude of the motion for the downstream deck is higher than for the upstream deck. A Fast Fourier Transform (FFT) was used to establish the vibration frequencies of both decks. The calculations show that the first frequency of the decks is the same and equals 3.14 Hz, as shown in Figure 5-3. Rotations of the decks around local center lines are presented in Figure 5-4. Slightly higher amplitude of the rotational angle is reported for the downstream deck, which is 1 degree compared to 0.7 degrees for the upstream deck. Rotational frequency is equal to 3.24 Hz, as shown in Figure 5-5. Plots of forces acting on the decks are shown in Figure 5-6. Table 5-1 combines the force values for the stationary twin deck and a twin deck suspended on elastic springs. The velocity field is recorded and examined in more detail after the activation of the springs. Table 5-2 shows how the velocity changes around the decks in consecutive time steps. The velocity vector field is plotted on the middle plane. Table 5-3 illustrates velocity streamlines changing in time. The seeds for the streamlines lay on the domain inlet at the same height. Figure 5-1: Twin deck model on elastic springs Figure 5-1: Twin deck model on elastic springs Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 39 (a) (b) Figure 5-2: Vertical translations of the decks in time, (a) during the entire simulation, (b) d last 5 seconds of the simulation time. 5.1. Configuration 1: both decks suspended on springs -0.05 -0.04 -0.03 -0.02 -0.01 0 30 40 50 60 70 80 90 100 Vertical translation [m] Simulation time [sec] Deck 1 Deck 2 -0.027 -0.026 -0.025 -0.024 -0.023 -0.022 95 96 97 98 99 100 Vertical translation [m] Simulation time [sec] Deck 1 Deck 2 (a) -0.05 -0.04 -0.03 -0.02 -0.01 0 30 40 50 60 70 80 90 100 Vertical translation [m] Simulation time [sec] (a) 100 (b) Figure 5-2: Vertical translations of the decks in time, (a) during the entire simulation, (b) during last 5 seconds of the simulation time. Page \| 40 Page \| 40 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Figure 5-3: Fast Fourier Transform of the vertical vibrations of the decks 0 0.002 0.004 0.006 0.008 0.01 0 1 2 3 4 5 Frequency [Hz] Deck 1 Deck 2 Figure 5-3: Fast Fourier Transform of the vertical vibrations of the decks 0 0.002 0.004 0.006 0.008 0.01 0 1 2 3 4 5 Frequency [Hz] Deck 1 Deck 2 Figure 5-3: Fast Fourier Transform of the vertical vibrations of the decks (a) (a) -1.5 -1 -0.5 0 0.5 1 1.5 30 40 50 60 70 80 90 100 Rotation angle [deg] Simulation time [sec] Deck 1 Deck 2 (a) -1.5 -1 -0.5 0 0.5 1 1.5 30 40 50 60 70 80 90 100 Rotation angle [deg] Simulation time [sec] Deck 1 Deck 2 Deck 1 Deck 2 Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 41 Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 42 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges (b) Figure 5-4: Rotations of the decks in time, (a) during the entire simulation, (b) during last 5 seconds of the simulation time -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 95 96 97 98 99 100 Rotation angle [deg] Simulation time [sec] Deck 1 Deck 2 (b) Figure 5-4: Rotations of the decks in time, (a) during the entire simulation, (b) during last 5 seconds of the simulation time -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 95 96 97 98 99 100 Rotation angle [deg] Simulation time [sec] Deck 1 Deck 2 (b) Figure 5-4: Rotations of the decks in time, (a) during the entire simulation, (b) during last 5 seconds of the simulation time Figure 5-5: Fast Fourier Transform of the rotational vibrations of the decks 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0 1 2 3 4 5 Frequency [Hz] Deck 1 Deck 2 Figure 5-5: Fast Fourier Transform of the rotational vibrations of the decks Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 42 Investigation of Aerodynamic Interference between Twin Deck Bridges (a) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 30 40 50 60 70 80 90 100 Drag force [N] (a) (b) ( ) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 30 40 50 60 70 80 90 100 Drag force [N] Simulation time [sec] Deck 1 Deck 2 -4 -3 -2 -1 0 1 2 3 4 30 40 50 60 70 80 90 100 Lift force [N] Simulation time [sec] Deck 1 Deck 2 Investigation of Aerodynamic Interference between Twin Deck Bridges Pa (b) (c) Figure 5-6: Forces acting on the decks in configuration 1, (a) drag, (b) lift and (c) mom 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 30 40 50 60 70 80 90 100 Drag force [N] Simulation time [sec] Deck 1 Deck 2 -4 -3 -2 -1 0 1 2 3 4 30 40 50 60 70 80 90 100 Lift force [N] Simulation time [sec] Deck 1 Deck 2 -0.25 -0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2 0.25 0.3 30 40 50 60 70 80 90 100 Moment [Nm] Simulation time [sec] Deck 1 Deck 2 (b) (c) 30 40 50 60 70 80 90 100 Simulation time [sec] Deck 1 Deck 2 -4 -3 -2 -1 0 1 2 3 4 30 40 50 60 70 80 90 100 Lift force [N] Simulation time [sec] Deck 1 Deck 2 0.2 0.25 0.3 (b) -4 -3 -2 -1 0 1 2 3 4 30 40 50 60 70 80 90 100 Lift force [N] Simulation time [sec] D k 1 D k 2 (b) (c) Deck 1 Deck 2 -0.25 -0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2 0.25 0.3 30 40 50 60 70 80 90 100 Moment [Nm] Simulation time [sec] Deck 1 Deck 2 (c) Figure 5-6: Forces acting on the decks in configuration 1, (a) drag, (b) lift and (c) moment Figure 5-6: Forces acting on the decks in configuration 1, (a) drag, (b) lift and (c) m Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 43 Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 43 Page \| 43 Table 5-1. Investigation of Aerodynamic Interference between Twin Deck Bridges Comparison of drag force, lift force and pitch moment for a stationary twin deck and a twin deck suspended on elastic springs Drag force [N] Lift force [N] Pitch moment [Nm] dynamic stationary dynamic stationary dynamic stationary upstream deck 0.414 0.413 0.462 0.455 0.077 0.086 downstream deck 0.179 0.178 0.107 0.108 -0.024 -0.02 Table 5-2: Velocity vector field on the middle plane in consecutive time steps Table 5-1. Comparison of drag force, lift force and pitch moment for a stationary twin deck and a twin deck suspended on elastic springs Drag force [N] Lift force [N] Pitch moment [Nm] dynamic stationary dynamic stationary dynamic stationary upstream deck 0.414 0.413 0.462 0.455 0.077 0.086 downstream deck 0.179 0.178 0.107 0.108 -0.024 -0.02 Table 5-2: Velocity vector field on the middle plane in consecutive time steps able 5-2: Velocity vector field on the middle plane in consecutive time steps Table 5-2: Velocity vector field on the middle plane in consecutive time steps Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 44 stigation of Aerodynamic Interference between Twin Deck Bridges Page Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 45 Table 5-3: Velocity streamlines in consecutive time steps Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 46 Table 5-3: Velocity streamlines in consecutive time steps Table 5-3: Velocity streamlines in consecutive time steps y p Page \| 46 Investigation of Aerodynamic Interference between Twin Deck Bridges vestigation of Aerodynamic Interference between Twin Deck Bridges Page Investigation of Aerodynamic Interference between Twin Deck Bridges 5.2. Configuration 2: the upstream deck is suspended on springs, the downstream deck is constrained In this configuration only the upstream deck is suspended on elastic springs. At the beginning of the simulation it is positioned 0.0246 m higher than the downstream deck to take into account the elongation of elastic springs. After the springs are activated, it starts to vibrate in the vertical direction around the equilibrium position (which is aligned with the downstream deck position) and rotate along the center line of its cross-section. The initial amplitude of 4.8 cm quickly decreases to a millimeter and the rotations at the end of the simulation oscillate between -0.238 deg and 0.238 deg. Resultant drag forces, lift forces and moments for both decks are presented in Figure 5-7. 5.2. Configuration 2: the upstream deck is suspended on springs, the downstream deck is constrained In this configuration only the upstream deck is suspended on elastic springs. At the beginning of the simulation it is positioned 0.0246 m higher than the downstream deck to take into account the elongation of elastic springs. After the springs are activated, it starts to vibrate in the vertical direction around the equilibrium position (which is aligned with the downstream deck position) and rotate along the center line of its cross-section. The initial amplitude of 4.8 cm quickly decreases to a millimeter and the rotations at the end of the simulation oscillate between -0.238 deg and 0.238 deg. Resultant drag forces, lift forces and moments for both decks are presented in Figure 5-7. Page \| 48 Page \| 48 Investigation of Aerodynamic Interference between Twin Deck Bridges (a) (b) (c) 0 0.2 0.4 0.6 0.8 1 30 40 50 60 70 80 90 100 Drag force [N] Simulation time [sec] Deck 1 Deck 2 -3 -2 -1 0 1 2 3 30 40 50 60 70 80 90 100 Lift force [N] Simulation time [sec] Deck 1 Deck 2 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 30 40 50 60 70 80 90 100 Moment [Nm] Simulation time [sec] (a) 0 0.2 0.4 0.6 0.8 1 30 40 50 60 70 80 90 100 Drag force [N] Si l ti ti [ ] (a) vestigation of Aerodynamic Interference between Twin Deck Bridges Pag (a) (b) (c) Figure 5-7: Forces and moment acting on the decks in configuration 2, (a) drag force, (b) force, (c) moment. 0 0.2 0.4 0.6 0.8 1 30 40 50 60 70 80 90 100 Drag force [N] Simulation time [sec] Deck 1 Deck 2 -3 -2 -1 0 1 2 3 30 40 50 60 70 80 90 100 Lift force [N] Simulation time [sec] Deck 1 Deck 2 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 30 40 50 60 70 80 90 100 Moment [Nm] Simulation time [sec] Deck 1 Deck 2 (b) 0 0.2 0.4 30 40 50 60 70 80 90 100 Drag fo Simulation time [sec] Deck 1 Deck 2 -3 -2 -1 0 1 2 3 30 40 50 60 70 80 90 100 Lift force [N] Simulation time [sec] Deck 1 Deck 2 (b) Deck 1 Deck 2 -3 -2 -1 0 1 2 3 30 40 50 60 70 80 90 100 Lift force [N] Simulation time [sec] Deck 1 Deck 2 (b) (c) Figure 5-7: Forces and moment acting on the decks in configuration 2, (a) drag force, (b) lift force, (c) moment. Investigation of Aerodynamic Interference between Twin Deck Bridges -3 -2 -1 30 40 50 60 70 80 90 100 Lift fo Simulation time [sec] Deck 1 Deck 2 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 30 40 50 60 70 80 90 100 Moment [Nm] Simulation time [sec] Deck 1 Deck 2 (c) Figure 5-7: Forces and moment acting on the decks in configuration 2, (a) drag force, (b) Deck 1 Deck 2 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 30 40 50 60 70 80 90 100 Moment [Nm] Simulation time [sec] Deck 1 Deck 2 (c) Figure 5-7: Forces and moment acting on the decks in configuration 2, (a) drag force, (b) lift force, (c) moment. Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 49 Page \| 49 5.3. Configuration 3: the upstream deck is constrained, the downstream deck is suspended on springs The force components are also established for a third configuration, in which the upstream deck is constrained and the downstream deck is attached to springs, allowing for vertical motion and rotation about the deck center axis. The initial vertical position of the downstream deck is 0.0246 m higher than the upstream deck. After the springs are activated, it starts to vibrate in the vertical direction around the equilibrium position (which is aligned with the upstream deck position). Figure 5-8 summarizes the force and moment history during the simulation. (a) (b) 0 0.2 0.4 0.6 0.8 1 30 40 50 60 70 80 90 100 Drag force [N] Simulation time [sec] Deck 1 Deck 2 -3 -2 -1 0 1 2 3 30 40 50 60 70 80 90 100 Lift force [N] Simulation time [sec] Deck 1 Deck 2 (a) (a) 0 0.2 0.4 0.6 0.8 1 30 40 50 60 70 80 90 100 Drag force [N] (b) Deck 1 Deck 2 -3 -2 -1 0 1 2 3 30 40 50 60 70 80 90 100 Lift force [N] Simulation time [sec] Deck 1 Deck 2 (b) Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 50 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 50 (c) (c) Figure 5-8: Forces and moment acting on the decks in configuration 3, (a) drag force, (b) lift force, (c) moment -0.3 -0.2 -0.1 0 0.1 0.2 0.3 30 40 50 60 70 80 90 100 Moment [Nm] Simulation time [sec] Deck 1 Deck 2 Figure 5-8: Forces and moment acting on the decks in configuration 3, (a) drag force, (b) lift force, (c) moment 5.4. Comparison of results for configurations 1, 2, and 3 Figure 5-9 and Figure 5-10 compare the force and moment between the stationary model and the time averaged force values obtained from three DFBI models. The averaging is done over the last ten seconds of the simulation, meaning between 90 seconds and 100 seconds of simulation time. The upstream deck does not experience significant changes in drag forces. The difference in drag forces is small, the differences are less than 2%. The difference in lift forces is up to 4%. The differences in pitch moments reach a maximum of 12%. The drag force acting on the downstream deck changes the value by only 1%. The difference in the lift force and moment acting on the downstream deck is more significant. The lift force changes between -7% (in configuration 2) to 19% (configuration 3). The biggest difference in pitch moment is between the base case and configuration 2, where pitch moment increased by 25%. Page \| 51 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges (a) (b) (c) 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 Drag force [N] stationary decks configuration 1 configuration 2 configuration 3 0 0.1 0.2 0.3 0.4 0.5 Lift force [N] stationary decks configuration 1 configuration 2 configuration 3 0.06 0.08 0.1 [Nm] (a) (b) (c) Figure 5-9: Comparison of force values acting on deck 1 at different configuration 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 Drag force [N] stationary decks configuration 1 configuration 2 configuration 3 0 0.1 0.2 0.3 0.4 0.5 Lift force [N] stationary decks configuration 1 configuration 2 configuration 3 0 0.02 0.04 0.06 0.08 0.1 Moment [Nm] stationary decks configuration 1 configuration 2 configuration 3 (a) 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 Drag force [N] stationary decks configuration 1 configuration 2 configuration 3 (a) stationary decks configuration 1 configuration 2 configuration 3 (b) 0 0.1 0.2 0.3 0.4 0.5 Lift force [N] stationary decks configuration 1 configuration 2 configuration 3 (b) stationary decks configuration 1 configuration 2 configuration 3 (c) 0 0.02 0.04 0.06 0.08 0.1 Moment [Nm] stationary decks configuration 1 configuration 2 configuration 3 (c) stationary decks configuration 1 configuration 2 configuration 3 Figure 5-9: Comparison of force values acting on deck 1 at different configurations Figure 5-9: Comparison of force values acting on deck 1 at different configurations Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 52 Page \| 52 (a) (b) 0 0.05 0.1 0.15 0.2 Drag force [N] stationary decks configuration 1 configuration 2 configuration 3 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 Lift force [N] stationary decks configuration 1 configuration 2 configuration 3 (a) (b) 0 0.05 0.1 0.15 0.2 Drag force [N] stationary decks configuration 1 configuration 2 configuration 3 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 Lift force [N] stationary decks configuration 1 configuration 2 configuration 3 (a) (b) (c) Figure 5-10: Comparison of force values acting on deck 2 at different configuration 0 0.05 0.1 0.15 0.2 Drag force [N] stationary decks configuration 1 configuration 2 configuration 3 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 Lift force [N] stationary decks configuration 1 configuration 2 configuration 3 -0.03 -0.025 -0.02 -0.015 -0.01 -0.005 0 Moment [Nm] stationary decks configuration 1 configuration 2 configuration 3 a) 0 0.05 0.1 0.15 0.2 Drag force [N] stationary decks configuration 1 configuration 2 configuration 3 (a) stationary decks configuration 1 configuration 2 configuration 3 (b) (b) 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 Lift force [N] stationary decks configuration 1 configuration 2 configuration 3 stationary decks configuration 1 configuration 2 configuration 3 (c) -0.03 -0.025 -0.02 -0.015 -0.01 -0.005 0 Moment [Nm] stationary decks configuration 1 configuration 2 configuration 3 (c) stationary decks configuration 1 configuration 2 configuration 3 Figure 5-10: Comparison of force values acting on deck 2 at different configurations Figure 5-10: Comparison of force values acting on deck 2 at different configurations Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 53 Page \| 53 6. Validation of the CFD modeling approach A study of the aerodynamic characteristics of the Stonecutters Bridge, described in chapter 1, is used as a validation reference of the modeling approach for analyzing interference between the decks of twin-deck bridges. The model was solved using unsteady Reynolds Averaged Navier- Stokes equations with a k-epsilon turbulence model and a trimmed cell computational mesh. The bridge cross section is presented in Figure 6-1 with full-scale dimensions. Figure 6-2 shows the adopted sign convention for the angle of attack, α, and force components (FD – drag force, FL – lift force, M – pitch moment), as well as for the approach flow velocity U. The main dimensions of the prototype are as follows: the total width of the twin deck is B=53.3 m, width of a single deck is C=19.5 m, and deck height is H=3.93 m. Figure 6-1: Stonecutters bridge cross-section. All dimensions are given in meters. Figure 6-2: Sign convention 53.3 19.5 1.25 5.35 4.5 6.25 3.4 2.61 0.94 0.92 1.04 3.93 1.4 C L BCHCL α FL FD M U Figure 6-1: Stonecutters bridge cross-section. All dimensions are given in meters. 53.3 19.5 1.25 5.35 4.5 6.25 3.4 2.61 0.94 0.92 1.04 3.93 1.4 C L BCHCL Fi 6 1 St tt b id ti All di i i i t 53.3 19.5 1.25 5.35 4.5 6.25 3.4 2.61 0.94 0.92 1.04 3.93 1.4 C L BCHCL 53.3 Figure 6-1: Stonecutters bridge cross-section. All dimensions are given in meters. Figure 6-2: Sign convention α FL FD M U Figure 6-2: Sign convention α FL FD M U M FL Figure 6-2: Sign convention The study covers computation of static forces and corresponding coefficients (a) at varying flow velocity in parallel flow, (b) at varying direction of the approach flow for a chosen flow velocity, and (c) at velocities at which vortex induced vibrations were observed. A set of computations was performed at a zero-degree angle of incidence and different approach flow velocities to assess the sensitivity of aerodynamic force coefficients to Reynolds number. The range of Reynolds number for these cases ranged from 15,000 to 62,000 (based on the height of the decks, H). Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges 6. Validation of the CFD modeling approach The force coefficients are calculated as follows: The force coefficients are calculated as follows: The force coefficients are calculated as follows: 𝐶𝐷= 𝐹𝐷 0.5𝜌𝑈2𝐵𝐿 , (1) (1) 𝐶𝐷= 𝐹𝐷 0.5𝜌𝑈2𝐵𝐿 , Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 54 Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 54 Page \| 54 𝐶𝐿= 𝐹𝐿 0.5𝜌𝑈2𝐵𝐿 , (2) 𝐶𝑀= 𝑀 0.5𝜌𝑈2𝐵2𝐿2 , (3) (2) (3) with B, the total width of the bridge, and taken as the characteristic length, 𝐿 is a unit length. Figure 6-3 presents a plot of the sensitivity of the force coefficients to Reynolds number change at zero angle of attack. The drag and moment coefficients differ for the lowest Re, but overall they are independent of flow velocity. The lift coefficient goes asymptotic later than drag force and pitch moment. Its absolute value keeps increasing up to Re=47,000, and tends to become constant for higher values of Re. These results are consistent with those reported in [22]. Figure 6-4, Figure 6-5, and Figure 6-6 illustrate the drag, lift and pitch moment coefficient obtained in CFD simulations and tests versus attack angles. They show a good agreement between the simulation and experimental results found in the literature [25], except for drag force and pitch moment at -10 degrees, and lift force and pitch moment at 10 degrees. This is not surprising because at larger angles of attack the variance of experimental measurements is higher. Figure 6-3: Sensitivity of force coefficients to Reynolds number -0.15 -0.1 -0.05 0 0.05 0.1 1.E+04 2.E+04 3.E+04 4.E+04 5.E+04 6.E+04 7.E+04 Force coefficient [-] Reynolds number [-] CD CL CM Figure 6-3: Sensitivity of force coefficients to Reynolds number Page \| 55 Investigation of Aerodynamic Interference between Twin Deck Bridges Figure 6-4: The drag coefficient at different angles of attack Figure 6-5: The lift coefficient at different angles of attack 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 -10 -8 -6 -4 -2 0 2 4 6 8 10 Drag coefficient [-] Angle of attack [deg] experiment CFD -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 -10 -8 -6 -4 -2 0 2 4 6 8 10 Lift coefficient [-] Angle of attack [deg] experiment CFD Figure 6-4: The drag coefficient at different angles of attack 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 -10 -8 -6 -4 -2 0 2 4 6 8 10 Drag coefficient [-] Angle of attack [deg] experiment CFD 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 -10 -8 -6 -4 -2 0 2 4 6 8 10 Drag coefficient [-] Figure 6-4: The drag coefficient at different angles of attack Figure 6-5: The lift coefficient at different angles of attack -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 -10 -8 -6 -4 -2 0 2 4 6 8 10 Lift coefficient [-] Angle of attack [deg] experiment CFD Figure 6-5: The lift coefficient at different angles of attack -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 -10 -8 -6 -4 -2 0 2 4 6 8 10 Lift coefficient [-] Angle of attack [deg] experiment CFD Figure 6-5: The lift coefficient at different angles of attack Page \| 56 Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 56 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Figure 6-6: The pitch moment coefficient at different angles of attack -0.15 -0.1 -0.05 0 0.05 0.1 0.15 -10 -8 -6 -4 -2 0 2 4 6 8 10 Moment coefficient [-] Angle of attack [deg] experiment CFD Figure 6-6: The pitch moment coefficient at different angles of attack At the design stage of the Stonecutters Bridge it was discovered that the bridge was prone to vortex induced vibrations at certain values of Reynold numbers, which are Re=12,000 and Re= 91,000 [25]. Velocity and vorticity fields are examined around the decks at these two values of Reynolds numbers. Also, the drag, lift force, and moment coefficients, as well as Strouhal number are calculated. Reference [18] presents, among others things, results of a comparison of Strouhal number values at different central gap widths. The Strouhal number was established in two ways: from the spectra of its wake flow and integrated lift force. The corresponding values for the selected test configuration (selected gap, 14.3 m) are equal to 0.23 and 0.27, respectively. Figure 6-7 presents the velocity fields and Figure 6-8 presents the vorticity fields on a vertical plane cut through the bridge decks at Re=12,000 and Re=91,000 respectively. A clear trail of twin vortices shows behind the decks at Re=12,000. For the higher Re, the oscillations are not as pronounced. In both cases, the downstream deck is located in the wake of the upstream deck and is subjected to the oscillating vortices separating from the upstream deck. Page \| 57 Investigation of Aerodynamic Interference between Twin Deck Bridges (a) (b) Figure 6-7: Velocity fields on a vertical plane at (a) Re=1.2e4, (b) 9.1e4 (a) (a) Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 58 (b) Figure 6-7: Velocity fields on a vertical plane at (a) Re=1.2e4, (b) 9.1e4 (a) (b) (b) Figure 6-7: Velocity fields on a vertical plane at (a) Re=1.2e4, (b) 9.1e4 Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 58 Figure 6-7: Velocity fields on a vertical plane at (a) Re=1.2e4, (b) 9.1e4 (a) Figure 6-7: Velocity fields on a vertical plane at (a) Re=1.2e4, (b) 9.1e4 (a) (a) Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 58 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 58 Page \| 58 (b) Figure 6-8: Vorticity fields on a vertical plane at (a) Re=1.2e4, (b) 9.1e4 (b) Figure 6-8: Vorticity fields on a vertical plane at (a) Re=1.2e4, (b) 9.1e4 (b) Figure 6-8: Vorticity fields on a vertical plane at (a) Re=1.2e4, (b) 9.1e4 The static force coefficients are calculated according to the formulas presented at the beginning of the chapter, equations (1), (2), and (3). The Strouhal number is calculated as follows: The static force coefficients are calculated according to the formulas presented at the beginning of the chapter, equations (1), (2), and (3). The Strouhal number is calculated as follows: 𝑆𝑡= 𝑓𝑠𝐻 𝑈, (4) 𝑆𝑡= 𝑓𝑠𝐻 𝑈, (4) where 𝑓𝑠 is the vortex shedding frequency. The vortex shedding frequency was derived from the period of the oscillating lift force and is equal 0.00312 Hz at Re=12,000, and 0.0223Hz at Re=91,000. where 𝑓𝑠 is the vortex shedding frequency. The vortex shedding frequency was derived from the period of the oscillating lift force and is equal 0.00312 Hz at Re=12,000, and 0.0223Hz at Re=91,000. Table 6-1: Force coefficients and Strouhal number obtained from CFD simulations Re = 1.2e4 Re = 9.1e4 CFD Ref. [18], [25] CFD Ref. [18], [25] CD 0.051 0.048 0.045 0.050 CL -0.125 -0.048 -0.115 -0.110 CM 0.003 0.017 0.003 0.009 St 0.24 0.23 0.22 0.27 Table 6-1: Force coefficients and Strouhal number obtained from CFD simulations The combined computational results of force coefficients and Strouhal number obtained from CFD simulations are presented in Table 6-1. Investigation of Aerodynamic Interference between Twin Deck Bridges The drag coefficient, as well as the lift coefficient, decrease in absolute value for higher Reynolds number, whereas the moment coefficient doesn’t experience any difference. Compared to the numbers found in the literature ([18], [25]) for Re =1.2e4, the drag, lift and moment coefficients differ by 6%, 160% and -82%, respectively. For the higher value of Reynolds number, Re=9.1e4, the coefficients differ by -10%, 4.5%, and 67%. The Strouhal numbers for the two considered Reynolds numbers differ by 4% and -19%, respectively. The combined computational results of force coefficients and Strouhal number obtained from CFD simulations are presented in Table 6-1. The drag coefficient, as well as the lift coefficient, decrease in absolute value for higher Reynolds number, whereas the moment coefficient doesn’t experience any difference. Compared to the numbers found in the literature ([18], [25]) for Re =1.2e4, the drag, lift and moment coefficients differ by 6%, 160% and -82%, respectively. For the higher value of Reynolds number, Re=9.1e4, the coefficients differ by -10%, 4.5%, and 67%. The Strouhal numbers for the two considered Reynolds numbers differ by 4% and -19%, respectively. 7. Conclusions The main goal of the present study is to assess the capabilities of 3D CFD computations in a parametric study of a twin deck bridge. The aerodynamic forces and response of a rigid section model with a generic cross-section was investigated. Gap-to-width ratios, wind speed, and wind direction were considered using three-dimensional CFD simulations. Static and dynamic responses of the decks were characterized, including computation of steady state aerodynamic forces and pitching moment in URANS as well as LES simulations to capture the formation of large eddies. Mesh density sensitivity tests revealed that there is strong influence on the results depending on the type of the mesh and wall Y+ value. There are also differences between the unsteady URANS solver with k-ε turbulence model and LES computations. A comparison of results led to a conclusion that in stationary computations it will be beneficial to use URANS and LES models in combination with a polyhedral mesh that is denser around the decks, so that wall Y+ is close to one. Polyhedral meshes are relatively easy and efficient to build and they contain significantly fewer cells than a hexahedral mesh for the same accuracy. A low wall Y+ number was chosen to resolve the boundary layer accurately. In dynamic simulations only the URANS model was used in order to save on computational time and resources. The observed pressure and velocity fields for a single deck serve as a reference for twin-deck models. The study shows that forces acting on the upstream deck in a tandem configuration, regardless of the spacing between decks, are similar to the single deck results. The values of lift and moment on the upstream deck are higher than for a single deck and they are almost constant regardless the gap size. The downstream deck experiences less drag because it is shielded to some extent by the upstream deck. This effect diminishes as the gap increases. For instance, when gap- to-width ratio equals 0.13, the upstream deck is subjected to a 0.415 N drag force, whereas the downstream deck is subjected to 0.148 N drag force. At gap-to-width ratio of 0.79, the upstream deck is subjected to a 0.421 N drag force, whereas the downstream deck is subjected to 0.241 N drag force. The drag force acting on the downstream deck increased by 63%. The lift force also increases as the gap between the twin decks gets bigger. Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 59 Page \| 59 7. Conclusions The lift force acting on the upstream deck equals 0.472 N and on the downstream deck 0.11 N, when the gap-to-width ratio is 0.13. When the gap-to-width ratio equals 0.79, then the lift force acting on the upstream deck equals 0.451 N and on the downstream deck 0.135 N. Therefore, the lift force acting on the downstream deck increased by 23%. The pitch moment does not vary significantly for the upstream deck, staying almost constant and equal to 0.086 Nm. The absolute value of the pitch moment for the downstream deck increases from -0.016 Nm at L/B=0.13, to -0.032 Nm at L/B=0.79, which gives a 100% increase. The force changes due to varying angle of attack were computed and combined in graphs. The character of the curves representing drag forces acting on both decks is similar. The drag force is always positive and it increases with the increase of the absolute angle of attack, reaching a maximum at the largest absolute angle value. The lift force changes sign depending on the angle of attack. This tendency occurs for the one deck model as well as for the twin deck model. For negative angles it assumes negative values, goes through zero for angles in the range from -5 deg to 0 deg, and reaches positive values for positive angles. The highest value of the lift force was obtained for the downstream deck at the biggest angle. It is almost two times higher for the Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 60 Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 60 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 60 60 downstream deck than the upstream deck. Values of the pitch moment are small compared to the force values for the selected range of the flow directions. The pitch moment for the upstream deck reaches a maximum for parallel flow and decreases as the absolute value of the angle of attack increases. The values of the pitch moment for downstream deck are negative for most of the considered angles of incidence. The aerodynamic response of the decks was investigated with the use of the Dynamic Fluid Body Interaction (DFBI) solver. The URANS solver with the k-ε turbulence model was used to solve the fluid flow. The model configurations that were taken into account are: both decks are stationary, both decks are suspended on springs; the upstream deck was suspended on springs and the downstream deck was constrained; and the upstream deck was constrained and the downstream deck was suspended on springs. The first two cases are the most common in testing of bridge sections. The latter, are usually not considered. The configuration with stationary decks was used as a reference case in the analysis. The configuration with both decks on springs was analyzed in more detail. Vertical motion and rotation of the decks were monitored and recorded. A Fast Fourier Transform was used to establish the frequencies of the vertical motion and rotation of the decks, which are 3.14 Hz and 3.24 Hz, respectively, for both decks. The oscillations differed with regards to the amplitude, which was higher for the downstream deck. The drag and lift forces and pitch moment were recorded for both decks in each configuration. The drag forces acting on the upstream deck did not experience significant changes between the different configurations of decks on springs. The highest difference is between the stationary decks and configuration 3, where the upstream deck is constrained but not the downstream deck, with the relative difference less than 1%. The lift forces for dynamic simulations increase, compared to the static computations. The biggest difference is 4% for configuration 2, where the downstream deck is constrained but the upstream deck is not. The most significant relative difference of was noticed in the pitch moments. They decreased by 11% in configuration 3, and by 12% in configurations 1 and 2. Investigation of Aerodynamic Interference between Twin Deck Bridges The influence on the forces acting on the downstream deck are much more pronounced. The drag force stays at the same level, with the differences only up to 1%. The lift force experiences a change between -7% (configuration 2) to 19% (configuration 3). The highest lift force occurs when the upstream deck is constrained and the downstream is not (configuration 3). The reason appears to be that when the upstream deck is constrained, the downstream deck moves to more exposed locations in the wake of the upstream deck. When both decks are unconstrained they tend to move in tandem, with the same frequency. The pitch moments increase by up to 25% in the configuration 2, compared to the stationary decks. As the comparison of results for different deck configurations shows, the computational model needs to represent the model tested in the laboratory as closely as possible. For different combinations of restraints on the decks, the response of the upstream and downstream decks is different, therefore when comparing with laboratory measurements where dynamic response is enabled, the same type of dynamic response needs to be included in the model. It is worth noticing that there was no significant difference in computational time between the simulations with one or two DFBI bodies, so there is no additional cost other than setting up the model in including the dynamic response of both decks. Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 61 In conclusion, an extensive study was performed on the influence of air flow around a twin deck bridge section model. The CFD software used provides an array of features that are useful in aerodynamic simulations, assessment of effects of parameter variation and, visualization of results to gain insight into the flow and pressure around twin deck bridge. Modeling of the dynamic response of the decks is also possible using the dynamic fluid body interaction capabilities of the software to analyze the system with the decks mounted on elastic springs, which is very similar to a method of wind tunnel testing of dynamic response. 8. Acknowledgements The funding for this project came from the Wind Research Program at the Turner-Fairbank Highway Research Center, through Interagency Agreement Number DTFH61-14-X-300002 between DOT and DOE, and the work was performed under DOE’s contract with UChicago Argonne, LLC, contract no. DE-AC02-06-CH11357. 9. References (2014). "Vortex induced vibrations of a bridge deck: Dynamic response and surface pressure distribution." Journal of Wind Engineering and Industrial Aerodynamics 133, no. 0: 160-168 [9] [9] Li, Hui, Shujin Laima, and Haiquan Jing. (2014). "Reynolds number effects on aerodynamic characteristics and vortex-induced vibration of a twin-box girder." Journal of Fluids and Structures 50: 358-375 [10] [10] Seo, Ju-Won, Ho-Kyung Kim, Jin Park, et al. (2013). "Interference effect on vortex- induced vibration in a parallel twin cable-stayed bridge." Journal of Wind Engineering and Industrial Aerodynamics 116: 7-20. [11] [11] Corriols, Abraham Sanchez, and Guido Morgenthal. (2014). "Vortex-induced vibrations on cross sections in tandem arrangement." Structural Engineering International: Journal of the International Association for Bridge and Structural Engineering (IABSE) 24, no. 1: 20-26 [12] Laima, Shujin, Hui Li, Wenli Chen, et al. (2013). "Investigation and control of vortex- induced vibration of twin box girders." Journal of Fluids and Structures 39: 205-221. [13] Wang, Qi, Haili Liao, and Mingshui Li. (2013). "Study on wind resistant performance of multiple spans cable-stayed bridge." 6th European and African Conference on Wind Engineering, EACWE 2013, July 7, 2013 - July 13, 2013: International Association for Wind Engineering (IAWE). [14] Nieto, F., S. Hernandez, I. Kusano, et al. (2014). "Assessment of the aerodynamic response of bridge decks by means of 2D Reynolds averaged Navier-Stokes simulations." 10th International Conference on Advances in Fluid Mechanics, AFM 2014, July 1, 2014 - July 3, 2014. WIT Transactions on Engineering Sciences 82: 407-417 [15] Lee, Min-Jae, Si-Chul Kim, Young-Hwa Seo, et al. (2012). "The Yi Sun-sin Bridge: Innovative solutions for suspension bridges." Structural Engineering International: Journal of the International Association for Bridge and Structural Engineering (IABSE) 22, no. 1: 32-35 [16] Liu, Zhiwen, Steve C. S. Cai, and Zhengqing Chen. (2010). "Numerical Simulation of Aerodynamic Interference Effects on Aerostatic Coefficients of Two Bluff Sections in Tandem." The Fifth International Symposium on Computational Wind Engineering (CWE2010), Chapel Hill, North Carolina, USA May 23-27, 2010 [17] Meng, Xiaoliang, Ledong Zhu, and Zhenshan Guo. (2011). "Aerodynamic interference effects and mitigation measures on vortex-induced vibrations of two adjacent cable-stayed bridges." Frontiers of Architecture and Civil Engineering in China 5, no. 4: 510-517 [18] Nieto, F., S. Hernandez, I. Kusano, et al. (2014). "Assessment of the aerodynamic response of bridge decks by means of 2D Reynolds averaged Navier-Stokes simulations." 10th International Conference on Advances in Fluid Mechanics, AFM 2014, July 1, 2014 - July 3, 2014. 9. References [1] Wang, Jie, Jinyun Zhao, and Jianxin Liu. (2012). "Experimental study of aerodynamic interference effects on double thin-walled hollow pier in Tandem arrangement." 4th International Conference on Technology of Architecture and Structure, ICTAS 2011, September 22, 2011 - September 24, 2011. Advanced Materials Research 368-373: 1517- 1520, Trans Tech Publications [2] Trein, Cristiano Augusto, Hiromichi Shirato, and Masaru Matsumoto. (2015). "On the effects of the gap on the unsteady pressure characteristics of two-box bridge girders." Engineering Structures 82, no. 0: 121-133 [3] Yang, Yongxin, Teng Wu, Yaojun Ge, et al. (2015). "Aerodynamic Stabilization Mechanism of a Twin Box Girder with Various Slot Widths." Journal of Bridge Engineering 20, no. 3 [3] Yang, Yongxin, Teng Wu, Yaojun Ge, et al. (2015). "Aerodynamic Stabilization Mechanism of a Twin Box Girder with Various Slot Widths." Journal of Bridge Engineering 20, no. 3 [4] Laima, Shujin, and Hui Li. (2015). "Effects of gap width on flow motions around twin-box girders and vortex-induced vibrations." Journal of Wind Engineering and Industrial [4] Laima, Shujin, and Hui Li. (2015). "Effects of gap width on flow motions around twin-box girders and vortex-induced vibrations." Journal of Wind Engineering and Industrial Aerodynamics 139: 37-49. [5] [5] Yang, YongXin, Rui Zhou, and Yao Jun Ge. (2014). "Aerodynamic Flutter Control of Parallel Cable-Stayed Bridges." IABSE Symposium Report, IABSE Madrid Symposium: Engineering for Progress, Nature and People 3190-3197, International Association for Bridge and Structural Engineering. [6] [6] Chen, Wen-Li, Hui Li, and Hui Hu. (2014). "An experimental study on the unsteady vortices and turbulent flow structures around twin-box-girder bridge deck models with different gap ratios." Journal of Wind Eng. and Industrial Aerodynamics 132: 27-36 [6] Chen, Wen-Li, Hui Li, and Hui Hu. (2014). "An experimental study on the unsteady vortices and turbulent flow structures around twin-box-girder bridge deck models with different gap ratios." Journal of Wind Eng. and Industrial Aerodynamics 132: 27-36 [7] [7] Li, Hui, Shujin Laima, and Haiquan Jing. (2014). "Reynolds number effects on aerodynamic characteristics and vortex-induced vibration of a twin-box girder." Journal of Fluids and Structures 50: 358-375 [7] Li, Hui, Shujin Laima, and Haiquan Jing. (2014). "Reynolds number effects on aerodynamic characteristics and vortex-induced vibration of a twin-box girder." Journal of Fluids and Structures 50: 358-375 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 62 Page \| 62 [8] [8] Belloli, M., F. Fossati, S. Giappino, et al. 9. References WIT Transactions on Engineering Sciences 82: 407-417, WITPress [19] Meng, Xiaoliang, Ledong Zhu, and Zhenshan Guo. (2011). "Aerodynamic interference effects and mitigation measures on vortex-induced vibrations of two adjacent cable-stayed bridges." Frontiers of Architecture and Civil Engineering in China 5, no. 4: 510-517 [19] Meng, Xiaoliang, Ledong Zhu, and Zhenshan Guo. (2011). "Aerodynamic interference effects and mitigation measures on vortex-induced vibrations of two adjacent cable-stayed bridges." Frontiers of Architecture and Civil Engineering in China 5, no. 4: 510-517 [20] Kwok, K. C. S., X. R. Qin, C. H. Fok, et al. (2012). "Wind-induced pressures around a sectional twin-deck bridge model: Effects of gap-width on the aerodynamic forces and vortex shedding mechanisms." J. of Wind Eng. and Industrial Aerodynamics 110: 50-61 [20] Kwok, K. C. S., X. R. Qin, C. H. Fok, et al. (2012). "Wind-induced pressures around a sectional twin-deck bridge model: Effects of gap-width on the aerodynamic forces and vortex shedding mechanisms." J. of Wind Eng. and Industrial Aerodynamics 110: 50-61 Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 63 Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 63 Page \| 63 [21] Qin, X. R., K. C. S. Kwok, C. H. Fok, et al. (2007). "Wind-induced self-excited vibrations of a twin-deck bridge and the effects of gap-width." Wind and Structures, An International Journal 10, no. 5: 463-479 [22] Fok, Chin Hong. (2006). "Aerodynamic characteristics of long-span twin-deck bridges." Thesis (M.Phil. Civil Engineering), Hong Kong University of Science and Technology [23] Fok, C.H., K.C.S. Kwok, P.A. Hitchcock, et al. (2006). "Effects of Gap width on a Twin- deck Bridge: Part 1 – Vortex Shedding Excitation." Proceedings of 12th AWES Wind Engineering Workshop, Darwin, Australia [24] Fok, C.H., K.C.S. Kwok, P.A. Hitchcock, et al. (2006). "Effects of Gap-width on a Twin Deck Bridge: Part 2 – Aerodynamic Admittance Functions." Proceedings of 12th AWES Wind Engineering Workshop, Darwin, Australia [25] Nieto F., Kusano I., Hernandez S., Jurado J.A. (2010) “CFD analysis of the vortex- shedding response of a twin-box deck cable-stayed bridge” 5th International Symposium on Computational Wind Engineering, Chapel Hill, North Carolina, USA, May 23-27, 2010 [26] Hui M., Ding Q.S. (2006) “Flutter analysis of Stonecutters bridge” Wind and Structures [25] Nieto F., Kusano I., Hernandez S., Jurado J.A. (2010) “CFD analysis of the vortex- shedding response of a twin-box deck cable-stayed bridge” 5th International Symposium on Computational Wind Engineering, Chapel Hill, North Carolina, USA, May 23-27, 2010 shedding response of a twin-box deck cable-stayed bridge 5th International Symposium on Computational Wind Engineering, Chapel Hill, North Carolina, USA, May 23-27, 2010 [26] Hui M., Ding Q.S. (2006) “Flutter analysis of Stonecutters bridge” Wind and Structures 9: 125-146 [26] Hui M., Ding Q.S. (2006) “Flutter analysis of Stonecutters bridge” Wind and Structures 9: 125-146 [27] LS-PrePost online documentation www.lstc.com/lspp/index.shtml [27] LS-PrePost online documentation www.lstc.com/lspp/index.shtml Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 64 Energy Systems Division Argonne National Laboratory 9700 South Cass Avenue Argonne, IL 60439-4815 www.anl.gov Argonne National Laboratory is a U.S. Department of Energy laboratory managed by UChicago Argonne, LLC Argonne National Laboratory is a U.S. Department of Energy laboratory managed by UChicago Argonne, LLC Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 64 Investigation of Aerodynamic Interference between Twin Deck Bridges Page \| 65 Investigation of Aerodynamic Interference between Twin Deck Bridges Investigation of Aerodynamic Interference between Twin Deck Bridges
https://openalex.org/W4205250418	https://www.frontiersin.org/articles/10.3389/fmicb.2021.817200/pdf	English	null	Pseudotyped Vesicular Stomatitis Virus-Severe Acute Respiratory Syndrome-Coronavirus-2 Spike for the Study of Variants, Vaccines, and Therapeutics Against Coronavirus Disease 2019	Frontiers in microbiology	2,022	cc-by	13,836	REVIEW published: 14 January 2022 doi: 10.3389/fmicb.2021.817200 REVIEW published: 14 January 2022 doi: 10.3389/fmicb.2021.817200 Pseudotyped Vesicular Stomatitis Virus-Severe Acute Respiratory Syndrome-Coronavirus-2 Spike for the Study of Variants, Vaccines, and Therapeutics Against Coronavirus Disease 2019 Marcela Salazar-García1,2, Samyr Acosta-Contreras2, Griselda Rodríguez-Martínez2, Armando Cruz-Rangel3, Alejandro Flores-Alanis4, Genaro Patiño-López5 and Victor M. Luna-Pineda2,5* Marcela Salazar-García1,2, Samyr Acosta-Contreras2, Griselda Rodríguez-Martínez2, Armando Cruz-Rangel3, Alejandro Flores-Alanis4, Genaro Patiño-López5 and Victor M. Luna-Pineda2,5* Keywords: vesicular stomatitis virus, SARS-CoV-2, Biosafety Level 3, pseudotyped viruses, pseudovirus, glycoprotein, ppVSV1G-SARS-CoV-2 S Edited by: Miguel Angel Martinez, IrsiCaixa, Spain An alternative to these BSL-3/-4 laboratories is to use a pseudotyped virus that can be handled in a BSL-2 laboratory for study purposes. Recombinant Vesicular Stomatitis Virus (VSV) can be generated with complementary DNA from complete negative-stranded genomic RNA, with deleted G glycoprotein and, instead, incorporation of other fusion protein, like SARS-CoV-2 Spike (S protein). Accordingly, it is called pseudotyped VSV-SARS-CoV-2 S. In this review, we have described the generation of pseudotyped VSV with a focus on the optimization and application of pseudotyped VSV-SARS-CoV-2 S. The application of this pseudovirus has been addressed by its use in neutralizing antibody assays in order to evaluate a new vaccine, emergent SARS-CoV-2 variants (delta and omicron), and approved vaccine efﬁcacy against variants of concern as well as in viral fusion-focused treatment analysis that can be performed under BSL-2 conditions. Specialty section: This article was submitted to Virology, a section of the journal Frontiers in Microbiology Received: 17 November 2021 Accepted: 20 December 2021 Published: 14 January 2022 Edited by: Miguel Angel Martinez, IrsiCaixa, Spain Edited by: Miguel Angel Martinez, IrsiCaixa, Spain 1 Laboratorio de Biología del Desarrollo y Teratogénesis Experimental, Hospital Infantil de México “Federico Gómez”, Mexico City, Mexico, 2 Laboratorio de Investigación en COVID-19, Hospital Infantil de México “Federico Gómez”, Mexico City, Mexico, 3 Laboratorio de Bioquímica de Enfermedades Crónicas, Instituto Nacional de Medicina Genómica, Mexico City, Mexico, 4 Departamento de Microbiología y Parasitología, Facultad de Medicina, Universidad Nacional Autónoma de México, Mexico City, Mexico, 5 Unidad de Investigación en Inmunología y Proteómica, Hospital Infantil de México “Federico Gómez”, Mexico City, Mexico Reviewed by: Daniel Smrz, Charles University in Prague, Czechia Melinda Ann Brindley, University of Georgia, United States Reviewed by: Daniel Smrz, Charles University in Prague, Czechia Melinda Ann Brindley, University of Georgia, United States Correspondence: Victor M. Luna-Pineda luna.pineda@hotmail.com Correspondence: Victor M. Luna-Pineda luna.pineda@hotmail.com *Correspondence: Victor M. Luna-Pineda luna.pineda@hotmail.com World Health Organization (WHO) has prioritized the infectious emerging diseases such as Coronavirus Disease (COVID-19) in terms of research and development of effective tests, vaccines, antivirals, and other treatments. Severe Acute Respiratory Syndrome- Coronavirus-2 (SARS-CoV-2), the etiological causative agent of COVID-19, is a virus belonging to risk group 3 that requires Biosafety Level (BSL)-3 laboratories and the corresponding facilities for handling. An alternative to these BSL-3/-4 laboratories is to use a pseudotyped virus that can be handled in a BSL-2 laboratory for study purposes. Recombinant Vesicular Stomatitis Virus (VSV) can be generated with complementary DNA from complete negative-stranded genomic RNA, with deleted G glycoprotein and, instead, incorporation of other fusion protein, like SARS-CoV-2 Spike (S protein). Accordingly, it is called pseudotyped VSV-SARS-CoV-2 S. In this review, we have described the generation of pseudotyped VSV with a focus on the optimization and application of pseudotyped VSV-SARS-CoV-2 S. The application of this pseudovirus has been addressed by its use in neutralizing antibody assays in order to evaluate a new vaccine, emergent SARS-CoV-2 variants (delta and omicron), and approved vaccine efﬁcacy against variants of concern as well as in viral fusion-focused treatment analysis that can be performed under BSL-2 conditions. World Health Organization (WHO) has prioritized the infectious emerging diseases such as Coronavirus Disease (COVID-19) in terms of research and development of effective tests, vaccines, antivirals, and other treatments. Severe Acute Respiratory Syndrome- Coronavirus-2 (SARS-CoV-2), the etiological causative agent of COVID-19, is a virus belonging to risk group 3 that requires Biosafety Level (BSL)-3 laboratories and the corresponding facilities for handling. Salazar-García M, Acosta-Contreras S, Rodríguez-Martínez G, Cruz-Rangel A, Flores-Alanis A, Patiño-López G and Luna-Pineda VM (2022) Pseudotyped Vesicular Stomatitis Virus-Severe Acute Respiratory Syndrome-Coronavirus-2 Spike for the Study of Variants, Vaccines, and Therapeutics Against Coronavirus Disease 2019. Front. Microbiol. 12:817200. doi: 10.3389/fmicb.2021.817200 PSEUDOTYPED VESICULAR STOMATITIS VIRUS: HISTORY Pseudotyped viruses are useful tools for studying the function of viral fusion proteins (Moore et al., 2004; Whitt, 2010; Mendenhall et al., 2012). PVs have been used in phenotypic mixing since the 1970s, in which two encapsulated viruses “share” coat proteins while having distinct genetic material (RNA or DNA) (Choppin and Compans, 1970; Závada, 1972; Huang et al., 1974). Thus, a temperature-sensitive (ts) mutant of VSV was reported, which was deﬁcient in the synthesis of its G protein at nonpermissive temperatures (Schnitzer et al., 1979). Several PVs were supplemented with foreign viral glycoproteins (GPs) using VSV strain ts045, including VSV-Avian sarcoma viruses, VSV-Rous sarcoma virus, VSV-Murine leukemia virus, VSV- Murine oncoviruses and -murine cytomegalovirus, and Rous- associated virus 1 is a VSV-Avian retrovirus (Weiss et al., 1977; Lodish and Weiss, 1979; Schnitzer and Gonczol, 1979; Weiss and Bennett, 1980). Furthermore, an infectious defective interfering (DI) particle was characterized as a VSV strain with faulty replication and ampliﬁcation of its RNA genome but proper viral packaging (Pattnaik and Wertz, 1990). These studies demonstrated RNAs produced from non-viral origins could be packaged into VSV particles. When co-expressed with the other VSV proteins, a full negative-stranded genomic RNA from a cDNA clone of a VSV DI RNA was replicated, transcribed, and packed into infectious particles (Pattnaik et al., 1992; Stillman et al., 1995). Full-length positive-sense RNA, complementary to the VSV genome, can be produced using the bacteriophage T7 RNA polymerase. Production of the full-length anti-genome along with proteins required for RNA replication (N, P, and L) enable the recovery of replication-competent recombinant (r)VSV from DNA plasmids (Lawson et al., 1995; Whelan et al., 1995). Recovery of rVSVs lacking the glycoprotein open reading frame from the genome (ppVSV1G) is accomplished by supplying the VSV G in trans and they are named ppVSV1G-G (Schnell et al., 1996). Several reporter genes have been cloned into ppVSV1G permitting various experimental read-outs, and these included green and red ﬂuorescent protein (GFP/RFP/DsRed), Many nations across the globe lack the infrastructure and resources needed to research these emerging and re- emerging pathogens. There are just seven BSL-3 laboratories in Mexico, and they are typically overworked. INTRODUCTION study virus entry (Millet et al., 2019). PV, sometimes known as “pseudoviruses” or “pseudoparticles” (pp), are ampliﬁcation- defective viruses capable of just one cycle of replication that can infect host cells in the same way as wild-type viruses do (Takada et al., 1997). PVs are primarily derived from retroviruses [HIV and Murine Leukemia Virus (MLV)] and rhabdoviruses (VSV) and are used to investigate the function of viral fusion proteins in enveloped viruses such as lifecycle initiation, host and cellular tropism, interspecies transmission, viral pathogenesis, and host cell entry pathways (Moore et al., 2004; Whitt, 2010; Mendenhall et al., 2012). VSV is a negative polarity enveloped RNA virus with a genome size of 11 kb that contains ﬁve main viral proteins: nucleoprotein (N), phosphoprotein (P), matrix protein (M), glycoprotein (G), and large polymerase protein (L). VSV has been frequently utilized as an enveloped virus for the creation of eﬃcient PV harboring a foreign virus’s surface protein. Infectious diseases develop and re-emerge regularly, triggering epidemics and pandemics (Morens et al., 2008). Human Immunodeﬁciency Virus/Acquired Immunodeﬁciency Syndrome (HIV/AIDS; 1981), Nipah virus (1999), Severe Acute Respiratory Syndrome (SARS; 2002), Middle East Respiratory Syndrome (MERS; 2012), and Coronavirus Disease (COVID-19; 2019) are examples of newly emerging infectious diseases while re-emerging infectious diseases have reappeared in new locations, such as West Nile in the United States and Russia (Morens and Fauci, 2020). The World Health Organization’s (WHO) Research and Development Blueprint Initiative has prioritized infection diseases such as COVID-19, Crimean-Congo hemorrhagic fever, Ebola virus and Marburg virus diseases, Lassa fever, MERS and SARS, Nipah and henipaviral diseases, Rift Valley fever, and Zika for the development of eﬀective tests, vaccines, antivirals, and other treatments (Kieny et al., 2016). To be sure, the pathogens that cause these diseases are classiﬁed as risk group 3 (high individual risk and low risk to the community) or risk group 4 (high individual risk and high risk to the community, without treatment), which necessitates the use of Biosafety Level (BSL)-3 and BSL-4 laboratories and facilities for handling and propagation (Artika and Ma’roef, 2017). Citation: Salazar-García M, Acosta-Contreras S, Rodríguez-Martínez G, Cruz-Rangel A, Flores-Alanis A, Patiño-López G and Luna-Pineda VM (2022) Pseudotyped Vesicular Stomatitis Virus-Severe Acute Respiratory Syndrome-Coronavirus-2 Spike for the Study of Variants, Vaccines, and Therapeutics Against Coronavirus Disease 2019. Front. Microbiol. 12:817200. doi: 10.3389/fmicb.2021.817200 January 2022 \| Volume 12 \| Article 817200 1 Frontiers in Microbiology \| www.frontiersin.org VSV1G-SARS-CoV-2 S Pseudoparticles and Their Uses Salazar-García et al. 1https://www.gob.mx/salud/acciones-y-programas/instituto-de-diagnostico-y- referencia-epidemiologicos-indre 2https://www.biomedicas.unam.mx/servicios/laboratorio-de-alta-seguridad- bsl-3/ 3https://www.gob.mx/senasica/acciones-y-programas/laboratorio-de-la- comision-mexico-estados-unidos-para-la-prevencion-de-la-ﬁebre-aftosa-y- otras-enfermedades-exoticas-de-los-animales-cpa 4https://www.incmnsz.mx/imagenes/siteLaboratoriodeMicrobiologia/ laboratoriobsl3.html 5https://www.gob.mx/salud/iner 6https://ciatej.mx/ 7https://www.udem.edu.mx/es/ciencias-de-la-salud/noticia/cuenta-con- recertiﬁcacion-internacional-laboratorio-de-bioseguridad PSEUDOTYPED VESICULAR STOMATITIS VIRUS: HISTORY secreted Embryonic Alkaline Phosphatase (SEAP), and ﬁreﬂy luciferase (fLuc), generating ppVSV1G-reporter (Takada et al., 1997; Fukushi et al., 2008; Tani et al., 2010; Muik et al., 2012). terminal domain contains two heptad repeats (HR), a single-pass transmembrane motif, and a cytoplasmic tail (CT) (Muñoz- Barroso et al., 1999; Lee et al., 2008; Hastie et al., 2017; Pallesen et al., 2017; Yuan et al., 2017; Wrapp et al., 2020). The membrane-associated RING-CH (MARCH)-8, a RING (really interesting new gene)-ﬁnger E3 ubiquitin ligase, has been reported to downregulate human transmembrane proteins, including the enveloped viral glycoproteins SARS-CoV-2 spike (S), HIV-1 Env, and EboV-GP (Tada et al., 2015). The CTs found in these viral glycoproteins are made up of Lys residues that can vary in quantity and serve as targets for MARCH- mediated ubiquitination (Figure 3A). Interestingly, expression of MARCH8 in the virus-producing cells reduced the levels of viral glycoprotein and compromised infection of cells was obtained by replacing Lys residues with Ala (K to A) in the CTs of these glycoproteins (Lun et al., 2021). Although S glycoprotein CTs of Coronaviruses (CoVs) features cysteine-rich motifs (six conserved residues) play an important role in S glycoprotein function, these cysteines are palmitoylated and their substitution with Ala (Cys-to-A) aﬀect the S-mediated cell fusion of these viruses (Figure 3; Chang et al., 2000; Petit et al., 2007). PSEUDOTYPED VESICULAR STOMATITIS VIRUS: HISTORY From these, the Biosafety Level 3 Laboratory from Institute for Epidemiological Diagnosis and Reference1, Biosecurity Unit from Institute of Biomedical Research from UNAM2, Animal Health Laboratory from National Service of Agrifood Health, Safety and Quality3, National Laboratory for Maximum Biological Safety from National Institute of Medical Sciences and Nutrition “Salvador Zubirán,”4 and Emerging and Non-Emerging Pathogens Research Tower from National Institute of Respiratory Diseases5 are located in Mexico City, whereas the other two BSL-3: the Center for Research and Assistance in Technology and Design of the State of Jalisco6 and the University of Monterrey/Autonomous University of Nuevo León7 are located in Guadalajara, Jalisco and Nuevo León, Monterrey, respectively. However, because Mexico has no BSL-4 laboratories, research with WHO-designated priority diseases such as Ebola is hampered (Figure 1). y To enable the study of BSL-3/-4 pathogens under BSL-2 laboratory conditions, one can use pseudotyped virus (PV) to January 2022 \| Volume 12 \| Article 817200 Frontiers in Microbiology \| www.frontiersin.org 2 VSV1G-SARS-CoV-2 S Pseudoparticles and Their Uses Salazar-García et al. FIGURE 1 \| Mexican BSL-3 laboratories. (a) BSL-3 Laboratory from the Institute for Epidemiological Diagnosis and Reference (Mexico City); (b) The Biosecurity Unit from Institute of Biomedical Research (UNAM, Mexico City); (c) Animal Health Laboratory from National Service of Agrifood Health, Safety, and Quality (Mexico City); (d) National Laboratory for Maximum Biological Safety from National Institute of Medical Sciences and Nutrition “Salvador Zubirán” (Mexico City); (e) Emerging and Non-Emerging Pathogens Research Tower from National Institute of Respiratory Diseases (Mexico City); (f) Center for Research and Assistance in Technology and Design of the State of Jalisco (Guadalajara City), and (g) University of Monterrey/Autonomous University of Nuevo León (Monterrey City). FIGURE 1 \| Mexican BSL-3 laboratories. (a) BSL-3 Laboratory from the Institute for Epidemiological Diagnosis and Reference (Mexico City); (b) The Biosecurity Unit from Institute of Biomedical Research (UNAM, Mexico City); (c) Animal Health Laboratory from National Service of Agrifood Health, Safety, and Quality (Mexico City); (d) National Laboratory for Maximum Biological Safety from National Institute of Medical Sciences and Nutrition “Salvador Zubirán” (Mexico City); (e) Emerging and Non-Emerging Pathogens Research Tower from National Institute of Respiratory Diseases (Mexico City); (f) Center for Research and Assistance in Technology and Design of the State of Jalisco (Guadalajara City), and (g) University of Monterrey/Autonomous University of Nuevo León (Monterrey City). Frontiers in Microbiology \| www.frontiersin.org IMPORTANCE OF THE GLYCOPROTEIN CYTOPLASMIC TAIL IN THE ASSEMBLY OF PSEUDOTYPED VESICULAR STOMATITIS VIRUSES The amino acid alignment was performed with the Jalview v2.11.1.4 using CLUSTAL W. (C) Amino acid of HIV, Lassa, and Ebola virus glycoproteins CT domains. Underline in Lassa virus indicates the ERRS motif. The HIV CT domain was trimmed to 60 and 40 amino acids in the N and C terminal, respectively. The red squares indicate Lys that could be implicated in efﬁcient infectivity, while the blue squares indicate the tyrosine-dependent internalization signals (Yxx8 motif, where 8 is F, I, L, M, or V). FIGURE 3 \| Amino acid sequences of viral glycoproteins class I CT domains. (A) Vesicular stomatitis virus CT domain Indian strain. (B) Amino acid sequence alignment of CoVs Spike protein in the CT domains. CRM, cysteine-rich motif; CRD, charge-rich domain; ERRS, endoplasmatic reticulum retrieval signal (KxHxx). The amino acid alignment was performed with the Jalview v2.11.1.4 using CLUSTAL W. (C) Amino acid of HIV, Lassa, and Ebola virus glycoproteins CT domains. Underline in Lassa virus indicates the ERRS motif. The HIV CT domain was trimmed to 60 and 40 amino acids in the N and C terminal, respectively. The red squares indicate Lys that could be implicated in efﬁcient infectivity, while the blue squares indicate the tyrosine-dependent internalization signals (Yxx8 motif, where 8 is F, I, L, M, or V). Gn, and a substitution (S1094L) in the stem region of Gc following three serial passages in Vero cells, that have the highest viral spread likely for relocalization of Gn/Gc from the Golgi complex to the cell surface (Slough et al., 2019). Overall, all ﬁndings emphasize the relevance of CT of these glycoproteins in the production of virus-like particles and virion, including pseudotyped viruses. Pseudotyped VSV virions have been used to assess cellular tropism, glycoprotein function, receptor recognition, and neutralization antibody assay in viruses from risk groups 3 and 4 (Table 1), including Crimean-Congo hemorrhagic fever virus, Ebola virus, Marburg virus, Lassa signal, as well as a tyrosine-dependent localization signal (YxxF or YxxI motif) that interacts with the CoVsM protein for virions incorporation (Figure 3; Lontok et al., 2004; McBride et al., 2007; Winter et al., 2008; Shirato et al., 2011). IMPORTANCE OF THE GLYCOPROTEIN CYTOPLASMIC TAIL IN THE ASSEMBLY OF PSEUDOTYPED VESICULAR STOMATITIS VIRUSES Enveloped viruses have fusion proteins that enable attachment and fusion into host cells (Barrett and Dutch, 2020). These viral fusion proteins are classiﬁed structurally as class I (e.g., HIV Env Glycoprotein), class II (e.g., Rift Valley fever virus glycoprotein C), and class III (e.g., VSV G glycoprotein), with all of them exhibiting both pre- and post-fusion static conformations (Blumenthal et al., 2012; Dessau and Modis, 2013; Kim et al., 2017). Lassa, Ebola, HIV, MERS, SARS, and SARS- CoV-2 viruses all feature class I viral fusion glycoproteins that have two domains, the C- and N-terminal domains located between the furin-like protease cleavage site (Figure 2). In their pre- and post-fusion states, they form homotrimers, and their C Coronaviruses S CT comprises a dilysine (KKxx-COOH) or a dibasic (KxHxx-COOH) endoplasmic reticulum retrieval January 2022 \| Volume 12 \| Article 817200 Frontiers in Microbiology \| www.frontiersin.org 3 VSV1G-SARS-CoV-2 S Pseudoparticles and Their Uses Salazar-García et al. FIGURE 2 \| Tridimensional structures of Vesicular Stomatitis Virus G protein and the main class I viral fusion proteins in pre-fusion static conformations. (A) Vesicular Stomatitis Virus (VSV) class III fusion glycoprotein (PDB: 6TIT) and representative class I viral fusion proteins: (B) SARS-CoV-2 S (PDB: 6VXX), (C) HIV glycoprotein (GP) 160 (PDB: 6ULC), (D) Lassa virus GP (PDB: 6P91), and (E) Ebola virus GP (PDB: 6QD7). They form homotrimers with two domains, the C- and N-terminal domains (Magenta, green, and cyan represent each monomer). N terminal contains the receptor binding site and C terminal domain contains two heptad repeats (HR), a single-pass transmembrane motif, and a cytoplasmic tail (CT). FIGURE 3 \| Amino acid sequences of viral glycoproteins class I CT domains. (A) Vesicular stomatitis virus CT domain Indian strain. (B) Amino acid sequence alignment of CoVs Spike protein in the CT domains. CRM, cysteine-rich motif; CRD, charge-rich domain; ERRS, endoplasmatic reticulum retrieval signal (KxHxx). The amino acid alignment was performed with the Jalview v2.11.1.4 using CLUSTAL W. (C) Amino acid of HIV, Lassa, and Ebola virus glycoproteins CT domains. Underline in Lassa virus indicates the ERRS motif. The HIV CT domain was trimmed to 60 and 40 amino acids in the N and C terminal, respectively. The red squares indicate Lys that could be implicated in efﬁcient infectivity, while the blue squares indicate the tyrosine-dependent internalization signals (Yxx8 motif, where 8 is F, I, L, M, or V). IMPORTANCE OF THE GLYCOPROTEIN CYTOPLASMIC TAIL IN THE ASSEMBLY OF PSEUDOTYPED VESICULAR STOMATITIS VIRUSES FIGURE 2 \| Tridimensional structures of Vesicular Stomatitis Virus G protein and the main class I viral fusion proteins in pre-fusion static conformations. (A) Vesicular Stomatitis Virus (VSV) class III fusion glycoprotein (PDB: 6TIT) and representative class I viral fusion proteins: (B) SARS-CoV-2 S (PDB: 6VXX), (C) HIV glycoprotein (GP) 160 (PDB: 6ULC), (D) Lassa virus GP (PDB: 6P91), and (E) Ebola virus GP (PDB: 6QD7). They form homotrimers with two domains, the C- and N-terminal domains (Magenta, green, and cyan represent each monomer). N terminal contains the receptor binding site and C terminal domain contains two heptad repeats (HR), a single-pass transmembrane motif, and a cytoplasmic tail (CT). FIGURE 2 \| Tridimensional structures of Vesicular Stomatitis Virus G protein and the main class I viral fusion proteins in pre-fusion static conformations. (A) Vesicular Stomatitis Virus (VSV) class III fusion glycoprotein (PDB: 6TIT) and representative class I viral fusion proteins: (B) SARS-CoV-2 S (PDB: 6VXX), (C) HIV glycoprotein (GP) 160 (PDB: 6ULC), (D) Lassa virus GP (PDB: 6P91), and (E) Ebola virus GP (PDB: 6QD7). They form homotrimers with two domains, the C- and N-terminal domains (Magenta, green, and cyan represent each monomer). N terminal contains the receptor binding site and C terminal domain contains two heptad repeats (HR), a single-pass transmembrane motif, and a cytoplasmic tail (CT). FIGURE 3 \| Amino acid sequences of viral glycoproteins class I CT domains. (A) Vesicular stomatitis virus CT domain Indian strain. (B) Amino acid sequence alignment of CoVs Spike protein in the CT domains. CRM, cysteine-rich motif; CRD, charge-rich domain; ERRS, endoplasmatic reticulum retrieval signal (KxHxx). The amino acid alignment was performed with the Jalview v2.11.1.4 using CLUSTAL W. (C) Amino acid of HIV, Lassa, and Ebola virus glycoproteins CT domains. Underline in Lassa virus indicates the ERRS motif. The HIV CT domain was trimmed to 60 and 40 amino acids in the N and C terminal, respectively. The red squares indicate Lys that could be implicated in efﬁcient infectivity, while the blue squares indicate the tyrosine-dependent internalization signals (Yxx8 motif, where 8 is F, I, L, M, or V). FIGURE 3 \| Amino acid sequences of viral glycoproteins class I CT domains. (A) Vesicular stomatitis virus CT domain Indian strain. (B) Amino acid sequence alignment of CoVs Spike protein in the CT domains. CRM, cysteine-rich motif; CRD, charge-rich domain; ERRS, endoplasmatic reticulum retrieval signal (KxHxx). IMPORTANCE OF THE GLYCOPROTEIN CYTOPLASMIC TAIL IN THE ASSEMBLY OF PSEUDOTYPED VESICULAR STOMATITIS VIRUSES Although these particles lack a glycoprotein in their genome, they are coated in VSV-G in trans that result in ppVSV1G- G (G-complemented particles). In a new passage of cells, it may incorporate the SARS-CoV-2 S protein onto the virus’s surface by cotransfection with the S gene-containing plasmid, producing ppVSV1G-SARS-CoV-2-S particles or pseudotyped VSV-SARS-CoV-2-S (Lawson et al., 1995; Whelan et al., 1995; Schnell et al., 1996). IMPORTANCE OF THE GLYCOPROTEIN CYTOPLASMIC TAIL IN THE ASSEMBLY OF PSEUDOTYPED VESICULAR STOMATITIS VIRUSES CoVs S CT with a nonsense mutation-generated 21- or 24-amino acid deletions (conserved KxHxx motif) showed the highest viral spread and the appearance of non-syncytium-forming infectious centers, likely driving rVSV-SARS-CoV-2 S adaptation for the eﬃcient spread in tissue culture (Case et al., 2020b; Dieterle et al., 2020). In addition, rVSV-Hantaan virus Gn and Gc glycoproteins also acquired a substitution in CT (I532K) of January 2022 \| Volume 12 \| Article 817200 Frontiers in Microbiology \| www.frontiersin.org 4 VSV1G-SARS-CoV-2 S Pseudoparticles and Their Uses Salazar-García et al. TABLE 1 \| Application of pseudotyped Vesicular Stomatitis Virus (VSV). Virus Viral protein Research area and application Reporter References Crimean-Congo hemorrhagic fever Virus GP Vaccine, viral entry mechanism, and neutralizing assay Luciferase Suda et al., 2016; Rodriguez et al., 2019 Ebola virus GP Vaccine and drug testing GFP Takada et al., 1997; Geisbert and Feldmann, 2011; Lennemann et al., 2017; Saito et al., 2020 Marburg virus GP Vaccine and drug testing GFP Geisbert and Feldmann, 2011; Zhang et al., 2017; Saito et al., 2020 Lassa virus GP Entry and receptor mechanism, and neutralization assays GFP Kunz et al., 2005; Hastie et al., 2017 Nipah virus G/F Fusion mechanism and neutralization assays GFP and SEAP Kaku et al., 2009, 2012; Tamin et al., 2009; Contreras et al., 2021 Rift Valley virus GP Serological assays Luciferase Bukbuk et al., 2014 MERS Spike Vaccine, neutralization assays, and receptor evaluation Luciferase and GFP Fukuma et al., 2015; Liu et al., 2018; Lester et al., 2019 SARS-CoV Spike Vaccine, entry mechanism, and neutralization assays GFP Fukushi et al., 2006, 2008; Ge et al., 2006; Kapadia et al., 2008 SARS-CoV-2 Spike Neutralization assays, entry mechanism, treatment testing, vaccine, and vaccine efﬁcacy GFP, luciferase, and SEAP Condor Capcha et al., 2020; Gasbarri et al., 2020; Collier et al., 2021; Malherbe et al., 2021; Tolah et al., 2021; Verma et al., 2021 GP, glycoprotein; GFP, green ﬂuorescent protein; SEAP, secreted alkaline phosphatase; F, fusion; G, attachment. TABLE 1 \| Application of pseudotyped Vesicular Stomatitis Virus (VSV). GP, glycoprotein; GFP, green ﬂuorescent protein; SEAP, secreted alkaline phosphatase; F, fusion; G, attachment. fever virus, Nipah virus, Rift Valley virus, MERS-CoV, and SARS-CoV-2 virus (Kunz et al., 2005; Fukushi et al., 2008; Kaku et al., 2012; Bukbuk et al., 2014; Suda et al., 2016; Lennemann et al., 2017; Lester et al., 2019; Saito et al., 2020; Zettl et al., 2020). 8https://coronavirus.jhu.edu/map.html 9https://datos.covid-19.conacyt.mx/ 9https://datos.covid-19.conacyt.mx/ Vesicular Stomatitis Virus-1G Pseudoparticles Packing In primary transfection, ﬁve plasmids are used to accomplish ppVSV1G packaging: (1) pVSV-1G-reporter containing all VSV antigenome directed by the T7 promoter, excepting the GP gene, which was deleted and replaced by a reporter gene-GFP, -RFP/DsRed, -SEAP, or -fLuc; (2) pVSV-G containing the VSV GP gene directed by the T7 promoter; (3) pVSV-L containing the VSV polymerase (L) gene directed by T7 promoter; (4) pVSV-N containing VSV nucleocapsid (N) gene directed by T7 promoter; and (5) pVSV-P containing VSV phosphoprotein (F) gene directed by T7 promoter. During cellular passages, ppVSV1G-G is generated with pCAGGS-G containing VSV GP gene directed by pol II promoter as chicken beta-actin promoter (Figure 4A). Pneumonia caused by the SARS-CoV-2 virus was recognized as COVID-19 by the WHO and is now considered a pandemic, with over 200 million cases and over four million fatalities globally8. More than three million COVID-19 infections have been reported in Mexico, with over 250 thousand deaths9. Because of the SARS-CoV-2 virus’s high infectivity, toxicity, and lack of therapies, BSL-3 laboratories are necessary for its handling in drug testing, neutralizing antibodies, and authorized vaccination eﬀectiveness (Kaufer et al., 2020). As a result, most research laboratories have been unable to conduct SARS-CoV-2 research; however, the use of ppVSV1G-SARS-CoV-2 S provides a feasible option for studying this virus at BSL-2 facilities (Plescia et al., 2021). To be able to transcribe the VSV antigenome and the N, P, G, and L genes, the HEK293T cell line must ﬁrst be infected with the vaccinia virus or transfected with pT7pol (expressing the bacteriophage T7 RNA polymerase). The VSV polymerase complex (VSV L and P proteins) and the envelope nucleoprotein N produce genome from the antigenome RNA produced by T7. The matrix (M) protein is encoded in the VSV RNA genome and is produced after the genome is made and transcription occurs. Because the genome lacks glycoprotein, the cell line must be transfected with pCAGGS-G to produce G in trans, and The generation of ppVSV1G-SARS-CoV-2 S may be classiﬁed as (a) ppVSV1G packaging and (b) SARS-CoV-2 S protein production (Figure 4). The ppVSV1G is a viral particle that can be packaged in vitro by transfection of six plasmids into the bacteriophage T7 RNA polymerase-harboring cell. January 2022 \| Volume 12 \| Article 817200 Frontiers in Microbiology \| www.frontiersin.org 5 VSV1G-SARS-CoV-2 S Pseudoparticles and Their Uses Salazar-García et al. Vesicular Stomatitis Virus-1G Pseudoparticles Packing 4 \| Scheme of the generation of pseudotyped vesicular stomatitis virus-severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) S. (A) Packaging V. (B) Optimized conditions for cloning Spike gene. (C) ppVSV1G-SARS-CoV-2 S assembly in HEK 293T cells, and infection assay in Vero E6 cells by G-SARS-CoV-2 S. URE 4 \| Scheme of the generation of pseudotyped vesicular stomatitis virus-severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) S. (A) Packaging FIGURE 4 \| Scheme of the generation of pseudotyped vesicular stomatitis virus-severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) S. (A) Packaging of ppVSV. (B) Optimized conditions for cloning Spike gene. (C) ppVSV1G-SARS-CoV-2 S assembly in HEK 293T cells, and infection assay in Vero E6 cells by ppVSV1G-SARS-CoV-2 S. make infectious G pseudotyped ppVSV1G-G virions (Figure 4A; Whitt, 2010). In addition, transfecting non-susceptible 293T cells with ACE2, Furin, and TMPRSS2 greatly enhanced the entry of ppVSV- SARS-CoV-2-S particles, supporting the fact that ACE2, Furin, and TMPRSS2 are required for optimal spike infectivity of kidney cells (Condor Capcha et al., 2020; Johnson et al., 2020; Xiong et al., 2020). upstream residues improves recognition and increases protein levels (Havranek et al., 2020). Severe Acute Respiratory Syndrome-Coronavirus-2 isolates from the initial human cases in Wuhan had an S protein with the D614 form; however, the viruses that are currently circulating in the human population have the G614 form. SARS-CoV-2 viruses with the G614 mutation in the S protein have greater infectious titers in vitro than those with the D614 mutation (Korber et al., 2020). Another R682Q mutation in the S protein has been characterized as a fast adaptation of SARS-CoV-2 after successive passaging in Vero E6 cells (Ogando et al., 2020). Both the D614G and R682Q alterations improve VSV-SARS- CoV-2 pseudotyping, with the S protein containing 19 deleted residues into the C terminus (19) (Johnson et al., 2020). CT alterations of various viral envelope GPs, including truncation and point mutation in this tail, may be necessary to facilitate appropriate integration into the ppVSV1G-SARS-CoV-2 S. The greater incorporation into ppVSV1G-SARS-CoV-2 S is most likely related to a change in cellular localization that produces additional S at sites of assembly (Zingler and Littman, 1993; Mammano et al., 1997; Fukushi et al., 2006; Slough et al., 2019). When compared to WT Spike, ﬁrst research utilizing shortened SARS-CoV-2 S protein with a deletion (1) of 8 to 39 amino acids in CT demonstrated a high titer of ppVSV1G-SARS-CoV-2 S in 119 and 126 (Giroglou et al., 2004). When the deletions 119 and 121 into CT were examined, additional investigations revealed a high titer of ppVSV1G-SARS-CoV-2 S with increased cell-to-cell fusion (Case et al., 2020b; Dieterle et al., 2020; Havranek et al., 2020; Johnson et al., 2020; Schmidt et al., 2020). If necessary, the SARS-CoV-2 S protein can be fused with a C terminal 3XFLAG tag to detect full-length S without interfering with fusion, surface expression, translation, and VSV incorporation of SARS-CoV-2 S protein (Havranek et al., 2020). All ﬁndings have proposed the cloning of Met1-S-121 with usage codons for eﬃcient production of SARS-CoV-2 S protein and ppVSV assembly into the optimal cell line (Figure 4B). make infectious G pseudotyped ppVSV1G-G virions (Figure 4A; Whitt, 2010). The obtained ppVSV viral particles are used to infect a cell line previously transfected with the plasmid containing the optimized S gene that will provide S protein for its assembly onto the virus’s surface. make infectious G pseudotyped ppVSV1G-G virions (Figure 4A; Whitt, 2010). make infectious G pseudotyped ppVSV1G-G virions (Figure 4A; Whitt, 2010). S protein comprises a signal peptide, an N-terminal S1 domain, and a C terminal S2 domain and has 1,274 amino acids. Replacing rare codons with abundant cognate tRNA in the cytosol can aﬀect protein translation from mRNA, resulting in guanine and cytosine enrichment as a process of sequence optimization with higher steady-state mRNA levels in vitro and protein expression in vivo (Kudla et al., 2006; Mauro and Chappell, 2014). Furthermore, the SARS-CoV-2 S protein comprises a short signal peptide with weak Sec61 recognition, but the inclusion of the nine S protein comprises a signal peptide, an N-terminal S1 domain, and a C terminal S2 domain and has 1,274 amino acids. Replacing rare codons with abundant cognate tRNA in the cytosol can aﬀect protein translation from mRNA, resulting in guanine and cytosine enrichment as a process of sequence optimization with higher steady-state mRNA levels in vitro and protein expression in vivo (Kudla et al., 2006; Mauro and Chappell, 2014). Furthermore, the SARS-CoV-2 S protein comprises a short signal peptide with weak Sec61 recognition, but the inclusion of the nine Production of Severe Acute Respiratory Syndrome-Coronavirus-2 S Protein: Optimized S Gene-Containing Plasmid The SARS-CoV-2 S protein interacts with the host cell receptor, allowing viral and cellular membranes to fuse. The SARS-CoV-2 January 2022 \| Volume 12 \| Article 817200 Frontiers in Microbiology \| www.frontiersin.org 6 VSV1G-SARS-CoV-2 S Pseudoparticles and Their Uses Salazar-García et al. (African green monkey kidney cells), Calu-3 (human lung cells), Caco-2 (human colon cells), MDBK (cattle kidney cells), MDCKII (Dog kidney cells), A549 (human lung cells), BEAS- 2B (human bronchus cells), and NCI-H1299 (human lung cells) have been evaluated for ppVSV1G-SARS-CoV-2 S entry (Hoﬀmann et al., 2020). Preliminary investigations revealed that Vero cells were highly susceptible to ppVSV1G-SARS-CoV- 2 S, along with Caco-2 and Calus-3, while other cell lines tested were not eﬃcient for ppVSV1G-SARS-CoV-2 S entry (Hoﬀmann et al., 2020). Vero E6 cells were also initially utilized in cell-culture-based infection models for SARS-CoV-1 studies (Keyaerts et al., 2005; Yamate et al., 2005). Several studies have compared diﬀerent cell lines for ppVSV1G-SARS-CoV- 2-S entry eﬃciency. When ppVSV-SARS-CoV-2-S was titrated in both Vero E6 and MA104 (Monkey African Green kidney) cell lines, the virus entry was increased (Case et al., 2020b). Optimal Cell Line to Evaluate of Pseudotyped Vesicular Stomatitis Virus-Severe Acute Respiratory Syndrome-Coronavirus-2 Spike Entry Optimal Cell Line to Evaluate of Pseudotyped Vesicular Stomatitis Virus-Severe Acute Respiratory Syndrome-Coronavirus-2 Spike Entry By attaching to a cellular receptor, like the angiotensin-converting enzyme 2 (ACE2), the SARS-CoV-2 S protein promotes viral entry into target cells (Datta et al., 2020; Wrapp et al., 2020). The lungs and other tissues, including the nasal and oral mucosa, vasculature, kidney, heart, gastrointestinal tract, pancreas, and brain, express ACE2 (Gembardt et al., 2005). Thus, cell lines from human and animal origin, such as 293T (human kidney cells), BHK-21 (Syrian hamster kidney cells), Huh-7 (human liver cells), LLC-PK1 (pig kidney cells), MRC-5 (human lung cells), MyDauLu/47.1 [Daubenton’s bat (Myotis daubentonii) lung cells], NIH/3T3 (Mouse embryonic ﬁbroblast cells), RhiLu/1.1 [Halcyon horseshoe bat (Rhinolophus alcyone) lung cells], Vero PSEUDOVIRUS NEUTRALIZATION ASSAY: SAFETY, SPEED, AND SCALABLE Although the enzyme-linked immunosorbent test is extensively used for detecting SARS-CoV-2 speciﬁc antibodies, it does not oﬀer information on virus-neutralizing antibody titers (Roy et al., 2020). The neutralization test is a technique for determining the presence of neutralizing antibodies. The traditional viral neutralization experiment necessitates the use of a live SARS- CoV-2 virus, which must be handled in a BSL-3 facility. The downside of using a live viral test is that it is labor-intensive, taking up to 4 days to complete (Ogando et al., 2020). To address these challenges, the pseudovirus neutralization assay (PVNA) is a viable option since it can be conducted under BSL-2 settings and is a safe, rapid, and scalable test. ppVSV1G-SARS-CoV- 2 S can be kept for an extended period (≥6 months) with negligible titer loss at −20 or −80, 4◦C (4 weeks), or even at room temperature (1 week). Furthermore, ppVSV-SARS-CoV- 1-S may withstand many freeze-thaw cycles without decreasing their infectivity (Wright et al., 2009; Molesti et al., 2014). Another advantage of using PVNA is that most ppVSV1G-SARS-CoV-2 S contain a marker gene that can be detected by a ﬂuorescence or luminescence signal, with a linear correlation between the value of ﬂuorescence or chemiluminescence and the number of infected ppVSV1G-SARS-CoV-2 S, allowing for easier and more accurate quantiﬁcation (Case et al., 2020b; Nie et al., 2020a). Calculation of Tissue Culture Infectious Dose or Pseudotyped Virus Neutralization Doses The Reed-Muench method is used to determine the 50% tissue culture infectious dose (TCID50) or pseudotyped virus neutralization doses (PVND50) of the ppVSV1G-SARS-CoV- 2 S, which includes calculating the proportional distance (PD) between dilutions above and below the 50% endpoint (Reed, 2010; Gauger and Vincent, 2014; Manenti et al., 2020). The dilution factor is deﬁned as the fold diﬀerence between two inoculum titers that are above and below a 50% response rate. Based on the results of Figure 5, the equation must be applied as follows: Material and Equipment Microplate luminometer is a sensitive, ready-to-use device for gene reporter, cell-based, and biochemical experiments in 96- well plates with luminous responses (Alam and Cook, 1990). Fluorescent imaging and quantitation may be assessed using January 2022 \| Volume 12 \| Article 817200 Frontiers in Microbiology \| www.frontiersin.org 7 VSV1G-SARS-CoV-2 S Pseudoparticles and Their Uses Salazar-García et al. a variety of microscopes that must automatically scan 96-well plates and include quick focusing, image collection, and big data processing (Case et al., 2020b). Flow cytometers can also be used to assess ﬂuorescent-based techniques. ppVSV1G-SARS-CoV-2 S titration is performed in 96-well plates with 10-fold serial dilution until a total of nine dilutions with six duplicates are obtained (Figure 5; Crawford et al., 2020). a variety of microscopes that must automatically scan 96-well plates and include quick focusing, image collection, and big data processing (Case et al., 2020b). Flow cytometers can also be used to assess ﬂuorescent-based techniques. ppVSV1G-SARS-CoV-2 S titration is performed in 96-well plates with 10-fold serial dilution until a total of nine dilutions with six duplicates are obtained (Figure 5; Crawford et al., 2020). For some groups, there is a good connection between the experimental data acquired using authentic SARS-CoV-2 and the PVNA. Neutralizing activity was detected with a remarkable connection between the two tests, as demonstrated by Spearman’s correlation with R values ranging from 0.76 to 0.939 and p values of <0.001 (Case et al., 2020b; Dieterle et al., 2020; Hyseni et al., 2020; Nie et al., 2020a; Xiong et al., 2020; Zettl et al., 2020; Tolah et al., 2021). Furthermore, once a ppVSV1G-SARS-CoV-2 S stock is produced, the PVNA may be completed in 1 day with a GFP or RLU readout of 7.5 h, whereas the live SARS-CoV-2 assay takes 30 h (Case et al., 2020b; Nie et al., 2020b). TCID50 mL = inverse of ID50 inoculum volume in mL ×numerical factor to reach 1 mL (4) ×numerical factor to reach 1 mL (4) PSEUDOTYPED VESICULAR STOMATITIS VIRUS-SEVERE ACUTE RESPIRATORY SYNDROME-CORONAVIRUS-2 SPIKE AND ITS USE Respiratory Syndrome Coronavirus 2 A SARS-CoV-2 variant is identiﬁed by alterations in receptor binding, decreased antibody neutralization (post-infection or post-vaccination), decreased treatment eﬀectiveness, or even a possible inﬂuence on diagnostics, as well as increased anticipated transmissibility or disease severity (WHO). ppVSV1G-SARS- CoV-2 S particles can rapidly be produced with the changes in newly deﬁne variants and tested for alterations in neutralizing capacity and entry eﬃciencies. PD = (50%) −(% of infected at dilution next below at 50%) (% of infected next above at 50%) −(% of infected next below at 50%) PD = (50%) −(% of infected at dilution next below at 50%) (% of infected next above at 50%) −(% of infected next below at 50%) (1) 2) 3) 4) p y y South Africa was the ﬁrst to describe the SARS-CoV-2 B.1.351 variety (Beta), which was characterized by N501Y, K417N, and E484K alterations. This variant is more transmissible and resistant to neutralizing antibodies. ppVSV1G-SARS-CoV- 2 S containing the S protein (B.1.351)-associated mutations can quickly infect cell cultures and tolerate neutralizing action of monoclonal antibodies against SARS-CoV-2 S RBD (Kim et al., 2021). 2) 3) 4) y y South Africa was the ﬁrst to describe the SARS-CoV-2 B.1.351 variety (Beta), which was characterized by N501Y, K417N, and E484K alterations. This variant is more transmissible and resistant to neutralizing antibodies. ppVSV1G-SARS-CoV- 2 S containing the S protein (B.1.351)-associated mutations can quickly infect cell cultures and tolerate neutralizing action of monoclonal antibodies against SARS-CoV-2 S RBD (Kim et al., 2021). ID50 = 10ID50 TCID50 mL = inverse of ID50 inoculum volume in mL Neutralization Assay and Correlation y Sera samples must be diluted before being combined with ppVSV1G-SARS-CoV-2 S 325 to 1,300 TCID50/mL and incubated at 37◦C for 1 h. Pre-incubated ppVSV1G-SARS-CoV- 2 S are used to infect cell lines for 24–72 h at 37◦C and 5% CO2, and the quantiﬁcation of ppVSV1G-SARS-CoV-2 S infecting the target cells is estimated by measuring the production of luciferase or ﬂuorescence as stated in the section “Pseudotyped Vesicular Stomatitis Virus-Severe Acute Respiratory Syndrome- Coronavirus-2 Spike titration” (Figure 5B). The luciferase activity is determined by relative light units (RLU), whereas the ﬂuorescence is determined by the number of GFP-positive cells. Percent neutralization must be adjusted by assuming uninfected cells to be 100% neutralized and infected cells with ppVSV1G-SARS-CoV-2 S alone to be 0% neutralized (Crawford et al., 2020). The PVND50 (50% ppVSV1G-SARS- CoV-2 S neutralizing doses) is determined as described in the section “Pseudotyped Vesicular Stomatitis Virus-Severe Acute Respiratory Syndrome-Coronavirus-2 Spike Titration” (Reed, 2010; Gauger and Vincent, 2014; Manenti et al., 2020). The SARS-CoV-2 B.1.427/B.1.429 variant (epsilon), a novel variant with S13I, W152C, and L452R amino acid changes, was reported for the ﬁrst time in California (34 nations, beginning May 2021) with enhanced transmissibility and infectivity. When compared to the D614G change, ppVSV1G- SARS-CoV-2 S bearing the L452R change showed a 6.7- to 22.5-fold higher infectivity in cell cultures and lung organoids, while the W152C change showed just a little increase in infection in these cells. Furthermore, ppVSV1G-SARS-CoV- 2 S with S13I, W152C, and L452R changes demonstrated considerable resistance to neutralization by post-infection (4.0- to 6.7-fold) and vaccination-elicited antibodies (2- fold) as well as monoclonal antibodies (Deng et al., 2021; McCallum et al., 2021). In the New York region, the SARS-CoV-2 B.1.526 variation (Iota), deﬁned by L5F, T95I, D253G, E484K, or S477N, D614G, and A701V change, was identiﬁed. ppVSV1G-SARS-CoV-2 S harboring L5F, T95I, D253G, E484K, D614G, and A701V, as well as L5F, T95I, D253G, S477N, D614G, and Q957R changes, January 2022 \| Volume 12 \| Article 817200 Frontiers in Microbiology \| www.frontiersin.org Frontiers in Microbiology \| www.frontiersin.org 8 VSV1G-SARS-CoV-2 S Pseudoparticles and Their Uses Salazar-García et al. URE 5 \| The design of titration and pseudovirus neutralization assay (PVNA). (A,B) Infection, titration and PVNA was performed in a 96-well plate with 10-fold dilution of ppVSV1G-SARS-CoV-2 S until a total of 8 or 10 dilutions and 6 or 8 replicates were obtained. Neutralization Assay and Correlation Neutralization assay was also performed in a 96-well with 6 dilutions and 2 replicates; however, the sera samples were previously diluted and then mixed with ppVSV1G-SARS-CoV-2 S 325 to 1,300 TCID50/mL ncubated for 1 h at 37◦C. (C) In titration, the cells containing the luciferase were lysed to perform the luciferase assay and a 96-well-plate luminometer was . In the ﬂuorescence method, the DAPI-stained cells containing GFP were analyzed by ﬂuorescence microscope, and the images were analyzed with alized software. An alternative method is by using a ﬂow cytometer equipped with a 96-well-autosampler. In PVNA, the luciferase activity was determined by elative light units (RLU) and the ﬂuorescence by the number of GFP-positive cells. Percent neutralization must be normalized considering uninfected cells as % neutralization and infected cells with ppVSV1G-SARS-CoV-2 S alone as 0% neutralization. (D) The 50% tissue culture infectious dose (TCID50) and/or the ppVSV1G-SARS-CoV-2 S neutralizing doses (PVND50) were calculated according to the Reed-Muench method and reference table. FIGURE 5 \| The design of titration and pseudovirus neutralization assay (PVNA). (A,B) Infection, titration and PVNA was performed in a 96-well plate with 10-fold serial dilution of ppVSV1G-SARS-CoV-2 S until a total of 8 or 10 dilutions and 6 or 8 replicates were obtained. Neutralization assay was also performed in a 96-well plate with 6 dilutions and 2 replicates; however, the sera samples were previously diluted and then mixed with ppVSV1G-SARS-CoV-2 S 325 to 1,300 TCID50/mL and incubated for 1 h at 37◦C. (C) In titration, the cells containing the luciferase were lysed to perform the luciferase assay and a 96-well-plate luminometer was used. In the ﬂuorescence method, the DAPI-stained cells containing GFP were analyzed by ﬂuorescence microscope, and the images were analyzed with specialized software. An alternative method is by using a ﬂow cytometer equipped with a 96-well-autosampler. In PVNA, the luciferase activity was determined by the relative light units (RLU) and the ﬂuorescence by the number of GFP-positive cells. Percent neutralization must be normalized considering uninfected cells as 100% neutralization and infected cells with ppVSV1G-SARS-CoV-2 S alone as 0% neutralization. (D) The 50% tissue culture infectious dose (TCID50) and/or the 50% ppVSV1G-SARS-CoV-2 S neutralizing doses (PVND50) were calculated according to the Reed-Muench method and reference table. FIGURE 5 \| The design of titration and pseudovirus neutralization assay (PVNA). Neutralization Assay and Correlation far, six very eﬀective COVID-19 vaccines have been approved for human use: Moderna mRNA1273, BioNTech BNT162b2, Janssen Ad26.COV2.S, Gamaleya’s Sputnik V, AstraZeneca’s AZD1222, and CoronaVac. All of these vaccines, albeit in various forms, are based on the Spike protein. Because VOC contains polymorphisms on the S gene, there is an urgent need to evaluate vaccination eﬀectiveness against prevalent SARS-CoV-2 VOC in all geographic areas. As a result, ppVSV1G-SARS-CoV-2 S have played an important role in determining vaccination eﬀectiveness for SARS-CoV-2 VOC. ppVSV1G-SARS-CoV-2 S, in particular, has been frequently used in assessing neutralizing antibody titers of vaccinated persons to evaluate if vaccinations provide enough protection against SARS-CoV-2 VOC infection (Table 2; Collier et al., 2021; Ikegame et al., 2021; McCallum et al., 2021; Muik et al., 2021; Shen et al., 2021a,b; Wang G.-L. et al., 2021; Wang P. et al., 2021). far, six very eﬀective COVID-19 vaccines have been approved for human use: Moderna mRNA1273, BioNTech BNT162b2, Janssen Ad26.COV2.S, Gamaleya’s Sputnik V, AstraZeneca’s AZD1222, and CoronaVac. All of these vaccines, albeit in various forms, are based on the Spike protein. Because VOC contains polymorphisms on the S gene, there is an urgent need to evaluate vaccination eﬀectiveness against prevalent SARS-CoV-2 VOC in all geographic areas. As a result, ppVSV1G-SARS-CoV-2 S have played an important role in determining vaccination eﬀectiveness for SARS-CoV-2 VOC. ppVSV1G-SARS-CoV-2 S, in particular, has been frequently used in assessing neutralizing antibody titers of vaccinated persons to evaluate if vaccinations provide enough protection against SARS-CoV-2 VOC infection (Table 2; Collier et al., 2021; Ikegame et al., 2021; McCallum et al., 2021; Muik et al., 2021; Shen et al., 2021a,b; Wang G.-L. et al., 2021; Wang P. et al., 2021). Recently, a new SARS-CoV-2 VOC was identiﬁed in November 2021 and named as Omicron (B.1.1.529) variant by the WHO10. This variant was ﬁrst detected in Botswana (Gauteng Province) on November 11, 2021, and 3 days later in South Africa. The Omicron variant is characterized by 26– 32 changes in the S protein, particularly within the RBD, as well as three deletions and one insertion in the S protein, along with mutations outside of the S protein [Network for Genomic Surveillance in South Africa (NGS-SA)]. Many of these mutations are either known or predicted to contribute not only to increase infectivity and transmissibility, but also to confer therapeutic and neutralization resistance (Schmidt et al., 2021). Indeed, Pulliam et al. Neutralization Assay and Correlation (2021), reported that reinfections in South Africa have increased as Omicron has spread. Therefore, studies comparing the SARS-CoV-2 Omicron variant vs. ppVSV1G-SARS-CoV-2 S are urgently needed to get a better understanding of how the immune system is reacting to this variant. Overall, the data show that SARS-CoV-2 VOC can reduce neutralization potency in vaccinated people’s sera. To note, the SARS-CoV-2 beta variant exhibited the greatest decrease in PVNAs in sera from persons who had been vaccinated with Moderna, Novavax, or PﬁzerBioNTech (Table 2). Despite the apparently worrying results obtained in the neutralization tests, it is important to consider that vaccines also induce cell mediated immunity; thus, even when neutralization levels decrease, vaccine eﬀectiveness is still high. As a result, monitoring the neutralizing activity evoked by vaccine sera will be required to assess whether a vaccination update is required to limit the establishment and spread of new SARS- CoV-2 variations, such as the delta and omicron variants. Furthermore, ppVSV1G-SARS-CoV-2 S are being utilized in the development of a vaccine against SARS-CoV-2. Thus, various VSV-SARS-CoV-2 vaccines have been shown in animal models to be eﬀective in both generating neutralizing antibodies at high titers and protecting against the SARS-CoV-2 challenge We propose comparing the SARS-CoV-2 B.1.617.2 (delta) variant to ppVSV1G-SARS-CoV-2 S with L452R, D614G, and P681R amino acid changes. This variety, which is the most common SARS-CoV-2 strain in Mexico and across the world, necessitates immediate monitoring for vaccination eﬀectiveness. In addition, given the fast spread of the Omicron variant, we also propose to compare the SARS-CoV-2 Omicron variant to ppVSV1G-SARS-CoV-2 S. aNeutralization assay (IC50 fold reduction compared to WT) in VOC, ND, not determined. Neutralization Assay and Correlation (A,B) Infection, titration and PVNA was performed in a 96-well plate with 10-fold serial dilution of ppVSV1G-SARS-CoV-2 S until a total of 8 or 10 dilutions and 6 or 8 replicates were obtained. Neutralization assay was also performed in a 96-well plate with 6 dilutions and 2 replicates; however, the sera samples were previously diluted and then mixed with ppVSV1G-SARS-CoV-2 S 325 to 1,300 TCID50/mL and incubated for 1 h at 37◦C. (C) In titration, the cells containing the luciferase were lysed to perform the luciferase assay and a 96-well-plate luminometer was used. In the ﬂuorescence method, the DAPI-stained cells containing GFP were analyzed by ﬂuorescence microscope, and the images were analyzed with specialized software. An alternative method is by using a ﬂow cytometer equipped with a 96-well-autosampler. In PVNA, the luciferase activity was determined by the relative light units (RLU) and the ﬂuorescence by the number of GFP-positive cells. Percent neutralization must be normalized considering uninfected cells as 100% neutralization and infected cells with ppVSV1G-SARS-CoV-2 S alone as 0% neutralization. (D) The 50% tissue culture infectious dose (TCID50) and/or the 50% ppVSV1G-SARS-CoV-2 S neutralizing doses (PVND50) were calculated according to the Reed-Muench method and reference table. January 2022 \| Volume 12 \| Article 817200 Frontiers in Microbiology \| www.frontiersin.org 9 VSV1G-SARS-CoV-2 S Pseudoparticles and Their Uses Salazar-García et al. showed reduced neutralization by post-infection- (2- to 4.5-fold) and vaccine-elicited antibodies (2- to 6-fold) (West et al., 2021). far, six very eﬀective COVID-19 vaccines have been approved for human use: Moderna mRNA1273, BioNTech BNT162b2, Janssen Ad26.COV2.S, Gamaleya’s Sputnik V, AstraZeneca’s AZD1222, and CoronaVac. All of these vaccines, albeit in various forms, are based on the Spike protein. Because VOC contains polymorphisms on the S gene, there is an urgent need to evaluate vaccination eﬀectiveness against prevalent SARS-CoV-2 VOC in all geographic areas. As a result, ppVSV1G-SARS-CoV-2 S have played an important role in determining vaccination eﬀectiveness for SARS-CoV-2 VOC. ppVSV1G-SARS-CoV-2 S, in particular, has been frequently used in assessing neutralizing antibody titers of vaccinated persons to evaluate if vaccinations provide enough protection against SARS-CoV-2 VOC infection (Table 2; Collier et al., 2021; Ikegame et al., 2021; McCallum et al., 2021; Muik et al., 2021; Shen et al., 2021a,b; Wang G.-L. et al., 2021; Wang P. et al., 2021). Coronavirus Disease Vaccine The ongoing development of SARS-CoV-2 variants of concern (VOC) across the world emphasizes the need of monitoring the eﬀectiveness of approved vaccinations for human use. So, TABLE 2 \| Summary of post-vaccine sera evaluated for neutralization potency by using pseudotyped VSV-Severe Acute Respiratory Syndrome-Coronavirus-2 (SARS-CoV-2) Spike variants of concern (VOC). Vaccine Company Spike construct Number of samples Time of sample collection B.1.1.7a (Alpha) P.1a (Gamma) B.1.351a (Beta) B.1.429a (Epsilon) References BNT162b2 BNT162b2 Pﬁzer/BioNTech Pﬁzer/BioNTech 2P 2P 37 21 3 weeks after 1st boost 3 weeks after 2nd boost 3.2-fold decrease 1.9-fold decrease ND ND ND ND ND ND Collier et al., 2021 mRNA-1273 NVX-CoV2373 Moderna Novavax 2P 3Q-2P 29 28 28 days after 2nd boost 2 weeks after 2nd boost 1-3-fold decrease 1-3 fold decrease ND ND ND ND ND ND Shen et al., 2021b BBIBP-CorV CoronaVac Sinopharm Sinovac Native Native 25 25 2-3 weeks after 2nd boost 2-3 weeks after 2nd boost Unchanged 0.7-fold change ND ND 2.5-fold change 3.3-fold change ND ND Wang G.-L. et al., 2021 mRNA-1273 NVX-CoV2372 Moderna Novavax 2P 3Q-2P 26 23 28 days after 2nd boost 14 days after 2nd boost ND ND ND ND 9.7-fold decrease 14.5-fold decrease 2- fold decrease 2.5-fold decrease Shen et al., 2021a Sputnik V Gamaleya Native 12 1 month after 2nd boost Unchanged 2.1-fold decrease 6.1-fold decrease ND Ikegame et al., 2021 mRNA-1273 BNT162b2 Moderna Pﬁzer/BioNTech 2P 2P 12 10 15 days after 2nd boost 7 days after 2nd boost Unchanged Unchanged ND ND 12.4-fold decrease 10.3-fold decrease ND ND Wang P. et al., 2021 BNT162b2 BNT162b2 Pﬁzer/BioNTech Pﬁzer/BioNTech 2P 2P 26 (23-55 year-old) 14 (57-73- year-old) 29 days after 2nd boost 43 days after 2nd boost 0.78-fold decrease 0.83-fold decrease ND ND ND ND ND ND Muik et al., 2021 mRNA-1273 BNT162b2 Moderna Pﬁzer/BioNTech 2P 2P 15 18 7-27 days after 2nd boost 7-27 days after 2nd boost ND 1.3-fold reduction ND 1.7-fold decrease ND 3.2-fold decrease 2.2-fold decrease 2.9-fold decrease McCallum et al., 2021 aNeutralization assay (IC50 fold reduction compared to WT) in VOC, ND, not determined. January 2022 \| Volume 12 \| Article 817200 Frontiers in Microbiology \| www.frontiersin.org 10 VSV1G-SARS-CoV-2 S Pseudoparticles and Their Uses Salazar-García et al. and 4 viruses such as SARS-CoV-2. It can be used in BSL-2 facilities for studies of host and cellular tropism, interspecies transmission, viral pathogenesis, and host cell entry pathways. Coronavirus Disease Therapeutics Some COVID-19 treatment drugs are directed against the SARS- CoV-2 S protein and its receptor ACE2, which are found on the membranes of diﬀerent human cells. Through ACE2 attachment, SARS-CoV-2 causes cell-membrane fusion, facilitating viral entrance (Hoﬀmann et al., 2020). Although preclinical research for eﬀective drugs against SARS-CoV-2 necessitates the use of live SARS-CoV-2, ppVSV1G-SARS-CoV-2 S can be utilized for S protein-focused therapy evaluation by blocking or down- regulating ACE2 or preventing viral fusion. g g p g MEK inhibitors (VS-6766, trametinib, and selumetinib) have been used to reduce ACE2 cellular expression as a method to prevent early SARS-CoV-2 infection. ppVSV1G- SARS-CoV-2 S and human bronchial epithelial, small airway epithelial, and lung cancer cells were utilized to assess infectivity alterations caused by MEK inhibitors (Zhou et al., 2020). Furthermore, PVs were used to test the creation of fusion inhibitor peptides against SARS-CoV-2, and micromolar concentrations of peptides suppressed ppVSV1G-SARS-CoV- 2 S infection by inhibiting viral fusion (Kandeel et al., 2021). Demethylzeylasteral, another inhibiting drug, can interact with hACE2 and the RBD of the SARS-CoV-2 S protein, thus when tested in ppVSV1G-SARS-CoV-2 S, it inhibited ppVSV1G- SARS-CoV-2 S entrance into 293T cells (Zhu et al., 2020). Polyunsaturated-3 fatty acids limit SARS-CoV-2 binding and cellular entrance, while linolenic and eicosapentaenoic acids prevent ppVSV1G-SARS-CoV-2 S penetration (Goc et al., 2021). Inhalable nano catchers containing hACE2 are a proposal for SARS-CoV-2 suppression, which was tested with ppVSV1G- SARS-CoV-2 S and shown a strong capability for infection inhibition in a hACE2-expressing mouse model (Zhang et al., 2021). Furthermore, when tested with ppVSV1G-SARS-CoV- 2 S and live SARS-CoV-2, hACE2-Fc inhibited Vero E6 cells (Case et al., 2020b). AUTHOR CONTRIBUTIONS VL-P and MS-G: conceptualization, resources, project administration, and funding acquisition. AC-R and AF-A: software and formal analysis. SA-C, GP-L, GR-M, AC-R, and VL-P: investigation. VL-P, GP-L, and GR-M: writing—original draft preparation. VL-P, GP-L, and MS-G: writing—review and editing. SA-C, AC-R, and AF-A: visualization. VL-P: supervision. All authors have read and agreed to the published version of the manuscript. VL-P and MS-G: conceptualization, resources, project administration, and funding acquisition. AC-R and AF-A: software and formal analysis. SA-C, GP-L, GR-M, AC-R, and VL-P: investigation. VL-P, GP-L, and GR-M: writing—original draft preparation. VL-P, GP-L, and MS-G: writing—review and editing. SA-C, AC-R, and AF-A: visualization. VL-P: supervision. All authors have read and agreed to the published version of the manuscript. FUNDING This review was funded by the Public Federal Funds from HIMFG with grant numbers HIM/2020/029 SSA. 1664, HIM/2020/060 SSA. 1686, and HIM/2021/007. Coronavirus Disease Vaccine ppVSV1G-SARS-CoV-2 S can be supplemented with SARS- CoV-2 S protein via changes that improve expression eﬃciency, and we propose the cloning of Met1-S-121 with usage codons. Once the S gen has been reﬁned and cloned, we propose using site-directed mutagenesis to generate novel variations of the SARS-CoV-2 S protein. ppVSV1G-G and ppVSV1G- SARS-CoV-2 S particles must be produced in HEK 293T cells, and infection assays with ppVSV1G-SARS-CoV-2 S in Vero E6 or ACE2- and Furin-transfected HEK 293T cells are suggested. The ppVSV-SARS-CoV-1-S are stable once produced, and they may be kept with low activity loss for up to 6 months (−20 or −80◦C), 4 weeks (4◦C), and just 1 week (room temperature) with up to four freeze-thaw cycles. PVNA, surprisingly, is a safe (BSL-2), eﬃcient (7.5 h), and scalable (r ≥0.9 and p ≤0.001) method that may be utilized to evaluate novel SARS-CoV-2 variants, post-infection- and vaccine-elicited neutralizing antibodies, and S protein-based COVID-19 therapies. ppVSV1G-SARS-CoV-2 S can be detected by a ﬂuorescence or luminescence signal with a linear correlation between the value of ﬂuorescence or chemiluminescence and the number of infective ppVSV1G-SARS-CoV-2 S, and these particles can be easily calculated using the TCID50/mL or PVID50 equations, which are included in this review. We are the ﬁrst Mexican organization to employ ppVSV1G- SARS-CoV-2 S for preclinical assessment of post-infection- neutralizing antibodies, evaluation of a possible COVID-19 vaccine and decontamination equipment, and evaluation of vaccinated volunteers against SARS-CoV-2 delta variants. (Case et al., 2020a; Yahalom-Ronen et al., 2020; Lu et al., 2021; Malherbe et al., 2021). Although the evidence provided by these vector vaccines is encouraging, it is still early in the research process, and additional study is needed before moving forward with human trials. CONCLUSION There are only seven BSL-3 laboratories in Mexico, four of which are in Mexico City. Because of the scarcity of BSL-3 facilities, research into existing and newly emerging infectious diseases are hampered. Furthermore, there is no BSL- 4 laboratory in Mexico, making it impossible to research the viruses that are classiﬁed as risk group 4 and are designated priority diseases by the WHO. This is an invitation to the Mexican government to develop research policies and infrastructure for the construction of BSL3 and BSL4 facilities. ppVSV1G-SARS-CoV-2 S is an excellent choice for studies of host and cellular tropism, interspecies transmission, viral pathogenesis, and host cell entry pathways with risk group 3 Frontiers in Microbiology \| www.frontiersin.org REFERENCES doi: 10.1016/j. peptides.2005.01.009 Case, J. B., Rothlauf, P. W., Chen, R. E., Liu, Z., Zhao, H., Kim, A. S., et al. (2020b). Neutralizing antibody and soluble ACE2 inhibition of a replication-competent VSV-SARS-CoV-2 and a clinical isolate of SARS-CoV-2. Cell Host Microbe 28, 475–485.e5. doi: 10.1016/j.chom.2020.06.021 Giroglou, T., Cinatl, J., Rabenau, H., Drosten, C., Schwalbe, H., Doerr, H. W., et al. (2004). Retroviral vectors pseudotyped with severe acute respiratory syndrome coronavirus S protein. J. Virol. 78, 9007–9015. doi: 10.1128/JVI.78.17.9007- 9015.2004 Chang, K. W., Sheng, Y., and Gombold, J. L. (2000). Coronavirus-induced membrane fusion requires the cysteine-rich domain in the spike protein. Virology 269, 212–224. doi: 10.1006/viro.2000.0219 Choppin, P. W., and Compans, R. W. (1970). Phenotypic mixing of envelope proteins of the parainﬂuenza virus SV5 and vesicular stomatitis virus. J. Virol. 5, 609–616. doi: 10.1128/JVI.5.5.609-616.1970 Goc, A., Niedzwiecki, A., and Rath, M. (2021). Polyunsaturated ω-3 fatty acids inhibit ACE2-controlled SARS-CoV-2 binding and cellular entry. Sci. Rep. 11:5207. doi: 10.1038/s41598-021-84850-1 Collier, D. A., De Marco, A., Ferreira, I. A. T. M., Meng, B., Datir, R. P., Walls, A. C., et al. (2021). Sensitivity of SARS-CoV-2 B.1.1.7 to mRNA vaccine-elicited antibodies. Nature 593, 136–141. doi: 10.1038/s41586-021-03412-7 Hastie, K. M., Zandonatti, M. A., Kleinfelter, L. M., Heinrich, M. L., Rowland, M. M., Chandran, K., et al. (2017). Structural basis for antibody-mediated neutralization of Lassa virus. Science 356, 923–928. doi: 10.1126/science. aam7260 Condor Capcha, J. M., Lambert, G., Dykxhoorn, D. M., Salerno, A. G., Hare, J. M., Whitt, M. A., et al. (2020). Generation of SARS-CoV-2 spike pseudotyped virus for viral entry and neutralization assays: a 1-week protocol. Front. Cardiovasc. Med. 7:618651. doi: 10.3389/fcvm.2020.618651 Havranek, K. E., Jimenez, A. R., Acciani, M. D., Lay Mendoza, M. F., Reyes Ballista, J. M., Diaz, D. A., et al. (2020). SARS-CoV-2 spike alterations enhance pseudoparticle titers and replication-competent VSV-SARS-CoV-2 virus. Viruses 12:1465. doi: 10.3390/v12121465 Contreras, E. M., Johnston, G. P., Buchholz, D. W., Ortega, V., Monreal, I. A., Zamora, J. L. R., et al. (2021). Roles of cholesterol in early and late steps of the nipah virus membrane fusion cascade. J. Virol. 95, e02323–20. doi: 10.1128/JVI. 02323-20 Hoﬀmann, M., Kleine-Weber, H., Schroeder, S., Krüger, N., Herrler, T., Erichsen, S., et al. (2020). SARS-CoV-2 cell entry depends on ACE2 and TMPRSS2 and is blocked by a clinically proven protease inhibitor. Cell 181, 271–280.e8. doi: 10.1016/j.cell.2020.02.052 Crawford, K. H. D., Eguia, R., Dingens, A. S., Loes, A. N., Malone, K. REFERENCES for detection of neutralizing antibody responses to SARS-CoV. J. Med. Virol. 78, 1509–1512. doi: 10.1002/jmv.20732 for detection of neutralizing antibody responses to SARS-CoV. J. Med. Virol. 78, 1509–1512. doi: 10.1002/jmv.20732 Alam, J., and Cook, J. L. (1990). Reporter genes: application to the study of mammalian gene transcription. Anal. Biochem. 188, 245–254. doi: 10.1016/ 0003-2697(90)90601-5 j Fukushi, S., Watanabe, R., and Taguchi, F. (2008). Pseudotyped vesicular stomatitis virus for analysis of virus entry mediated by SARS coronavirus spike proteins. Methods Mol. Biol. 454, 331–338. doi: 10.1007/978-1-59745-181-9_23 Artika, I. M., and Ma’roef, C. N. (2017). Laboratory biosafety for handling emerging viruses. Asian Pac. J. Trop. Biomed. 7, 483–491. doi: 10.1016/j.apjtb.2017.01.020 Gasbarri, M., V’kovski, P., Torriani, G., Thiel, V., Stellacci, F., Tapparel, C., et al. (2020). SARS-CoV-2 inhibition by sulfonated compounds. Microorganisms 8:E1894. doi: 10.3390/microorganisms8121894 Barrett, C. T., and Dutch, R. E. (2020). Viral membrane fusion and the transmembrane domain. Viruses 12:E693. doi: 10.3390/v12070693 Gauger, P. C., and Vincent, A. L. (2014). Serum virus neutralization assay for detection and quantitation of serum-neutralizing antibodies to inﬂuenza A virus in swine. Methods Mol. Biol. 1161, 313–324. doi: 10.1007/978-1-4939- 0758-8_26 Blumenthal, R., Durell, S., and Viard, M. (2012). HIV entry and envelope glycoprotein-mediated fusion. J. Biol. Chem. 287, 40841–40849. doi: 10.1074/ jbc.R112.406272 Ge, J., Wen, Z., Wang, X., Hu, S., Liu, Y., Kong, X., et al. (2006). Generating vesicular stomatitis virus pseudotype bearing the severe acute respiratory syndrome coronavirus spike envelope glycoprotein for rapid and safe neutralization test or cell-entry assay. Ann. N. Y. Acad. Sci. 1081, 246–248. doi: 10.1196/annals.1373.030 Bukbuk, D. N., Fukushi, S., Tani, H., Yoshikawa, T., Taniguchi, S., Iha, K., et al. (2014). Development and validation of serological assays for viral hemorrhagic fevers and determination of the prevalence of Rift Valley fever in Borno State, Nigeria. Trans. R. Soc. Trop. Med. Hyg. 108, 768–773. doi: 10.1093/trstmh/ tru163 Geisbert, T. W., and Feldmann, H. (2011). Recombinant vesicular stomatitis virus– based vaccines against Ebola and Marburg virus infections. J. Infect. Dis. 204, S1075–S1081. doi: 10.1093/infdis/jir349 Case, J. B., Rothlauf, P. W., Chen, R. E., Kafai, N. M., Fox, J. M., Smith, B. K., et al. (2020a). Replication-competent vesicular stomatitis virus vaccine vector protects against SARS-CoV-2-mediated pathogenesis in mice. Cell Host Microbe 28, 465–474.e4. doi: 10.1016/j.chom.2020.07.018 Gembardt, F., Sterner-Kock, A., Imboden, H., Spalteholz, M., Reibitz, F., Schultheiss, H.-P., et al. (2005). Organ-speciﬁc distribution of ACE2 mRNA and correlating peptidase activity in rodents. Peptides 26, 1270–1277. REFERENCES D., Wolf, C. R., et al. (2020). Protocol and reagents for pseudotyping lentiviral particles with SARS-CoV-2 Spike protein for neutralization assays. Viruses 12:E513. doi: 10.3390/v12050513 Huang, A. S., Palma, E. L., Hewlett, N., and Roizman, B. (1974). Pseudotype formation between enveloped RNA and DNA viruses. Nature 252, 743–745. doi: 10.1038/252743a0 Hyseni, I., Molesti, E., Benincasa, L., Piu, P., Casa, E., Temperton, N. J., et al. (2020). Characterisation of SARS-CoV-2 lentiviral pseudotypes and correlation between pseudotype-based neutralisation assays and live virus- based micro neutralisation assays. Viruses 12:E1011. doi: 10.3390/v1209 1011 Datta, P. K., Liu, F., Fischer, T., Rappaport, J., and Qin, X. (2020). SARS-CoV- 2 pandemic and research gaps: understanding SARS-CoV-2 interaction with the ACE2 receptor and implications for therapy. Theranostics 10, 7448–7464. doi: 10.7150/thno.48076 Deng, X., Garcia-Knight, M. A., Khalid, M. M., Servellita, V., Wang, C., Morris, M. K., et al. (2021). Transmission, infectivity, and neutralization of a spike L452R SARS-CoV-2 variant. Cell 184, 3426–3437.e8. doi: 10.1016/j.cell.2021. 04.025 Ikegame, S., Siddiquey, M. N. A., Hung, C.-T., Haas, G., Brambilla, L., Oguntuyo, K. Y., et al. (2021). Neutralizing activity of Sputnik V vaccine sera against SARS-CoV-2 variants. Nat. Commun. 12:4598. doi: 10.1038/s41467-021-2 4909-9 Dessau, M., and Modis, Y. (2013). Crystal structure of glycoprotein C from Rift Valley fever virus. Proc. Natl. Acad. Sci. U.S.A. 110, 1696–1701. doi: 10.1073/ pnas.1217780110 Johnson, M. C., Lyddon, T. D., Suarez, R., Salcedo, B., LePique, M., Graham, M., et al. (2020). Optimized pseudotyping conditions for the SARS-COV-2 spike glycoprotein. J. Virol. 94, e01062–20. doi: 10.1128/JVI.01062-20 Dieterle, M. E., Haslwanter, D., Bortz, R. H., Wirchnianski, A. S., Lasso, G., Vergnolle, O., et al. (2020). A replication-competent vesicular stomatitis virus for studies of SARS-CoV-2 spike-mediated cell entry and its inhibition. Cell Host Microbe 28, 486–496.e6. doi: 10.1016/j.chom.2020.06.020 Kaku, Y., Noguchi, A., Marsh, G. A., Barr, J. A., Okutani, A., Hotta, K., et al. (2012). Second generation of pseudotype-based serum neutralization assay for Nipah virus antibodies: sensitive and high-throughput analysis utilizing secreted alkaline phosphatase. J. Virol. Methods 179, 226–232. doi: 10.1016/j. jviromet.2011.11.003 Fukuma, A., Tani, H., Taniguchi, S., Shimojima, M., Saijo, M., and Fukushi, S. (2015). Inability of rat DPP4 to allow MERS-CoV infection revealed by using a VSV pseudotype bearing truncated MERS-CoV spike protein. Arch. Virol. 160, 2293–2300. doi: 10.1007/s00705-015-2506-z Kaku, Y., Noguchi, A., Marsh, G. A., McEachern, J. A., Okutani, A., Hotta, K., et al. (2009). ACKNOWLEDGMENTS We want to acknowledge the research department of HIMFG for the support given in this review. January 2022 \| Volume 12 \| Article 817200 Frontiers in Microbiology \| www.frontiersin.org 11 VSV1G-SARS-CoV-2 S Pseudoparticles and Their Uses Salazar-García et al. REFERENCES Lancet 388, 460– 461. doi: 10.1016/S0140-6736(16)31152-7 samples. J. Med. Virol. 92, 2096–2104. doi: 10.1002/jmv.25986 Mauro, V. P., and Chappell, S. A. (2014). A critical analysis of codon optimization in human therapeutics. Trends Mol. Med. 20, 604–613. doi: 10.1016/j.molmed. 2014.09.003 Kim, I. S., Jenni, S., Stanifer, M. L., Roth, E., Whelan, S. P. J., van Oijen, A. M., et al. (2017). Mechanism of membrane fusion induced by vesicular stomatitis virus G protein. Proc. Natl. Acad. Sci. U.S.A. 114, E28–E36. doi: 10.1073/pnas. 1618883114 McBride, C. E., Li, J., and Machamer, C. E. (2007). The cytoplasmic tail of the severe acute respiratory syndrome coronavirus spike protein contains a novel endoplasmic reticulum retrieval signal that binds COPI and promotes interaction with membrane protein. J. Virol. 81, 2418–2428. doi: 10.1128/JVI. 02146-06 Kim, Y. J., Jang, U. S., Soh, S. M., Lee, J.-Y., and Lee, H.-R. (2021). The impact on infectivity and neutralization eﬃciency of SARS-CoV-2 lineage B.1.351 pseudovirus. Viruses 13:633. doi: 10.3390/v13040633 McCallum, M., Bassi, J., De Marco, A., Chen, A., Walls, A. C., Di Iulio, J., et al. (2021). SARS-CoV-2 immune evasion by the B.1.427/B.1.429 variant of concern. Science 373, 648–654. doi: 10.1126/science.abi7994 Korber, B., Fischer, W. M., Gnanakaran, S., Yoon, H., Theiler, J., Abfalterer, W., et al. (2020). Tracking changes in SARS-CoV-2 spike: evidence that D614G increases infectivity of the COVID-19 virus. Cell 182, 812–827.e19. doi: 10. 1016/j.cell.2020.06.043 Mendenhall, A., Lesnik, J., Mukherjee, C., Antes, T., and Sengupta, R. (2012). Packaging HIV- or FIV-based lentivector expression constructs and transduction of VSV-G pseudotyped viral particles. J. Vis. Exp. 63:e3171. doi: 10.3791/3171 j Kudla, G., Lipinski, L., Caﬃn, F., Helwak, A., and Zylicz, M. (2006). High guanine and cytosine content increases mRNA levels in mammalian cells. PLoS Biol. 4:e180. doi: 10.1371/journal.pbio.0040180 Millet, J. K., Tang, T., Nathan, L., Jaimes, J. A., Hsu, H.-L., Daniel, S., et al. (2019). Production of pseudotyped particles to study highly pathogenic coronaviruses in a biosafety level 2 setting. J. Vis. Exp. doi: 10.3791/59010 Kunz, S., Rojek, J. M., Perez, M., Spiropoulou, C. F., and Oldstone, M. B. A. (2005). Characterization of the interaction of Lassa fever virus with its cellular receptor alpha-dystroglycan. J. Virol. 79, 5979–5987. doi: 10.1128/JVI.79.10.5979-5987. 2005 in a biosafety level 2 setting. J. Vis. Exp. doi: 10.3791/59010 Molesti, E., Wright, E., Terregino, C., Rahman, R., Cattoli, G., and Temperton, N. J. (2014). Multiplex evaluation of inﬂuenza neutralizing antibodies with potential applicability to in-ﬁeld serological studies. J. Immunol. REFERENCES Res. 2014:457932. doi: 10.1155/2014/457932 Lawson, N. D., Stillman, E. A., Whitt, M. A., and Rose, J. K. (1995). Recombinant vesicular stomatitis viruses from DNA. Proc. Natl. Acad. Sci. U.S.A. 92, 4477– 4481. doi: 10.1073/pnas.92.10.4477 Lee, J. E., Fusco, M. L., Hessell, A. J., Oswald, W. B., Burton, D. R., and Saphire, E. O. (2008). Structure of the Ebola virus glycoprotein bound to an antibody from a human survivor. Nature 454, 177–182. doi: 10.1038/nature07082 Moore, M. J., Dorfman, T., Li, W., Wong, S. K., Li, Y., Kuhn, J. H., et al. (2004). Retroviruses pseudotyped with the severe acute respiratory syndrome coronavirus spike protein eﬃciently infect cells expressing angiotensin- converting enzyme 2. J. Virol. 78, 10628–10635. doi: 10.1128/JVI.78.19.10628- 10635.2004 Lennemann, N. J., Herbert, A. S., Brouillette, R., Rhein, B., Bakken, R. A., Perschbacher, K. J., et al. (2017). Vesicular Stomatitis Virus Pseudotyped with Ebola Virus glycoprotein serves as a protective, noninfectious vaccine against Ebola virus challenge in mice. J. Virol. 91, e00479–17. doi: 10.1128/JVI.00 479-17 Morens, D. M., and Fauci, A. S. (2020). Emerging pandemic diseases: how we got to COVID-19. Cell 182, 1077–1092. doi: 10.1016/j.cell.2020.08.021 Morens, D. M., Folkers, G. K., and Fauci, A. S. (2008). Emerging infections: a perpetual challenge. Lancet Infect. Dis. 8, 710–719. doi: 10.1016/S1473-3099(08) 70256-1 Lester, S., Harcourt, J., Whitt, M., Al-Abdely, H. M., Midgley, C. M., Alkhamis, A. M., et al. (2019). Middle East respiratory coronavirus (MERS-CoV) spike (S) protein vesicular stomatitis virus pseudoparticle neutralization assays oﬀer a reliable alternative to the conventional neutralization assay in human seroepidemiological studies. Access Microbiol. 1:e000057. doi: 10.1099/acmi.0. 000057 Muik, A., Dold, C., Geiß, Y., Volk, A., Werbizki, M., Dietrich, U., et al. (2012). Semireplication-competent Vesicular Stomatitis virus as a novel platform for oncolytic virotherapy. J. Mol. Med. (Berl.) 90, 959–970. doi: 10.1007/s00109- 012-0863-6 Liu, R., Wang, J., Shao, Y., Wang, X., Zhang, H., Shuai, L., et al. (2018). A recombinant VSV-vectored MERS-CoV vaccine induces neutralizing antibody and T cell responses in rhesus monkeys after single dose immunization. Antiviral Res. 150, 30–38. doi: 10.1016/j.antiviral.2017.12.007 Muik, A., Wallisch, A.-K., Sänger, B., Swanson, K. A., Mühl, J., Chen, W., et al. (2021). Neutralization of SARS-CoV-2 lineage B.1.1.7 pseudovirus by BNT162b2 vaccine-elicited human sera. Science 371, 1152–1153. doi: 10.1126/ science.abg6105 Lodish, H. F., and Weiss, R. A. (1979). Selective isolation of mutants of Vesicular Stomatitis virus defective in production of the viral glycoprotein. J. Virol. 30, 177–189. REFERENCES A neutralization test for speciﬁc detection of Nipah virus antibodies using Pseudotyped Vesicular Stomatitis virus expressing green ﬂuorescent protein. J. Virol. Methods 160, 7–13. doi: 10.1016/j.jviromet.2009.04.037 Fukushi, S., Mizutani, T., Saijo, M., Kurane, I., Taguchi, F., Tashiro, M., et al. (2006). Evaluation of a novel vesicular stomatitis virus pseudotype-based assay January 2022 \| Volume 12 \| Article 817200 Frontiers in Microbiology \| www.frontiersin.org 12 VSV1G-SARS-CoV-2 S Pseudoparticles and Their Uses Salazar-García et al. Kandeel, M., Yamamoto, M., Tani, H., Kobayashi, A., Gohda, J., Kawaguchi, Y., et al. (2021). Discovery of new fusion inhibitor peptides against SARS-CoV- 2 by targeting the Spike S2 subunit. Biomol. Ther. (Seoul) 29, 282–289. doi: 10.4062/biomolther.2020.201 Lun, C. M., Waheed, A. A., Majadly, A., Powell, N., and Freed, E. O. (2021). Mechanism of viral glycoprotein targeting by membrane-associated RING-CH proteins. mBio 12, e219–e221. doi: 10.1128/mBio.00219-21 Malherbe, D. C., Kurup, D., Wirblich, C., Ronk, A. J., Mire, C., Kuzmina, N., et al. (2021). A single dose of replication-competent VSV-vectored vaccine expressing SARS-CoV-2 S1 protects against virus replication in a hamster model of severe COVID-19. NPJ Vaccines 6:91. doi: 10.1038/s41541-021- 00352-1 Kapadia, S. U., Simon, I. D., and Rose, J. K. (2008). SARS vaccine based on a replication-defective recombinant vesicular stomatitis virus is more potent than one based on a replication-competent vector. Virology 376, 165–172. doi: 10.1016/j.virol.2008.03.002 Kaufer, A. M., Theis, T., Lau, K. A., Gray, J. L., and Rawlinson, W. D. (2020). Laboratory biosafety measures involving SARS-CoV-2 and the classiﬁcation as a Risk Group 3 biological agent. Pathology 52, 790–795. doi: 10.1016/j.pathol. 2020.09.006 Mammano, F., Salvatori, F., Indraccolo, S., De Rossi, A., Chieco-Bianchi, L., and Göttlinger, H. G. (1997). Truncation of the human immunodeﬁciency virus type 1 envelope glycoprotein allows eﬃcient pseudotyping of Moloney Murine Leukemia virus particles and gene transfer into CD4+ cells. J. Virol. 71, 3341–3345. doi: 10.1128/JVI.71.4.3341-3345.1997 3341–3345. doi: 10.1128/JVI.71.4.3341-3345.1997 Keyaerts, E., Vijgen, L., Maes, P., Neyts, J., and Van Ranst, M. (2005). Growth kinetics of SARS-coronavirus in Vero E6 cells. Biochem. Biophys. Res. Commun. 329, 1147–1151. doi: 10.1016/j.bbrc.2005.02.085 Manenti, A., Maggetti, M., Casa, E., Martinuzzi, D., Torelli, A., Trombetta, C. M., et al. (2020). Evaluation of SARS-CoV-2 neutralizing antibodies using of a CPE-based Colorimetric live virus micro-neutralization assay in human serum samples. J. Med. Virol. 92, 2096–2104. doi: 10.1002/jmv.25986 Kieny, M. P., Rottingen, J.-A., and Farrar, J. (2016). The need for global R&D coordination for infectious diseases with epidemic potential. REFERENCES doi: 10.1177/003335491012500105 Diagnostics (Basel) 11:994. doi: 10.3390/diagnostics11060994 Verma, D., Church, T. M., and Swaminathan, S. (2021). Epstein-barr virus lytic replication induces ACE2 expression and enhances SARS-CoV-2 pseudotyped virus entry in epithelial cells. J. Virol. 95:e0019221. doi: 10.1128/JVI.00192-21 Rodriguez, S. E., Cross, R. W., Fenton, K. A., Bente, D. A., Mire, C. E., and Geisbert, T. W. (2019). Vesicular Stomatitis virus-based vaccine protects mice against Crimean-Congo hemorrhagic fever. Sci. Rep. 9:7755. doi: 10.1038/s41598-019- 44210-6 Wang, G.-L., Wang, Z.-Y., Duan, L.-J., Meng, Q.-C., Jiang, M.-D., Cao, J., et al. (2021). Susceptibility of circulating SARS-CoV-2 variants to neutralization. N. Engl. J. Med. 384, 2354–2356. doi: 10.1056/NEJMc2103022 Roy, V., Fischinger, S., Atyeo, C., Slein, M., Loos, C., Balazs, A., et al. (2020). SARS- CoV-2-speciﬁc ELISA development. J. Immunol. Methods 484–485:112832. doi: 10.1016/j.jim.2020.112832 Wang, P., Nair, M. S., Liu, L., Iketani, S., Luo, Y., Guo, Y., et al. (2021). Antibody resistance of SARS-CoV-2 variants B.1.351 and B.1.1.7. Nature 593, 130–135. doi: 10.1038/s41586-021-03398-2 Saito, T., Maruyama, J., Nagata, N., Isono, M., Okuya, K., Takadate, Y., et al. (2020). A surrogate animal model for screening of Ebola and Marburg Glycoprotein-targeting drugs using pseudotyped Vesicular Stomatitis viruses. Viruses 12:E923. doi: 10.3390/v12090923 Weiss, R. A., and Bennett, P. L. (1980). Assembly of membrane glycoproteins studied by phenotypic mixing between mutants of vesicular stomatitis virus and retroviruses. Virology 100, 252–274. doi: 10.1016/0042-6822(80)90518-8 Schmidt, F., Weisblum, Y., Muecksch, F., Hoﬀmann, H.-H., Michailidis, E., Lorenzi, J. C. C., et al. (2020). Measuring SARS-CoV-2 neutralizing antibody activity using pseudotyped and chimeric viruses. J. Exp. Med. 217:e20201181. doi: 10.1084/jem.20201181 Weiss, R. A., Boettiger, D., and Murphy, H. M. (1977). Pseudotypes of avian sarcoma viruses with the envelope properties of vesicular stomatitis virus. Virology 76, 808–825. doi: 10.1016/0042-6822(77)90261-6 Schmidt, F., Weisblum, Y., Rutkowska, M., Poston, D., DaSilva, J., Zhang, F., et al. (2021). High genetic barrier to SARS-CoV-2 polyclonal neutralizing antibody escape. Nature 600, 512–516. doi: 10.1038/s41586-021-04005-0 West, A. P., Wertheim, J. O., Wang, J. C., Vasylyeva, T. I., Havens, J. L., Chowdhury, M. A., et al. (2021). Detection and characterization of the SARS-CoV-2 lineage B.1.526 in New York. Nat. Commun. 12:4886. doi: 10.1038/s41467-021-2 5168-4 Schnell, M. J., Buonocore, L., Kretzschmar, E., Johnson, E., and Rose, J. K. (1996). Foreign glycoproteins expressed from recombinant vesicular stomatitis viruses are incorporated eﬃciently into virus particles. Proc. Natl. Acad. Sci. U.S.A. 93, 11359–11365. doi: 10.1073/pnas.93.21.11359 Whelan, S. P., Ball, L. A., Barr, J. N., and Wertz, G. T. REFERENCES doi: 10.1128/JVI.30.1.177-189.1979 Muñoz-Barroso, I., Salzwedel, K., Hunter, E., and Blumenthal, R. (1999). Role of the membrane-proximal domain in the initial stages of human immunodeﬁciency virus type 1 envelope glycoprotein-mediated membrane fusion. J. Virol. 73, 6089–6092. doi: 10.1128/JVI.73.7.6089-6092.1999 Lontok, E., Corse, E., and Machamer, C. E. (2004). Intracellular targeting signals contribute to localization of Coronavirus spike proteins near the virus assembly site. J. Virol. 78, 5913–5922. doi: 10.1128/JVI.78.11.5913-5922.2004 Nie, J., Li, Q., Wu, J., Zhao, C., Hao, H., Liu, H., et al. (2020a). Establishment and validation of a pseudovirus neutralization assay for SARS-CoV-2. Emerg. Microbes Infect. 9, 680–686. doi: 10.1080/22221751.2020.1743767 Lu, M., Zhang, Y., Dravid, P., Li, A., Zeng, C., Kc, M., et al. (2021). A methyltransferase-defective VSV-based SARS-CoV-2 vaccine candidate provides complete protection against SARS-CoV-2 infection in hamsters. J. Virol. 95:e0059221. doi: 10.1128/JVI.00592-21 Nie, J., Li, Q., Wu, J., Zhao, C., Hao, H., Liu, H., et al. (2020b). Quantiﬁcation of SARS-CoV-2 neutralizing antibody by a pseudotyped virus-based assay. Nat. Protoc. 15, 3699–3715. doi: 10.1038/s41596-020-0394-5 January 2022 \| Volume 12 \| Article 817200 Frontiers in Microbiology \| www.frontiersin.org 13 VSV1G-SARS-CoV-2 S Pseudoparticles and Their Uses Salazar-García et al. Ogando, N. S., Dalebout, T. J., Zevenhoven-Dobbe, J. C., Limpens, R. W. A. L., van der Meer, Y., Caly, L., et al. (2020). SARS-coronavirus-2 replication in Vero E6 cells: replication kinetics, rapid adaptation and cytopathology. J. Gen. Virol. 101, 925–940. doi: 10.1099/jgv.0.001453 Slough, M. M., Chandran, K., and Jangra, R. K. (2019). Two Point mutations in old world hantavirus glycoproteins aﬀord the generation of highly infectious recombinant Vesicular Stomatitis virus vectors. mBio 10, e02372–18. doi: 10. 1128/mBio.02372-18 Stillman, E. A., Rose, J. K., and Whitt, M. A. (1995). Replication and ampliﬁcation of novel vesicular stomatitis virus minigenomes encoding viral structural proteins. J. Virol. 69, 2946–2953. doi: 10.1128/JVI.69.5.2946-2953.1995 Pallesen, J., Wang, N., Corbett, K. S., Wrapp, D., Kirchdoerfer, R. N., Turner, H. L., et al. (2017). Immunogenicity and structures of a rationally designed prefusion MERS-CoV spike antigen. Proc. Natl. Acad. Sci. U.S.A. 114, E7348–E7357. doi: 10.1073/pnas.1707304114 Suda, Y., Fukushi, S., Tani, H., Murakami, S., Saijo, M., Horimoto, T., et al. (2016). Analysis of the entry mechanism of Crimean-Congo hemorrhagic fever virus, using a Vesicular Stomatitis virus pseudotyping system. Arch. Virol. 161, 1447–1454. doi: 10.1007/s00705-016-2803-1 Pattnaik, A. K., Ball, L. A., LeGrone, A. W., and Wertz, G. W. (1992). Infectious defective interfering particles of VSV from transcripts of a cDNA clone. Cell 69, 1011–1020. REFERENCES doi: 10.1016/0092-8674(92)90619-n Pattnaik, A. K., and Wertz, G. W. (1990). Replication and ampliﬁcation of defective interfering particle RNAs of vesicular stomatitis virus in cells expressing viral proteins from vectors containing cloned cDNAs. J. Virol. 64, 2948–2957. doi: 10.1128/JVI.64.6.2948-2957.1990 Tada, T., Zhang, Y., Koyama, T., Tobiume, M., Tsunetsugu-Yokota, Y., Yamaoka, S., et al. (2015). MARCH8 inhibits HIV-1 infection by reducing virion incorporation of envelope glycoproteins. Nat. Med. 21, 1502–1507. doi: 10. 1038/nm.3956 Takada, A., Robison, C., Goto, H., Sanchez, A., Murti, K. G., Whitt, M. A., et al. (1997). A system for functional analysis of Ebola virus glycoprotein. Proc. Natl. Acad. Sci. U.S.A. 94, 14764–14769. doi: 10.1073/pnas.94.26.14764 Petit, C. M., Chouljenko, V. N., Iyer, A., Colgrove, R., Farzan, M., Knipe, D. M., et al. (2007). Palmitoylation of the cysteine-rich endodomain of the SARS- coronavirus spike glycoprotein is important for spike-mediated cell fusion. Virology 360, 264–274. doi: 10.1016/j.virol.2006.10.034 Tamin, A., Harcourt, B. H., Lo, M. K., Roth, J. A., Wolf, M. C., Lee, B., et al. (2009). Development of a neutralization assay for Nipah virus using pseudotype Tamin, A., Harcourt, B. H., Lo, M. K., Roth, J. A., Wolf, M. C., Lee, B., et al. (2009). Development of a neutralization assay for Nipah virus using pseudotype particles. J. Virol. Methods 160, 1–6. doi: 10.1016/j.jviromet.2009.02.025 Plescia, C. B., David, E. A., Patra, D., Sengupta, R., Amiar, S., Su, Y., et al. (2021). SARS-CoV-2 viral budding and entry can be modeled using BSL-2 level virus-like particles. J. Biol. Chem. 296:100103. doi: 10.1074/jbc.RA120.016148 (2009). Development of a neutralization assay for Nipah virus using pseudotype particles. J. Virol. Methods 160, 1–6. doi: 10.1016/j.jviromet.2009.02.025 Tani, H., Shiokawa, M., Kaname, Y., Kambara, H., Mori, Y., Abe, T., et al. (2010). Involvement of ceramide in the propagation of Japanese encephalitis virus. J. Virol. 84, 2798–2807. doi: 10.1128/JVI.02499-09 Pulliam, J. R. C., van Schalkwyk, C., Govender, N., von Gottberg, A., Cohen, C., Groome, M. J., et al. (2021). Increased risk of SARS-CoV-2 reinfection associated with emergence of the Omicron variant in South Africa. medRxiv [Preprint]. doi: 10.1101/2021.11.11.21266068 Tolah, A. M. K., Sohrab, S. S., Tolah, K. M. K., Hassan, A. M., El-Kafrawy, S. A., and Azhar, E. I. (2021). Evaluation of a pseudovirus neutralization assay for SARS-CoV-2 and correlation with live virus-based micro neutralization assay. Diagnostics (Basel) 11:994. doi: 10.3390/diagnostics11060994 Reed, N. G. (2010). The history of ultraviolet germicidal irradiation for air disinfection. Public Health Rep. 125, 15–27. REFERENCES A single dose of recombinant VSV-1G-spike vaccine provides protection against SARS-CoV-2 challenge. Nat. Commun. 11:6402. doi: 10. 1038/s41467-020-20228-7 relevant to SARS-CoV-2 infection. Oncotarget 11, 4201–4223. doi: 10.18632/ oncotarget.27799 Zhu, Z.-L., Qiu, X.-D., Wu, S., Liu, Y.-T., Zhao, T., Sun, Z.-H., et al. (2020). Blocking Eﬀect of demethylzeylasteral on the interaction between human ACE2 Protein and SARS-CoV-2 RBD protein discovered using SPR technology. Molecules 26:57. doi: 10.3390/molecules2601 0057 Yamate, M., Yamashita, M., Goto, T., Tsuji, S., Li, Y.-G., Warachit, J., et al. (2005). Establishment of Vero E6 cell clones persistently infected with severe acute respiratory syndrome coronavirus. Microbes Infect. 7, 1530–1540. doi: 10.1016/ j.micinf.2005.05.013 Zingler, K., and Littman, D. R. (1993). Truncation of the cytoplasmic domain of the simian immunodeﬁciency virus envelope glycoprotein increases env incorporation into particles and fusogenicity and infectivity. J. Virol. 67, 2824– 2831. doi: 10.1128/JVI.67.5.2824-2831.1993 Yuan, Y., Cao, D., Zhang, Y., Ma, J., Qi, J., Wang, Q., et al. (2017). Cryo- EM structures of MERS-CoV and SARS-CoV spike glycoproteins reveal the dynamic receptor binding domains. Nat. Commun. 8:15092. doi: 10.1038/ ncomms15092 Conﬂict of Interest: The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Závada, J. (1972). Pseudotypes of vesicular stomatitis virus with the coat of murine Leukaemia and of Avian Myeloblastosis viruses. J. Gen. Virol. 15, 183–191. doi: 10.1099/0022-1317-15-3-183 Zettl, F., Meister, T. L., Vollmer, T., Fischer, B., Steinmann, J., Krawczyk, A., et al. (2020). Rapid quantiﬁcation of SARS-CoV-2-neutralizing antibodies using propagation-defective Vesicular Stomatitis virus pseudotypes. Vaccines (Basel) 8:E386. doi: 10.3390/vaccines8030386 Publisher’s Note: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their aﬃliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Publisher’s Note: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their aﬃliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Zhang, H., Zhu, W., Jin, Q., Pan, F., Zhu, J., Liu, Y., et al. (2021). REFERENCES (1995). Eﬃcient recovery of infectious vesicular stomatitis virus entirely from cDNA clones. Proc. Natl. Acad. Sci. U.S.A. 92, 8388–8392. doi: 10.1073/pnas.92.18.8388 Schnitzer, T. J., Dickson, C., and Weiss, R. A. (1979). Morphological and biochemical characterization of viral particles produced by the tsO45 mutant of vesicular stomatitis virus at restrictive temperature. J. Virol. 29, 185–195. doi: 10.1128/JVI.29.1.185-195.1979 Whitt, M. A. (2010). Generation of VSV pseudotypes using recombinant 1G- VSV for studies on virus entry, identiﬁcation of entry inhibitors, and immune responses to vaccines. J. Virol. Methods 169, 365–374. doi: 10.1016/j.jviromet. 2010.08.006 Schnitzer, T. J., and Gonczol, E. (1979). Phenotypic mixing between murine oncoviruses and murine cytomegalovirus. J. Gen. Virol. 43, 691–695. doi: 10. 1099/0022-1317-43-3-691 Winter, C., Schwegmann-Wessels, C., Neumann, U., and Herrler, G. (2008). The spike protein of infectious bronchitis virus is retained intracellularly by a tyrosine motif. J. Virol. 82, 2765–2771. doi: 10.1128/JVI.02064-07 Wrapp, D., Wang, N., Corbett, K. S., Goldsmith, J. A., Hsieh, C.-L., Abiona, O., et al. (2020). Cryo-EM structure of the 2019-nCoV spike in the prefusion conformation. Science 367, 1260–1263. doi: 10.1126/science.abb2507 Shen, X., Tang, H., McDanal, C., Wagh, K., Fischer, W., Theiler, J., et al. (2021a). SARS-CoV-2 variant B.1.1.7 is susceptible to neutralizing antibodies elicited by ancestral spike vaccines. Cell Host Microbe 29, 529–539.e3. doi: 10.1016/j.chom. 2021.03.002 Wright, E., McNabb, S., Goddard, T., Horton, D. L., Lembo, T., Nel, L. H., et al. (2009). A robust lentiviral pseudotype neutralisation assay for in-ﬁeld serosurveillance of Rabies and Lyssaviruses in Africa. Vaccine 27, 7178–7186. doi: 10.1016/j.vaccine.2009.09.024 Shen, X., Tang, H., Pajon, R., Smith, G., Glenn, G. M., Shi, W., et al. (2021b). Neutralization of SARS-CoV-2 Variants B.1.429 and B.1.351. N. Engl. J. Med. 384, 2352–2354. doi: 10.1056/NEJMc2103740 Xiong, H.-L., Wu, Y.-T., Cao, J.-L., Yang, R., Liu, Y.-X., Ma, J., et al. (2020). Robust neutralization assay based on SARS-CoV-2 S-protein-bearing vesicular stomatitis virus (VSV) pseudovirus and ACE2-overexpressing BHK21 cells. Emerg. Microbes Infect. 9, 2105–2113. doi: 10.1080/22221751.2020.1815589 Shirato, K., Maejima, M., Matsuyama, S., Ujike, M., Miyazaki, A., Takeyama, N., et al. (2011). Mutation in the cytoplasmic retrieval signal of porcine epidemic diarrhea virus spike (S) protein is responsible for enhanced fusion activity. Virus Res. 161, 188–193. doi: 10.1016/j.virusres.2011.07.019 January 2022 \| Volume 12 \| Article 817200 Frontiers in Microbiology \| www.frontiersin.org 14 VSV1G-SARS-CoV-2 S Pseudoparticles and Their Uses Salazar-García et al. Yahalom-Ronen, Y., Tamir, H., Melamed, S., Politi, B., Shifman, O., Achdout, H., et al. (2020). January 2022 \| Volume 12 \| Article 817200 Frontiers in Microbiology \| www.frontiersin.org REFERENCES Inhalable nanocatchers for SARS-CoV-2 inhibition. Proc. Natl. Acad. Sci. U.S.A. 118:e2102957118. doi: 10.1073/pnas.2102957118 Copyright © 2022 Salazar-García, Acosta-Contreras, Rodríguez-Martínez, Cruz- Rangel, Flores-Alanis, Patiño-López and Luna-Pineda. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Zhang, L., Li, Q., Liu, Q., Huang, W., Nie, J., and Wang, Y. (2017). A bioluminescent imaging mouse model for Marburg virus based on a pseudovirus system. Hum. Vaccin. Immunother. 13, 1811–1817. doi: 10.1080/ 21645515.2017.1325050 Zhou, L., Huntington, K., Zhang, S., Carlsen, L., So, E.-Y., Parker, C., et al. (2020). MEK inhibitors reduce cellular expression of ACE2, pERK, pRb while stimulating NK-mediated cytotoxicity and attenuating inﬂammatory cytokines January 2022 \| Volume 12 \| Article 817200 Frontiers in Microbiology \| www.frontiersin.org 15
https://openalex.org/W4200519033	https://hal.univ-lorraine.fr/hal-03489461/document	English	null	Conditioning of poultry manure ash for subsequent phosphorous separation and assessment for a process design	Sustainable materials and technologies	2,022	cc-by	7,534	To cite this version: Rodolfomarin Rivera, Alexandre Chagnes, Michel Cathelineau, Marie-Christine Boiron. Conditioning of poultry manure ash for subsequent phosphorous separation and assessment for a process design. Sustainable Materials and Technologies, 2022, 31, pp.e00377. 10.1016/j.susmat.2021.e00377. hal- 03489461 Conditioning of poultry manure ash for subsequent phosphorous separation and assessment for a process design Rodolfomarin Rivera, Alexandre Chagnes, Michel Cathelineau, Marie-Christine Boiron Distributed under a Creative Commons Attribution 4.0 International License A R T I C L E I N F O Keywords: Ash Hydrometallurgy Leaching Phosphorus Sulfuric acid Keywords: Ash Hydrometallurgy Leaching Phosphorus Sulfuric acid Phosphorus is a strategic resource that cannot be substituted in the foods and pharmaceutical industries. It is an irreplaceable macronutrient for crops, cells, humans and animals. The supply risk can lead to geopolitical problems, and it is urgent to encourage the development of environmentally sustainable recovery technologies of phosphorus from various alternative materials. Poultry manure ash produced inside fluidized bed reactor could constitute an alternative resource of phosphorus to phosphate ores. Therefore, the phosphorus recovery in two representative ash samples has been investigated by combining leaching and solvent extraction under optimal conditions. The mineralogy of these samples has been studied in detail to get relevant information for designing the phosphorus extraction process. Although the relative abundances of phases vary between the two samples, phosphorous is mainly present as a hydroxyapatite mineral and in similar amounts in all particle fractions. Potassium, magnesium, sodium are located in other mineral phases (K-Na-Mg sulfates and K-(Na)-chlorides) mixed with the two main Ca-phases, calcite and hydroxyapatite. The poultry manure ashes were digested by dilute sulfuric acid at liquid-to-solid ratios ranging from L/S = 10 mL/g to L/S = 5 mL/g. About 90 wt% of phosphorous in the poultry manure ashes was leached under the optimal sulfuric acid concentration with a liquid-to-solid ratio of 5 mL/g. * Corresponding author. E-mail address: alexandre.chagnes@univ-lorraine.fr (A. Chagnes). Available online 10 December 2021 2214-9937/© 2021 Elsevier B.V. All rights reserved. https://doi.org/10.1016/j.susmat.2021.e00377 Received 2 June 2021; Received in revised form 15 November 2021; Accepted 7 December 2021 Conditioning of poultry manure ash for subsequent phosphorous separation and assessment for a process design RodolfoMarin Rivera, Alexandre Chagnes , Michel Cathelineau, Marie-Christine Boiron Universit´e de Lorraine, CNRS, GeoRessources, F- 54000 Nancy, France Corresponding author. E-mail address: alexandr HAL Id: hal-03489461 https://hal.univ-lorraine.fr/hal-03489461v1 Submitted on 17 Dec 2021 L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. Distributed under a Creative Commons Attribution 4.0 International License Sustainable Materials and Technologies 31 (2022) e00377 Contents lists available at ScienceDirect 1. Introduction In this context, the utilisation of ashes produced as by- products of P-rich residue incineration, such as wastewater treatment sludge, animal manure or animal by-products, biomass ash has been proposed as a sustainable alternative source of phosphorus [2,14,32]. These ashes can have a phosphorus content between 5 and 10 wt% (11–23 wt% P2O5), an order of magnitude less than that present in minerals, and potassium content, which is slightly lower than that pre sent in minerals found in phosphate rocks (~ 2 wt%). The recovery of phosphorus from these materials has the advantage of an eventual decrease in the consumption of natural resources and the environmental and economic impact related to phosphate rocks mining besides elimi nating pathogens and toxic organic substances [10,32]. The presence of impurities, such as Cu, Zn, Al, Si, Fe, Mg, can be significantly lower (< 1 wt%) compared to naturally occurring phosphate rocks [17,20,21,26]. However, certain materials can contain non-negligible amounts of heavy metals such as Pb or Cd, which must be managed to avoid disseminating in the environment by adding stabilizers [32]. Such features can facili tate phosphorus extraction and other subsequent purification processes. Notice that incineration of residues and the production of ashes rich in phosphorus is readily available, but they are often stockpiled and entail a significant management cost. Phosphorus and potassium should eventually be first separated from the solid matrix to enhance the val orisation of incineration ashes. Dilute sulfuric acid (H2SO4) and hydro chloric acid (HCl) are commonly used to leach phosphorus raw materials and produce phosphoric acid [3,8,15,18,31,39]. However, H2SO4 is particularly interesting as a leaching reagent for the production of phosphoric acid from raw material containing apatite as calcium released to the leaching solution is drastically reduced by calcium sul fate precipitation [31]. Meanwhile, phosphorus can be recovered from the leaching solution by precipitation of struvite phosphate mineral (NH4MgPO4⋅6H2O) or as hydroxyapatite (CaHPO4⋅2H2O) [7,8,20,39]. Although struvite crystallization is the most mature process for phos phorous recovery from aqueous streams [23], the production of H3PO4 from waste residues is still under development. This investigation aims to evaluate phosphorus effective leaching from poultry manure ashes produced in a fluidised bed furnace. Firstly, particle size distribution, chemical composition, mineralogy, and morphology were determined to assess the experimental conditions for leaching. Secondly, ash digestion with dilute H2SO4 was studied by considering different acid concentrations and liquid-to-solid ratios. 1. Introduction The optimal conditions for phosphorus recovery from poultry manure ashes were established (high dissolution yield of Phosphorus with the lowest possible acid concentration). and it is urgent to encourage the development of environmentally sus tainable recovery technologies of phosphorus from various alternative materials. In this context, the utilisation of ashes produced as by- products of P-rich residue incineration, such as wastewater treatment sludge, animal manure or animal by-products, biomass ash has been proposed as a sustainable alternative source of phosphorus [2,14,32]. These ashes can have a phosphorus content between 5 and 10 wt% (11–23 wt% P2O5), an order of magnitude less than that present in minerals, and potassium content, which is slightly lower than that pre sent in minerals found in phosphate rocks (~ 2 wt%). The recovery of phosphorus from these materials has the advantage of an eventual decrease in the consumption of natural resources and the environmental and economic impact related to phosphate rocks mining besides elimi nating pathogens and toxic organic substances [10,32]. The presence of impurities, such as Cu, Zn, Al, Si, Fe, Mg, can be significantly lower (< 1 wt%) compared to naturally occurring phosphate rocks [17,20,21,26]. However, certain materials can contain non-negligible amounts of heavy metals such as Pb or Cd, which must be managed to avoid disseminating in the environment by adding stabilizers [32]. Such features can facili tate phosphorus extraction and other subsequent purification processes. Notice that incineration of residues and the production of ashes rich in phosphorus is readily available, but they are often stockpiled and entail a significant management cost. Phosphorus and potassium should eventually be first separated from the solid matrix to enhance the val orisation of incineration ashes. Dilute sulfuric acid (H2SO4) and hydro chloric acid (HCl) are commonly used to leach phosphorus raw materials and produce phosphoric acid [3,8,15,18,31,39]. However, H2SO4 is particularly interesting as a leaching reagent for the production of phosphoric acid from raw material containing apatite as calcium released to the leaching solution is drastically reduced by calcium sul fate precipitation [31]. Meanwhile, phosphorus can be recovered from the leaching solution by precipitation of struvite phosphate mineral (NH4MgPO4⋅6H2O) or as hydroxyapatite (CaHPO4⋅2H2O) [7,8,20,39]. Although struvite crystallization is the most mature process for phos phorous recovery from aqueous streams [23], the production of H3PO4 from waste residues is still under development. 2. Combustion installation – Production of poultry manure ash in fluidized bed reactor 1. Introduction continues growing, an increasing need for phosphate fertilisers is ex pected in the upcoming years. However, as the phosphorus grade in phosphate ores continues decreasing, extraction and processing costs also increase [6,28]. Phosphorous (P) is a finite resource and an essential nutrient for living beings. It mainly occurs at mineable grades and tonnages in phosphate sedimentary rock at around 20–50 wt% P2O5 (9–22 wt% P), with Morocco and Western Sahara controlling 77% of the global reserves [4,33,38]. The geological reserves have estimated about 69 million tonnes worldwide, mainly as sedimentary phosphate deposits contain ing fluorapatite and detrital quartz, clays, and carbonate cements (calcite and dolomite). Igneous phosphate deposits constitute the second type of ore where apatite is the predominant phosphorus-phase, such as carbonatite-alkaline suites, or felsic “Kiruna-type” apatite-magnetite intrusions [19,33]. i Apatite and some rarer phosphates are the primary phosphate rocks used in the production of phosphoric acid (H3PO4) by the so-called wet process, i.e. chemical reaction of the phosphate rock with a mineral acid ([41,42,37,40]). H3PO4 is the essential chemical used in the fertilizer and food industry, but it has also demonstrated promising results for metal surface treatment [34,35]. However, the large quantity of impu rities and the gradual reduction of phosphorus concentration in phos phate rocks is detrimental for the economy of current H3PO4 production routes. Mining phosphate rocks production has increased significantly in the last decade, and in 2019, the world phosphate production reached 240,000 t. Phosphate rock is mainly used for the production of mineral fertilisers such as diammonium-phosphate (DAP) and triple superphos phate (TSP), which on average accounts for 74% of the global inorganic phosphate consumption [13]. As the worldwide demand for crops At present, Europe does not produce any raw phosphate ores, and it depends on imports of all raw phosphate and phosphate fertilisers from abroad. At the same time, modern agriculture continues relying on non- renewable resources. Consequently, the European Union (EU) has considered phosphate rocks as critical raw materials (EU Report COM/ 2014/0297). Phosphorus supply risk can lead to geopolitical problems, Sustainable Materials and Technologies 31 (2022) e00377 R. Rivera et al. and it is urgent to encourage the development of environmentally sus tainable recovery technologies of phosphorus from various alternative materials. 2. Combustion installation – Production of poultry manure ash in fluidized bed reactor Laying hens, manure and ashes were collected from Güres Power Plant (Turkey). Poultry manure was received from laying hens kept in cages where the mixture of excreta and urine, broken eggshells and feathers fell into shallow and finally conveyed out for collection. The manure taken from the poultry farm was about 70 wt% moisture con tent, but it was reduced to ca. 20 wt% by the Organic Rankine Cycle (ORC) [5]. Fig. 1 provides the flowsheet description of the poultry manure combustion system by the biomass power plant of Güres Group. Fig. 1 initially shows that at the fuel feeding system (point 1), chicken manure with 25–35 wt% moisture content was fed homoge neously in the fluidized bed gasifier. Chicken manure was gasified and burned with a minimum of 95% efficiency at 750 ◦C in a fluidized bed with partial air (point 2). The secondary air supply was used to achieve complete combustion, resulting in a combustion gas that remained at 850–950 ◦C for a minimum of 2.5 s (point 3). The ash-rich gas first passed through the radiation economizer and transferred heat to the thermal oil (point 4). Large particles were then caught under the radi ation economizer. The flue gas left at 330 ◦C, and the particles were trapped into the oil from the cassette type exchanger (point 5). Automatic cleaners were available on every cassette. The heat transfer of the metal surface temperature was kept below 350 ◦C. The fly ash in the gas from the oil economizer was held in the cyclone, and the cleaned combustion gas was sent to the air economizer (point 6). The air Fig. 1. Flowsheet description of poultry manure combustion system by the biomass power plant of Güres Group in Turkey [5]. (1): fuel feeding system, (2): fluidized bed, (3): combustion chamber, (4): thermal oil economizer, (5): cassette-type exchanger, (6): cyclone, (7) pre-heater, (8): wet scrubber, (9): ash silo, (10): Thermal oil, (11): Manure dryer system. Fig. 1. Flowsheet description of poultry manure combustion system by the biomass power plant of Güres Group in Turkey [5]. (1): fuel feeding system, (2): fluidized bed, (3): combustion chamber, (4): thermal oil economizer, (5): cassette-type exchanger, (6): cyclone, (7) pre-heater, (8): wet scrubber, (9): ash silo, (10): Thermal oil, (11): Manure dryer system. 2 Sustainable Materials and Technologies 31 (2022) e00377 R. Rivera et al. 2. Combustion installation – Production of poultry manure ash in fluidized bed reactor and sulphur contents were determined by infrared measurement after induction heating with iron, tungsten and tin accelerators on Horiba EMIA-320 V2. The content of organic carbon (Corg) was determined by hot etching with dilute hydrochloric acid. The measuring device was the same for caron and sulphur. X-ray powder diffraction (XRD, D8 Discover Bruker) and Scanning Electron Microscope (SEM) were used for mineralogical and chemical determinations. economizer pre-heated the primary and secondary fresh air to reduce waste gas losses to a minimum. The heated fresh air provided gasifica tion in the fluidized bed and stabilized the combustion reactions at the combustion side (point 7). The flue gas was passed through a wet scrubber, reducing the amount of dust below 50 mg/Nm3 (point 8). The ash from the fuel was collected from many different locations and recovered in the ash silos (point 9). The ash containing high potassium and phosphorus contents was regarded as an organo-mineral fertilizer raw material. The ORC system was used in 19% of the thermal oil energy for electricity generation, while 81% was used for closed cycle hot water in manure drying and poultry heating (point 10). The manure drying system used waste heat energy to dry the wet manure. It also ensures the necessary fuel for the co-generation system (point 11). Ash digestion with mineral acid was carried out with H2SO4 (95–97%, Sigma-Aldrich) at liquid-to-solid (L/S) ratios of 10 and 5 in centrifugal tubes (15 mL capacity) under constant agitation using a laboratory shaker (Gerhardt incubator shaker THO500/1) at 200 rpm for 24 h. After agitation, the tubes were centrifuged at 4000 rpm for 4 min to achieve efficient solid-liquid phase separation. The Pregnant Leach Solution (PLS) was filtered using a syringe filter (pore size of 0.45 μm) and diluted with 2 vol% nitric acid (HNO3) to determine the con centrations of Si, Al, Mg, Ca, Na, K, Fe, Mn and P using a Microwave Plasma-Atomic Emission Spectrometry (MP-AES, Agilent 4210). The pH measurements were performed on each PLS before diluting the sample using a pH-meter pHenomenal® pH 1100 L with a pH-electrode pHenomenal® 221. The corresponding dissolution yields of elements were calculated according to Eq. (1). % = Amount of element in the PLS Total amount of metal in the sample 3. Materials and methods Two types of poultry manure ash were studied: (1) FAECO obtained from the economiser fly ash (point 4 in Fig. 1), and (2) FACYC obtained from the underflow of the cyclone (point 6 in Fig. 1). Both samples were further dried at 105 ◦C for 24 h (moisture content 20–30 wt%) upon arrival in the lab. The concentrations of Mg, Ca, K, Na, P, Fe, Al, Si, and Mn were determined after total digestion by Inductively Couple Plasma Optical Emission Spectroscopy (ICP-OES). A Karl Fischer titration with endpoint detection potentiometry (with two impressed-current indica tor electrodes) was used to determine the bulk water content. The samples were calcined at 1000 ◦C in a tube furnace under a nitrogen atmosphere, and the released water vapour was transported into the titration vessel for analysis (fully automatic system). The total carbon % = Amount of element in the PLS Total amount of metal in the sample (1) 3 (a) (b) 0.1 1 10 100 1000 0 20 40 60 80 100 Particle size, µm Cumulated distribution, vol.% FACYC FAECO < 25 25-45 45-75 75-150 0 10 20 30 40 Particle size, µm Composition, wt.% Mg Ca Na K P FAECO < 25 25-45 45-75 75-150 0 10 20 30 40 Particle size, µm Composition, wt.% P Mg Ca Na K FACYC Fig. 2. (a) Particle size distribution (in vol%) and (b) distribution of major elements (P, Mg, Ca, Na and K) in the different fractions for FAECO and FACYC particles. (a) 0.1 1 10 100 1000 0 20 40 60 80 100 Particle size, µm Cumulated distribution, vol.% FACYC FAECO (a) (b) < 25 25-45 45-75 75-150 0 10 20 30 40 Particle size, µm Composition, wt.% Mg Ca Na K P FAECO < 25 25-45 45-75 75-150 0 10 20 30 40 Particle size, µm Composition, wt.% P Mg Ca Na K FACYC (b) < 25 25-45 45-75 75-150 0 10 20 30 40 Particle size, µm Composition, wt.% P Mg Ca Na K FACYC (b) Fig. 2. (a) Particle size distribution (in vol%) and (b) distribution of major elements (P, Mg, Ca, Na and K) in the different fractions for FAECO and FACYC particles. Fig. 2. (a) Particle size distribution (in vol%) and (b) distribution of major elements (P, Mg, Ca, Na and K) in the different frac 3 Fig. 2. 4.1. Characterization of poultry manure ashes Fig. 2a shows the particle size distribution of both two poultry manure ash samples. FAECO's sample (D90 = 44.7 μm, D50 = 17.5 μm, D10 = 6.4 μm) was composed of finer particles than FACYC (D90 = 157 μm, D50 = 68.2 μm, D10 = 24.3 μm). Yet, their chemical compositions are slightly different. The chemical composition is very homogeneous, whatever the considered size fraction (Fig. 2b). Ca is the major component in both two samples (26–36 wt%), followed by P (4–8 wt%) and K (6–11 wt%) (Table 1). The highest concentration of Ca was observed in FACYC, while the highest concentration of K was found in FAECO. Other elements, such as Si, Al, Fe and Mn, were found with concentrations ≤1 wt%. FAECO ash contains 5.6 ppm, 1.7 ppm, 15.9 ppm, 4.3 ppm and 0.9 ppm of As, Cd, Cr, Pb and U, respectively. Likewise, FACYC contains 1.2 ppm, 0.6 ppm, 13.4 ppm, 1.4 ppm and 0.5 ppm of As, Cd, Cr, Pb and U, respectively. Furthermore, Zn and Cu concentrations in FAECO are equal to 279 and 189 ppm, and their concentrations in FAECYC are equal to 2433 and 1113 ppm, respectively. Heavy metals present in the FAECO and FACIT ashes will be undoubtedly dissolved in the leach solution. Still, their concentration will be sufficiently low to avoid any the implementation of costly purification steps. These impurities will be transferred in res idues, but the concentration will be lower than those found in tailings from phosphate mining and processing. Fig. 4(b) describes the normalized mineralogical compositions of the two samples, which were determined based on the chemical composi tion depicted in Table 1, as well as XRD and SEM-EDX analyses. Hydroxyapatite is the main mineral phase in both samples (ca. 38 ± 2 wt%), followed by calcite (19–23 wt%). The calculated amount of portlandite is higher in FACYC than in FAECO due to the relatively high content of OH (expressed in the bulk H2O content in the Table) and Ca (see Table 1). Its content in FACYC decreases with the decrease of par ticle size. Such a trend could occur presumably due to the gradient temperature in the cyclone. The aphthitalite content ranges between 11 and 15 wt%. However, due to the high K-content, the FAECO sample contains more sylvite (12 ± 1 wt%) and arcanite (1.0 ± 0.6 wt%, especially in small particle sizes). 3. Materials and methods (a) Particle size distribution (in vol%) and (b) distribution of major elements (P, Mg, Ca, Na and K) in the different fractions for FAECO and FACYC particles. 3 R. Rivera et al. Sustainable Materials and Technologies 31 (2022) e00377 μm) and consist mainly of clusters of irregularly shaped particles together with some spherical grains (Fig. 3(ii)). Other elements were present as chlorides (K, Na), sulfates (K, Na, Mg), hydroxides (Ca) or oxides (Si). Particles have an irregular shape except rounded grains dominated by apatite (Fig. 3(i)-a and (ii)-c). Chloride compounds (mostly KCl) deposited on the edge of particles are euhedral (Fig. 3(i)-b). Rod-shaped silica particles are elongated and constitute isolated parti cles. K, Mg, Na, and traces of SO4- and Cl-based compounds were also found onto the particle surface in both samples as mixed phases of less than a micron in size (Fig. 3(i) and (ii)). Likewise, the deposition of K-, SO4- and Cl-based compounds were also found in some spots. These mineralogical phases were detected by XRD analysis (Fig. 3(ii)-a) and confirmed by Raman spectroscopy. The main compounds containing Ca are hydroxyapatite (Ca5(PO4)3(OH)), lime (CaO), calcite (CaCO3), and portlandite (Ca(OH)2). Furthermore, S content in both two samples is significant (see Table 1) and corresponds to potassium bearing sulfate (Fig. 3(i) and (ii)) as aphthitalite (K2.25Na1.75(SO4)2) and arcanite (K2SO4), which XRD did not detect. 4.1. Characterization of poultry manure ashes Therefore, this material is of great interest to produce phosphoric acid as it does not contain high amounts of heavy metals and transition metals, unlike phosphate rocks. It could also be a source of zinc as the concentration is around 1000–2000 ppm. However, the grade is much lower than the average grade of most Zn ores, which is about 8–10%, e.g. around 100 times higher than the measured concentrations. Therefore, the exploitation of such phosphate resources may make the purification/ extraction step of a hydrometallurgical process easier as heavy metal concentrations are low and phosphate content reaches 6%. 4.2. Ash digestion with a mineral acid Phosphorus extraction from ashes can be performed by ash digestion using appropriate reactive such as acids. For instance, Fahimi et al. [16] compared several technologies to recover phosphorus from sewage sludge ashes based on thermo-chemical treatment, wet chemical leaching, and the combination of wet chemical leaching and thermo- chemical treatment. These authors proposed a new approach to directly evaluate the sustainability of the technologies to recover phosphorus from the secondary waste streams towards the phosphorus extraction from phosphate rock utilizing the sulfuric acid-based wet process. It is generally observed that temperature, resins for cation/ anion exchange, or some leaching reagents play a crucial role in the sustainability of technology besides the origin of the phosphorus resource. In the case of phosphorus recovery from sewage sludge ashes, these authors demonstrated no general rule to design sustainable tech nologies for phosphorus extraction from ashes. Still, the origin, the na ture and the composition of the feed material are key factors of the environmental impact of a process besides the unit operations imple mented in the process. Nevertheless, in all cases, thermo-reductive based dry processes such as Recophos Inducarb and Thermphos appear to be less sustainable than the sulfuric acid-based wet process due to the en ergy need for inductively thermal heating [16]. Fig. 3 displays SEM-EDX images (Scanning Electron Microscopy- Energy Dispersive X-Ray Spectroscopy) of FAECO and FAECYC poultry manure ash samples. The FAECYC and FAECO are composed of particles sized up to 60 μm and a few 10 μm, respectively (images in Fig. 3(i)-a and (i)-b). Fig. 3 shows the presence of calcite (CaCO3) and hydroxyapatite (Ca5(PO4)3(OH)). In the case of FACYC (Fig. 3(i)), Ca-phosphate was precipitated as a rim covering calcite (Fig. 3 (i)-a). Small hydroxyapatite grains were also found in low particle size fractions (20–80 μm). In that case, particles have an irregular shape but tend to be more rounded when particles size is lower than 100 μm (Fig. 3(i)-c). FAECO's particles are thinner (5 to 30 Table 1 Table 1 Concentration of the major elements in the poultry manure ashes (SD: Standard Deviation). Element FAECO FACYC wt% SD wt% SD Si 0.7 ± 0.1 1.0 ± 0.6 Al 0.1 ± 0.0 0.1 ± 0.02 Mg 3.3 ± 0.1 2.7 ± 0.4 Ca 23.1 ± 0.9 28.3 ± 1.2 Na 1.6 ± 0.04 1.3 ± 0.1 K 9.0 ± 0.6 4.8 ± 0.8 Fe 0.4 ± 0.3 0.2 ± 0.01 Mn 0.1 ± 0.0 0.1 ± 0.02 P 6.3 ± 0.3 6.2 ± 0.2 CO2 34.9 ± 8.7 54.1 ± 10.9 H2O 9.6 ± 2.4 22.4 ± 4.5 FeO 0.8 ± 0.2 0.4 ± 0.1 S 12.1 ± 3.0 8.2 ± 1.6 Corg 0.2 ± 0.1 0.2 ± 0.04 Concentration of the major elements in the poultry manure ashes (SD: Standard Deviation). Therefore, in the present work, the dissolution of elements from FACYC and FAECO poultry manure ashes was evaluated after ash digestion with dilute H2SO4 as a leaching reagent. The sludge pH ob tained at two different L/S ratios and acid concentration is shown in Fig. 4(a). Both samples resulted in high alkalinity (initial pH 11.7–12.2), but while increasing the acid concentration, FACYC exhibited slightly higher pH-values than FAECO (Fig. 5(a)). The release of hydroxyl anions from hydroxyapatite and portlandite buffers the slurry's pH, particularly in FACYC. Also, different changes in L/S-ratios demonstrated a clear difference in the variation at pH during leaching (see Fig. 5(a)). L/S- ratios <10 mL/g allow increasing the consumption of solid material to 4 Sustainable Materials and Technologies 31 (2022) e00377 R. Rivera et al. Fig. 3. SEM images in backscattered electron (BSE) mode of (i) FAECYC and (ii) FAECO poultry manure ash sample. a: General view of particles, b-d: zoom-in particles showing the different morphologies (Ap: apatite, Cal: calcite). Fig. 3. SEM images in backscattered electron (BSE) mode of (i) FAECYC and (ii) FAECO poultry manure ash sample. a: General view of particles, b-d: zoom-in particles showing the different morphologies (Ap: apatite, Cal: calcite). (b) 0 10 20 30 40 50 Calcite Hydroxyapatite Portlandite Lime Aphthitalite Arcanite Sylvite Periclase Silica glass Distribution, wt.% < 25 25-45 45-75 75-150 FAECO (a) (b) 0 10 20 30 40 50 60 70 0 1000 2000 3000 Position 2 (Copper) a.u. Table 1 FAECO H H H H S S Pe Pe C S P 0 10 20 30 40 50 60 70 0 1000 2000 3000 Position 2 (Copper) a.u. FACYC C H H L L P P Pe Pe L C S S: Sylvite (KCl) P: Portlandite (Ca(OH)2) H: Hydroxyapatite (Ca5(OH)(PO4)3) C: Calcite (CaCO3) Pe: Periclase (MgO) L: Lime (CaO) 0 10 20 30 40 50 Calcite Hydroxyapatite Portlandite Lime Aphthitalite Arcanite Sylvite Periclase Silica glass Distribution, wt.% < 25 25-45 45-75 75-150 FAECO 0 10 20 30 40 50 Calcite Hydroxyapatite Portlandite Lime Aphthitalite Arcanite Sylvite Periclase Silica glass Distribution, wt.% < 25 25-45 45-75 75-150 FACYC Fig. 4. (a) XRD pattern (a.u.: arbitrary unit) and (b) Normalized mineralogical compositions (to 100 wt%) of FAECO-FACYC poultry manure ashes. (a) (b) 0 10 20 30 40 50 60 70 0 1000 2000 3000 a.u. FAECO H H H H S S Pe Pe C S P S: Sylvite (KCl) P: Portlandite (Ca(OH)2) H: Hydroxyapatite (Ca5(OH)(PO4)3) C: Calcite (CaCO3) Pe: Periclase (MgO) L: Lime (CaO) 0 10 20 30 40 50 Calcite Hydroxyapatite Portlandite Lime Aphthitalite Arcanite Sylvite Periclase Silica glass < 25 25-45 45-75 75-150 FAECO (a) 0 10 20 30 40 50 60 70 0 1000 2000 3000 Position 2 (Copper) a.u. FAECO H H H H S S Pe Pe C S P 0 10 20 30 40 50 60 70 0 1000 2000 3000 Position 2 (Copper) a.u. FACYC C H H L L P P Pe Pe L C S S: Sylvite (KCl) P: Portlandite (Ca(OH)2) H: Hydroxyapatite (Ca5(OH)(PO4)3) C: Calcite (CaCO3) Pe: Periclase (MgO) L: Lime (CaO) (b) (a) 0 10 20 30 40 50 Calcite Hydroxyapatite Portlandite Lime Aphthitalite Arcanite Sylvite Periclase Silica glass Distribution, wt.% < 25 25-45 45-75 75-150 FACYC .u.: arbitrary unit) and (b) Normalized mineralogical compositions (to 100 wt%) of FAECO-FACYC poultry manure ashes. g. 4. (a) XRD pattern (a.u.: arbitrary unit) and (b) Normalized mineralogical compositions (to 100 wt%) of FAECO-FACYC phosphorus concentration in the PLS was multiplied by two when the leaching was performed at L/S = 5 mL/g instead of L/S = 10 mL/g when using an acid concentration of 150 g/L for FAECO 175 g/L for FACYC. enhance the enrichment of P in solution. However, low L/S ratios can limit mass transfers due to an increase in pulp density. Table 1 Furthermore, low L/S-ratios can also enhance the co-dissolution of other impurities. An L/ S = 5 mL/g is commonly suggested in industrial applications for an optimal solid and liquid streaming handle. By reducing the L/S-ratio from 10 ml/g to 5 mL/g, phosphorus was enriched in the solution as the ash consumption was doubled. The co- dissolution of Na, Mg and K phases tended to occur earlier than hydroxy-apatite (P) not only because they are readily soluble in acidic media but also because they are the first to react with the acid as they are deposited onto the surface of phosphorus-rich particles (see Fig. 3(i) and (ii)). Fig. 5(b) and (c) show the effect of acid concentration on the disso lution rates at L/S-ratios of 5 mL/g and 10 mL/g . The dissolution of hydroxyapatite, the only phase to release phosphorus, increased with increasing acid concentration. However, the highest phosphorus disso lution, i.e. the maximum phosphorus concentration in the solution, [P]max, was reached at a defined acid concentration depending on the L/ S-ratio. Thus, with an L/S = 10 mL/g, the [P]max for both samples (5–6 g/L) were achieved with an acid concentration of 100 g/L. Meanwhile, The highest concentration of potassium in PLS (9–17 g/L) is obtained from FAECO digestion as potassium content in FAECO is higher than in FACYC, yielding a higher concentration in solution. Most of the Al, Fe 5 5 R. Rivera et al. Table 1 Sustainable Materials and Technologies 31 (2022) e00377 (a) (b) 0 50 100 150 200 250 300 0 2 4 6 8 10 12 14 Concentration of H2SO4, g/L pH FACYC FAECO L/S = 5 0 50 100 150 200 250 300 0 2 4 6 8 10 12 14 Concentration of H2SO4, g/L pH FACYC FAECO L/S = 10 0 25 50 75 100 125 150 175 200 225 250 275 0 20 40 60 80 100 Concentration of H2SO4, g/L Dissolution, wt.% Mg Ca Fe Mn K Si Al P 0 25 50 75 100 125 150 175 200 225 250 275 0 5 10 15 20 Concentration of H2SO4, g/L Concentration, g/L Na Si Mg P K 0 25 50 75 100 125 150 175 200 225 250 275 0 20 40 60 80 100 Concentration of H2SO4, g/L Dissolution, wt.% Mg Ca Fe Mn K Si Al P 0 25 50 75 100 125 150 175 200 225 250 275 0 5 10 15 20 Concentration of H2SO4, g/L Concentration, g/L Na Si Mg P K FACYC FAECO Fig. 5. Chemical dissolution of FACYC and FAECO poultry manure ash: (a) Effect of H2SO4 concentration on the pH in the chemical dissolution at two different liquid-to-solid (L/S) ratios; Effect of H2SO4 concentrations at (b) L/S = 5 mL/g and (c) L/S= 10 mL/g (c) (T = 25 ◦C, t = 24 h, 200 rpm). Na dissolution resulted in 100 wt% in the whole range of acid concentration. Concentrations of Ca, Al, Fe and Mn < 1 g/L are not shown to improve readability. Table 1 (a) 0 50 100 150 200 250 300 0 2 4 6 8 10 12 14 Concentration of H2SO4, g/L pH FACYC FAECO L/S = 10 0 50 100 150 200 250 300 0 2 4 6 8 10 12 14 Concentration of H2SO4, g/L pH FACYC FAECO L/S = 5 (a) Concentration of H2SO4, g/L 0 25 50 75 100 125 150 175 200 225 250 275 0 20 40 60 80 100 Concentration of H SO g/L Dissolution, wt.% Mg Ca Fe Mn K Si Al P FACYC (b) 0 25 50 75 100 125 150 175 200 225 250 275 0 20 40 60 80 100 Concentration of H2SO4, g/L Dissolution, wt.% Mg Ca Fe Mn K Si Al P FAECO (b) Concentration of H2SO4, g/L Concentration of H2SO4, g/L 2 4, g 0 25 50 75 100 125 150 175 200 225 250 275 0 5 10 15 20 Concentration of H2SO4, g/L Concentration, g/L Na Si Mg P K 0 25 50 75 100 125 150 175 200 225 250 275 0 5 10 15 20 Concentration of H2SO4, g/L Concentration, g/L Na Si Mg P K Concentration of H2SO4, g/L Concentration of H2SO4, g/L Concentration of H2SO4, g/L Concentration of H2SO4, g/L Fig. 5. Chemical dissolution of FACYC and FAECO poultry manure ash: (a) Effect of H2SO4 concentration on the pH in the chemical dissolution at two different liquid-to-solid (L/S) ratios; Effect of H2SO4 concentrations at (b) L/S = 5 mL/g and (c) L/S= 10 mL/g (c) (T = 25 ◦C, t = 24 h, 200 rpm). Na dissolution resulted in 100 wt% in the whole range of acid concentration. Concentrations of Ca, Al, Fe and Mn < 1 g/L are not shown to improve readability. CaCO3 + H2SO4 + H2O→CaSO4∙H2O + H2CO3 (8) CaO + H2SO4 + H2O→CaSO4∙2H2O (9) 5 Conclusions and Mn phases were co-dissolved, but their concentrations were lower than 1 g/L. Meanwhile, more than 40 wt% of Si was co-dissolved at a relatively high acid concentration (>75 g/L H2SO4), but its solution concentration did not surpass 5 g/L. Presumably, the dissolution of silica-rich particles (Si) is related to the dissolution of Na rich glass. (8) (9) 5. Conclusions The chemical decomposition of hydroxyapatite in acidic media may occur in several steps through the intermediate formation of Ca3(PO4)2 and CaHPO4 (reactions (4)–(6)) [11,12]. Further reaction of these compounds with H2SO4 led to CaSO4 and H3PO4, as described by re actions (6)–(7). Likewise, other Ca minerals, such as calcite and lime, also tend to form CaSO4.2H2O (reactions (8) and (9)), which exhibits low solubility and, consequently, precipitates [31]. Despite relatively high Ca content in samples, the Ca concentration in the PLS remains low due to gypsum precipitation (CaSO4.2H2O). The leaching of poultry manure ashes produced in a fluidised bed furnace was investigated to extract phosphorus. FAECO's ash particles were formed at higher pressure and temperature than FACYC's ash particles, which led to the production of ashes with different particle sizes. Hydroxyapatite is the main mineral phase in both two samples and the unique phosphorus bearer. Ca is present as calcite, portlandite and hydroxyapatite. These phases are mixed with alkali metals sulfates (K, Na (Mg): arcanite-aphthitalite series), chlorides (K (Na): predominant sylvite), and minor periclase. Elemental analyses showed the presence of high content of zinc and relatively high content of copper besides the presence of heavy metals such as Pb, Cr, U, as well as As. Nevertheless, the concentration of these metals remains very low compared to typical concentrations in phosphate rocks. As a conclusion, poultry manure ash as a phosphorus source is of great interest from an economic and envi ronmental viewpoint since no complex purification operation will be Ca5(PO4)3OH + H2SO4 + H2O→Ca5(PO4)3⋅(H2O)+ + H3O+ + SO2− 4 (4) Ca5(PO4)3OH + H2SO4 + H2O→Ca5(PO4)3⋅(H2O)+ + H3O+ + SO2− 4 (4) (4) Ca5(PO4)3OH + H2SO4 + H2O→Ca5(PO4)3⋅(H2O)+ + H3O+ + SO2− 4 (4) 2Ca5(PO4)3⋅(H2O)+→3Ca3(PO4)2 + Ca2+ + 2H2O (5) Ca3(PO4)2 + H2SO4→CaSO4 + 2CaHPO4 (6) CaHPO4 + H2SO4→CaSO4 + H3PO4 (7) Ca5(PO4)3OH + H2SO4 + H2O→Ca5(PO4)3⋅(H2O)+ + H3O+ + SO2− 4 (4) 2Ca5(PO4)3⋅(H2O)+→3Ca3(PO4)2 + Ca2+ + 2H2O (5) Ca3(PO4)2 + H2SO4→CaSO4 + 2CaHPO4 (6) CaHPO4 + H2SO4→CaSO4 + H3PO4 (7) Ca5(PO4)3OH + H2SO4 + H2O→Ca5(PO4)3⋅(H2O)+ + H3O+ + SO2− 4 (4) 2Ca5(PO4)3⋅(H2O)+→3Ca3(PO4)2 + Ca2+ + 2H2O (5) Ca3(PO4)2 + H2SO4→CaSO4 + 2CaHPO4 (6) CaHPO4 + H2SO4→CaSO4 + H3PO4 (7) (5) (6) Ca3(PO4)2 + H2SO4→CaSO4 + 2CaHPO4 (7) CaHPO4 + H2SO4→CaSO4 + H3PO4 6 R. Rivera et al. [2] A. Amann, O. Zoboli, J. Krampe, H. Rechberger, M. Zessner, L. Egle, Environmental impacts of phosphorus recovery from municipal wastewater, Resour. Conserv. Recycl. 130 (2018) 127–139, https://doi.org/10.1016/j.resconrec.2017.11.002. Author statement Funding acquisition: M.C; Supervision: A.C.; Methodology: R.M.R, A. C. and M.C.; Investigation: R.M.R; Formal analysis: R.M.R, M.C., M.C.B.; Writing-original draft: R.M.R., A.C.; Writing-review & editing: R.M.R, A. C., M.C., M.C.B [8] M. Darwish, A. Aris, M.H. Puteh, M.N.H. Jusoh, A. Abdul Kadir, Waste bones ash as an alternative source of P for struvite precipitation, J. Environ. Manag. 203 (2017) 861–866, https://doi.org/10.1016/j.jenvman.2016.02.033. [10] S. Donatello, C.R. Cheeseman, Recycling and recovery routes for incinerated sewage sludge ash (ISSA): a review, Waste Manag. 33 (2013) 2328e2340, https:// doi.org/10.1016/j.wasman.2013.05.024. [11] S.V. Dorozhkin, Inorganic chemistry of the dissolution phenomenon: the dissolution mechanism of calcium apatites at the atomic (ionic) level, Comments Inorg. Chem. 20 (1999) 285–299, https://doi.org/10.1080/ 02603599908021447. 5. Conclusions During ash digestion, as expected, Na and K were entirely dissolved. Meanwhile, the dissolution of phosphorus increased with increasing acid concentration. Thus, the phosphorus concentration in solution at an L/S = 5 mL/g was twice the P-concentration obtained at L/S = 10 mL/g when using a slightly high acid concentration (150–175 g/L H2SO4). The high content of K in FAECO yielded a higher concentration in solution (9–17 g/L), particularly at L/S = 5 mL/g. Fe and Mn also tended to be co- dissolved, but their concentrations in solution were lower than 1 g/L. [4] M.A. De Boer, L. Wolzak, J.C. Slootweg, Phosphorus : Reserves, production, and applications, in: H. Ohtake, S. Tsuneda (Eds.), Phosphorous Recovery and Recycling, Springer Nature Singapore Pte Ltd., Singapore, 2019, pp. 75–100. [ ] C b i G Gü l k O il i l f l i i h b bbli [5] E. Cebeci, G. Gürlek, T.O. Yilmaz, Disposal of laying manure with bubbling fluidized bed reactor in combined heat and power system, in: I. Dincer, C. O. Colpan, M.A. Ezan, A. Tokuç (Eds.), 7th Global Conference on Global Warming, Izmir, 2018. [6] D. Cordell, J.O. Drangert, S. White, The story of phosphorus: global food security and food for thought, Glob. Environ. Chang. 19 (2009) 292–305, https://doi.org/ 10.1016/j.gloenvcha.2008.10.009. [7] J. Dai, W.T. Tang, Y.S. Zheng, H.R. Mackey, H.K. Chui, M.C.M. van Loosdrecht, G. H. Chen, An exploratory study on seawater-catalysed urine phosphorus recovery (SUPR), Water Res. 66 (2014) 75–84, https://doi.org/10.1016/j. watres.2014.08.008. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. [12] S.V. Dorozhkin, Dissolution mechanism of calcium apatites in acids: a review of literature, World J. Methodol. 2 (2012) 1, https://doi.org/10.5662/wjm.v2.i1.1. [13] R.J. van Enk, L.K. Acera, R.D. Schuiling, P. Ehlert, J.G. de Wilt, R.J.F. van Haren, The Phosphate Balance : Current Developments and Future Outlook, InnovationNetwork, Utrecht, 2011. Acknowledgements [14] A. Fahimi, F. Bilo, A. Assi, R. Dalipi, S. Federici, A. Guedes, B. Valentim, H. Olgun, G. Ye, B. Bialecka, L. Fiameni, L. Borgese, M. Cathelineau, M.C. Boiron, G. Predeanu, E. Bontempi, Poultry litter ash characterisation and recovery, Waste Manag. 111 (2020) 10–21, https://doi.org/10.1016/j.wasman.2020.05.010. This work was carried out under the framework of the ERAMIN project DEASPHOR (Design of a product of Substitution of phosphate rocks) funded by ADEME (AGDI_161873_01_111). The Authors thank the Güres Group, especially Dr. Hayati Olgum, for providing the poultry manure ashes. Dr. Bruno Valentim sent us sieved material prepared in collaboration with Ario Fahimi, who has undertaken a preliminary characterization of the ashes. [15] L. Fang, J. Shan Li, M.Z. Guo, C.R. Cheeseman, D.C.W. Tsang, S. Donatello, C. S. Poon, Phosphorus recovery and leaching of trace elements from incinerated sewage sludge ash (ISSA), Chemosphere 193 (2018) 278–287, https://doi.org/ 10.1016/j.chemosphere.2017.11.023. [16] A. Fahimi, S. Federici, L.E. Depero, B. Valentim, I. Vassura, F. Ceruti, L. Cutaia, E. Bontempi, Evaluation of the sustainability of technologies to recover phosphorus from sewage sludge ash based on embodied energy and CO2 footprint, J. Clean. Prod. 289 (2021) (2021), 125762, https://doi.org/10.1016/j. jclepro 2020 125762 5. Conclusions Sustainable Materials and Technologies 31 (2022) e00377 g ( ) 0 25 50 75 100 125 150 175 200 225 250 275 0 20 40 60 80 100 Concentration of H2SO4, g/L Dissolution, wt.% Ca Fe Mn Si Al P K Mg 0 25 50 75 100 125 150 175 200 225 250 275 0 5 10 15 20 Concentration of H2SO4, g/L Concentration, g/L Mg P K Na Si 0 25 50 75 100 125 150 175 200 225 250 275 0 20 40 60 80 100 Concentration of H2SO4, g/L Dissolution, wt.% Mg Ca Fe Mn P K Si Al 0 25 50 75 100 125 150 175 200 225 250 275 0 5 10 15 20 Concentration of H2SO4, g/L Concentration, g/L Mg P K Na SI FACYC FAECO (c) Fig. 5. (continued). 0 25 50 75 100 125 150 175 200 225 250 275 0 20 40 60 80 100 Concentration of H2SO4, g/L Dissolution, wt.% Ca Fe Mn Si Al P K Mg FACYC ( 0 25 50 75 100 125 150 175 200 225 250 275 0 20 40 60 80 100 Concentration of H2SO4, g/L Dissolution, wt.% Ca Fe Mn Si Al P K Mg 0 25 50 75 100 125 150 175 200 225 250 275 0 5 10 15 20 Concentration of H2SO4, g/L Concentration, g/L Mg P K Na Si FACYC ( 0 25 50 75 100 125 150 175 200 225 250 275 0 20 40 60 80 100 Concentration of H2SO4, g/L Dissolution, wt.% Mg Ca Fe Mn P K Si Al FAECO (c) 0 25 50 75 100 125 150 175 200 225 250 275 0 20 40 60 80 100 Concentration of H2SO4, g/L Dissolution, wt.% Mg Ca Fe Mn P K Si Al 0 25 50 75 100 125 150 175 200 225 250 275 0 5 10 15 20 Concentration of H2SO4, g/L Concentration, g/L Mg P K Na SI FAECO c) (c) Concentration of H2SO4, g/L Concentration of H2SO4, g/L Concentration of H2SO4, g/L Fig. 5. (continued). Fig. 5. (continued). requested after ash digestion and waste treatment will be facilitated. [3] M. Atienza-Martínez, G. Gea, J. Arauzo, S.R.A. Kersten, A.M.J. Kootstra, Phosphorus recovery from sewage sludge char ash, Biomass Bioenergy 65 (2014) 42–50, https://doi.org/10.1016/j.biombioe.2014.03.058. requested after ash digestion and waste treatment will be facilitated. References [17] S.M. Heilmann, J.S. Molde, J.G. Timler, B.M. Wood, A.L. Mikula, G.V. Vozhdayev, E.C. Colosky, K.A. Spokas, K.J. Valentas, Phosphorus reclamation through hydrothermal carbonization of animal manures, Environ. Sci. Technol. 48 (2014) 10323–10329, https://doi.org/10.1021/es501872k. 7 R. Rivera et al. Sustainable Materials and Technologies 31 (2022) e00377 [18] K.J. Hong, N. Tarutani, Y. Shinya, T. Kajiuchi, Study on the recovery of phosphorus from waste-activated sludge incinerator ash, J. Environ. Sci. Heal. Part A 40 (2005) 617–631, https://doi.org/10.1081/ESE-200046614. [33] P. Pt´aˇcek, Phosphate rocks, in: Apatites and their Synthetic Analogues - Synthesis, Structure, Properties and Applications, 2016, pp. 335–381, https://doi.org/ 10.5772/62214. p g [19] E. Jonsson, V.R. Troll, K. H¨ogdahl, C. Harris, F. Weis, K.P. Nilsson, A. Skelton, Magmatic origin of giant “Kiruna-type” apatite-iron-oxide ores in Central Sweden, Sci. Rep. 3 (2013) 1–8, https://doi.org/10.1038/srep01644. [34] R. S´anchez-Tovar, M.T. Monta˜n´es, J. García-Ant´on, A. Guenbour, Galvanic corrosion of the base AISI 316l/micro-plasma arc welded AISI 316l in polluted phosphoric acid media at different temperatures, Int. J. Electrochem. Sci. 6 (2011) 5550–5564. [20] K. Kaikake, T. Sekito, Y. Dote, Phosphate recovery from phosphorus-rich solution obtained from chicken manure incineration ash, Waste Manag. 29 (2009) 1084–1088, https://doi.org/10.1016/j.wasman.2008.09.008. [35] I. Santana, A. Pepe, E. Jimenez-Pique, S. Pellice, S. Cer´e, Silica-based hybrid coatings for corrosion protection of carbon steel. Part I: effect of pretreatment with phosphoric acid, Surf. Coat. Technol. 236 (2013) 476–484, https://doi.org/ 10.1016/j.surfcoat.2012.07.086. p g j [21] T. Komiyama, A. Kobayashi, M. Yahagi, The chemical characteristics of ashes from cattle, swine and poultry manure, J. Mater. Cycles Waste Manag. 15 (2013) 106–110, https://doi.org/10.1007/s10163-012-0089-2. [37] M. Takhim, Process for the Production of High Purity Phosphoric Acid. US 2011/ 0200509 A1, 2008. [23] B. Li, I.A. Udugama, S.S. Mansouri, W. Yu, S. Baroutian, K.V. Gernaey, B.R. Young, An exploration of barriers for commercializing phosphorus recovery technologies, J. Clean. Prod. 229 (2019) 1342–1354, https://doi.org/10.1016/j. jclepro.2019.05.042. [38] U.S. Geological Survey, Mineral Commodity Summaries 2020, United States Geological Survey, Virginia, 2020. [39] H. Xu, P. He, G. Wang, L. Shao, Phosphorus recovery as struvite from sewage sludge ash, J. Environ. Sci. (China) 24 (2012) 1533–1538, https://doi.org/ 10.1016/S1001-0742(11)60969-8. i [26] L. Luyckx, G.H.J. de Leeuw, J. Van Caneghem, Characterization of poultry litter ash in view of its valorization, Waste Biomass Valorization. (2019), https://doi. org/10.1007/s12649-019-00750-6. i [40] C. Ye, J. Li, Wet process phosphoric acid purification by solvent extraction using N- octanol and tributylphosphate mixtures, J. Chem. Technol. Biotechnol. 88 (2013) 1715–1720, https://doi.org/10.1002/jctb.4023. References i g [28] M.C. Mew, G. Steiner, B. Geissler, Phosphorus supply chain-scientific, technical, and economic foundations: a transdisciplinary orientation, Sustain. 10 (2018), https://doi.org/10.3390/su10041087. [41] A. Hannachi, D. Habaili, C. Chtara, A. Ratel, Purification of wet process phosphoric acid by solvent extraction with TBP and MIBK mixtures, Sep. Purif. Technol. 55 (2007) 212–216, https://doi.org/10.1016/j.seppur.2006.12.014. p g [31] L.M. Ottosen, G.M. Kirkelund, P.E. Jensen, Extracting phosphorous from incinerated sewage sludge ash rich in iron or aluminum, Chemosphere 91 (2013) 963–969, https://doi.org/10.1016/j.chemosphere.2013.01.101. [42] S. Khorfan, O. Shino, A. Wahoud, Extraction of H3PO4 from wet phosphoric acid by nC4 – nC7 alcohols, Period. Polytech. Ser. Chem. Eng. 45 (2001) 139–148. [32] M. Pasquali, A. Zanoletti, L. Benassi, S. Federici, L.E. Depero, E. Bontempi, Stabilized biomass ash as a sustainable substitute for commercial P-fertilizers, Land Degrad. Dev. 29 (7) (2018) 2199–2207, https://doi.org/10.1002/ldr.2915.
https://openalex.org/W4387460873	https://bmcmedinformdecismak.biomedcentral.com/counter/pdf/10.1186/s12911-023-02311-3	English	null	Associations between blood pressure control and clinical events suggestive of nutrition care documented in electronic health records of patients with hypertension	BMC medical informatics and decision making	2,023	cc-by	7,752	Abstract Background Clinical events suggestive of nutrition care found in electronic health records (EHRs) are rarely explored for their associations with hypertension outcomes. Methods Longitudinal analysis using structured EHR data from primary care visits at a health system in the US from December 2017—December 2020 of adult patients with hypertension (n = 4,237) tested for associations between last visit blood pressure (BP) control (≤ 140 Systolic BP and ≤ 90 Diastolic BP) and ≥ 1 nutrition care clinical event operationalized as (overweight or obesity (BMI > 25 or 30, respectively) diagnoses, preventive care visits, or pro- vision of patient education materials (PEM)). Descriptive statistics and longitudinal targeted maximum likelihood estimation (LTMLE) models were conducted to explore average treatment effects (ATE) of timing and dose response from these clinical events on blood pressure control overall and by race. Results The median age was 62 years, 29% were male, 52% were Black, 25% were from rural areas and 50% had controlled BP at baseline. Annual documentation of overweight/obesity diagnoses ranged 3.0–7.8%, preventive care visits ranged 6.2–15.7%, and PEM with dietary and hypertension content were distributed to 8.5–28.8% patients. LTMLE models stratified by race showed differences in timing, dose, and type of nutrition care. Black patients who had nutrition care in Year 3 only compared to none had lower odds for BP control (ATE -0.23, 95% CI: -0.38,-0.08, p = 0.003), preventive visits in the last 2 years high higher odds for BP control (ATE 0.31, 95% CI: 0.07,0.54, p = 0.01), and early or late PEMs had lower odds for BP control (ATE -0.08, 95% CI: -0.15,-0.01, p = 0.03 and ATE -0.23, 95% CI: -0.41,-0.05, p = 0.01, respectively). Conclusions In this study, clinical events suggestive of nutrition care are significantly associated with BP control, but are infrequent and effects differ by type, timing, and patient race. Preventive visits appear to have the most effect; additional research should include examining clinical notes for evidence of nutrition care among different popula- tions, which may uncover areas for improving nutrition care for patients with chronic disease. Keywords Chronic disease, Hypertension, Preventive care, Electronic health records, Nutrition April R. Williams1* , Maria D. Thomson1 and Erin L. Britton1 April R. Williams1* , Maria D. Thomson1 and Erin L. Britton1 This is a U.S. Government work and not under copyright protection in the US; foreign copyright protection may apply 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permit- ted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecom- mons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/ zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Background *Correspondence: April R. Williams April.Williams.PhD@gmail.com 1 Health Behavior and Policy, Virginia Commonwealth University, 830 E Main Street, Richmond, VA 23219, USA High blood pressure (BP), or hypertension, affects nearly 1 in 3 US adults [1], with approximately half of cases under poor control, increasing the risk of death from a sudden heart attack or stroke [2]. A variety of clinical High blood pressure (BP), or hypertension, affects nearly 1 in 3 US adults [1], with approximately half of cases under poor control, increasing the risk of death from a sudden heart attack or stroke [2]. A variety of clinical BMC Medical Informatics and Decision Making BMC Medical Informatics and Decision Making Williams et al. BMC Medical Informatics and Decision Making (2023) 23:208 https://doi.org/10.1186/s12911-023-02311-3 (2023) 23:208 Study design and setting A single, safety-net academic medical center with six primary care clinic sites in the Mid-Atlantic US serves a diverse patient population of both rural and urban, and underrepresented patients. Deidentified EHR data were obtained for clinical visits (n = 43,246) between December 2017 and December 2020 for (n= 4,237) adult patients, age 18 to 85 years at the date of their last BP measurement, who had their first visit prior to 1/1/2019 with a recorded BP measurement, at least two primary care visits during the study period and had a previ- ous diagnosis of hypertension. Records of patients with chronic kidney disease, pregnancy, or who were in hos- pice or long-term care were excluded per the clinical quality measure for controlling high BP [16]. Patient data included demographics: age, sex, racial/ethnicity cat- egories, rural geography indicated by Rural Urban Con- tinuum codes 4–9 [17], and insured status; and clinical This is a U.S. Government work and not under copyright protection in the US; foreign copyright protection may apply 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permit- ted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecom- mons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/ zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Williams et al. BMC Medical Informatics and Decision Making (2023) 23:208 Page 2 of 9 factors relevant to hypertension: BMI, comorbidities [18], hypertension medications (see Supplemental Materials) [19], number of visits, and indicators for clinical events suggestive of nutrition care and BP control [16]. events suggestive of nutrition care that occur in primary care settings may have a positive effect on patients’ die- tary patterns, and therefore promote blood pressure control [3]. These could be in the form of the awareness brought to the patient through an obesity or overweight diagnosis [4], providers or staff spending time counseling a patient about the importance of diet to prevent, treat, or manage a chronic disease during preventive care visits [5, 6], or patient education materials (PEMs) containing nutrition information relevant to chronic disease man- agement provided in an after-visit summary [7]. Operationalizing nutrition care Th l d d Three commonly provided services in primary care hypertension management were used as a proxy for nutrition care, justified by the assumption that nutrition- related information was exchanged with the patient. These nutrition care events were exclusive of the final visit and identified by diagnosis and billing codes for overweight/obesity diagnoses and preventive visits, and a flag that diet-related education materials were provided to the patient within the study period. Clinical guide- lines for the treatment of overweight/obesity include dietary behavior management [20, 21], and the diagnosis itself may inherently offer an opportunity as a “teachable moment” behavior change intervention [4]. Preventive visits that include evaluation for and management of chronic diseases like hypertension were chosen for the comprehensive counseling and guidance provided to patients to reduce risk factors that include diet [22, 23]. Providing patients with printed education materials is a common and passive method for counseling patients about dietary recommendations. Little is known about the rates of nutrition care events documented for hypertension management or their relationship with BP control of patients. Given this, the objectives of this study were to assess of rates of nutri- tion care events identified in structured EHR data and to test whether the timing and dose of these were associated with BP control for adult patients in an academic health system. The hypothesis tested was that documented nutrition care, controlling for patient demographic and health characteristics, is associated with BP control. Blood pressure controlh The outcome of interest was BP control (yes/no), defined as ≤ 140 mm Hg systolic BP (SBP) and ≤ 90 mm Hg dias- tolic BP (DBP) per the National Committee for Quality Assurance (NCQA) quality measure for “Controlling High Blood Pressure”. This outcome was calculated at the patients’ first visit and annually based on the last meas- ured BP in the calendar year. BP control values were carried forward if they were not updated in subsequent visits. Providing nutrition care may improve patient self- management and hypertension outcomes overall [8, 9], but many primary care clinicians and staff face substan- tial time barriers to conducting and documenting these activities in electronic health records [10, 11]. Rates of nutrition care events are generally low [5, 12], and their documentation in structured EHR data is largely unknown [13, 14]. A lack of documentation in the EHR may result in limited communication within the care team about patients’ nutrition statuses and challenges in evaluating care quality, which may be a missed opportu- nity to improve patient outcomes [15]. Analyses D i i Descriptive statistics (frequencies, proportions, medians, and ranges) were calculated for all patients’ and strati- fied samples of Black and white patients’ demographic characteristics and hypertension-relevant indicators, proportions of patients with controlled BP, and rates of documented nutrition care events. Longitudinal Targeted Maximum Likelihood Estimate (LTMLE) [24, 25] is a doubly-robust, semiparametric approach that combines sequential g-computation and inverse probability weighting to estimate an effect of a longitudinal treatment (in this case, clinical events sug- gestive of nutrition care) respecting its temporal relation- ship with an outcome (BP control). The LTMLE package [26] for R was used to estimate average treatment effects Williams et al. BMC Medical Informatics and Decision Making (2023) 23:208 Williams et al. BMC Medical Informatics and Decision Making (2023) 23:208 Page 3 of 9 (ATE) of nutrition care events (i. any nutrition care event; ii. overweight or obesity diagnoses; iii. preventive care visits; and iv. provision of patient education materi- als) on patients’ BP control (yes/no) over the study period and stratified among data for Black patients and white patients. See Fig. 1 for an explanation of the data shap- ing process used for the models. The models adjusted for fixed covariates: age, sex, race/ethnicity, geography (rural/urban), comorbidities; and time-varying covari- ates: number of hypertension medications prescribed and number of primary care visits annually during the study period. The models were stratified by race as a way to separately examine [27] the strength of association between controlled BP and nutrition care events among Black patient and white patient groups. Sensitivity anal- yses were conducted with insured status and BMI, as well as elimination of Year 3 (2020) due to the substan- tial reduction in recorded primary care visits due to the COVID-19 pandemic. Most patients in the sample were overweight or obese (n = 3,843; 90.7%) for at least one recorded BMI during the study period. The median total number of anti-hyper- tensive prescriptions per patient over the study period was 6 (range:0,50). About a third (n = 1,550; 36.6%) had one or more comorbidities, and less than half (n = 2,103; 49.6%) had controlled BP at their first recorded BP dur- ing the study period. Table 2 shows a summary of documented nutrition care event rates by year among all the patients in the sam- ple, as well as rates stratified by Black and white patient groups. Analyses D i i The median number of nutrition care events documented in the EHR data was 1(range:0,16) per patient during the study period, with substantially fewer in Year 3 (2020). Distribution of PEMs was the most fre- quent nutrition care event documented each year (range overall: n = 533 (16.4%) to n = 1,855 (43.8%); for Black patients: n = 125 (7.6%) to n = 617 (28.1%); and for white patients: n = 153 (9.5%) to n = 602 (29.5%)). Overweight/ obesity diagnoses were recorded for far fewer patients annually across the study period (range overall: n = 98 (3.0%) to n = 329 (7.8%); for Black patients: n = 61 (3.7%) to n = 210 (10.3%); and for white patients: n = 37 (2.3%) to n = 125 (6.1%)). Ranges of documented preventive visits were overall: n = 203 (6.2%) to n = 664 (15.7%); for Black patients: n = 85 (5.1%) to n = 319 (14.5%); and white patients: n = 118 (7.4%) to n = 345 (16.9%). Results are presented as average treatment effect parameter estimate differences (ATE), 95% confidence intervals, and p-values for each LTMLE model with alpha 0.05. Comorbidities were calculated with HCUP Elixhauser software [18] program using SAS Enterprise Edition 3.7 (SAS Institute, Cary, North Carolina, USA). All remaining analyses were conducted using R Statisti- cal Software (Version 4.0.3; R Foundation for Statistical Computing, 2020). This study was approved as exempt by the Institutional Review Board at Virginia Common- wealth University. Timing of and repeated nutrition care events effects on BP control Estimates of average treatment effects (ATE) on BP control for any nutrition care event, at least one of pre- ventive care visits, obesity/overweight diagnosis, or provision of PEMs at different timing intervals – Year 1 (Early), Year 3 (Late) – compared to none of these clinical visits are presented in Table 3. Any nutrition care events recorded in year 3, compared with none, were inversely associated with BP control for the over- all sample and for Black patients (ATE -0.12, 95% CI: Results Descriptive statistics for patients’ characteristics and clinical factors are found in Table 1. The overall median age of patients in the EHR data sample (n = 4,237) was 62 years (range:18,85), nearly a third (n = 1,217; 28.7%) were male, half (n = 2,198; 51.9%) were Black/African American, 25.4% (n = 1,077) were from rural areas, and 7.9% (n = 333) were covered by Medicaid insurance. Fig. 1 Temporal order of EHR visit data coarsened into yearly intervals. The baseline variables (W) included race (African American or Black/White), rural, sex, age2, # comorbidities were recorded at the first visit of the study window. Each follow-up interval included time-varying covariates (L,; # hypertension medications and # primary care visits), an indicator of having received any nutrition care events during the interval (At); and an indicator of blood pressure control (Yt) recorded at the last visit of each calendar year. The censoring variable (Ct) indicates a patient did not return for follow-up Fig. 1 Temporal order of EHR visit data coarsened into yearly intervals. The baseline variables (W) included race (African American or Black/White), rural, sex, age2, # comorbidities were recorded at the first visit of the study window. Each follow-up interval included time-varying covariates (L,; # hypertension medications and # primary care visits), an indicator of having received any nutrition care events during the interval (At); and an indicator of blood pressure control (Yt) recorded at the last visit of each calendar year. The censoring variable (Ct) indicates a patient did not return for follow-up Williams et al. Results BP was calculated at the patients’ first visit and annually based on the last measured BP in the calendar year g in rural areas was determined using patient’s zip code corresponding to RUCC code (4–9) g Patients with controlled hypertension (≤ 140 mm Hg SBP and ≤ 90 mm Hg DBP) per the National Committee for Quality Assurance (NCQ “Controlling High Blood Pressure” [22] were flagged (yes/no). BP was calculated at the patients’ first visit and annually based on the last m p = 0.03 and ATE -0.23, 95% CI: -0.41,-0.05, p = 0.01, respectively). p = 0.03 and ATE -0.23, 95% CI: -0.41,-0.05, p = 0.01, respectively). -0.24,-0.01, p = 0.03 and ATE -0.23, 95% CI: -0.38,-0.08, p = 0.003, respectively). Having a nutrition care event in both years 1 and 3 compared to none was posi- tively associated with BP control (ATE 0.16, 95% CI: 0.00,0.32, p = 0.046) among Black patients, however. Results BMC Medical Informatics and Decision Making (2023) 23:208 Page 4 of 9 Table 1 Patient demographics and hypertension risk factors Table 1 Patient demographics and hypertension risk factors a b f b d l Total Patientsa (n = 4,237) Black Patients(n = 2,198) White Patients(n = 2,037) Median Age (range) 62 (18,85) 59 (18,85) 66 (18,85) Male n (%) 1,215 (28.7) 487 (22.2) 728 (35.7) Raceb n (%) African American or Black 2,198 (51.9) - - White 2,039 (48.1) - - Ruralc n (%) 1,077 (25.4) 314 (14.3) 763 (37.4) Medicaid n (%) 333 (7.9) 256 (11.6) 77 (53.8) Overweight or Obese BMId (%) 3,843 (90.7) 2,037 (92.7) 1,806 (88.5) Median Hypertension Medicationse (range) 6 (0,50) 7 (0,50) 6 (0,31) One or More Comorbiditiesf n (%) 1,550 (36.6) 896 (40.8) 652 (32.1) Median Number of Visits (range) 9 (0,91) 9 (0,84) 9 (0,91) Controlled Blood Pressureg n (%) Baseline 2,103 (49.6) 1,051 (47.8) 1,052 (51.6) Year 1 2,437 (57.5) 1,230 (56.0) 1,207 (59.2) Year 2 2,345 (55.4) 1,172 (53.3) 1,173 (57.5) Year 3 1,920 (45.3) 943 (42.9) 977 (47.9) b Includes both Hispanic and non-Hispanic African American/Black and Whites c Patients residing in rural areas was determined using patient’s zip code corresponding to RUCC code (4–9) d Overweight or obese categories derived from BMI is calculated as the maximum recorded BMI across the study period per patient If BMI was not available, the patient’s height and maximum weight across the study period were used to calculate BMI using the standard formula (weight (kg) / height (cm) / height (cm)] × 10,000) e Hypertension medications were the total prescribed across the study period per patient. Medications were identified using the HEDIS reference list for hypertension medications [28] and reviewed with two physicians boarded in internal medicine prior to making calculations f Comorbidities calculated as mode of Elixhauser Comorbidity Index across the study period per patient. The index was calculated using the Healthcare Cost and Utilization Project Methods Series Comorbidity Software. The comorbidity index is a total sum of diagnosed chronic disease comorbidities found in 31 categories defined in the software [21]. g Patients with controlled hypertension (≤ 140 mm Hg SBP and ≤ 90 mm Hg DBP) per the National Committee for Quality Assurance (NCQA) quality measure for “Controlling High Blood Pressure” [22] were flagged (yes/no). Discussion Services were flagged (yes/no) based on the list of International Classification of Diseases version 10 (ICD-10) diagnosis and Current Procedural Terminology (CPT) codes b Number of patients who received at least one form of nutrition care across the study period c Number of patients who received at least one patient education material (PEM) across the study period d Number of patients who were diagnosed with overweight or obesity (ICD-10: Z68.XX; Z71.3; R63.5; E66.XX; E88.81) across the study period e Number of patients who were provided with a preventive care visit (CPT: 99,381–99,387, 99,391–99,397) that includes counseling about chronic disease risk factors and management strategies across the study period f Proportions of nutrition care services for years 2 and 3 are based on patients who were not censored: Year 1: n = 3,941; Year 2: n = 3,254 Total Patients (n = 4,237) Black Patients (n = 2,198) White Patients (n = 2,039) Median Nutrition Carea (range) 1 (0, 16) 1 (0, 16) 1 (0, 15) Nutrition Careb n (%) Year 1 1,855 (43.8%) 950 (43.2%) 905 (44.4%) Year 2 1,538 (39.0%) 831 (40.6%) 707 (37.3%) Year 3 533 (16.4%) 251 (15.2%) 282 (17.6%) Preventive Visitse n (%) Year 1 664 (15.7%) 319 (14.5%) 345 (16.9%) Year 2 573 (14.5%) 268 (13.1%) 305 (16.1%) Year 3 203 (6.2%) 85 (5.1%) 118 (7.4%) Overweight/Obesity Diagnosisd n (%) Year 1 329 (7.8%) 204 (9.3%) 125 (6.1%) Year 2 295 (7.5%) 210 (10.3%) 85 (4.5%) Year 3 98 (3.0%) 61 (3.7%) 37 (2.3%) Patient Education Materialsc n (%) Year 1 1,219 (28.8%) 617 (28.1%) 602 (29.5%) Year 2 965 (24.5%) 530 (25.9%) 435 (23.0%) Year 3 278 (8.5%) 125 (7.6%) 153 (9.5%) Missing Blood Pressuref Year 2 296 (7.0%) 151 (6.9%) 145 (7.1%) Year 3 983 (23.2%) 547 (24.9%) 436 (21.4%) Table 2 Rates of nutrition care events In this study, an overweight or obesity diagnosis was only associated with controlled BP among white patients when recorded in both years 2 and 3 compared with no record or diagnosis. However, BMI persists as a clini- cal tool for risk assessment, and the diagnosis, as previ- ously mentioned, could play an important role in patient awareness of a health risk. Mixed findings about BMI associations with hypertension management is seen in a recent study of BP control rates among adults with hypertension. Discussion Foti and colleagues found those with over- weight or obesity had higher rates of BP control com- pared with those who had lower BMI. This, the authors suggested, might be because the lower BMI patients may be less aware of their hypertension and be treated less often or intensely [30]. Interestingly, rates of obesity and overweight diagnoses were not concordant with BMI cal- culated from patient chart data in the present study. The disconnect between BMI and documented obesity diag- noses has been shown elsewhere; one study of EHR data found that while more than half (52%) of patients in the sample had a BMI ≥ 30.0 qualifying them for an obesity diagnosis, very few (5.6%) had obesity recorded in their health record problem list [31]. One factor contributing to the lack of documenting or addressing overweight or obesity with patients may be clinicians are ill-prepared to discuss the topic [32]. Few patients in the present study had at least one preventive visit that included a discussion of nutrition- related risk factors documented in their EHR record. Clinical guidelines suggest nutrition counseling is included in lifestyle treatment for chronic diseases for which diet is a risk factor. Healthy People 2020 suggested a goal to “increase the proportion of physician office visits made by patients with a diagnosis of cardiovascular dis- ease, diabetes, or hyperlipidemia that include counseling or education related to diet or nutrition,” [33] from 11.5% to 12.7% (a 10% increase) by 2020 as measured by the National Ambulatory Medical Care Survey (NAMCS). Overall rates reached over 20% by 2015, which may be why the objective was eliminated altogether for Healthy People 2030. However, rates assessed by the present study and Healthy People remain objectively low for a service that may be universally useful if it were achievable to provide such service to all patients, as preventive vis- its have long been shown to help improve patient health outcomes. Overall low rates of documented nutrition care events found in this study support the wider call to address barriers faced by clinicians and staff such as time pressures and limited provider education in pro- viding nutrition care in primary care settings [10, 11]. To understand how these rates might increase, these clinical activities must be measured, and to be meas- ured, they must be documented. Discussion EHR data was used in this study to examine associations between BP control and nutrition care events identified through documentation of clinical activities that imply nutrition or diet information was communicated with patients who have hypertension. Visits reported in the EHR data were at primary care clinics part of a health sys- tem that serves a diverse population of rural and under- represented racial/ethnic patients. Among the 4,237 patients with hypertension, rates of nutrition care events were generally low, although number of documented nutrition care events, preventive care visits where guide- lines suggest counseling about dietary risk factors take place were found to be associated with improved odds for BP control. However, there were disparities in BP control and rates of clinical events suggestive of nutrition care recorded by race. p g p Preventive visits in years 2 and 3 versus none among Black patients had a positive effect on BP control (ATE 0.31, 95% CI: 0.07,0.54, p = 0.01). Similarly, preventive visits in only year 3 versus non among white patients had a positive effect on BP control (ATE 0.19, 95% CI: 0.01,0.38, p = 0.04). Obesity/overweight diagnoses during both years 2 and 3 versus none among white patients had the greatest effect on BP control in the study (ATE 0.41, 95% CI: 0.38,0.43, p < 0.001). Patient Education materials early (year 1) or late (year 3) ver- sus none both had negative and significant associa- tions with BP control for the full sample (ATE -0.06, 95% CI: -0.11,-0.01, p = 0.016 and ATE -0.15, 95% CI: -0.28,-0.02, p = 0.02, respectively). This was also true among Black patients (ATE -0.08, 95% CI: -0.15,-0.01, Williams et al. BMC Medical Informatics and Decision Making (2023) 23:208 Page 5 of 9 Page 5 of 9 Table 2 Rates of nutrition care events a Number of nutrition care services documented per patient across the study period. Discussion E(Y(0,0,0))b Early E(Y(1,0,0)) -0.05, (-0.10,0.01; 0.09) 0.04, (-0.03,0.11; 0.264) -0.01, (-0.10,0.08; 0.83) -0.06, (-0.11,-0.01; 0.016) Late E(Y(0,0,1)) -0.12, (-0.24,-0.01; 0.03) 0.12, (-0.03,0.26; 0.11) 0.01, (-0.14,0.15; 0.92) -0.15, (-0.28,-0.02; 0.02) Early + Late E(Y(1,0,1)) 0.08, (-0.04,0.20; 0.19) -0.08, (-0.23,0.08; 0.35) -0.02, (-0.19,0.15; 0.84) 0.09, (-0.04,0.22; 0.185) All E(Y(1,1,1)) 0.00, (-0.08,0.09; 0.93) -0.05, (-0.22,0.11; 0.53) 0.09, (-0.28,0.45; 0.64) 0.01, (-0.12,0.15; 0.826) More Early E(Y(1,1,0)) -0.02, (-0.07,0.03; 0.40) 0.06, (-0.04,0.16; 0.22) 0.09, (-0.06,0.25; 0.25) -0.01, (-0.08,0.05; 0.684) More Late E(Y(0,1,1)) -0.03, (-0.16,0.09; 0.60) 0.15, (-0.06,0.36; 0.15) -0.1, (-0.65,0.45; 0.73) -0.15, (-0.35,0.04; 0.118) Estimates vs. E(Y(0,0,0)) among Black Patients Early E(Y(1,0,0)) -0.07, (-0.14,0.01; 0.09) 0.08, (-0.02,0.18; 0.10) 0, (-0.13,0.13; 0.99) -0.08, (-0.15,-0.01; 0.03) Late E(Y(0,0,1)) -0.23, (-0.38,-0.08; 0.003) 0.15, (-0.06,0.35; 0.15) -0.05, (-0.24,0.13; 0.58) -0.23, (-0.41,-0.05; 0.01) Early + Late E(Y(1,0,1)) 0.16, (0.00,0.32; 0.046) -0.07, (-0.29,0.16; 0.56) 0.05, (-0.17,0.27; 0.64) 0.15, (-0.03,0.34; 0.11) All E(Y(1,1,1)) -0.05, (-0.17,0.07; 0.45) -0.02, (-0.30,0.26; 0.89) -0.09, (-0.61,0.44; 0.75) -0.04, (-0.22,0.15; 0.70) More Early E(Y(1,1,0)) -0.06, (-0.14,0.02; 0.13) 0.09, (-0.06,0.25; 0.25) 0.06, (-0.13,0.24; 0.56) -0.03, (-0.13,0.06; 0.51) More Late E(Y(0,1,1)) 0.06, (-0.11,0.22; 0.48) 0.31, (0.07,0.54; 0.01) -0.16, (-0.66,0.34; 0.52) -0.15, (-0.42,0.12; 0.28) Estimates vs. E(Y(0,0,0)) among white Patients Early E(Y(1,0,0)) -0.04, (-0.12,0.04; 0.35) -0.03, (-0.13,0.07; 0.55) -0.05, (-0.19,0.10; 0.51) -0.04, (-0.11,0.03; 0.30) Late E(Y(0,0,1)) 0.01, (-0.15,0.17; 0.87) 0.19, (0.01,0.38; 0.04) 0.12, (-0.15,0.39; 0.38) -0.07, (-0.25,0.11; 0.46) Early + Late E(Y(1,0,1)) -0.01, (-0.18,0.17; 0.95) -0.03, (-0.38,0.31; 0.85) -0.18, (-0.86,0.49; 0.59) 0.00, (-0.20,0.21; 0.98) All E(Y(1,1,1)) 0.05, (-0.10,0.20; 0.50) -0.06, (-0.28,0.16; 0.57) 0.21, (-0.26,0.68; 0.37) 0.07, (-0.11,0.26; 0.45) More Early E(Y(1,1,0)) 0.00, (-0.08,0.08; 0.99) 0.03, (-0.11,0.17; 0.70) 0.23, (-0.02,0.48; 0.07) 0.00, (-0.10,0.09; 0.95) More Late E(Y(0,1,1)) -0.15, (-0.34,0.03; 0.11) 0.00, (-0.34,0.34; 0.99) 0.41, (0.38,0.43; < 0.001) -0.23, (-0.51,0.04; 0.10) Blood Pressure ControlRisk Difference (RD) (95% CI; p-value) Blood Pressure ControlRisk Difference (RD) (95% CI; p-value) Any Nutrition Care Events Preventive Care Visits Overweight/Obesity Diagnoses Patient Education Materials further understanding and improvements are needed. Others have researched and found delays and underuti- lization of health services by Black patients due to a vari- ety of reasons [34] including socioeconomic barriers [35], medical mistrust [36] and perceived racism [37]. were inversely and significantly associated with BP con- trol in this study, they were the most commonly provided nutrition care event. Timely research is needed to under- stand the effects of communicating nutrition-related chronic disease prevention and management in primary care. Discussion Stange and colleagues argued health systems measure what is valued [29], and this study highlights a few important opportunities where nutrition is addressed with patients: timing and frequency (dose) of diagnoses of overweight or obe- sity diagnoses, preventive care visits, and provision of patient education materials. Of note, preventive care visits were less common for Black patients compared to white patients. Given BP con- trol was less likely among Black patients, yet preventive care visits in years 2 and 3 appeared to have the strongest positive and significant effect on BP control compared to no preventive care visits. This suggests a disparity in ser- vice uptake that may be a missed opportunity for which Williams et al. BMC Medical Informatics and Decision Making (2023) 23:208 Page 6 of 9 Table 3 Estimated average treatment effects of nutrition care eventsa on blood pressure control Blood Pressure ControlRisk Difference (RD) (95% CI; p-value) Table 3 Estimated average treatment effects of nutrition care eventsa on blood pressure control Table 3 Estimated average treatment effects of nutrition care eventsa on blood pressure control a Longitudinal Targeted Maximum Likelihood Estimate (LTMLE) model: Fixed covariates: Race (African American or Black / White), Rural, Sex, Age2, # Comorbidities; Time-varying covariates: # hypertension medications at baseline, end Year 1, end Year 2, end Year 3; # primary care visits end Year 1, end Year 2, end Year 3; Independent Variable: > = 1 Nutrition Care Events at end Year 1, end Year 2, end Year 3; Outcome Variable: Blood Pressure Control (yes/no); Censoring indicator for missing outcome at end Year 2 and end Year 3 b Notation example: E(Y(1,0,0)) vs. E(Y(0,0,0)): Expected risk difference for at least one nutrition care event during Year 1, but none in Years 2 or 3, compared to none across the study period Blood Pressure ControlRisk Difference (RD) (95% CI; p-value) Any Nutrition Care Events Preventive Care Visits Overweight/Obesity Diagnoses Patient Education Material Estimates vs. a Longitudinal Targeted Maximum Likelihood Estimate (LTMLE) model: Fixed covariates: Race (African American or Black / White), Rural, Sex, Age2, # Comorbidities; Time-varying covariates: # hypertension medications at baseline, end Year 1, end Year 2, end Year 3; # primary care visits end Year 1, end Year 2, end Year 3; Independent Variable: > = 1 Nutrition Care Events at end Year 1, end Year 2, end Year 3; Outcome Variable: Blood Pressure Control (yes/no); Censoring indicator for missing outcome at end Year 2 and end Year 3 b Notation example: E(Y(1,0,0)) vs. E(Y(0,0,0)): Expected risk difference for at least one nutrition care event during Year 1, but none in Years 2 or 3, compared to none across the study period Funding g There were no funding sources for this study. Supplementary Information The online version contains supplementary material available at https://doi. org/10.1186/s12911-023-02311-3. The online version contains supplementary material available at https://doi. org/10.1186/s12911-023-02311-3. Our study has important limitations. Using EHR data for research has known challenges due to variability of data entry [43]. As a single health system study, findings may not be generalizable, clinics part of the health system served diverse patient populations that represent a higher risk for health disparities. Measurement bias needs to be considered in the study findings, given there may have been some proportion of clinical visits in virtual format at the onset of COVID. We attempted to correct for as much bias as possible with inclusion criteria described in the Methods and conducted a sensitivity analysis in mod- els that excluded year 3 (2020) and found similar general results. Furthermore, the proxy measures for nutrition care used in this study were not manually confirmed with chart review and would be an interesting future study. These services are critical to chronic disease manage- ment and challenges related to their documentation and delivery to patients remain globally applicable. Conclusionsh The present study utilized EHR data to provide a descrip- tion of nutrition care delivery and examination of its association with patient BP outcomes within a single health system. Overall, documentation of received nutri- tion care events was low, and preventive care visits, but not overweight/obesity diagnosis or delivery of patient education materials, was associated with patients’ BP control. Additionally, disparities were identified in rates of nutrition care events and odds for BP control by race. Further research is needed to explore ways to improve documentation and equity of nutrition care. Enhancing EHR workflows, education for providers and staff about nutrition care, and improving effective communication and collaboration within the clinical team may all be sys- tem-level targets for improving nutrition care and hyper- tension outcomes. Acknowledgements g Data for this project described was supported in part by CTSA award No. UL1TR002649 from the National Center for Advancing Translational Sciences. Its contents are solely the responsibility of the authors and do not necessarily represent official views of the National Center for Advancing Translational Sciences or the National Institutes of Health. The authors would also like to acknowledge and thank Alan Dow for his assistance in identifying the hypertension medications in this study. Availability of data and materials The datasets generated and/or analyzed during the current study are not publicly available due to the data use agreement from source health records. Source code and analytic files are available from the corresponding author on reasonable request. Competing interests None of the authors have competing interests. Received: 24 October 2022 Accepted: 26 September 2023 Received: 24 October 2022 Accepted: 26 September 2023 Authors’ contributions CRediT Statement: AW: Conceptualization, Methodology, Software, Validation, Formal Analysis, Writing—Original Draft, Writing—Review & Editing, Project Administration; MT: Conceptualization, Writing—Review & Editing, Supervi- sion; EB: Methodology, Software, Validation, Formal Analysis, Writing—Review & Editing, Supervision. References 1. Ostchega Y, Fryar CD, Nwankwo T, Nguyen DT. Hypertension prevalence among adults aged 18 and over: United States, 2017–2018. NCHS Data Brief. 2020;364:1–8. Discussion Due to the fast rate at which technology use in health care has advanced, health services researchers face additional challenges in efficient dissemination and eval- uation of EHR process implementations [42]. Leveraging data entered into the patient’s EHR is a way to support the care team in prioritizing appropri- ate preventive care depending upon patients’ individual needs. Krist, et. al. developed an EHR intervention that incorporated automated, tailored, patient-centered mes- saging for preventive care. In a summary of the initial six months of its use (November 2010 through May 2011), clinics were successful in using the tool to support cli- nicians to counsel patients about health behaviors and customize prevention and treatment plans for patients [38]. Such initiatives have been instituted within various health systems and clinics nationally over the past dec- ade to support clinicians through the use of risk calcula- tors [39], automated prompts for clinicians and patients [40], and provision of printed or electronic patient educa- tion materials [41]. Although patient education materials Findings of this study offer new contributions and sug- gestions for additional work needed within the burgeon- ing area of health services research that seeks to improve preventive services to support patients with chronic disease. First, due to our limited established structures to measure or examine nutrition-related activities in primary care, this study provides a framework of proxy measures for preventive nutrition care that may be found in the voluminous EHR data. Next, this study applied longitudinal analyses in stratified samples of underrep- resented racial and ethnic patients, with significant find- ings to support conducting further detailed explorations Williams et al. BMC Medical Informatics and Decision Making (2023) 23:208 Page 7 of 9 of causes and ways forward for improvement. Although not part of the present study, additional stratification by comorbidities may be conducted to assess differences in treatment plans and nutrition care for complex patients. To gain more insight into nutrition care services and relevant patient outcomes, future research should use a larger data sample or data from a group of health sys- tems, and additional data sources e.g., clinical notes for preventive counseling; and claims data for clinical refer- rals to dietitians are other avenues to offer a more robust picture of successful processes for conducting and docu- menting nutrition care delivery that might reveal targets for improvements. References References 1. Ostchega Y, Fryar CD, Nwankwo T, Nguyen DT. Hypertension prevalence among adults aged 18 and over: United States, 2017–2018. NCHS Data Brief. 2020;364:1–8. Abbreviations HER Electronic health records BP Blood pressure Ethics approval and consent to participate Ethics approval and consent to participate In compliance with the World Medical Association Declaration of Helsinki on Ethical Principles for Medical Research Involving Human Subjects, the need for consent in this study was waived by the Institutional Review Board at Virginia Commonwealth University (#HM20020501). Consent for publication Not applicable. Competing interests None of the authors have competing interests. Consent for publication Not applicable. Competing interests None of the authors have competing interests. Abbreviations l BP Williams et al. BMC Medical Informatics and Decision Making (2023) 23:208 Page 8 of 9 2. Whelton PK, Carey RM, Aronow WS, et al. 2017 ACC/AHA/AAPA/ABC/ ACPM/AGS/APhA/ ASH/ASPC/NMA/PCNA guideline for the prevention, detection, evaluation, and management of high blood pressure in adults a report of the American College of Cardiology/American Heart Associa- tion Task Force on Clinical pr. Hypertension. 2018;71(6):E13–115. https:// doi.org/10.1161/HYP.0000000000000065. 2. Whelton PK, Carey RM, Aronow WS, et al. 2017 ACC/AHA/AAPA/ABC/ ACPM/AGS/APhA/ ASH/ASPC/NMA/PCNA guideline for the prevention, detection, evaluation, and management of high blood pressure in adults a report of the American College of Cardiology/American Heart Associa- tion Task Force on Clinical pr. Hypertension. 2018;71(6):E13–115. https:// doi.org/10.1161/HYP.0000000000000065. Statement. JAMA. 2018;320(11):1163. https://doi.org/10.1001/jama.2018. 13022. Statement. JAMA. 2018;320(11):1163. https://doi.org/10.1001/jama.2018. 13022. 22. Diet in the treatment and prevention of hypertension - UpToDate. https://www.uptodate.com/contents/diet-in-the-treatment-and-preve ntion-of-hypertension. Accessed 16 Aug 2019. 23. Casey DE, Thomas RJ, Bhalla V, et al. 2019 AHA/ACC clinical performance and quality measures for adults with high blood pressure: a report of the American college of cardiology/American heart association task force on performance measures. Circ Cardiovasc Qual Outcomes. 2019;12(11):e000057. https://doi.org/10.1161/HCQ.0000000000000057. 3. Lacey K, Pritchett E. Nutrition care process and model: ADA adopts road map to quality care and outcomes management. J Am Diet Assoc. 2003;103(8):1061–72. https://doi.org/10.1016/S0002-8223(03)00971-4. 3. Lacey K, Pritchett E. Nutrition care process and model: ADA adopts road map to quality care and outcomes management. J Am Diet Assoc. 2003;103(8):1061–72. https://doi.org/10.1016/S0002-8223(03)00971-4. 4. Coa KI, Smith KC, Klassen AC, et al. Capitalizing on the “teachable moment” to promote healthy dietary changes among cancer survi- vors: the perspectives of health care providers. Support Care Cancer. 2015;23(3):679–86. https://doi.org/10.1007/s00520-014-2412-z. 4. Coa KI, Smith KC, Klassen AC, et al. Capitalizing on the “teachable moment” to promote healthy dietary changes among cancer survi- vors: the perspectives of health care providers. Support Care Cancer. 2015;23(3):679–86. https://doi.org/10.1007/s00520-014-2412-z. 24. Petersen M, Schwab J, Gruber S, Blaser N, Schomaker M, van der Laan M. Targeted maximum likelihood estimation for dynamic and static longitudinal marginal structural working models. J Causal Inference. 2014;2(2):147–85. https://doi.org/10.1515/jci-2013-0007. 5. Grabovac I, Smith L, Stefanac S, et al. Health Care Providers’ Advice on Lifestyle Modification in the US Population: Results from the NHANES 2011–2016. Am J Med. 2019;132(4):489-497.e1. https://doi.org/10.1016/j. amjmed.2018.11.021. 5. Grabovac I, Smith L, Stefanac S, et al. Health Care Providers’ Advice on Lifestyle Modification in the US Population: Results from the NHANES 2011–2016. Am J Med. 2019;132(4):489-497.e1. https://doi.org/10.1016/j. amjmed.2018.11.021. 25. Abbreviations l Schnitzer ME, van der Laan MJ, Moodie EEM, Platt RW. LTMLE with Clus- tering. In: van der Laan MJ, Rose S, eds. Targeted Learning in Data Science: Causal Inference for Complex Longitudinal Studies. Springer Series in Statistics. Springer International Publishing; 2018:233–251. https://doi. org/10.1007/978-3-319-65304-4_15. 6. Williams AR, Wilson-Genderson M, Thomson MD. A cross-sectional analysis of associations between lifestyle advice and behavior changes in patients with hypertension or diabetes: NHANES 2015–2018. Prev Med. 2021;145:106426. https://doi.org/10.1016/j.ypmed.2021.106426. 26. Lendle SD, Schwab J, Petersen ML, van der Laan MJ. ltmle: an R package implementing targeted minimum loss-based estimation for longitudinal data. J Stat Softw. 2017;81:1–21. https://doi.org/10.18637/jss.v081.i01. 7. Lyles CR, Gupta R, Tieu L, Fernandez A. After-visit summaries in primary care: mixed methods results from a literature review and stakeholder interviews. Fam Pract. 2019;36(2):206–13. https://doi.org/10.1093/fampra/ cmy045. 27. Ward JB, Gartner DR, Keyes KM, Fliss MD, McClure ES, Robinson WR. How do we assess a racial disparity in health? Distribution, interaction, and interpretation in epidemiological studies. Ann Epidemiol. 2019;29:1–7. https://doi.org/10.1016/j.annepidem.2018.09.007. 8. Kravetz JD, Walsh RF. Team-based hypertension management to improve blood pressure control. J Prim Care Community Health. 2016;7(4):272–5. https://doi.org/10.1177/2150131916645580. 28. HEDIS MY 2020 Medication List Directory. https://store.ncqa.org/hedis- my-2020-medication-list-directory.html. Accessed 18 June 2021 29. Stange KC, Etz RS, Gullett H, et al. Metrics for assessing improvements in primary health care. Annu Rev Public Health. 2014;35:423–42. https://doi. org/10.1146/annurev-publhealth-032013-182438. 9. Kaboli PJ, Howren MB, Ishani A, Carter B, Christensen AJ, Vander Weg MW. Efficacy of Patient Activation Interventions With or Without Financial Incentives to Promote Prescribing of Thiazides and Hypertension Control. JAMA Netw Open. 2018;1(8):e185017. https://doi.org/10.1001/jaman etworkopen.2018.5017. g p 30. Foti K, Hardy ST, Chang AR, Selvin E, Coresh J, Muntner P. BMI and blood pressure control among United States adults with hypertension. J Hyper- tens. 2022;40(4):741–8. https://doi.org/10.1097/HJH.0000000000003072. 10. Kushner RFF. Barriers to Providing Nutrition Counseling by Physicians: A Survey of Primary Care Practitioners. Prev Med. 1995;24(6):546–52. https://doi.org/10.1006/PMED.1995.1087. 31. Mattar A, Carlston D, Sariol G, et al. The prevalence of obesity documenta- tion in primary care electronic medical records: Are we acknowledging the problem? Appl Clin Inform. 2017;8(1):67–79. https://doi.org/10.4338/ ACI-2016-07-RA-0115. 11. Bardach SH, Schoenberg NE. Primary care physicians’ prevention counseling with patients with multiple morbidity. Qual Health Res. 2012;22(12):1599–611. https://doi.org/10.1177/1049732312458183. 32. Antognoli EL, Seeholzer EL, Gullett H, Jackson B, Smith S, Flocke SA. Primary Care Resident Training for Obesity, Nutrition, and Physical Activity Counseling: A Mixed-Methods Study. Health Promot Pract. 2017;18(5):672–80. https://doi.org/10.1177/1524839916658025. 12. Abbreviations l Aspry KE, Linda Van Horn C, Chair Jo Ann Carson VS, et al. Medical Nutri- tion Education, Training, and Competencies to Advance Guideline-Based Diet Counseling by Physicians. Circulation. 2018;137(1):821–41. https:// doi.org/10.1161/CIR.0000000000000563 33. Nutrition and Weight Status \| Healthy People 2020. https://www.healt hypeople.gov/2020/topics-objectives/topic/nutrition-and-weight-status. https://www.healthypeople.gov/2020/topics-objectives/topic/nutrition- and-weight-status. Accessed 16 Mar 2020. 13. Hosomura N, Goldberg SI, Shubina M, Zhang M, Turchin A. Electronic documentation of lifestyle counseling and glycemic control in patients with diabetes. Diabetes Care. 2015;38(7):1326–32. https://doi.org/10. 2337/dc14-2016. 34. Powell W, Richmond J, Mohottige D, Yen I, Joslyn A, Corbie-Smith G. Medical mistrust, racism, and delays in preventive health screening among African-American men. Behav Med Wash DC. 2019;45(2):102–17. https://doi.org/10.1080/08964289.2019.1585327. 14. Bae J, Hockenberry JM, Rask KJ, Becker ER. Evidence that electronic health records can promote physician counseling for healthy behaviors. Health Care Manage Rev. 2017;42(3):258–68. https://doi.org/10.1097/HMR.00000 00000000108. 35. Ravenell JE, Whitaker EE, Johnson WE. According to him: barri- ers to healthcare among African-American men. J Natl Med Assoc. 2008;100(10):1153–60. https://doi.org/10.1016/s0027-9684(15)31479-6. 15. Watterson JL, Rodriguez HP, Aguilera A, Shortell SM. Ease of use of electronic health records and relational coordination among primary care team members. Health Care Manage Rev. 2020;45(3):267–75. https://doi. org/10.1097/HMR.0000000000000222. 36. Hammond WP, Matthews D, Mohottige D, Agyemang A, Corbie-Smith G. Masculinity, medical mistrust, and preventive health services delays among community-dwelling African-American Men. J Gen Intern Med. 2010;25(12):1300–8. https://doi.org/10.1007/s11606-010-1481-z. g 16. CQF. 2019 HEDIS ® Measures Controlling High Blood Pressure (CBP) HEDIS Measure Description. www.ncqa.org. Accessed 21 Dec 2019. 37. Cheatham CT, Barksdale DJ, Rodgers SG. Barriers to health care and health-seeking behaviors faced by Black men. J Am Acad Nurse Pract. 2008;20(11):555–62. https://doi.org/10.1111/j.1745-7599.2008.00359.x. 17. USDA ERS - Rural-Urban Continuum Codes. https://www.ers.usda.gov/ data-products/rural-urban-continuum-codes.aspx. Accessed 23 Oct 2019. p p 18. HCUP Methods Series Comorbidity Software Documentation. https:// wwwhcup us ahrq gov/reports/methods/2004 1 jsp 18. HCUP Methods Series Comorbidity Software Documentation. https:// www.hcup-us.ahrq.gov/reports/methods/2004_1.jsp. 38. Krist AH, Peele E, Woolf SH, et al. Designing a patient-centered personal health record to promote preventive care. BMC Med Inform Decis Mak. 2011;11(1):1–11. https://doi.org/10.1186/1472-6947-11-73. 19. HEDIS MY 2020 Medication List Directory. https://store.ncqa.org/hedis- my-2020-medication-list-directory.html. Accessed 18 June 2021. 39. McGuinness JE, Zhang TM, Cooper K, et al. Extraction of Electronic Health Record Data using Fast Healthcare Interoperability Resources for Automated Breast Cancer Risk Assessment. AMIA Annu Symp Proc. 2022;2021:843–52. 20. Obesity in adults: Overview of management - UpToDate. https://www- uptodate-com.proxy.library.vcu.edu/contents/obesity-in-adults-overv iew-of-management?search=obesity%20management&source=search_ result&selectedTitle=1~150&usage_type=default&display_rank=1. Accessed 16 July 2022. 40. Krist AH. Abbreviations l Electronic Health Record Innovations for Healthier Patients and Happier Doctors. J Am Board Fam Med JABFM. 2015;28(3):299–302. https://doi.org/10.3122/jabfm.2015.03.150097. y 21. US Preventive Services Task Force, Curry SJ, Krist AH, et al. Behavioral weight loss interventions to prevent obesity-related morbidity and mortality in adults: US Preventive services task force recommendation 21. US Preventive Services Task Force, Curry SJ, Krist AH, et al. Behavioral weight loss interventions to prevent obesity-related morbidity and mortality in adults: US Preventive services task force recommendation Page 9 of 9 Williams et al. BMC Medical Informatics and Decision Making (2023) 23:208 41. Schooley B, Singh A, Hikmet N, Brookshire R, Patel N. Integrated digital patient education at the bedside for patients with chronic conditions: observational study. JMIR MHealth UHealth. 2020;8(12):e22947. https:// doi.org/10.2196/22947. g 42. Brownson RC, Eyler AA, Harris JK, Moore JB, Tabak RG. Getting the word out: new approaches for disseminating public health science. J Public Health Manag Pract. 2018;24(2):102–11. https://doi.org/10.1097/PHH. 0000000000000673. 42. Brownson RC, Eyler AA, Harris JK, Moore JB, Tabak RG. Getting the word out: new approaches for disseminating public health science. J Public Health Manag Pract. 2018;24(2):102–11. https://doi.org/10.1097/PHH. 0000000000000673. 43. Casey JA, Schwartz BS, Stewart WF, Adler NE. Using electronic health records for population health research: a review of methods and applica- tions HHS public access. Annu Rev Public Health. 2016;37:61–81. https:// doi.org/10.1146/annurev-publhealth-032315-021353. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from:
W2008186762.txt	https://www.ocl-journal.org/articles/ocl/pdf/2001/04/ocl200184p333.pdf	fr		Evolution des paramètres lipidiques sanguins chez l’homme, secondaire à l’introduction de lin, riche en acide alpha-linolénique (n-3), dans l’alimentation d’animaux destinés à la consommation humaine	Oilseeds and fats, crops and lipids/OCL. Oilseeds & fats crops and lipids	2,001	cc-by	2,562	Evolution des paramètres lipidiques sanguins chez l'homme, secondaire à l'introduction de lin, riche en acide alpha-linolénique (n-3), dans l'alimentation d'animaux destinés à la consommation humaine Development of blood lipid parameters in humans following the introduction of linseed, rich in linolenic acid (n-3), into the food of animals reared for human consumption Oléagineux, Corps Gras, Lipides. Volume 8, Numéro 4, 333-5, Juillet - Août 2001, Dossier : "Aliments fonctionnels et lipides" Auteur(s) : Pierre WEILL, Bernard SCHMITT, Philippe LEGRAND, Laboratoire de biochimie Ensa-Inra, 65, rue de Saint-Brieuc, 35042 Rennes Cedex. Résumé : L'introduction de graines de lin extrudées dans l'alimentation animale permet de modifier le régime de l'homme et en particulier le profil des acides gras des lipides sanguins dans le sens d'un enrichissement en acides gras n-3 et en acides gras conjugués. Ceci ouvre des perspectives intéressantes en termes de prévention pour se rapprocher des régimes de type crétois, sans changement des habitudes alimentaires. Mots-clés : nutrition animale, nutrition humaine, graine de lin, acide alpha-linolénique, acides gras conjugués (CLA). Summary : Consuming foodstuffs from animals fed linseed-added diets induces significant modifications of consumers' plasma and erythrocyte fatty acid parameters and in particular an increase in C18:3 n-3, comparable to that noted under the "Cretan" diet. Those modifications are also accompanied by a sharp increase in CLA. Those results open interesting prospects in terms of prevention and can be achieved without any change in consumers' eating habits. Keywords : animal nutrition, human nutrition, linseed, alpha-linolenic acid, conjugated linoleic acid (CLA). ARTICLE Il est couramment admis que l'alimentation des pays occidentaux est déficitaire en acides gras (n-3), notamment en C18:3 (n-3), et excédentaire en acides gras (n-6), plus particulièrement en C18:2 (n-6). Les résultats d'études épidémiologiques [1, 2], cliniques [3] et interventionnelles [4-6] confirment l'aggravation du risque de mortalité cardiovasculaire et du cancer liée à ce déséquilibre alimentaire. Ainsi, les recommandations nutritionnelles en matière d'alimentation lipidique préconisent une baisse du rapport C18:2 (n-6)/C18:3 (n-3) associée à une augmentation de l'apport d'AGPI (n-3) [7]. Article disponible sur le site http://www.ocl-journal.org ou http://dx.doi.org/10.1051/ocl.2001.0333 Différentes solutions ont été proposées afin d'augmenter l'apport d'AGPI (n-3) alimentaire : consommation de poisson permettant un apport d'AGPI (n-3) à très longues chaînes, utilisation d'huile de colza riche en C18:3 (n-3) ou apport de végétaux riches en C18:3 (n-3). Parmi ceux-ci, la graine de lin tient une place particulière. Particulièrement riche en C18:3 (n-3) - jusque 70 % de ses huiles - au point qu'elle a donné son nom à l'acide alpha linolénique, son utilisation sous forme de graines broyées et bouillies dans l'alimentation des animaux était courante en Europe jusqu'au début du xxe siècle [8, 9]. Plusieurs études ont été réalisées avec un apport de graine ou d'huile de lin dans les régimes de l'homme. Cette pratique est très difficile à mettre en œuvre étant donné le haut potentiel de peroxydation de l'huile de lin [10]. Par ailleurs, la graine de lin, consommée crue, contient des composés cyanogènes toxiques pour l'homme à fortes doses [11]. Partant de ce constat, il paraissait intéressant de contourner la difficulté en introduisant du lin extrudé, donc détoxifié, dans l'alimentation d'animaux destinés à la consommation humaine. On sait en effet que les graisses d'origine animale constituent environ la moitié de l'apport lipidique de l'alimentation humaine dans les pays industrialisés [12] et que leur composition varie en fonction de l'alimentation des animaux [13]. Or, les méthodes d'élevage modernes ont profondément modifié la composition de l'alimentation du bétail. Alors que le C18:3 (n-3) ou acide alpha linolénique (ALA) est un composant majeur des membranes des végétaux en croissance, notamment de l'herbe et des algues en période de synthèse chlorophyllienne (au même titre que le pourpier des Crétois) [14], le remplacement des fourrages verts par le maïs et le soja s'est traduit par un apport accru de C18:2 (n-6) ou acide linoléique (LA) dont il est le principal composant des réserves lipidiques [15]. Ainsi, en éloignant irréversiblement l'herbe de l'alimentation des animaux, l'agriculture moderne a, par voie de conséquence, considérablement diminué l'apport de C18:3 (n-3) dans l'assiette du consommateur. Afin de restaurer un apport suffisant de C18:3 (n-3), les mesures interventionnelles de prévention des maladies cardio-vasculaires et du cancer impliquent des changements de comportements alimentaires souvent importants. Ceux-ci sont souvent impossibles à mettre en œuvre à grande échelle et dans la durée, compte tenu des résistances culturelles importantes. S'il est prouvé que le régime crétois représente le meilleur modèle alimentaire dans le cadre de la prévention primaire et secondaire des maladies cardio-vasculaires, il est peu probable qu'un tel modèle puisse s'imposer à grande échelle dans les pays du Nord de l'Europe et dans les pays anglo-saxons qui sont pourtant ceux où la prévalence de ces maladies est la plus importante. Nous avons donc testé la double hypothèse suivante : - une modification de l'alimentation animale, exclusivement végétale et enrichie au lin extrudé, entraîne une modification du profil des acides gras au niveau des produits issus de ces élevages incorporation des (n-3) aux triglycérides de réserve, élongation/désaturation et incorporation aux phospholipides des viandes et des œufs ; - la consommation de ces produits par l'homme améliore le profil lipidique de celui-ci - mesuré par la composition des acides gras des triglycérides plasmatiques et des phospholipides des hématies - et pourrait constituer ainsi une mesure de prévention des maladies cardiovasculaires sans changement radical des habitudes alimentaires antérieures. Déroulement de l'étude Pour tester ces hypothèses, nous avons mené une étude randomisée en double aveugle et cross-over impliquant 75 volontaires sains, en deux périodes expérimentales de 35 jours séparées par une période de wash-out de 18 jours. Au préalable, des essais zootechniques ont comparé les performances d'animaux (vaches, porcs, bovins, poules, agneaux et poulets) nourris soit avec du lin extrudé (5 % de la ration), soit avec du soja (5 % de la ration), toutes choses égales par ailleurs. Les produits de ces animaux (lait, viandes et œufs) ont été prélevés pour distribution aux volontaires comme seule source de matière grasse animale pendant toutes les périodes expérimentales. Les graines de lin utilisées pour l'alimentation des animaux « essai » étaient issues de cinq variétés et avaient subi un traitement de cuisson-extrusion pour aboutir aux caractéristiques suivantes : - C18:3 n-3 > 230 g/kg MS ; - MG libre/MG totale > 75 % ; - HCN < 10 mg/kg MS ; - peroxydes < 20 mEq O2/kg MS. Les volontaires (sujets sains, jeunes - 34 ans - et normolipidiques) ont consommé un régime « normal » : 2 020 Kcal par jour en moyenne, dont 33 % de l'apport énergétique sous forme de lipides, et 70 % de ces lipides sous forme de graisses animales issues de l'essai. Les consommations de MG animale représentent 54 g par jour et par volontaire dont 32 g de MG laitières, 10 g par les œufs et 12 g par les viandes (porc, poulet, bœuf et agneau). À chaque prise de sang (J-21, J+15, J+35, j+53, J+88, J+102), nous avons réalisé un bilan lipidique ainsi qu'un profil d'acides gras sur les MG du sérum et des hématies (après séparation par centrifugation). Enfin, des tests hédoniques impliquant un jury de 50 consommateurs indépendants par couple de produits (essai et témoin) sont réalisés pour les 11 produits servant à l'approvisionnement des volontaires (3 lots de beurres, 1 lot d'œufs et 7 lots de viandes). Modification des produits animaux Dans tous les produits, les taux de C18:3 (n-3) augmentent de façon significative (en moyenne + 200 %). Le rapport C18:2 (n-6)/C18:3 (n-3) ainsi que le rapport des AG (n-6)/(n-3) diminuent de façon significative. On constate également que : - les taux de C16:0 diminuent alors que le C18:0 augmente ; - le taux de C18:1 diminue dans tous les produits (sauf le lait) et en particulier dans les œufs ; - les AG (n-3) dérivés (EPA, DHA) augmentent dans tous les produits animaux alors que l'acide arachidonique (AA) C20:4 (n-6) diminue ; - les graisses des ruminants « essai » sont deux fois plus riches en acides linoléiques conjugués (CLA), constitués essentiellement par l'acide ruménique C18:2 cis 9 trans 11. Modification des régimes des volontaires Les volontaires des deux groupes « essai » et « témoin » ont consommé la même quantité d'œufs, de beurre et de viandes pendant les deux périodes. Les différences constatées ne peuvent provenir que de l'alimentation des animaux d'élevage. Or, cette modification de l'alimentation animale génère des différences importantes selon le type de régime, dans la nature des acides gras ingérés. Le régime des volontaires « essai » est plus riche en C18:3 n-3 (+ 0,9 g/jour soit + 120 %) et s'approche ainsi des ANC. L'apport des dérivés n-3 augmente lui de 0,15 g/j (+ 130 %). Le taux de CLA double (0,4 versus 0,2 g/j ), et le rapport des n-3 sur n-6 diminue de 15 à 7 pour les précurseurs, et de 14 à 6 pour les dérivés. Modification des paramètres lipidiques du sérum Les AG libérés après hydrolyse des lipides alimentaires participent à plusieurs voies métaboliques possibles : oxydation, élongation et/ou désaturation, captation tissulaire, notamment par le système nerveux et synthèse des triglycérides et des phospholipides membranaires. La comparaison des profils d'acides gras des sérums des deux groupes est particulièrement intéressante : - le taux de 18:3 n-3 passe de 0,44 % (groupe témoin) à 0,93 % (p < 0,0001), ce qui est une valeur particulièrement élevée ; - le taux de CLA passe lui de 0,28 % à 0,42 % (p < 0,0001) : les n-3 dérivés augmentent (+ 23 %, p > 0,001) et le rapport n-6/n-3 diminue de 15 à 10. Modifications de la composition lipidique des hématies L'étude de la composition des acides gras constitutionnels des phospholipides de l'hématie présente un grand intérêt. En effet, cette composition est le résultat d'une synthèse cellulaire précédant le stade de l'érythrocyte, et non le résultat d'un échange direct entre la paroi de l'hématie et les AGL circulants. Elle traduit bien ainsi les modifications structurelles induites par le régime alimentaire sur le métabolisme cellulaire général. Malgré le protocole en cross-over, et la période de wash-out un peu courte, on observe des modifications significatives de la composition lipidique des hématies : le taux des dérivés longues chaînes augmente (+ 10 %, p < 0,01), particulièrement le C20:5 n-3 (EPA) (+ 32 %, p < 0,001), et le rapport n-6/n-3 diminue significativement (3,8 vs 4,2, p < 0,001). Abréviations utilisées AA : acide arachidonique AG : acide gras AGMI : acide gras mono-insaturé AGPI : acide gras poly-insaturé AGS : acide gras saturé ALA : acide alpha-linolénique ANC : apports nutritionnels conseillés CLA : acides linoléiques conjugués DHA : acide docosahexaénoïque DPA : acide docosapentaénoïque EPA : acide éicosapentaénoïque LA : acide linoléique LC : longues chaînes MS : matière sèche PL : phospholipides TG : triglycérides CONCLUSION Le lin extrudé introduit en faibles quantités en alimentation animale modifie de façon importante le profil lipidique des œufs, laits et viandes obtenus. Ces modifications permettent chez l'homme de doubler, voire tripler l'apport de C18:3 n-3 et de ses dérivés longues chaînes n-3 pour se rapprocher des recommandations des ANC à « isoconsommation » de produits animaux. Ce régime induit ensuite chez l'homme des modifications du profil d'AG des sérums, caractérisées par une augmentation significative des principaux acides gras n-3. Le taux de C18:3 n-3 sérique dépasse les valeurs obtenues dans les études sur le régime crétois. De plus, il est très intéressant d'observer qu'une petite modification de l'alimentation des animaux peut modifier, chez l'homme, la composition des AG des phospholipides des hématies, en particulier les très longues chaînes n-3 dont le C20:5 n-3 précurseur des prostaglandines de la série 3. En conclusion, il semble intéressant d'introduire le lin extrudé dans l'alimentation animale puisque cela permet de modifier le régime de l'homme et le profil des acides gras des lipides sanguins dans le sens d'un enrichissement en acides gras de la série n-3 à la fois en précurseur alpha-linolénique et en ses dérivés à très longues chaînes. Pour les principaux AG, les modifications observées dans les régimes, les sérums et les hématies des groupes expérimentaux vont en effet dans le sens des recommandations actuelles des nutritionnistes. La comparaison des profils lipidiques obtenus dans le groupe « essai » avec les résultats de certaines études épidémiologiques ou d'intervention, notamment la comparaison avec les données du « régime crétois », permet d'espérer, à terme, des bénéfices santé du même ordre que ceux décrits dans ces études. L'intérêt de notre étude réside aussi dans le fait que les modifications sont obtenues sans changement des habitudes alimentaires des consommateurs, puisque seuls différaient les régimes des animaux. Enfin, dans l'hypothèse où ce type d'alimentation était proposé au consommateur, nous avons mesuré l'acceptabilité des produits animaux sous forme de « tests hédoniques » (questionnaire de préférence). Les résultats indiquent une préférence significative pour l'ensemble des produits « essai » (p < 0,05). REFERENCES 1. KEYS A, MENOTTI A, ARAVANIS C, et al. (1984). The seven countries study: 2,289 deaths in 15 years. J Prev Med, 13 : 141-54. 2. KARDINAL AFM, KOK KJ, RINGSTAD, et al. (1993). Antioxidants in adipose tissue and risk in myocardial infarction: the Euramic study. Lancet 342 : 1379-84. 3. DOLECEK A (1992). Epidemiological evidence of relationships between dietary polyunsaturated fatty acids and mortality in the multiple risk factor intervention trial. Proc Soc Exper Biol Med 2000 : 177-82. 4. DE LORGER LM, RENAUD S, MAMELLE N, et al. (1994). Mediterranean alpha-linolenic acid rich-diet in the secondary prevention of coronary hearth desease. Lancet, 343 : 1454-9. 5. LAVILLONNIÈRE F, BOUGNOUX P (1999). Conjugated Linoleic Acid (CLA) and the risk of breast cancer. In : YURAWECZ MP, MOSSOBA MM, KRAMER JKG, PARIZA MW, NELSON GJ, eds. Advances in conjugated linoleic acid research, volume 1, chapter 10. Champaign, Illinois : AOCS Press. 6. BELURY MA, VAN DEN HEUVEL JP (1999). Modulation of diabetes by conjugated linoleic acid. In : YURAWECZ MP, MOSSOBA MM, KRAMER JKG, PARIZA MW, NELSON GJ, eds. Advances in conjugated linoleic acid research, volume 1, chapter 32. Champaign, Illinois : AOCS Press. 7. LEGRAND P, BOURRE JM, DESCOMPS B, DURAND G, RENAUD S (2000). Apports nutritionnels conseillés pour la population française, 3e édition. AFSSA, CNERNA-GNRS, Tec et Doc : Lavoisier Ed, 63-82. 8. DEBRY HG (1923). Le vétérinaire moderne. Wargnies : Éditions Broquet. 9. DECHAMBRE P (1907). La vache laitière. Paris VI : Librairie des Sciences Agricoles. 10. BHATTY RS (1995). Nutrient composition of whole flaseed and flaxseed meal. In : CUNNANE SC, THOMPSON LU, eds. Flaxseed in human nutrition, chapter 2. Champaign, Illinois : AOCS Press. 11. MAZZA G, OOMAH BD (1995). Flaxseed, dietary liber and cyanogens. In : CUNNANE SC, THOMPSON LU, eds. Flaxseed in human nutrition, chapter 4. Champaign, Illinois : AOCS Press. 12. GIRARD JP, RANDRIAMANARVO M, DENOYER C (1986). Leur rôle dans le déterminisme des qualités de la viande, du tissu adipeux et des produits carnés. In : Les lipides animaux dans la filière viande, tome 2, vol. 39. Paris : éd. Apria. 13. MORAND-FEHR P, CHILLIARD Y, BAS P (1986). Répercussions de l'apport de matières grasses dans la ration sur la production et la composition du lait de ruminant. Bull Tech CRZV Theix Inra, 64 : 5972. 14 SIEGENTHALER PA, MURATA N (1998). Lipids in photosynthesis: structure, function and genetics. Dordrecht, Boston : Kluver academic publishers. 15. JARRIGE R, et al. (1988). Alimentation des bovins, ovins et caprins. Paris : Inra.
https://openalex.org/W3211549619	https://www.mdpi.com/1996-1944/14/22/6818/pdf?version=1637313830	English	null	A Numerical Investigation into the Effect of Homogeneity on the Time-Dependent Behavior of Brittle Rock	Materials	2,021	cc-by	11,390	materials materials Keywords: creep; homogeneity; stress level; steady creep rate; dilatancy; failure pattern Article A Numerical Investigation into the Effect of Homogeneity on the Time-Dependent Behavior of Brittle Rock o-Zhe Chen 1,2, Zhu-Shan Shao 2,3,, Dong-Dong Jin 2,3, Zhe Zhang 1,2 and Dong-Bo Zhou 1,2 1 School of Civil Engineering, Xi’an University of Architecture & Technology, Xi’an 710055, China; chenhaozhe515@foxmail.com (H.-Z.C.); zzhe0315@xauat.edu.cn (Z.Z.); zdb@xauat.edu.cn (D.-B.Z.) 2 Shaanxi Key Lab of Geotechnical and Underground Space Engineering, Xi’an University of Architecture & Technology, Xi’an 710055, China; jindongdong@xauat.edu.cn 3 School of Science, Xi’an University of Architecture & Technology, Xi’an 710055, China Correspondence: shaozhushan@xauat.edu.cn 1 School of Civil Engineering, Xi’an University of Architecture & Technology, Xi’an 710055, China; chenhaozhe515@foxmail.com (H.-Z.C.); zzhe0315@xauat.edu.cn (Z.Z.); zdb@xauat.edu.cn (D.-B.Z.) 2 Shaanxi Key Lab of Geotechnical and Underground Space Engineering, y gy * Correspondence: shaozhushan@xauat.edu.cn Abstract: To investigate the brittle creep failure process of rock material, the time-dependent prop- erties of brittle rocks under the impact of homogeneity are analyzed by the numerical simulation method, RFPA-Creep (2D). Deformation is more palpable for more homogeneous rock material under the uniaxial creep loading condition. At a low stress level, diffusion creep may occur and transition to dislocation creep with increasing applied stress. The law for increasing creep strain with the homogeneity index under a constant conﬁned condition is similar to the uniaxial case, and dislocation creep tends to happen with increasing conﬁning pressure for the same homogeneity index. The dilatancy index reaches its maximum at a high stress level when rock approaches failure, and the evolution of the dilatancy index with the homogeneity index under the same conﬁning pressure is similar to the uniaxial case and is more marked than that under the unconﬁned condition. Both uniaxial and triaxial creep failure originate from the ductile damage accumulation inside rock. The dominant shear-type failure is exhibited by uniaxial creep and the conventional compression case presents the splitting-based failure mode. Under conﬁning pressure, the creep failure pattern is prone to shear, which is more notable for the rock with higher homogeneity. Citation: Chen, H.-Z.; Shao, Z.-S.; Jin, D.-D.; Zhang, Z.; Zhou, D.-B. A Numerical Investigation into the Effect of Homogeneity on the Time-Dependent Behavior of Brittle Rock. Materials 2021, 14, 6818. https://doi.org/10.3390/ma14226818 1. Introduction Lockner and Madden, 1991b managed to depict the characteristics of accelerating creep phase and predict the proper stress sensitivity of creep rate by the developed numerical multiple-crack interaction model [26]. According to the precious progressive damage model [27], Amitrano and Helmstetter, 2006 further established the time-independent model and gained the spacial distribution of damage of rock at different stress levels by numerical analysis [8]. As a valid numerical analysis tool, RFPA (realistic failure process analysis) comprehensively considers the nonlinearity of the failure process and the heterogeneity of rock material, and it introduces the multiple critical factors such as temperature, moisture, etc., and a few applications on the study of rock time-dependency have been achieved [28–31]. Li et al., 2008 observed the creep failure process of rock during unconﬁned loading by introducing the constitutive model for microscopic elements under the time effect [32]. DIC (digital image correlation) technology [19]. Although the initiation of the numerical method in measuring the mechanical behavior of rock material is relatively recent, it has been widely adopted and constantly developed by numerous scholars and engineers including the ﬁnite difference methods (FDM), ﬁnite element methods (FEM), boundary element methods (BEM) and meshless methods [20–25], etc., and a few numerical studies which have focused on the creep property of rock. Lockner and Madden, 1991b managed to depict the characteristics of accelerating creep phase and predict the proper stress sensitivity of creep rate by the developed numerical multiple-crack interaction model [26]. According to the precious progressive damage model [27], Amitrano and Helmstetter, 2006 further established the time-independent model and gained the spacial distribution of damage of rock at different stress levels by numerical analysis [8]. As a valid numerical analysis tool, RFPA (realistic failure process analysis) comprehensively considers the nonlinearity of the failure process and the heterogeneity of rock material, and it introduces the multiple critical factors such as temperature, moisture, etc., and a few applications on the study of rock time-dependency have been achieved [28–31]. Li et al., 2008 observed the creep failure process of rock during unconﬁned loading by introducing the constitutive model for microscopic elements under the time effect [32]. 2. Model Descriptions and Setup 2.1. Brief Description of RFPA 2D (Creep) 1. Introduction Therefore, in order to investigate the mechanism of brittle creep failure of rock ma- terial, with the aid of the numerical simulation method, RFPA-Creep (2D), the effect of homogeneity (the homogeneity indexes m of 1.5, 2, 2.5, 3 and 5) on time-dependent proper- ties of brittle rock under uniaxial and triaxial creep loading is analyzed in this paper. The relationship between stress level and steady creep rate, the characteristics of dilatancy, the damage evolution and failure pattern of rock are discussed. 1. Introduction Excavations of rock mass at great depth for mining, tunneling, etc. could be accom- panied by server dynamic disasters such as rockbursts, which commonly take place in a sudden manner during excavations [1–3]. In most cases, the rheology of surrounding rock mass manifests as obvious creep for the delayed rockburst under the action of a strong time effect [4], especially when a deep rock mass is subjected to high in situ stresses, and the delayed duration of the rockburst events ranges from several hours to days, even months [5–7]. With the evolution of time-dependency, the creep failure of hard and brittle surrounding rock easily triggers the time-delayed rockburst, threatening the safety and stability during construction of deep rock engineering. Received: 8 October 2021 Accepted: 8 November 2021 Published: 11 November 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. It is known that creep of brittle rocks can be deﬁned as the time-dependent irreversible deformation which occurs during the action of a constant applied stress lower than short- term strength [8,9]. Currently, the various research approaches for time-dependent behavior of brittle rocks are mainly divided into three aspects: theory, experiment and numerical simulation [10–14]. Brantut et al., 2012 and Li et al., 2019 explored the relationship between brittle creep behavior and crack growth on a microscopic scale by extending the same micromechanical theoretical model [15–17]. Shi et al., 2018 analyzed the damage evolution inside a cuboid-shaped sandstone specimen reﬂected by the real-time recorded spacial development of AE activities during the creep loading test [18]. Tang, 2013 obtained the surface creep deformation of rock specimen based on strain ﬁelds that are measured using Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). https://www.mdpi.com/journal/materials Materials 2021, 14, 6818. https://doi.org/10.3390/ma14226818 Materials 2021, 14, 6818 2 of 17 DIC (digital image correlation) technology [19]. Although the initiation of the numerical method in measuring the mechanical behavior of rock material is relatively recent, it has been widely adopted and constantly developed by numerous scholars and engineers including the ﬁnite difference methods (FDM), ﬁnite element methods (FEM), boundary element methods (BEM) and meshless methods [20–25], etc., and a few numerical studies which have focused on the creep property of rock. 2.1. Brief Description of RFPA 2D (Creep) The two-dimensional ﬁnite mode, RFPA 2D, is a numerical simulation tool introduc- ing the homogeneity index, which can be used to analyze progressive internal damage evolution until the macroscopic failure of brittle rock material, and the time effect is fur- ther considered by RFPA-Creep. From a microscopic angle, it is assumed that the model medium is composed of rectangular elements on the same scale and that the statistical distribution of elemental mechanical properties obeys a Weibull distribution [33]: ϕ(α) = m α0 α α0 m−1 exp − α α0 m (1) (1) where α is a mechanical property where strength and elastic modulus are set as the same distribution; α0 is a mean value of the corresponding parameter and m refers to the homogeneity index of rock. The rock material becomes more homogeneous with a larger m (Figure 1) [28]. The mechanical behavior of microscopic elements is modeled by the approach on damage mechanics. The initiation of micro-scale failure occurs after the stress state of an element meets a strength criterion such as the Coulomb criterion. The elastic modulus of an element is assumed to degrade gradually during damage evolution and can be described by: (2) E = (1 −D)E0 (2) where D is the damage variable; E and E0 are the elastic modulus of the degraded and original element, respectively. In this model, the compressive stress is set as the positive and the tensile stress is negative (Figure 2). Both shear and tensile failure patterns are considered where the former mode appears when the compressive stress of an element satisﬁes the Mohr–Coulomb failure criterion: σ = σ1 −σ3 1 + sinϕ 1 −sinϕ ≥σc (3) (3) Materials 2021, 14, 6818 3 of 17 where σ1 and σ3 represent the major and minor principal stress, respectively; σc is the compressive strength and ϕ is the internal friction angle. An element fails in tensile pattern when the minor principal stress reaches and exceeds the tensile strength σt, which is expressed as follows: EVIEW 3 of 18 where σ1 and σ3 represent the major and minor principal stress, respectively; σc is the compressive strength and ϕ is the internal friction angle. An element fails in tensile pattern when the minor principal stress reaches and exceeds the tensile strength σt, which is expressed as follows: VIEW 3 of 18 (4) σ3 ≤−σt (4) Figure 1. 2.1. Brief Description of RFPA 2D (Creep) Mechanical property distribution for five various homogeneous specimens (both strength and elastic modulus follow the same distribution). Figure 1. Mechanical property distribution for ﬁve various homogeneous specimens (both strength and elastic modulus follow the same distribution). Figure 1. Mechanical property distribution for five various homogeneous specimens (both strength and elastic modulus follow the same distribution). The mechanical behavior of microscopic elements is modeled by the approach on damage mechanics. The initiation of micro‐scale failure occurs after the stress state of an element meets a strength criterion such as the Coulomb criterion. The elastic modulus of an element is assumed to degrade gradually during damage evolution and can be de‐ scribed by:   0 1 E D E   (2) where D is the damage variable; E and E0 are the elastic modulus of the degraded and original element, respectively. In this model, the compressive stress is set as the positive and the tensile stress is negative (Figure 2). Both shear and tensile failure patterns are considered where the for‐ mer mode appears when the compressive stress of an element satisfies the Mohr–Cou‐ lomb failure criterion: Figure 1. Mechanical property distribution for five various homogeneous specimens (both strengt and elastic modulus follow the same distribution). Figure 1. Mechanical property distribution for ﬁve various homogeneous specimens (both strength and elastic modulus follow the same distribution). negative (Figure 2). Both shear and tensile failure patterns are considered where the fo mer mode appears when the compressive stress of an element satisfies the Mohr–Cou lomb failure criterion: Figure 1. Mechanical property distribution for five various homogeneous specimens (both strength and elastic modulus follow the same distribution). Figure 1. Mechanical property distribution for ﬁve various homogeneous specimens (both strength and elastic modulus follow the same distribution). negative (Figure 2). Both shear and tensile failure patterns are considered where the for mer mode appears when the compressive stress of an element satisfies the Mohr–Cou‐ lomb failure criterion: The mechanical behavior of microscopic elements is modeled by the approach on damage mechanics. The initiation of micro‐scale failure occurs after the stress state of an element meets a strength criterion such as the Coulomb criterion. 2.1. Brief Description of RFPA 2D (Creep) In RFPA‐Creep, the long‐term strength of rock is introduced, representing that as In RFPA-Creep, the long-term strength of rock is introduced, representing that as time goes by, the strength of rock material degrades under a constant stress level. The creep constitutive relationship for microscopic elements is described as follows (Figure 3) [32]: p g g p g time goes by, the strength of rock material degrades under a constant stress level. The creep constitutive relationship for microscopic elements is described as follows (Figure 3) [32]: σ = σ∞+ (σ0 −σ∞) exp(−Bt) (7)     Bt σ σ σ σ       exp 0 (7) (7)(7) where σ0 and σ∞separately represent the short-term and long-term strength of an element and B is the related attenuation coefﬁcient. where σ0 and σ∞ separately represent the short‐term and long‐term strength of an element and B is the related attenuation coefficient. where σ0 and σ∞separately represent the short-term and long-term strength of an element and B is the related attenuation coefﬁcient. where σ0 and σ∞ separately represent the short‐term and long‐term strength of an element and B is the related attenuation coefficient. Figure 3. Constitutive law for strength degradation of the micro element. Figure 3. Constitutive law for strength degradation of the micro element. Figure 3. Constitutive law for strength degradation of the micro element Figure 3. Constitutive law for strength degradation of the micro element. It has been found that the degradation mechanism of elastic modulus is similar to that of strength [34]. Thus, it is assumed that the elastic modulus of rock obeys the same l It has been found that the degradation mechanism of elastic modulus is similar to that of strength [34]. Thus, it is assumed that the elastic modulus of rock obeys the same law:    t B E E E E ’     exp 0 (8) E = E∞+ (E0 −E∞) exp −B′t (8) (8) (8)       p 0 (8) where E0 and E∞represent the short-term and long-term elastic modulus of an element, respectively, and B′ is the related attenuation coefﬁcient. Both B and B′ are assumed to be the same in this model. 2.1. Brief Description of RFPA 2D (Creep) The elastic modulus o an element is assumed to degrade gradually during damage evolution and can be de scribed by:   0 1 E D E   (2 where D is the damage variable; E and E0 are the elastic modulus of the degraded and original element, respectively. In this model, the compressive stress is set as the positive and the tensile stress i negative (Figure 2). Both shear and tensile failure patterns are considered where the for mer mode appears when the compressive stress of an element satisfies the Mohr–Cou lomb failure criterion: Figure 2. Elastic–brittle damage constitutive law of the micro element. Figure 2. Elastic–brittle damage constitutive law of the micro element. Figure 2. Elastic–brittle damage constitutive law of the micro element. Figure 2. Elastic–brittle damage constitutive law of the micro element. Figure 2. Elastic–brittle damage constitutive law of the micro element. Figure 2. Elastic–brittle damage constitutive law of the micro element. c 3 1 sin 1 sin 1 σ φ φ σ σ σ      (3) The damage variable of an element with a compressive and tensile state can be, respectively, described as [29]: sin 1 φ D = 0 1 −σcr E0ε (ε<εc0) (εc0 ≤εr) (5) D =    0 1 −σtr E0ε 1 (ε<εt0) (εtu<ε ≤εt0) (ε ≤εtu) (6) (5) (6) Materials 2021, 14, 6818 4 of 17 (6) where σcr and σtr are the residual compressive and tensile strengths, respectively; εc0 and εt0 are the threshold compressive and tensile strains, respectively; εtu is the ultimate tensile strain and εr represents the residual strain. p g p y εt0 are the threshold compressive and tensile strains, respectively; εtu is the ultimate tensile strain and εr represents the residual strain. It should be noted that the tensile criterion is given priority over the Mohr–Coulomb It should be noted that the tensile criterion is given priority over the Mohr–Coulomb failure criterion for elements in this model, owing to the short compressive strength generally far outweighing the tensile one for brittle rock material. The Mohr–Coulomb criterion is adopted when the tensile criterion is not met for an element. failure criterion for elements in this model, owing to the short compressive strength gen‐ erally far outweighing the tensile one for brittle rock material. The Mohr–Coulomb crite‐ rion is adopted when the tensile criterion is not met for an element. 2.1. Brief Description of RFPA 2D (Creep) Figure 4 exhibits the result of a uniaxial creep test for deep quartz sandstone taken from the Laobishan Tunnel at a low stress level set as 70.2 MPa and the corresponding numerical simulation. The specimen did not rupture and the evolution curve of axial strain with time nearly matched the numerical result. It can be observed that the homogeneity index m is 1, indicating that the rock specimen model is relatively heterogeneous. The signiﬁcant difference between quartz sandstone specimens in mineral ﬁrmness and distribution and the particle size also gives rise to the heterogeneity of rock material on a microscopic scale. In addition, the mean value is used for the mechanical property of the micro element, and the size and shape, such as rectangle or triangle, of the element in model is invariable and regular, which is in contrast to the rock entity material whose micro-particle form and internal structure are uncertain. This may explain the slight difference in deformation between experimental and numerical results. Hence, RFPA-Creep (2D) can be applied to the simulation of time-dependent behavior concerning the homogeneity of rock material. 5 of 17 g A‐Creep h Materials 2021, 14, 6818 Figure 4. Comparison between experimental and numerical results. Figure 4. Comparison between experimental and numerical results. d l Figure 4. Comparison between experimental and numerical results Figure 4. Comparison between experimental and numerical results. In this study (Table 1 and Figure 5), the creep model dimension was 120 mm × 50 mm and the mesh was discretized into 240 × 100 = 24,000 elements. The specimen geometry was 100 mm × 50 mm (200 × 100 = 20,000 elements), and the loading plates with extremely high stiffness and homogeneity were set on the top and bottom of the specimen during the uniaxial creep loading, respectively. The plane stress compression was performed on all specimen models. The homogeneity indexes m of 1.5, 2, 2.5, 3 and 5 were selected fo unconfined simulation conditions, and for triaxial cases with m = 2, 5, the confining pres sure Pc (σ3) was set as 1.5 and 5 MPa. 2.1. Brief Description of RFPA 2D (Creep) Based on the simulation results of the uniaxial com pression process (Figure 6), the applied stress levels were set as 0.55σc, 0.65σc, 0.75σc and 0.85σc before failure at the last constant loading stage, which was also used for the triaxia creep simulation In this study (Table 1 and Figure 5), the creep model dimension was 120 mm × 50 mm and the mesh was discretized into 240 × 100 = 24,000 elements. The specimen geometry was 100 mm × 50 mm (200 × 100 = 20,000 elements), and the loading plates with extremely high stiffness and homogeneity were set on the top and bottom of the specimen during the uniaxial creep loading, respectively. The plane stress compression was performed on all specimen models. The homogeneity indexes m of 1.5, 2, 2.5, 3 and 5 were selected for unconﬁned simulation conditions, and for triaxial cases with m = 2, 5, the conﬁning pressure Pc (σ3) was set as 1.5 and 5 MPa. Based on the simulation results of the uniaxial compression process (Figure 6), the applied stress levels were set as 0.55σc, 0.65σc, 0.75σc and 0.85σc before failure at the last constant loading stage, which was also used for the triaxial creep simulation. Table 1. Rock material parameters of the numerical model. Homogeneity Index (m) 1.5, 2, 2.5, 3, 5 (Uniaxial) 2, 5 (Triaxial) Mean compressive strength (σ0/MPa) 500 Mean elastic modulus (E0/MPa) 65,000 Poisson’s ratio (µ) 0.28 Friction angle (ψ/◦) 30 Ratio of compression and tension strength (σc/σt) 10 Coefﬁcient of residual strength 0.1 Attenuation coefﬁcient of strength 0.1 Attenuation coefﬁcient of elastic modulus 0.1 Ratio of long-term strength and short-term strength (σ∞/σc) 0.7 Table 1. Rock material parameters of the numerical model. Table 1. Rock material parameters of the numerical model. Table 1. Rock material parameters of the numerical model. 6 of 17 Materials 2021, 14, 6818 Figure 5. Stress–strain response and stages of rock during progressive fracture and classical creep behavior corresponding to the numerical model. Figure 5. Stress–strain response and stages of rock during progressive fracture and classical creep behavior corresponding to the numerical model. g p g g p g p p g to the numerical model. Table 1. Rock material parameters of the numerical model. 2.1. Brief Description of RFPA 2D (Creep) Homogeneity Index (m) 1.5, 2, 2.5, 3, 5 (Unia 2, 5 (Triaxial) Mean compressive strength (σ0/MPa) 500 Mean elastic modulus (E0/MPa) 65,000 Poisson’s ratio (μ) 0.28 Friction angle (ψ/°) 30 Ratio of compression and tension strength (σc/σt) 10 Coefficient of residual strength 0.1 Attenuation coefficient of strength 0.1 Attenuation coefficient of elastic modulus 0.1 Ratio of long‐term strength and short‐term strength (σ∞/σc) 0.7 Furthermore, in order to calibrate the numerical specimen model, the partially parameters of critical mechanical properties were set the same as those in the labor test or the setup in the model of RFPA 2D [28,32,35]. Figure 6. Uniaxial compressive simulation results for various homogeneity indexes. Figure 6. Uniaxial compressive simulation results for various homogeneity indexes. test or the setup in the model of RFPA 2D [28,32,35]. Figure 6. Uniaxial compressive simulation results for various homogeneity indexes. Figure 6. Uniaxial compressive simulation results for various homogeneity indexes. Furthermore, in order to calibrate the numerical specimen model, the partially input parameters of critical mechanical properties were set the same as those in the laboratory test or the setup in the model of RFPA 2D [28,32,35]. Furthermore, in order to calibrate the numerical specimen model, the partially input parameters of critical mechanical properties were set the same as those in the laboratory test or the setup in the model of RFPA 2D [28,32,35]. 2.1. Brief Description of RFPA 2D (Creep) Homogeneity Index (m) 1.5, 2, 2.5, 3, 5 (Uniax 2, 5 (Triaxial) Mean compressive strength (σ0/MPa) 500 Mean elastic modulus (E0/MPa) 65,000 Poisson’s ratio (μ) 0.28 Friction angle (ψ/°) 30 Ratio of compression and tension strength (σc/σt) 10 Coefficient of residual strength 0.1 Attenuation coefficient of strength 0.1 Attenuation coefficient of elastic modulus 0.1 Ratio of long‐term strength and short‐term strength (σ∞/σc) 0.7 Furthermore, in order to calibrate the numerical specimen model, the partially in parameters of critical mechanical properties were set the same as those in the labora test or the setup in the model of RFPA 2D [28,32,35]. lib t th Figure 5. Stress–strain response and stages of rock during progressive fracture and classical creep behavior corresponding to the numerical model. Figure 5. Stress–strain response and stages of rock during progressive fracture and classical creep behavior corresponding to the numerical model. Furthermore, in order to calibrate the numerical specimen model, the partially parameters of critical mechanical properties were set the same as those in the labor test or the setup in the model of RFPA 2D [28,32,35]. Table 1. Rock material parameters of the numerical model. Homogeneity Index (m) 1.5, 2, 2.5, 3, 5 (U 2, 5 (Triaxi Mean compressive strength (σ0/MPa) 500 Mean elastic modulus (E0/MPa) 65,000 Poisson’s ratio (μ) 0.28 Friction angle (ψ/°) 30 Ratio of compression and tension strength (σc/σt) 10 Coefficient of residual strength 0.1 Attenuation coefficient of strength 0.1 Attenuation coefficient of elastic modulus 0.1 Ratio of long‐term strength and short‐term strength (σ∞/σc) 0.7 Furthermore, in order to calibrate the numerical specimen model, the partia parameters of critical mechanical properties were set the same as those in the la test or the setup in the model of RFPA 2D [28,32,35]. Figure 6. Uniaxial compressive simulation results for various homogeneity indexes. Figure 6. Uniaxial compressive simulation results for various homogeneity indexes. Table 1. Rock material parameters of the numerical model. 3. Numerical Results It can be seen that the transition from the secondary to tertiary creep is not distinct when a sudden failure commenced with dramatic deformation. At the same time, the lat‐ eral strain of the specimen was more prominent than the axial strain during the fracturing process. Figure 7. Behavior of creep strain at the last stress level for various homogeneity indexes during uniaxial creep loading. Figure 7. Behavior of creep strain at the last stress level for various homogeneity indexes during uniaxial creep loading. Figure 7. Behavior of creep strain at the last stress level for various homogeneity indexes during uniaxial creep loading. Figure 7. Behavior of creep strain at the last stress level for various homogeneity indexes during uniaxial creep loading. The last constant applied stress that triggers the creep failure increased with homo geneity index m, suggesting that a higher failure strength of rock is enhanced by a higher homogeneity of structural composition. The corresponding axial strain also presented an increasing mode with m, which was more notable for the secondary creep, and the latera deformation displayed the same law as the axial one. The last constant applied stress that triggers the creep failure increased with homo- geneity index m, suggesting that a higher failure strength of rock is enhanced by a higher homogeneity of structural composition. The corresponding axial strain also presented an increasing mode with m, which was more notable for the secondary creep, and the lateral deformation displayed the same law as the axial one. With the condition where m is equal to 2, the variation for axial and lateral creep strains with time under the effect of confining pressure showed a similar tendency to the unconfined case (Figure 8). When the confining pressure increased, both the axial and lateral deformation increased with the deviatoric stress level, and the continuous duration for the last deviatoric stress stage was the longest (Pc = 5 MPa). With the condition where m is equal to 2, the variation for axial and lateral creep strains with time under the effect of conﬁning pressure showed a similar tendency to the unconﬁned case (Figure 8). When the conﬁning pressure increased, both the axial and lateral deformation increased with the deviatoric stress level, and the continuous duration for the last deviatoric stress stage was the longest (Pc = 5 MPa). VIEW 8 of 18 Figure 8. 3. Numerical Results During the uniaxial loading phase, the creep failure emerged when the last stress level reached 0.92σc for m = 1.5 (74 MPa), 2 (90 MPa) and 5 (166 MPa), and the applied loads when m was both both 2.5 (105 MPa) and 3 (124 MPa) were 0.95σc. In the triaxial creep condition, when the conﬁning pressure was 1.5 MPa, the specimen entered into the fracturing stage at applied stresses that were 96 MPa and 178 MPa (both σa = 0.91σc) for m = 2 and 5, respectively, and when Pc = 5 MPa, the last load levels were 109 MPa (m = 2) and 206 MPa (m = 5), and both exceeded 0.9σc. In addition, the duration for all specimens during the unchanging high stress level which led to the failure was within 2 d, except for the unconﬁned condition, with m = 3 lasting 2.4 d. Fi 6 U i i l i i l ti lt f i h it i d g Under the unconﬁned creep loading condition, the time-dependent evolution of axial and lateral strains for various m is shown in Figure 7. All the specimens exhibited Materials 2021, 14, 6818 7 of 17 t for the instantaneous elastic deformation after initial loading and then entered into the phase of primary (transient or attenuated) creep, whose duration was extremely short. The strain rate declined rapidly and stayed almost constant at the steady-state (secondary) creep stage. In the end, all the specimens ruptured with the prominent strain accompanied by the increasing strain rate within quite a short period, marking the tertiary (accelerating) creep stage. It can be seen that the transition from the secondary to tertiary creep is not distinct when a sudden failure commenced with dramatic deformation. At the same time, the lateral strain of the specimen was more prominent than the axial strain during the fracturing process. taneous elastic deformation after initial loading and then entered into the phase of pri‐ mary (transient or attenuated) creep, whose duration was extremely short. The strain rate declined rapidly and stayed almost constant at the steady‐state (secondary) creep stage In the end, all the specimens ruptured with the prominent strain accompanied by the in‐ creasing strain rate within quite a short period, marking the tertiary (accelerating) creep stage. 3. Numerical Results Behavior of creep strain at the last stress level for the constant homogeneity index (m = 2). 4 Di i Figure 8. Behavior of creep strain at the last stress level for the constant homogeneity index (m = 2). Figure 8. Behavior of creep strain at the last stress level for the constant homogeneity index (m = 2). Figure 8. Behavior of creep strain at the last stress level for the constant homogeneity index (m = 2). Materials 2021, 14, 6818 8 of 17 4. Discussions 9 of 17 Materials 2021, 14, 6818 Figure 10. Relationship between stress level and steady creep rate for various homogeneity indexes during uniaxial creep loading. Figure 10. Relationship between stress level and steady creep rate for various homogeneity indexes during uniaxial creep loading. Figure 10. Relationship between stress level and steady creep rate for various homogeneity indexes during uniaxial creep loading. Figure 10. Relationship between stress level and steady creep rate for various homogeneity indexes during uniaxial creep loading. The relationship between deviatoric stress level and axial steady creep rate exhibits a similar increasing law with m to the uniaxial case under the same confining condition. n increases with Pc for the same m (Figure 11), which indicates that the dislocation creep of rock may be more inclined to occur with the increasing confining pressure. The relationship between deviatoric stress level and axial steady creep rate exhibits a similar increasing law with m to the uniaxial case under the same conﬁning condition. n increases with Pc for the same m (Figure 11), which indicates that the dislocation creep of rock may be more inclined to occur with the increasing conﬁning pressure. VIEW 10 of 18 Figure 11. Relationship between stress level and steady creep rate for various homogeneity indexes during triaxial creep loading. Figure 11. Relationship between stress level and steady creep rate for various homogeneity indexes during triaxial creep loading. Figure 11. Relationship between stress level and steady creep rate for various homogeneity indexes during triaxial creep loading. Figure 11. Relationship between stress level and steady creep rate for various homogeneity indexes during triaxial creep loading. The stress level basically moves right by the confining pressure for the invariable homogeneity index, apparently indicating the increase in rock strength under the confin‐ ing condition. This phenomenon can be further explained by the following model [36], where an inclined crack of the length 2a0 with wing cracks can be approximated by a straight crack of the length 2l with a concentrated force F at the center (Figure 12). The stress intensity factor at the crack tips can be described as follows: The stress level basically moves right by the conﬁning pressure for the invariable homogeneity index, apparently indicating the increase in rock strength under the conﬁning condition. 4. Discussions 4.1. Evolution Laws of Stress Level vs. Strain Rate 4.1. Evolution Laws of Stress Level vs. Strain Rate The strain rate during the secondary creep phase is a critical index for the evaluation of long-term stability and creep amount of rock mass. The relationship between applied load level and steady creep rate can be quantitatively represented by: .ε = Aσn (9) (9) .ε = Aσn where A and n are constants, n = 3–8 for dislocation creep and n is approximately equivalent to 1 for diffusion creep. The parameter n is the slope of linear ﬁtting curve (the Logσ–Log .ε plot) converted from the original expression and is also used for the creep of rocks under compression, sometimes being called the stress corrosion index. As presented in Figures 9 and 10, the obtained magnitude order of strain rate reaches 10−8/s. All the correlation coefﬁcients R2 under the unconﬁned condition are greater than 0.8, which reﬂects that the corresponding relationships can be well depicted by the model. The ﬁtting relation between the last stress level and axial steady creep rate shows an increasing trend with homogeneity index m, indicating that the deformation is more signiﬁcant for more homogeneous rock material, which further impacts the formation of a damage zone at the accelerating creep stage. For all the specimens, n ranges from 1 to 2 and the maximum is 1.7336 for m = 5, suggesting that diffusion creep may occur at a low stress level and gradually transition to dislocation creep with the increase in failure strength that is caused by a higher homogeneity of rock. Moreover, for the constant homogeneity index, the steady creep rate increases with stress level and reaches the maximum at the last ﬁxed load that induces failure, which shows that the deformation of rock during the steady-state creep phase depends heavily on the applied stress level. EVIEW 9 of 1 Figure 9. Relationship between the last stress level and steady creep rate for various homogeneity indexes during uniaxial creep loading. Figure 9. Relationship between the last stress level and steady creep rate for various homogeneity indexes during uniaxial creep loading. Figure 9. Relationship between the last stress level and steady creep rate for various homogeneit indexes during uniaxial creep loading. Figure 9. Relationship between the last stress level and steady creep rate for various homogeneity indexes during uniaxial creep loading. 4. Discussions This phenomenon can be further explained by the following model [36], where an inclined crack of the length 2a0 with wing cracks can be approximated by a straight crack of the length 2l with a concentrated force F at the center (Figure 12). The stress intensity factor at the crack tips can be described as follows: l π σ l π F K 3 I   (10) KI = F √ πl −σ3 √ πl (10) (10 (10) where where   α α σ σ a F cos sin 2 2   (11) F = 2a0(σ1 −σ3) sin 2α cos α (11)   α α σ σ a F cos sin 2 2 (11) F = 2a0(σ1 −σ3) sin 2α cos α (11) (11) 10 of 17 Materials 2021, 14, 6818 Figure 12. Approximation of an inclined crack with wing cracks by a straight crack with a concen‐ trated force at the center. Figure 12. Approximation of an inclined crack with wing cracks by a straight crack with a concen- trated force at the center. Figure 12. Approximation of an inclined crack with wing cracks by a straight crack with a concen trated force at the center. Figure 12. Approximation of an inclined crack with wing cracks by a straight crack with a concen- trated force at the center. For simplicity, it is assumed that there is no friction between the inclined crack sur‐ faces. Equations (10) and (11) suggest that the stress intensity factor at the wing crack tip decreased by the confining pressure Pc (σ3) for the same axial‐applied stress σ1. For simplicity, it is assumed that there is no friction between the inclined crack sur- faces. Equations (10) and (11) suggest that the stress intensity factor at the wing crack tip decreased by the conﬁning pressure Pc (σ3) for the same axial-applied stress σ1. 4.2. Characteristics of Dilatancy The evolution law of dilatancy is an effective indicator of the instability and failure of brittle rocks. Time-dependent deformation comprises the elastic and inelastic stages where the dilatancy can be reﬂected by the inelastic volumetric deformation resulting from the development of microcracks in the interior of the rock under different load levels. It is assumed that both elastic modulus E and Poisson’s ratio ν are the same in all specimen models; the inelastic shear strain εie q and the inelastic volumetric strain εie v can be separately expressed as follows: εie q = εq −2 √ 2 3 1 + υ E (σ1 −σ3) (12) εie v = εv −1 −2υ E (σ1 + 2σ3) (13) εq = 2 √ 2 3 (ε1 + ε3) (14) εv = ε1 + 2ε3 (15) (12) (13) (14) (15) and for the uniaxial case (σ3 = 0 MPa): and for the uniaxial case (σ3 = 0 MPa): εie q = εq −2 √ 2 3 1 + υ E σ1 (16) εie v = εv −1 −2υ E σ1 (17) (16) εie v = εv −1 −2υ E σ1 (17) (17) where εq and εv represent the ﬁnal shear strain and ﬁnal volumetric strain, respectively. q The relationship between inelastic shear strain and inelastic volumetric strain can be further illustrated by the dilatancy index which is the slope of strain path deﬁning the ﬂow law of rocks during the creep process [11,35]: 1 DI = dεie q dεiev (18) (18) Materials 2021, 14, 6818 11 of 17 11 of 17 where dεie q and dεie v are the increment of inelastic shear strain and inelastic volumetric strain, and it should be noted that the former is always positive and, due to the expansion of rock, the latter is always negative. Thus, the absolute value of 1/DI is more concise for interpretation. At the last ﬁxed stress, the relationship between inelastic shear strain and inelastic volumetric strain with various m is presented in Figure 13. The strain path displays an almost linear form, suggesting that both the slipping along inclined cracks and the propagation of axial cracks may commence from the initiation of rupture. This is similar to the previously observed experimental phenomena [11,35]. EVIEW 12 of 1 Figure 13. Strain path in inelastic shear strain vs. volumetric strain space for various homogeneity indexes during uniaxial creep loading. Figure 13. 4.2. Characteristics of Dilatancy Strain path in inelastic shear strain vs. volumetric strain space for various homogeneity indexes during uniaxial creep loading. Figure 13. Strain path in inelastic shear strain vs. volumetric strain space for various homogeneit indexes during uniaxial creep loading. Figure 13. Strain path in inelastic shear strain vs. volumetric strain space for various homogeneity indexes during uniaxial creep loading. Figure 13. Strain path in inelastic shear strain vs. volumetric strain space for various homogenei indexes during uniaxial creep loading. Figure 13. Strain path in inelastic shear strain vs. volumetric strain space for various homogeneity indexes during uniaxial creep loading. As shown in Figure 13, the dilatancy index 1/DI increases with the homogeneity in dex m. The reason for this phenomenon is that larger increments of applied stress with higher m lead to larger increments of inelastic shear strain, and DI declines. For the con stant homogeneity index (m = 2) (Figure 14), a similar slight decreasing tendency of DI i also attributed to the increase in inelastic shear strain increments when the increment‐of stress level ranges from a low to high degree, which indicates that when the rock materia approaches failure, DI reaches the minimum at a certain high stress level. As shown in Figure 13, the dilatancy index 1/DI increases with the homogeneity index m. The reason for this phenomenon is that larger increments of applied stress with higher m lead to larger increments of inelastic shear strain, and DI declines. For the constant homogeneity index (m = 2) (Figure 14), a similar slight decreasing tendency of DI is also attributed to the increase in inelastic shear strain increments when the increment-of-stress level ranges from a low to high degree, which indicates that when the rock material approaches failure, DI reaches the minimum at a certain high stress level. pp g Figure 15 presents the variation of 1/DI with m under the impact of conﬁning pressure. For the same m, 1/DI gradually increases with Pc. The inelastic volumetric strain in the uniaxial case is smaller than that with Pc = 5 MPa and larger than that with Pc = 1.5 MPa, which may result from the close applied load level caused by quite a low conﬁned condition for hard and brittle rock specimen. 4.2. Characteristics of Dilatancy Furthermore, the evolution law of 1/DI with increasing m under unchanging conﬁning pressure is similar to the uniaxial case and is more marked than that under the unconﬁned condition. 4.3. Failure Pattern When the homogeneity index m is equal to 2, the evolution of creep strain at different stress levels according to a similar increment of 10% σc is displayed in Figure 16. The accelerating creep is not observed until the last stress level surpasses 90% σc (σa = 0.92σc), resulting in the ultimate rupture of the rock specimen, which basically agrees with the experimental result by Ma, 2004 [37]. 12 of 17 y of DI is ment‐of‐ Materials 2021, 14, 6818 Figure 14. Strain path in inelastic shear strain vs. volumetric strain space for the constant homoge‐ neity index (m = 2) during uniaxial creep loading. Figure 14. Strain path in inelastic shear strain vs. volumetric strain space for the constant homogene- ity index (m = 2) during uniaxial creep loading. REVIEW 13 of 18 Figure 14. Strain path in inelastic shear strain vs. volumetric strain space for the constant homoge‐ neity index (m = 2) during uniaxial creep loading. Figure 14. Strain path in inelastic shear strain vs. volumetric strain space for the constant homogene- ity index (m = 2) during uniaxial creep loading. EVIEW 13 of 18 Figure 15 presents the variation of 1/DI with m under the impact of confining pres‐ sure. For the same m, 1/DI gradually increases with Pc. The inelastic volumetric strain in the uniaxial case is smaller than that with Pc = 5 MPa and larger than that with Pc = 1.5 MPa, which may result from the close applied load level caused by quite a low confined condition for hard and brittle rock specimen. Furthermore, the evolution law of 1/DI with increasing m under unchanging confining pressure is similar to the uniaxial case and is more marked than that under the unconfined condition. Figure 15. Strain path in inelastic shear strain vs. volumetric strain space for various homogeneit indexes during triaxial creep loading. Figure 15. Strain path in inelastic shear strain vs. volumetric strain space for various homogeneity indexes during triaxial creep loading. Figure 15. Strain path in inelastic shear strain vs. volumetric strain space for various homogeneit indexes during triaxial creep loading. Figure 15. Strain path in inelastic shear strain vs. volumetric strain space for various homogeneity indexes during triaxial creep loading. 4.3. 4.3. Failure Pattern Failure Pattern When the homogeneity index m is equal to 2, the evolution of creep strain at differen stress levels according to a similar increment of 10% σc is displayed in Figure 16. The ac celerating creep is not observed until the last stress level surpasses 90% σc (σa = 0.92σc resulting in the ultimate rupture of the rock specimen, which basically agrees with th experimental result by Ma, 2004 [37]. With the generation of new cracks and the growth of original cracks inside rock specimens, internal damage continuously accumulates, which dominates progressive deformation, leading to a sudden failure. Corresponding to Figure 16, Figure 17 exhibits the damage accumulation process in the interior of rock material with an increasing applied load for m = 2. When σa reaches 0.55σc (54 MPa), a few new cracks appear and almost no new damage is generated, showing that the compaction on the microcracks and the specimen stays in a stable state after the self-adjustment of internal stress. At 74 MPa (0.75σc), the microcracks tend to aggregate and some local damage zones appear, which indicates that the specimen is starting to enter the unsteady stage. With the further accumulation of damage until σa = 0.92σc (90 MPa), the intensity of microcracks exceeds a critical density, triggering the accelerating creep, during which the microcracks extensively propagate and swiftly coalesce until the macroscopic failure region emerges. It can be found that at low loading stages, the local failure area ﬁrst appears after reaching the yield strength of the specimen owing to the low homogeneity of internal structure, inducing internal damage accumulation, which further results in the increase in anisotropy of the specimen, Materials 2021, 14, 6818 13 of 17 13 of 17 and the main fracture zone gradually appears and extends with elapsed time. At the same time, as the applied stress increases, the tensional stress along the axial direction after stress redistribution reaches the tensile strength, producing local axial splitting failure areas. e e e a o i g o a i i a i e e o 0% i i p aye i igu e 6 e a celerating creep is not observed until the last stress level surpasses 90% σc (σa = 0.92σc), resulting in the ultimate rupture of the rock specimen, which basically agrees with the experimental result by Ma, 2004 [37]. Figure 16. 4.3. Failure Pattern Behavior of creep strain with various stress levels for the constant homogeneity index (m = 2) during uniaxial creep loading. Figure 16. Behavior of creep strain with various stress levels for the constant homogeneity index (m = 2) during uniaxial creep loading. 021, 14, x FOR PEER REVIEW damage until σa = 0.92σc (90 MPa), the intensity of microcracks exceeds a critic triggering the accelerating creep, during which the microcracks extensively and swiftly coalesce until the macroscopic failure region emerges. It can be fou low loading stages, the local failure area first appears after reaching the yield s the specimen owing to the low homogeneity of internal structure, inducing inte age accumulation, which further results in the increase in anisotropy of the spec the main fracture zone gradually appears and extends with elapsed time. At time, as the applied stress increases, the tensional stress along the axial dire stress redistribution reaches the tensile strength, producing local axial splitting eas. Figure 16. Behavior of creep strain with various stress levels for the constant homogeneity index (m = 2) during uniaxial creep loading. Figure 16. Behavior of creep strain with various stress levels for the constant homogeneity index (m = 2) during uniaxial creep loading. time, as the applied stress increases, the tensional stress along the axial dire stress redistribution reaches the tensile strength, producing local axial splitting eas. With the generation of new cracks and the growth of original cracks inside rock spec imens, internal damage continuously accumulates, which dominates progressive defor mation, leading to a sudden failure. Corresponding to Figure 16, Figure 17 exhibits th damage accumulation process in the interior of rock material with an increasing applied load for m = 2. When σa reaches 0.55σc (54 MPa), a few new cracks appear and almost n new damage is generated, showing that the compaction on the microcracks and the spec imen stays in a stable state after the self‐adjustment of internal stress. At 74 MPa (0.75σc the microcracks tend to aggregate and some local damage zones appear, which indicate that the specimen is starting to enter the unsteady stage. With the further accumulation o Figure 17. Damage accumulation for the constant homogeneity index (m = 2) during uniaxial creep loading. The creep failure pattern for various homogeneity indexes under the uncon Figure 17. Damage accumulation for the constant homogeneity index (m = 2) during uniaxial creep loading. Figure 17. 4.3. Failure Pattern Damage accumulation for the constant homogeneity index (m = 2) during uniaxial creep loading. The creep failure pattern for various homogeneity indexes under the uncon Figure 17. Damage accumulation for the constant homogeneity index (m = 2) during uniaxial creep loading. ditions is shown in Figure 18a. Most rock specimens present the nearly penetra type planes such as the case that m = 2, 3, and for m = 1.5, 3 and 5, the domin mode is accompanied by partial splitting. The effect of homogeneity on failu after uniaxial creep loading is not conspicuous in this simulation due to the dif the internal heterogeneity of specimens on the microscopic scale. However, ba statistical homogeneity from the macroscopic angle, the specimen under con compression shows the splitting‐based mode compared to the creep conditio 18b), and both cases are similar to the failure pattern of marble described by Z 2012 [5], which suggests that the creep failure of rock is a progressive proce which the circulation is between internal damage accumulation and the exp weak yield areas, with the continuous evolution of microcracks until eventual r The creep failure pattern for various homogeneity indexes under the unconﬁned conditions is shown in Figure 18a. Most rock specimens present the nearly penetrated shear-type planes such as the case that m = 2, 3, and for m = 1.5, 3 and 5, the dominated shear mode is accompanied by partial splitting. The effect of homogeneity on failure pattern after uniaxial creep loading is not conspicuous in this simulation due to the difference in the internal heterogeneity of specimens on the microscopic scale. However, based on the statistical homogeneity from the macroscopic angle, the specimen under conventional compression shows the splitting-based mode compared to the creep condition (Figure 18b), and both cases are similar to the failure pattern of marble described by Zhao et al., 2012 [5], which suggests that the creep failure of rock is a progressive process during which the Materials 2021, 14, 6818 14 of 17 14 of 17 circulation is between internal damage accumulation and the expansion of weak yield areas, with the continuous evolution of microcracks until eventual rupture. 021, 14, x FOR PEER REVIEW circulation is between internal damage accumulation and the expansion of weak yield areas, with the continuous evolution of microcracks until eventual rupture. 4.3. Failure Pattern 021, 14, x FOR PEER REVIEW (a) Uniaxial creep condition (b) Conventional uniaxial condition ure 18. Damage accumulation for various homogeneity indexes under uniaxial creep (a) and conventional compr nditions (b). Under the impact of confining pressure, the creep failure type with m = shown in Figure 19. Both specimens under the confining pressure of 1.5 and 5 M Figure 18. Damage accumulation for various homogeneity indexes under uniaxial creep (a) and conventional compression conditions (b). Under the impact of conﬁning pressure, the creep failure type with m = 2 and 5 is shown in Figure 19 Both specimens under the conﬁning pressure of 1 5 and 5 MPa present (a) Uniaxial creep condition (a) Uniaxial creep condition (a) Uniaxial creep condition (b) Conventional uniaxial condition (b) Conventional uniaxial condition re 18. Damage accumulation for various homogeneity indexes under uniaxial creep (a) and conventional com itions (b). Figure 18. Damage accumulation for various homogeneity indexes under uniaxial creep (a) and conventional compression conditions (b). Under the impact of confining pressure, the creep failure type with m = shown in Figure 19. Both specimens under the confining pressure of 1.5 and 5 M several macroscopic shear failure regions when m = 2. For the higher homogen that is 5, the shear‐based form appears in the case where Pc = 1.5 and 5 MP rupture is mainly located in the upper region of the specimen, connected wit heterogeneity of rock under the time effect to some extent. The fracture of the s prone to shear pattern under the confined condition, which is more evident fo homogeneity of rock material. Under the impact of conﬁning pressure, the creep failure type with m = 2 and 5 is shown in Figure 19. Both specimens under the conﬁning pressure of 1.5 and 5 MPa present several macroscopic shear failure regions when m = 2. For the higher homogeneity index that is 5, the shear-based form appears in the case where Pc = 1.5 and 5 MPa, and the rupture is mainly located in the upper region of the specimen, connected with the local heterogeneity of rock under the time effect to some extent. The fracture of the specimen is prone to shear pattern under the conﬁned condition, which is more evident for a higher homogeneity of rock material. Materials 2021, 14, 6818 15 of 17 gure 19. In o 5. Concluding Remarks the numerical simulation approach, RFPA‐Creep (2D), is applied to the compar ysis on the corresponding time‐dependent behavior and failure mode for hom indexes m = 1.5, 2, 2.5, 3 and 5. The main conclusions of this study are as follow 1. During uniaxial creep loading, the deformation is more significant for m geneous rock material Diffusion creep may occur at low stress levels and In order to clarify the effect of homogeneity on the creep properties of brittle rock, the numerical simulation approach, RFPA-Creep (2D), is applied to the comparative analysis on the corresponding time-dependent behavior and failure mode for homogeneity indexes m = 1.5, 2, 2.5, 3 and 5. The main conclusions of this study are as follows: geneous rock material. Diffusion creep may occur at low stress levels and to dislocation creep with increasing applied loads. The increasing law for cr with the homogeneity index under an unaltered triaxial condition is sim uniaxial case and the dislocation creep may be more inclined to emerge wi ing confining pressure for rock with the same homogeneity. 2. The dilatancy index reaches the maximum at a certain high load level w failure happens, and the evolution of dilatancy index with homogeneity in 1. During uniaxial creep loading, the deformation is more signiﬁcant for more homoge- neous rock material. Diffusion creep may occur at low stress levels and transition to dislocation creep with increasing applied loads. The increasing law for creep strain with the homogeneity index under an unaltered triaxial condition is similar to the uni- axial case and the dislocation creep may be more inclined to emerge with increasing conﬁning pressure for rock with the same homogeneity. the same confining pressure is similar to the uniaxial case and is more than that under the unconfined condition. 3. Both the uniaxial and triaxial creep failure are based on the ductile damag lation inside rock. The dominant shear‐type failure is shown by the unia l di d h i l i h li 2. The dilatancy index reaches the maximum at a certain high load level when creep failure happens, and the evolution of dilatancy index with homogeneity index under the same conﬁning pressure is similar to the uniaxial case and is more prominent than that under the unconﬁned condition. loading manner and the conventional compression case presents the splitt failure form. 4.3. Failure Pattern Damage accumulation at the last stress level for various homogeneity indexes during triaxial creep loa 5. Concluding Remarks Figure 19. Damage accumulation at the last stress level for various homogeneity indexes during triaxial creep loading. re 19. Damage accumulation at the last stress level for various homogeneity indexes during triaxial creep loa 5. Concluding Remarks Figure 19. Damage accumulation at the last stress level for various homogeneity indexes during triaxial creep loading. Conﬂicts of Interest: The authors declare no conﬂict of interest. Conﬂicts of Interest: The authors declare no conﬂict of interest. References (In Chinese) [CrossRef] 6. Yang, Y.S.; Zhou, H.; Zhang, C.Q.; Zhang, K.; Yan, F. Experimental investigation on time-lag failure properties of marble under uniaxial compressive test. Chin. J. Rock Soil Mech. 2011, 32, 2714–2720. (In Chinese) [CrossRef] 6. Yang, Y.S.; Zhou, H.; Zhang, C.Q.; Zhang, K.; Yan, F. Experimental investigation on time-lag failure properties of marble under uniaxial compressive test. Chin. J. Rock Soil Mech. 2011, 32, 2714–2720. (In Chinese) [CrossRef] 7. Li, X.Z. Micromechanics-Based Study on the Progressive and Creep Failure Prop University of Architecture and Technology, Xi’an, Shaanxi, China, 2016. (In Chinese 7. Li, X.Z. Micromechanics-Based Study on the Progressive and Creep Failure Properties of Britt , y g p p University of Architecture and Technology, Xi’an, Shaanxi, China, 2016. (In Chinese) University of Architecture and Technology, Xi an, Shaanxi, China, 2016. (In Chinese) 8. Amitrano, D.; Helmstetter, A. Brittle creep, damage and time to failure in rocks. J. Geophs. Res-Sol. Ea. 2006, 111, B11201. [CrossRef] 8. Amitrano, D.; Helmstetter, A. Brittle creep, damage and time to failure in rocks. J. Geophs. Res-So [CrossRef] 9. Paraskevopoulou, C.; Matthew, P.; Mark, D.; Simon, L.; Tom, L.; Mark, J. Time-dependent behaviour of brittle rocks based on static load laboratory tests. Geotech. Geol. Eng. 2018, 36, 337–376. [CrossRef] y g 10. Fujii, Y.; Kiyama, T.; Ishijima, Y.; Kodama, J. Circumferential strain behavior during creep tests of brittle rocks. Int. J. Rock Mech. Min. Sci. 1999, 36, 323–337. [CrossRef] 11. Lin, Q.X.; Liu, Y.M.; Tham, L.G.; Tang, C.A.; Lee, P.K.K.; Wang, J. Time-dependent strength degradation of granite. Int. J. Rock Mech. Min. Sci. 2009, 46, 1103–1114. [CrossRef] 12. Nicolas, A.; Fortin, J.; Regnet, J.B.; Verberne, B.A.; Plümper, O.; Dimanov, A.; Spiers, C.J.; Guéguen, Y. Brittle and semibrittle creep of tavel limestone deformed at room temperature. J. Geophys. Res. 2017, 112, 4436–4459. [CrossRef] 13. Zhao, T.B.; Zhang, Y.B.; Zhang, Q.Q.; Tan, Y.L. Analysis on the creep response of bolted rock using bolted Eng. 2018, 14, 141–149. [CrossRef] g 14. Liu, J.P.; Xu, S.D.; Li, Y.H. Analysis of rock mass stability according to power–law attenuation characteristics of acoustic emission and microseismic activities. Tunn. Undergr. Sp. Tech. 2019, 83, 303–312. [CrossRef] g p 15. Ashby, M.F.; Sammis, C.G. The damage mechanics of brittle solids in compression. Pure Appl. Geophys. 1990, 133, 489–521. [CrossRef] 16. Brantut, N.; Baud, P.; Heap, M.J.; Meredith, P.G. Micromechanics of brittle creep in rocks. J. Geophys. Res. 2012, 117, B08412. References [CrossRef] 17. Li, X.Z.; Qu, X.L.; Qi, C.Z.; Shao, Z.S. A uniﬁed analytical method calculating brittle rocks deformation induced by crack growth. Int. J. Rock Mech. Min. Sci. 2019, 113, 134–141. [CrossRef] 18. Shi, G.C.; Yang, X.J.; Yu, H.C.; Zhu, C. Acoustic emission characteristics of creep fracture evolution in double-fracture ﬁne sandstone under uniaxial compression. Eng. Fract. Mech. 2018, 210, 13–28. [CrossRef] 19. Tang, Y. Time-Dependent Behavior of Rocks under Triaxial Compression Condition; Chongqing University: (In Chinese) 20. Saez, A.; Dominguez, J. Dynamic crack problems in three-dimensional transversely isotropic solids. Eng. Anal. Bound. Elem. 2001, 25, 203–210. [CrossRef] [ ] 21. Chen, S.; Yue, Z.Q.; Tham, L.G. Digital image based approach for three-dimensional mechanical analysis of heterogeneous rocks. Rock Mech. Rock. Eng. 2007, 40, 145–168. [CrossRef] 22. Dai, Q. Two- and three-dimensional micromechanical viscoelastic ﬁnite element modeling of stone-based materials with X-ray computed tomography images. Constr. Build. Mater. 2011, 25, 1102–1114. [CrossRef] 23. Zhuang, X.Y.; Zhu, H.H.; Augarde, C. An improved meshless Shepard and least squares method possessing the delta property and requiring no singular weight function. Comput. Mech. 2014, 53, 343–357. [CrossRef] g g p antuzzi, N.; Zhao, Y.; Liu, Y.B.; Peng, B.; Chen, J. Investigation of the Flow Properties of CBM Based on Network Modeling. Materials 2019, 12, 2387. [CrossRef] [PubMed] 24. Zhang, B.; Li, Y.; Fantuzzi, N.; Zhao, Y.; Liu, Y.B.; Peng, B.; Chen, J. Investigation of the Flow Stochastic Fracture Network Modeling. Materials 2019, 12, 2387. [CrossRef] [PubMed] g 25. Wang, H.; Li, Y.; Cao, S.G.; Fantuzzi, N.; Pan, R.K.; Tian, M.Y.; Liu, Y.B.; Yang, H.Y. Fracture toughness analysis of HCCD specimens of Longmaxi shale subjected to mixed mode I-II loading. Eng. Fract. Mech. 2020, 239, 107299. [CrossRef] p g j g g 26. Lockner, D.A.; Madden, T.R. A multiple-crack model of brittle fracture: 2. Time-dependent simulations. J. Geophys. Res. 1991, 96, 19643–19654. [CrossRef] 27. Amitrano, D.; Grasso, J.R.; Hantz, D. From diffuse to localized damage through elastic interaction. Geophys. Res. Lett. 1999, 26, 2109–2112. [CrossRef] 27. Amitrano, D.; Grasso, J.R.; Hantz, D. From diffuse to localized damage through elastic interaction. Geophys. Res. Lett. 1999, 26, 2109–2112. [CrossRef] 28. Tang, C.A.; Liu, H.; Lee, P.K.K.; Tsuiet, Y.; Tham, L.G. Numerical studies of the inﬂuence of microstructure on rock failure in uniaxial compression-Part I: Effect of heterogeneity. Int. J. Rock Mech. Min. Sci. 2000, 37, 555–569. [CrossRef] 28. In o 5. Concluding Remarks Under the confining pressure, the creep failure mode tends wards the shear, which is more obvious with a higher homogeneity of rock Author Contributions: Conceptualization, H.‐Z.C. and Z.‐S.S.; data curation, D.‐D.J., Z formal analysis, H.‐Z.C. and D.‐D.J.; funding acquisition, Z.‐S.S.; investigation, D.‐D.J., Z 3. Both the uniaxial and triaxial creep failure are based on the ductile damage accumu- lation inside rock. The dominant shear-type failure is shown by the uniaxial creep loading manner and the conventional compression case presents the splitting-based failure form. Under the conﬁning pressure, the creep failure mode tends more towards the shear, which is more obvious with a higher homogeneity of rock. H.‐Z.C.; writing—review and editing, H.‐Z.C. All authors have read and agreed to the version of the manuscript. Funding: This research was funded by the National Natural Science Foundation of C number No. 11872287, the Found of Shaanxi Key Research and Development Program, ber No. 2019ZDLGY01‐10. Author Contributions: Conceptualization, H.-Z.C. and Z.-S.S.; data curation, D.-D.J., Z.Z., D.-B.Z.; formal analysis, H.-Z.C. and D.-D.J.; funding acquisition, Z.-S.S.; investigation, D.-D.J., Z.Z., D.-B.Z.; methodology, H.-Z.C. and Z.-S.S.; resources, D.-D.J. and Z.Z.; writing—original draft preparation, H.-Z.C.; writing—review and editing, H.-Z.C. All authors have read and agreed to the published version of the manuscript. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: The data presented in this study are available on reque Funding: This research was funded by the National Natural Science Foundation of China, grant number No. 11872287, the Found of Shaanxi Key Research and Development Program, grant number No. 2019ZDLGY01-10. corresponding author. Institutional Review Board Statement: Not applicable. corresponding author. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: The data presented in this study are available on request from the corresponding author. Data Availability Statement: The data presented in this study are available on request from the corresponding author. 16 of 17 16 of 17 Materials 2021, 14, 6818 Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. Singh, S.P. The inﬂuence of rock properties on the occurrence and control of rockbursts. Min. Sci. Technol. 1987, 5, 11–18. [CrossRef] 2. Linkov, A.M. Rockbursts and the instability of rock masses. Int. J. Rock Mech. Min. Sci. Geomech. Abstr. 1996, 33, 727–732. [CrossRef] 1. Singh, S.P. The inﬂuence of rock properties on the occurrence and control of rockbursts. Min. Sci. Technol. 1987, 5, 11–18. [CrossRef] 2. Linkov, A.M. Rockbursts and the instability of rock masses. Int. J. Rock Mech. Min. Sci. Geomech. [CrossRef] 3. He, M.C.; Miao, J.L.; Feng, J.L. Rock burst process of limestone and its acoustic emission characteristics under true-triaxial unloading conditions. Int. J. Rock Mech. Min. Sci. 2010, 47, 286–298. [CrossRef] 3. He, M.C.; Miao, J.L.; Feng, J.L. Rock burst process of limestone and its acoustic emission characteristics under true-triaxial unloading conditions. Int. J. Rock Mech. Min. Sci. 2010, 47, 286–298. [CrossRef] g J 4. Chen, B.R.; Feng, X.T.; Ming, H.J.; Zhou, H.; Zeng, X.H.; Feng, G.L.; Xiao, Y.X. Evolution Law and Mechanism of Rockburst In Deep Tunnel: Time Delayed Rockburst. Chin. J. Rock Mech. Eng. 2012, 31, 561–569. (In Chinese) g 4. Chen, B.R.; Feng, X.T.; Ming, H.J.; Zhou, H.; Zeng, X.H.; Feng, G.L.; Xiao, Y.X. Evolution Law and Me D T l Ti D l d R kb t Chi J R k M h E 2012 31 561 569 (I Chi ) 4. Chen, B.R.; Feng, X.T.; Ming, H.J.; Zhou, H.; Zeng, X.H.; Feng, G.L.; Xiao, Y.X. Evolution Law and Mechanism of Rockburst In Deep Tunnel: Time Delayed Rockburst. Chin. J. Rock Mech. Eng. 2012, 31, 561–569. (In Chinese) 4. Chen, B.R.; Feng, X.T.; Ming, H.J.; Zhou, H.; Zeng, X.H.; Feng, G.L.; Xiao, Y.X. Evolution Law and Me Deep Tunnel: Time Delayed Rockburst. Chin. J. Rock Mech. Eng. 2012, 31, 561–569. (In Chinese) p y g 5. Zhao, X.J.; Chen, B.R.; Zhao, H.B.; Jie, B.H.; Ning, Z.F. Laboratory creep tests for time-dependent properties of a marble in Jinping II hydropower station. J. Rock Mech. Geotech. Eng. 2012, 4, 168–176. [CrossRef] 5. Zhao, X.J.; Chen, B.R.; Zhao, H.B.; Jie, B.H.; Ning, Z.F. Laboratory creep tests for time-dependent properties of a marble in Jinping II hydropower station. J. Rock Mech. Geotech. Eng. 2012, 4, 168–176. [CrossRef] 6. Yang, Y.S.; Zhou, H.; Zhang, C.Q.; Zhang, K.; Yan, F. Experimental investigation on time-lag failure p uniaxial compressive test. Chin. J. Rock Soil Mech. 2011, 32, 2714–2720. Conﬂicts of Interest: The authors declare no conﬂict of interest. References Tang, C.A.; Liu, H.; Lee, P.K.K.; Tsuiet, Y.; Tham, L.G. Numerical studies of the inﬂuence of microstructure on rock failure in uniaxial compression-Part I: Effect of heterogeneity. Int. J. Rock Mech. Min. Sci. 2000, 37, 555–569. [CrossRef] 17 of 17 17 of 17 Materials 2021, 14, 6818 29. Tang, C.A.; Tham, L.G.; Lee, P.K.K.; Yang, T.H.; Li, L.C. Coupled analysis of ﬂow, stress and damage (FSD) in rock failure. Int. J. Rock Mech. Min. Sci. 2002, 39, 477–489. [CrossRef] [ ] 30. Wang, S.Y.; Sloan, S.W.; Huang, M.L.; Tang, C.A. Numerical study of failure mechanism of serial and parallel rock pillars. Rock Mech. Rock Eng. 2011, 44, 179–198. [CrossRef] 30. Wang, S.Y.; Sloan, S.W.; Huang, M.L.; Tang, C.A. Numerical study of failure mechanism of serial and parallel rock pillars. Rock Mech. Rock Eng. 2011, 44, 179–198. [CrossRef] g [ ] 31. Wang, Y.; Li, X.; Zhou, R.Q.; Zheng, B.; Zhang, B.; Wu, Y.F. Numerical evaluation of the effect of fracture network connectivity in naturally fractured shale based on FSD model. Sci. Chin. Earth Sci. 2015, 59, 626–639. [CrossRef] g 31. Wang, Y.; Li, X.; Zhou, R.Q.; Zheng, B.; Zhang, B.; Wu, Y.F. Numerical evaluation of the effect of fracture network connectivity in naturally fractured shale based on FSD model. Sci. Chin. Earth Sci. 2015, 59, 626–639. [CrossRef] y 32. Li, L.C.; Xu, T.; Tang, C.A.; Zhu, L.K. Numerical simulation of creep induced progressive failure process of rock under uniaxial compression. Chin. J. Rock Soil Mech. 2008, 27, 2034–2042. (In Chinese) [CrossRef] y 32. Li, L.C.; Xu, T.; Tang, C.A.; Zhu, L.K. Numerical simulation of creep induced progressive failure process of rock under uniaxial compression. Chin. J. Rock Soil Mech. 2008, 27, 2034–2042. (In Chinese) [CrossRef] p ( ) 33. Weibull, W. A statistical distribution function of wide applicability. J. Appl. Mech. 1951, 18, 293–297. [CrossRef] 34. Xu, H.F. Time dependent behaviors of strength and elasticity modulus of weak rock. Chin. J. Rock Mech. Eng. 1997, 16, 246–251. (In Chinese) [CrossRef] 33. Weibull, W. A statistical distribution function of wide applicability. J. Appl. Mech. 1951, 18, 293–297. [CrossRef] 34. Xu, H.F. Time dependent behaviors of strength and elasticity modulus of weak rock. Chin. J. Rock Mech. Eng. 1997, 16, 246–251. (In Chinese) [CrossRef] , pp y J pp , , [ ] 34. Xu, H.F. Time dependent behaviors of strength and elasticity modulus of weak rock. Chin. J. Rock Mech. Eng. References 1997, 16, 246–251. (In Chinese) [CrossRef] 35. Liu, N.; Zhang, C.S.; Chu, W.J. Experimental research on time-dependent behavior of crack propagation in Jinping deep marble. Chin. J. Rock Soil Mech. 2012, 27, 2034–2042. (In Chinese) 35. Liu, N.; Zhang, C.S.; Chu, W.J. Experimental research on time-dependent behavior of crack propagation in Jinping deep marble. Chin. J. Rock Soil Mech. 2012, 27, 2034–2042. (In Chinese) 36. Horii, H.; Nemat-Nasser, S. Compression-induced microcrack growth in brittle solids-axial splitting and shear failure. J. Geophs. Res.-Sol. Ea. 1985, 90, 3105–3125. [CrossRef] 36. Horii, H.; Nemat-Nasser, S. Compression-in Res.-Sol. Ea. 1985, 90, 3105–3125. [CrossRef] 36. Horii, H.; Nemat-Nasser, S. Compression-induced microcrack growth in brittle solids-axial splitting and shear failure. J. Geophs. Res.-Sol. Ea. 1985, 90, 3105–3125. [CrossRef] Res.-Sol. Ea. 1985, 90, 3105–3125. [CrossRef] , , [ ] 37. Ma, L. Experimental Investigation of Time Dependent Behavior of Welded Tonopah Spring Tuff; ProQuest Information and Learning Company: Ann Arbor, MI, USA, 2004. 37. Ma, L. Experimental Investigation of Time Dependent Behavior of Welded Tonopah Spring Tuff; ProQuest Information and Learning Company: Ann Arbor, MI, USA, 2004.
https://openalex.org/W4361214490	https://www.nature.com/articles/s41598-023-32102-9.pdf	English	null	Impact of radiation dose reduction and iterative image reconstruction on CT-guided spine biopsies	Scientific reports	2,023	cc-by	8,974	Impact of radiation dose reduction and iterative image reconstruction on CT‑guided spine biopsies Karolin J. Paprottka 1, Karina Kupfer 1, Vivian Schultz 1, Meinrad Beer 3, Claus Zimmer 1,2, Thomas Baum 1, Jan S. Kirschke 1,2 & Nico Sollmann 1,2,3 OPEN This study aimed to systematically evaluate the impact of dose reduction on image quality and confidence for intervention planning and guidance regarding computed tomography (CT)-based intervertebral disc and vertebral body biopsies. We retrospectively analyzed 96 patients who underwent multi-detector CT (MDCT) acquired for the purpose of biopsies, which were either derived from scanning with standard dose (SD) or low dose (LD; using tube current reduction). The SD cases were matched to LD cases considering sex, age, level of biopsy, presence of spinal instrumentation, and body diameter. All images for planning (reconstruction: “IMR1”) and periprocedural guidance (reconstruction: “iDose4”) were evaluated by two readers (R1 and R2) using Likert scales. Image noise was measured using attenuation values of paraspinal muscle tissue. The dose length product (DLP) was statistically significantly lower for LD scans regarding the planning scans (SD: 13.8 ± 8.2 mGycm, LD: 8.1 ± 4.4 mGycm, p < 0.01) and the interventional guidance scans (SD: 43.0 ± 48.8 mGycm, LD: 18.4 ± 7.3 mGycm, p < 0.01). Image quality, contrast, determination of the target structure, and confidence for planning or intervention guidance were rated good to perfect for SD and LD scans, showing no statistically significant differences between SD and LD scans (p > 0.05). Image noise was similar between SD and LD scans performed for planning of the interventional procedures (SD: 14.62 ± 2.83 HU vs. LD: 15.45 ± 3.22 HU, p = 0.24). Use of a LD protocol for MDCT-guided biopsies along the spine is a practical alternative, maintaining overall image quality and confidence. Increasing availability of model-based iterative reconstruction in clinical routine may facilitate further radiation dose reductions. Advances in imaging technologies enable a better differentiation between benign and malignant bone lesions, detection of smaller bone lesions by radiologists, and a better diagnosis of different inflammatory processes1,2. However, the comparatively low specificity of imaging modalities for correct diagnosis still demands a histo- pathologic confirmation in indeterminate cases3–5. For obtaining a representative tissue sample, a biopsy via an open surgical access is still considered the reference standard6. However, this open surgical access also has disadvantages, such as high time expense, high cost, prolonged hospitalization, and risks of general anesthesia and major follow-up surgery (e.g., due to complications)7. j p g y ( g p ) As an alternative, imaging-guided percutaneous biopsy is a well-described and successful technique for yielding histopathologic diagnoses8,9. www.nature.com/scientificreports www.nature.com/scientificreports Material and methods Study cohort. We retrospectively reviewed CT-guided intervertebral disc or vertebral bone biopsies, which were performed with two different dose levels: SD and LD scans for planning and procedure performance. A general adjustment of our institutional CT protocols took place in October 2020, hence all LD scans that were included in our study were acquired between November 2020 and June 2021, while all SD scans were derived from the interval between January 2020 and September 2020. The adjustment of scanning parameters was based on recent simulation studies from data of the herein used scanner regarding feasibility of LD imaging for the purpose of various clinical applications16,17,21,22. p p pp Patients were consecutively included if they had multi-detector CT (MDCT) scanning available for a biopsy of the vertebral bone or intervertebral disc according to clinical indications (to diagnose unclear bone lesions/ suspected bone tumors or suspected inflammation/spondylodiscitis). Patients were identified in the hospital’s institutional digital picture archiving and communication system (PACS). All eligible patient cases with LD scans were matched according to sex (m/f), age (± 10 years), level of the procedure (cervical, thoracic, or lumbosacral), presence/absence of spinal instrumentation (metallic hardware causing artifacts in imaging data and making the access route to the target structure more demanding), and body diameter (< 20 cm, 20–25 cm, 25–30 cm, and > 30 cm). Patients were excluded if 1) non-diagnostic image quality was present in MDCT data (e.g., due to motion artifacts), or 2) the planned biopsy (including survey, planning, and procedure scans) was not accom- plished (e.g., due to incompliance of the patient and preliminary abortion of the exam). p ( g p p p y ) Overall, 96 cases were eligible and considered in this study (48 patients with SD scans and 48 matched patients with LD scans, 34 cases each with intervertebral disc biopsy and 14 cases with vertebral bone biopsy). Multi‑detector computed tomography. All scans included were performed with the patient in prone position using the same 128-slice MDCT scanner (Ingenuity Core 128; Philips Healthcare, Best, The Nether- lands). After performing the scout scan covering the planned site of biopsy according to previous diagnostic imaging, a planning scan of the region to be considered was performed (spot scanning). Impact of radiation dose reduction and iterative image reconstruction on CT‑guided spine biopsies Karolin J. Paprottka 1, Karina Kupfer 1, Vivian Schultz 1, Meinrad Beer 3, Claus Zimmer 1,2, Thomas Baum 1, Jan S. Kirschke 1,2 & Nico Sollmann 1,2,3 OPEN Specifically, it can yield an accuracy of over 94% in determining benign, inflammatory, or malignant etiologies10. In this regard, the Infectious Diseases Society of America officially recommends computed tomography (CT)-guided biopsies via a percutaneous approach as a first attempt in many cases of suspected discitis.t Guidance using CT is often considered for precise localization of a lesion before and during biopsy7,8,11,12. It gives the interventionalist not only a detailed view of the anatomic circumstances for biopsy planning and execution, but also allows for confirmation of the correct needle placement within the area of concern dur- ing the procedure. At many institutions, CT guidance is the primary method of choice for biopsies of osseous lesions within the vertebrae or with regard to suspected infections along the spine3,7,13, owing to high avail- ability, comparatively low cost, high spatial resolution, and relatively few procedure-limiting contraindications. Although CT guidance is frequently used for various procedures, there is concern over the amount of radiation 1Department of Diagnostic and Interventional Neuroradiology, School of Medicine, Klinikum rechts der Isar, Technical University of Munich, Munich, Germany. 2TUM‑Neuroimaging Center, Klinikum rechts der Isar, Technical University of Munich, Munich, Germany. 3Department of Diagnostic and Interventional Radiology, University Hospital Ulm, Ulm, Germany. email: Karolin.Paprottka@tum.de \| https://doi.org/10.1038/s41598-023-32102-9 Scientific Reports \| (2023) 13:5054 www.nature.com/scientificreports/ Table 1. Details of the scanning protocol and image reconstruction for the planning and periprocedural guidance scans using a standard-dose (SD) and low-dose (LD) protocol. Periprocedural guidance scan Planning scan Standard-dose (SD) imaging Low-dose (LD) imaging Standard-dose (SD) imaging Low-dose (LD) imaging Cycle time (in s) 2.4 6 No. of cycles 1 1 Rotation time (in s) 0.75 0.75 Tube voltage (in kV) 120 120 Tube current (in mA) 40–67 20–40 40–67 20–40 Estimated exposure (in mAs) ~30–60 ~15–20 ~30–60 ~15–20 Collimation width (in mm) 0.625 Windowing L = 750.0 ; W = 2500.0 Image reconstruction iDose4 iDose4 IMR1 IMR1 Table 1. Details of the scanning protocol and image reconstruction for the planning and periprocedural guidance scans using a standard-dose (SD) and low-dose (LD) protocol. Table 1. Details of the scanning protocol and image reconstruction for the planning and periprocedural guidance scans using a standard-dose (SD) and low-dose (LD) protocol. exposure for the patient. Radiation dose reduction is commonly used in diagnostic scans especially for pediatric patients14,15. Impact of radiation dose reduction and iterative image reconstruction on CT‑guided spine biopsies Karolin J. Paprottka 1, Karina Kupfer 1, Vivian Schultz 1, Meinrad Beer 3, Claus Zimmer 1,2, Thomas Baum 1, Jan S. Kirschke 1,2 & Nico Sollmann 1,2,3 OPEN Additionally, it is also a relevant issue for patients who receive multiple scans due to acute disease follow-up exams, chronic diseases, and screening purposes16–19. Hence, there are many studies concerning the safety and efficacy of spine biopsies5,7,20. However, with the wider availability of modern model-based iterative reconstruction techniques, even further radiation dose reduction could become possible when combined with systematic lowering of the tube current. Specifically, no studies with a matched-pair design are available regard- ing dose reductions for planning and performing CT-guided biopsies at the spine with a focus on image quality and confidence of the interventionalist regarding the procedure.h i g g p The purpose of our retrospective study was to demonstrate that a low-dose (LD) protocol for CT-guided percutaneous biopsies of the vertebral bone or intervertebral discs is a practical alternative to standard-dose (SD) approaches, maintaining overall image quality and confidence for the interventionalist for planning and performing a safe procedure at reduced ionizing radiation doses. Material and methods The acquired scan was used for the purpose of procedure planning, with the interventionalist first selecting the slice allowing for optimal visualization of the access route to the intervertebral disc or vertebral body. During the subsequently performed interventional procedure, sequential scanning was achieved for guidance and surveillance during needle placement using a foot pedal (intermittent scanning, three axial images per shot). By default, images were reconstructed with model-based iterative reconstruction (planning scans; IMR1, Philips Healthcare, Best, The Netherlands) or hybrid iterative reconstruction (periprocedural guidance scans; iDose4, Philips Healthcare, Best, The Netherlands). The parameters for the planning and periprocedural guidance scans are illustrated in Table 1. https://doi.org/10.1038/s41598-023-32102-9 Scientific Reports \| (2023) 13:5054 \| www.nature.com/scientificreports/ Figure 1. Lateral scout obtained for a planned lumbar intervertebral disc biopsy in a 47-year-old woman with suspected spondylodiscitis at level L4/5. The blue line indicates the antero-posterior body diameter (251.0 mm). Figure 1. Lateral scout obtained for a planned lumbar intervertebral disc biopsy in a 47-year-old woman with suspected spondylodiscitis at level L4/5. The blue line indicates the antero-posterior body diameter (251.0 mm). Table 2. Semi-quantitative scoring for image evaluation by two readers. Item Score 1 2 3 4 5 Overall image quality Very good to perfect Good to very good Medium Poor No value Overall artifacts None Minimal Prominent Major Severe Image contrast Very good to perfect Good to very good Medium Poor No value Determination of target structure Possible Unclear Not possible x x Confidence for intervention planning (planning scans before the infiltration) High Medium Low x x Confidence for intervention guidance (sequential scans during the infiltration) High Medium Low x x Table 2. Semi-quantitative scoring for image evaluation by two readers. Item Score 1 2 3 4 5 Overall image quality Very good to perfect Good to very good Medium Poor No value Overall artifacts None Minimal Prominent Major Severe Image contrast Very good to perfect Good to very good Medium Poor No value Determination of target structure Possible Unclear Not possible x x Confidence for intervention planning (planning scans before the infiltration) High Medium Low x x Confidence for intervention guidance (sequential scans during the infiltration) High Medium Low x x Table 2. Semi-quantitative scoring for image evaluation by two readers. Table 2. Semi-quantitative scoring for image evaluation by two readers. Material and methods Parameters obtained from SD and LD scanning included the dose length product (DLP), volumetric CT dose index (CTDIvol), number of scans required to perform the spine biopsy (periprocedural guidance scans via sequential scanning), and measurements of body diameter. The individual body diameter was measured in the lateral scout scan at the level of the planned intervention and was determined from skin-to-skin surface (Fig. 1)23. Image evaluation. After the biopsies, imaging data were transferred to PACS and evaluated with the stand- ard PACS viewer (IDS7; Sectra AB, Linköping, Sweden) by two radiologists (board-certified radiologist with 9 years of experience, reader 1 [R1], and resident in radiology with 2 years of experience, reader 2 [R2]). The readers semi-quantitatively evaluated overall image quality, overall artifacts, image contrast, determination of the target structure (intervertebral disc or vertebral body), and confidence for intervention planning (based on planning scans) and confidence for intervention guidance (based on the sequential scans during performance of the intervention), using 5-point or 3-point Likert scales (Table 2; Fig. 2). g p p g During evaluations, the readers were strictly blinded to the ratings of each other, and there was an interval of at least 3 weeks between the readings of data of different dose levels. Per image reading round (SD or LD scans), the images of the patients were presented in randomized order. Furthermore, the readers were unaware of the distinct protocol used for scanning per reading round. The readers were presented with images using bone and soft tissue windowing, and they were allowed to manually adjust windowing levels if wanted, starting with a standard output in the PACS viewer (window length = 750.0, width = 2500.0). In addition to semi-quantitative rating using Likert scales, quantitative evaluation was performed by measur- ing image noise at the level of the procedure. This was achieved by manually placing ~10 mm2 circular regions of interest (ROIs) and measuring the standard deviation (StDev) of the attenuation in Hounsfield units (HU) in the psoas muscles for lumbosacral and trapezius muscles for cervical interventions23,24. In each patient case, three separate measurements were performed (Fig. 3), and the obtained values of the three ROIs were averaged per patient. Statistical analysis. GraphPad Prism (version 6.0; GraphPad Software, San Diego, CA, USA) and SPSS (version 28.0; IBM SPSS Statistics for Windows, IBM Corp., Armonk, NY, USA) were used for statistical data analysis. Statistical analysis. GraphPad Prism (version 6.0; GraphPad Software, San Diego, CA, USA) and SPSS (version 28.0; IBM SPSS Statistics for Windows, IBM Corp., Armonk, NY, USA) were used for statistical data analysis. A p-value < 0.05 (two-sided) was considered statistically significant. Material and methods A p-value < 0.05 (two-sided) was considered statistically significant. Scientific Reports \| (2023) 13:5054 \| https://doi.org/10.1038/s41598-023-32102-9 www.nature.com/scientificreports/ Figure 2. Examples for planning scans for lumbosacral spine biopsies in different patients performed with a low-dose (LD) protocol for scanning, which were rated with perfect (A), good (B), medium (C), and poor (D) overall image quality. igure 2. Examples for planning scans for lumbosacral spine biopsies in different patients performed with a ow-dose (LD) protocol for scanning, which were rated with perfect (A), good (B), medium (C), and poor (D) verall image quality. Descriptive statistics were calculated for the scores assigned by each reader concerning the single items of image evaluation (using Likert scales) and attenuation measurements (in HU) as well as for patient demo- graphics, characteristics of interventions, and dose measurements. Weighted Cohen’s kappa (κ) was calculated to assess inter-reader agreement regarding scorings for overall image quality, overall artifacts, image contrast, determination of the target structure (intervertebral disc or vertebral body), and confidence for intervention planning and intervention guidance during the biopsy. To compare the scores from image evaluation using Likert scales between SD and LD scans, Wilcoxon signed-rank tests were performed per reader. Furthermore, Wilcoxon signed-rank tests were also conducted to compare measures of image noise between SD and LD scans, demographics, and dose characteristics. Ethical approval. This retrospective monocentric study involving human participants with a matched pairs design was approved by the ethics committee of the Faculty of Medicine at Technical University of Munich and was in accordance with the ethical standards of the institutional research committee and with the Declaration of Helsinki. Informed consent. The requirement for written informed consent was waived by the ethics committee of the Technical University of Munich due to the study’s retrospective design. Results Furthermore, no major periprocedural complications (e.g., bleeding) were reported for any of the biopsies performed either orsal spinal instrumentation) within the field of view chosen for the intervention. Suspected tumor or meta- tatic lesions made up the indication for biopsy in 11 patients of the LD group and in 12 patients of the SD group he remaining patients underwent biopsies due to suspected inflammatory processes (e.g., spondylodiscitis).hif dorsal spinal instrumentation) within the field of view chosen for the intervention. Suspected tumor or meta- static lesions made up the indication for biopsy in 11 patients of the LD group and in 12 patients of the SD group, the remaining patients underwent biopsies due to suspected inflammatory processes (e.g., spondylodiscitis).hif gl y g y There were no statistically significant differences between patients for SD versus LD scans regarding age, body diameter, or number of sequential scans needed during performance of the intervention (Table 3). Furthermore, no major periprocedural complications (e.g., bleeding) were reported for any of the biopsies performed either with the SD or LD protocol. Figures 4 and 5 depict exemplary patient cases. Semi‑quantitative evaluation. Image quality, image contrast, the determination of the target structure, and confidence for planning or intervention guidance were rated good to perfect for both SD and LD scans according to evaluations of both readers, without statistically significant differences between SD versus LD scans for these parameters (p > 0.05; Tables 4 and 5). Further, inter-reader agreement was at least substantial for the images from intervention planning (range of κ: 0.64–0.90) as well as from intervention guidance (range of κ: 0.72–0.88), except for confidence for intervention guidance with moderate agreement between readers (κ = 0.59; Tables 4 and 5). Radiation exposure. The DLP was statistically significantly lower for LD scans regarding the planning scans (SD: 13.8 ± 8.2 mGycm, LD: 8.1 ± 4.4 mGycm, p < 0.01) as well as the periprocedural guidance scans (SD: 43.0 ± 48.8 mGycm, LD: 18.4 ± 7.3 mGycm, p < 0.01; Fig. 6). Similarly, the CTDIvol was also statistically signifi- cantly lower for LD scans regarding the planning scans (SD: 2.1 ± 0.4 mGy, LD: 1.3 ± 0.8 mGy, p < 0.01) as well as the periprocedural guidance scans (SD: 32.4 ± 14.3 mGy, LD: 18.1 ± 7.4 mGy, p < 0.01). Image noise. Results Noise according to quantitative evaluation using muscle attenuation values was similar between SD and LD scans performed for planning of the interventional procedure (SD: 14.62 ± 2.83 HU vs. LD: 15.45 ± 3.22 HU, p = 0.24). Results Results Patient cohort. Overall, 96 patients were enrolled in this study (48 patients with SD imaging and 48 matched patients with LD imaging). According to matching criteria, in both groups 26 patients were female and 22 patients were male. Indication for intervertebral disc biopsy was made in 34 patients per group, while 14 patients per group underwent biopsy of a vertebral body lesion. Seven patients per group had implants (e.g., after https://doi.org/10.1038/s41598-023-32102-9 Scientific Reports \| (2023) 13:5054 \| www.nature.com/scientificreports/ Figure 3. Measurement of image noise for a planned lumbosacral biopsy. Three measurements within the psoas muscle were performed to derive attenuation values from manually placed circular regions of interest (ROIs). Figure 3. Measurement of image noise for a planned lumbosacral biopsy. Three measurements within the psoas muscle were performed to derive attenuation values from manually placed circular regions of interest (ROIs). Table 3. Cohort and procedure characteristics. SD LD p value Age (in years, mean ± StDev) 67.7 ± 12.9 69.5 ± 12.7 0.65 Body diameter on scout image (in cm, mean ± StDev) 27.7 ± 5.2 25.9 ± 4.2 0.44 Number of scans needed during the intervention (n) 13.1 ± 4.7 13.9 ± 6.4 0.19 Table 3. Cohort and procedure characteristics. SD LD p value Age (in years, mean ± StDev) 67.7 ± 12.9 69.5 ± 12.7 0.65 Body diameter on scout image (in cm, mean ± StDev) 27.7 ± 5.2 25.9 ± 4.2 0.44 Number of scans needed during the intervention (n) 13.1 ± 4.7 13.9 ± 6.4 0.19 Table 3. Cohort and procedure characteristics. SD LD p value Age (in years, mean ± StDev) 67.7 ± 12.9 69.5 ± 12.7 0.65 Body diameter on scout image (in cm, mean ± StDev) 27.7 ± 5.2 25.9 ± 4.2 0.44 Number of scans needed during the intervention (n) 13.1 ± 4.7 13.9 ± 6.4 0.19 Table 3. Cohort and procedure characteristics. dorsal spinal instrumentation) within the field of view chosen for the intervention. Suspected tumor or meta- static lesions made up the indication for biopsy in 11 patients of the LD group and in 12 patients of the SD group, the remaining patients underwent biopsies due to suspected inflammatory processes (e.g., spondylodiscitis). There were no statistically significant differences between patients for SD versus LD scans regarding age, body diameter, or number of sequential scans needed during performance of the intervention (Table 3). Discussion Lowering the tube current for MDCT can be a simple and effective method for reducing radiation exposure to both the patient and interventionalist regarding CT-guided spine biopsies. In our study, we were able to show that dose reduction for planning and periprocedural guidance scans for intervertebral disc and bone biopsies with MDCT is feasible and can be performed without clinically relevant drawbacks regarding image quality or confidence. The DLP and CTDIvol for LD scans were statistically significantly lower regarding the planning scans as well as for the periprocedural guidance scans. Overall image quality, image contrast, the determination of the target structure for biopsy, and confidence for planning or intervention guidance were rated good to perfect https://doi.org/10.1038/s41598-023-32102-9 Scientific Reports \| (2023) 13:5054 \| www.nature.com/scientificreports/ www.nature.com/scientificreports/ Figure 4. Exemplary patient cases for a L3 bone biopsy for a suspected bone tumor in a 77-year-old male using scanning with standard dose (SD; A) and a L1 bone biopsy in a 49-year-old woman with a suspected bone metastasis from known breast cancer using scanning with low dose (LD; B). The scans were rated with excellent image quality and high confidence. Figure 4. Exemplary patient cases for a L3 bone biopsy for a suspected bone tumor in a 77-year-old male using scanning with standard dose (SD; A) and a L1 bone biopsy in a 49-year-old woman with a suspected bone metastasis from known breast cancer using scanning with low dose (LD; B). The scans were rated with excellent image quality and high confidence. Figure 5. Exemplary patient cases for intervertebral disc biopsies using scanning with standard dose (SD; A, B) and scanning with low dose (LD; C, D). The upper row shows a L2/3 intervertebral disc biopsy in a 66-year- old woman (A) and a L5/S1 intervertebral disc biopsy in a 58-year-old woman with artifacts due to spinal instrumentation (B). The second row shows a L5/S1 intervertebral disc biopsy in a 48-year-old male patient (C) and a L4/5 intervertebral disc biopsy in an 80-year-old man with artifacts due to spinal instrumentation (D). All patients underwent intervertebral disc biopsy due to suspected spondylodiscitis. Figure 5. Exemplary patient cases for intervertebral disc biopsies using scanning with standard dose (SD; A, B) and scanning with low dose (LD; C, D). Discussion The upper row shows a L2/3 intervertebral disc biopsy in a 66-year- old woman (A) and a L5/S1 intervertebral disc biopsy in a 58-year-old woman with artifacts due to spinal instrumentation (B). The second row shows a L5/S1 intervertebral disc biopsy in a 48-year-old male patient (C) and a L4/5 intervertebral disc biopsy in an 80-year-old man with artifacts due to spinal instrumentation (D). All patients underwent intervertebral disc biopsy due to suspected spondylodiscitis. Figure 5. Exemplary patient cases for intervertebral disc biopsies using scanning with standard dose (SD; A, B) and scanning with low dose (LD; C, D). The upper row shows a L2/3 intervertebral disc biopsy in a 66-year- old woman (A) and a L5/S1 intervertebral disc biopsy in a 58-year-old woman with artifacts due to spinal instrumentation (B). The second row shows a L5/S1 intervertebral disc biopsy in a 48-year-old male patient (C) and a L4/5 intervertebral disc biopsy in an 80-year-old man with artifacts due to spinal instrumentation (D). All patients underwent intervertebral disc biopsy due to suspected spondylodiscitis. https://doi.org/10.1038/s41598-023-32102-9 Scientific Reports \| (2023) 13:5054 \| www.nature.com/scientificreports/ Table 4. Semi-quantitative scoring for intervention planning scans according to evaluations of two readers (R1 and R2) considering scanning with standard dose (SD) and low dose (LD). Intervention planning LD SD p value Overall image quality R1 1 (1–4) 2 (1–4) 0.89 R2 1 (1–4) 2 (1–4) 0.64 Kappa 0.90 0.81 Overall artifacts R1 1 (1–4) 1 (1–4) 0.99 R2 1 (1–4) 1 (1–4) 0.41 Kappa 0.87 0.90 Image contrast R1 1 (1–4) 2 (1–4) 0.17 R2 1 (1–4) 2 (1–3) 0.11 Kappa 0.85 0.72 Determination of target structure R1 1 (1–2) 1 (1–2) 0.99 R2 1 (1–2) 1 (1–2) 0.38 Kappa 0.90 0.77 Confidence for planning R1 1 (1–2) 1 (1–2) 0.25 R2 1 (1–3) 1 (1–3) 0.45 Kappa 0.64 0.74 Table 4. Semi-quantitative scoring for intervention planning scans according to evaluations of two readers (R1 and R2) considering scanning with standard dose (SD) and low dose (LD). Discussion Intervention planning LD SD p value Overall image quality R1 1 (1–4) 2 (1–4) 0.89 R2 1 (1–4) 2 (1–4) 0.64 Kappa 0.90 0.81 Overall artifacts R1 1 (1–4) 1 (1–4) 0.99 R2 1 (1–4) 1 (1–4) 0.41 Kappa 0.87 0.90 Image contrast R1 1 (1–4) 2 (1–4) 0.17 R2 1 (1–4) 2 (1–3) 0.11 Kappa 0.85 0.72 Determination of target structure R1 1 (1–2) 1 (1–2) 0.99 R2 1 (1–2) 1 (1–2) 0.38 Kappa 0.90 0.77 Confidence for planning R1 1 (1–2) 1 (1–2) 0.25 R2 1 (1–3) 1 (1–3) 0.45 Kappa 0.64 0.74 Table 4 S i tit ti i Intervention planning LD SD p value Overall image quality R1 1 (1–4) 2 (1–4) 0.89 R2 1 (1–4) 2 (1–4) 0.64 Kappa 0.90 0.81 Overall artifacts R1 1 (1–4) 1 (1–4) 0.99 R2 1 (1–4) 1 (1–4) 0.41 Kappa 0.87 0.90 Image contrast R1 1 (1–4) 2 (1–4) 0.17 R2 1 (1–4) 2 (1–3) 0.11 Kappa 0.85 0.72 Determination of target structure R1 1 (1–2) 1 (1–2) 0.99 R2 1 (1–2) 1 (1–2) 0.38 Kappa 0.90 0.77 Confidence for planning R1 1 (1–2) 1 (1–2) 0.25 R2 1 (1–3) 1 (1–3) 0.45 Kappa 0.64 0.74 Table 4. Semi-quantitative scoring for intervention planning scans according to evaluations of two readers (R1 and R2) considering scanning with standard dose (SD) and low dose (LD). 7 3:5054 \| https://doi.org/10.1038/s41598-023-32102-9 Table 5. Semi-quantitative scoring for periprocedural guidance scans according to evaluations of two readers (R1 and R2) considering scanning with standard dose (SD) and low dose (LD). Intervention guidance LD SD p value Overall image quality R1 1 (1–3) 1 (1–3) 0.70 R2 1 (1–4) 1 (1–3) 0.99 Kappa 0.82 0.85 Overall artifacts R1 1 (1–4) 1 (1–3) 0.99 R2 1 (1–4) 1 (1–4) 0.94 Kappa 0.86 0.80 Image contrast R1 1 (1–3) 1 (1–3) 0.82 R2 1 (1–3) 1 (1–3) 0.44 Kappa 0.76 0.72 Determination of target structure R1 1 (1–2) 1 (1–2) 0.63 R2 1 (1–2) 1 (1–2) 0.63 Kappa 0.79 0.88 Confidence for guidance R1 1 (1–2) 1 (1–2) 0.99 R2 1 (1–2) 1 (1–3) 0.25 Kappa 0.85 0.59 Table 5. Semi-quantitative scoring for periprocedural guidance scans according to evaluations of two readers (R1 and R2) considering scanning with standard dose (SD) and low dose (LD). Discussion Overall, noise according to quantitative evaluation using muscle attenuation values was similar between SD and LD scans performed for planning of the interventional procedures.ii 8 9 p p g p Imaging guidance for biopsy is a commonly used procedure in patients with findings in need for clarification8,9. Yet, concerns with CT guidance relate to potential consequences of ionizing radiation. Hence, there are many efforts to keep radiation exposure as low as reasonably achievable (ALARA)25–27. The options available for reach- ing the goal of low radiation dose in CT are manifold and primarily include adaptions in scanning parameters such as for tube current, tube voltage, slice thickness, patient coverage, number of acquisitions, or length of the procedure26. In the course of CT scanning protocol optimizations and introduction of model-based itera- tive reconstruction, we adjusted the CT protocol based on a former conventional SD protocol to provide LD scanning with reduced radiation exposure. Previously published in-vivo studies demonstrated the utility of LD techniques for a multitude of interventional procedures. Specifically, Meng et al. performed a study with focus on biopsies of lung lesions with a LD protocol (group 1: 120 kV, 200 mA; group 2: 120 kV, 10 mA) and revealed that a reduction of radiation dose and DLP were possible without a relevant loss of diagnostic yield28. Despite a considerable reduction of the radiation dose during CT-guided percutaneous lung biopsies by more than 90% (from DLP of 677.5 mGycm to 18.3 mGycm), Smith et al. found no relevant decrease in technical success or patient safety25. Furthermore, especially for pediatric CT scanning, many studies were able to show that CT- guided bone biopsies performed using techniques for lowering radiation exposure can also lead to acceptable image quality and provide similar diagnostic yield compared with SD protocols29–31. g q y p g y p p A substantial proportion of applied radiation results from performing pre- and post-biopsy scans as they are designed to optimize the visualization of soft tissues for needle guidance and to exclude biopsy-related compli- cations. A review by Sarti et al. showed that up to 90% of the total radiation dose during biopsies was caused by the helical planning scan32. In our study, we tried to solve this problem using two different methods. On the one hand, we lowered the applied radiation dose by reductions in tube current based on the former SD protocol. Discussion Intervention guidance LD SD p value Overall image quality R1 1 (1–3) 1 (1–3) 0.70 R2 1 (1–4) 1 (1–3) 0.99 Kappa 0.82 0.85 Overall artifacts R1 1 (1–4) 1 (1–3) 0.99 R2 1 (1–4) 1 (1–4) 0.94 Kappa 0.86 0.80 Image contrast R1 1 (1–3) 1 (1–3) 0.82 R2 1 (1–3) 1 (1–3) 0.44 Kappa 0.76 0.72 Determination of target structure R1 1 (1–2) 1 (1–2) 0.63 R2 1 (1–2) 1 (1–2) 0.63 Kappa 0.79 0.88 Confidence for guidance R1 1 (1–2) 1 (1–2) 0.99 R2 1 (1–2) 1 (1–3) 0.25 Kappa 0.85 0.59 Intervention guidance LD SD p value Overall image quality R1 1 (1–3) 1 (1–3) 0.70 R2 1 (1–4) 1 (1–3) 0.99 Kappa 0.82 0.85 Overall artifacts R1 1 (1–4) 1 (1–3) 0.99 R2 1 (1–4) 1 (1–4) 0.94 Kappa 0.86 0.80 Image contrast R1 1 (1–3) 1 (1–3) 0.82 R2 1 (1–3) 1 (1–3) 0.44 Kappa 0.76 0.72 Determination of target structure R1 1 (1–2) 1 (1–2) 0.63 R2 1 (1–2) 1 (1–2) 0.63 Kappa 0.79 0.88 Confidence for guidance R1 1 (1–2) 1 (1–2) 0.99 R2 1 (1–2) 1 (1–3) 0.25 Kappa 0.85 0.59 Table 5. Semi-quantitative scoring for periprocedural guidance scans according to evaluations of two readers (R1 and R2) considering scanning with standard dose (SD) and low dose (LD). Table 5. Semi-quantitative scoring for periprocedural guidance scans according to evaluations of two readers (R1 and R2) considering scanning with standard dose (SD) and low dose (LD). https://doi.org/10.1038/s41598-023-32102-9 Scientific Reports \| (2023) 13:5054 \| www.nature.com/scientificreports/ Figure 6. Scatter dot plots with mean ± standard deviation (StDev) for the dose-length product (DLP, in mGycm) of planning and periprocedural guidance scans using scanning with standard dose (SD) and low dose (LD). Figure 6. Scatter dot plots with mean ± standard deviation (StDev) for the dose-length product (DLP, in mGycm) of planning and periprocedural guidance scans using scanning with standard dose (SD) and low dose (LD). or SD and LD scans, respectively. Overall, noise according to quantitative evaluation using muscle attenuation alues was similar between SD and LD scans performed for planning of the interventional procedures.ii for SD and LD scans, respectively. www.nature.com/scientificreports/ www.nature.com/scientificreports/ biopsies performed at 30 mAs were similar compared to those for biopsies performed with the SD protocol26. Yet, in 13 cases where patients underwent biopsies with the new LD protocol, the masses were not clearly identi- fied, thus these procedures were completed at a higher dose26. The complication rate of the LD technique was comparable to that of the SD technique in their study26. These results are in accordance with our results where a reduction of the radiation dose led to a significant reduction of the DLP regarding the planning scans as well as the interventional guidance scans. Furthermore, the tube current in our study was even lower compared to that previous study. biopsies performed at 30 mAs were similar compared to those for biopsies performed with the SD protocol26. Yet, in 13 cases where patients underwent biopsies with the new LD protocol, the masses were not clearly identi- fied, thus these procedures were completed at a higher dose26. The complication rate of the LD technique was comparable to that of the SD technique in their study26. These results are in accordance with our results where a reduction of the radiation dose led to a significant reduction of the DLP regarding the planning scans as well as the interventional guidance scans. Furthermore, the tube current in our study was even lower compared to that previous study. p y All intervertebral disc and vertebral body biopsies in our study were performed under CT guidance. Other publications have reported that the use of CT fluoroscopy exposes patients to less radiation than conventional CT-guided procedures34,35. Yet, the use of CT fluoroscopy can expose the interventionalist to more radiation than using conventional CT techniques26. We believe that the use of LD techniques under CT fluoroscopic guidance may be the optimal way to decrease radiation exposure during procedures. McNamara and coworkers provided a systematic review and meta-analysis about image-guided biopsies conducted in patients with suspected discitis and investigated 14 biopsies performed under CT guidance, as well as 6 studies under fluoroscopic guidance13. They concluded that fluoroscopic guidance was associated with a higher yield at 55% compared with CT guid- ance at 44%, though the difference was not statistically significant13. gf y gi Recent techniques have set a focus on iterative image reconstruction models such as adaptive statistical iterative reconstruction or model-based iterative reconstruction16,17,22,36–38. www.nature.com/scientificreports/ Fourth, the reduced dose level was reached by lowering the tube current in combination with model- based iterative reconstruction, but without evaluating other modern approaches to limit radiation exposure, such as sparse sampling that has been successfully applied for spine CT in previous simulation studies16,17,21,42. Furthermore we did not explicitly assess which parameter adjustments contributed the most to radiation dose Yet, there are some limitations to this study. First, the study design was retrospective and the study was per- formed at a single academic institution, implicating that spine and intervertebral disc biopsies were conducted by different interventionalists with different education levels. Therefore, the reproducibility of the results using the new LD protocol cannot be fully assessed in this study. Second, the lack of quantifying artifacts specifi- cally associated with the metallic needle can be considered a limitation, which can lead to restrictions in overall image quality and thus can have impact on periprocedural guidance. Third, despite a different level of experience and education of the two readers (radiologist with 9 years and resident with 2 years of experience), the reading results were very good and similar with only small differences between the readers. This could be a hint for a LD protocol that could have been subject to even further decreases in radiation dose. On the other hand, this could also indicate that dose reduction in CT-guided spine biopsies is an applicable method in clinical practice that does not result in considerably reduced image quality or diagnostic confidence when performed with the herein presented protocol, neither for an experienced radiologist nor a resident with much less experience. Fur- thermore, it would be difficult to determine whether more experienced radiologists may have contributed to a greater efficiency of the procedure using a LD protocol. As a consequence, there is a potential bias for radiation exposure as well as procedural time and number of scans needed during intervention that is inherent to the study design. Fourth, the reduced dose level was reached by lowering the tube current in combination with model- based iterative reconstruction, but without evaluating other modern approaches to limit radiation exposure, such as sparse sampling that has been successfully applied for spine CT in previous simulation studies16,17,21,42. www.nature.com/scientificreports/ Furthermore, we did not explicitly assess which parameter adjustments contributed the most to radiation dose reductions, and this may need to be solved primarily by phantom or cadaver studies in which more than one scan could be achieved, given that radiation protection is not an issue. However, a recent review concluded that nowadays, considerable dose reductions in spinal CT for various indications can be realized with dose reductions around 50%43. Although some novel approaches during image acquisitions have not yet been implemented in commercially available MDCT systems (e.g., sparse sampling), such techniques may become of general interest in the near future44,45. Additionally, for improving image quality and reducing radiation dose, deep learning-based reconstruction approaches are promising tools41,46,47. p p g y pp p p Furthermore, we did not explicitly assess which parameter adjustments contributed the most to radiation dose reductions, and this may need to be solved primarily by phantom or cadaver studies in which more than one scan could be achieved, given that radiation protection is not an issue. However, a recent review concluded that nowadays, considerable dose reductions in spinal CT for various indications can be realized with dose reductions around 50%43. Although some novel approaches during image acquisitions have not yet been implemented in commercially available MDCT systems (e.g., sparse sampling), such techniques may become of general interest in the near future44,45. Additionally, for improving image quality and reducing radiation dose, deep learning-based reconstruction approaches are promising tools41,46,47. Conclusion We demonstrated that a LD imaging protocol combined with advanced image reconstruction for MDCT scan- ning during planning and performing intervertebral disc or vertebral body biopsies is a viable option, given that radiation exposure to the patient significantly decreased without considerably affecting image quality or the confidence for planning and guidance. Furthermore, despite a significantly lower total cumulative radiation exposure compared with regular-dose CT-guided biopsies, using a LD protocol did not alter the rate of compli- cations. We therefore encourage other centers to consider LD imaging protocols combined with model-based iterative reconstruction as an alternative to conventional protocols for CT-guided spine interventions. Discussion On the other hand, we only scanned the region of interest of the planned CT intervention. From this point of view, pre-biopsy imaging should be carefully reviewed and when possible focused on the definite region of biopsy, together with an optimized radiation dose. This approach is similar to a recent study investigating LD protocols for CT-guided periradicular infiltrations33. However, biopsies use hardware for puncture of the target region with comparatively large calibers (e.g., potential risk for more pronounced artifacts especially with LD protocols) that often need to be inserted deeper into tissue (e.g., close to the nerve root in periradicular infiltrations versus biopsies of vertebral bone), making distinct investigations for biopsies necessary in the light of patient safety and sufficient image quality. Further, in a study by Lucey and coworkers, 291 CT-guided interventional procedures without the use of CT fluoroscopy (for percutaneous biopsy, needle aspiration, and percutaneous catheter place- ment) were performed using a LD protocol with an exposure of 30 mAs and a tube voltage of 120–140 kV for the performance scan26. They found that by decreasing the effective tube current and exposure from 180–240 mAs to 30 mAs, radiation to the patient decreased six- to eightfold26. At the same time, the technical success rates of https://doi.org/10.1038/s41598-023-32102-9 Scientific Reports \| (2023) 13:5054 \| www.nature.com/scientificreports/ These algorithms are useful options for further dose reductions, not only in diagnostic CT but also in CT-guided interventions39–41. Yet, their avail- ability is currently limited to newer CT scanners, and if used in clinical practice, such approaches are primarily applied for diagnostic CT scanning and less for interventional CT scanning. Prospectively, a wider availability of iterative reconstruction software may translate into an increased comfort for the radiologist while evaluating images with LD regimens during CT-guided interventions and, therefore, potentially result in further reductions of the radiation dose applied to the patients. In our study, all planning scans were reconstructed using model- based iterative reconstruction algorithms, given that model-based iterative approaches may help to increase the visibility of anatomical details whilst facilitating LD protocols39–41. y g p Yet, there are some limitations to this study. First, the study design was retrospective and the study was per- formed at a single academic institution, implicating that spine and intervertebral disc biopsies were conducted by different interventionalists with different education levels. Therefore, the reproducibility of the results using the new LD protocol cannot be fully assessed in this study. Second, the lack of quantifying artifacts specifi- cally associated with the metallic needle can be considered a limitation, which can lead to restrictions in overall image quality and thus can have impact on periprocedural guidance. Third, despite a different level of experience and education of the two readers (radiologist with 9 years and resident with 2 years of experience), the reading results were very good and similar with only small differences between the readers. This could be a hint for a LD protocol that could have been subject to even further decreases in radiation dose. On the other hand, this could also indicate that dose reduction in CT-guided spine biopsies is an applicable method in clinical practice that does not result in considerably reduced image quality or diagnostic confidence when performed with the herein presented protocol, neither for an experienced radiologist nor a resident with much less experience. Fur- thermore, it would be difficult to determine whether more experienced radiologists may have contributed to a greater efficiency of the procedure using a LD protocol. As a consequence, there is a potential bias for radiation exposure as well as procedural time and number of scans needed during intervention that is inherent to the study design. Data availabilityh y The datasets used and analysed during the current study are available anonymized from the corresponding author on reasonable request. Received: 15 July 2022; Accepted: 22 March 2023 Received: 15 July 2022; Accepted: 22 March 2023 Received: 15 July 2022; Accepted: 22 March 2023 https://doi.org/10.1038/s41598-023-32102-9 Scientific Reports \| (2023) 13:5054 \| www.nature.com/scientificreports/ References T. et al. Effect of patient size on radiation dose for abdominal MDCT with automatic tube current modulation: phantom study. AJR Am. J. Roentgenol. 190(2), W100–W105 (2008). p y J J g ( ), ( ) 25. Smith, J. C. et al. Ultra-low-dose protocol for CT-guided lung biopsies. J. Vasc. Interv. Radiol. 22(4), 431–436 (2011). 26. Lucey, B. C. et al. CT-guided intervention with low radiation dose: feasibility and experience. AJR Am. J. Roentgenol 1187–1194 (2007).f 27. Bevelacqua, J. J. Practical and effective ALARA. Health Phys. 98(Suppl 2), S39-47 (2010). f 28. Meng, X. X. et al. Comparison of lung lesion biopsies between low-dose CT-guided and conventional CT-guided techniques. Radiol. 54(8), 909–915 (2013). 29. Kalra, M. K. et al. Strategies for CT radiation dose optimization. Radiology 230(3), 619–628 (2004). 30. Natali, G. L. et al. Paediatric musculoskeletal interventional radiology. Br. J. Radiol. 89(1057), 20150369 (2016). al. Paediatric musculoskeletal interventional radiology. Br. J. Radio 31. Patel, A. S. et al. Radiation dose reduction in pediatric CT-guided musculoskeletal procedures. Pediatr. Radiol. 43(10), 130 (2013). 32. Sarti, M., Brehmer, W. P. & Gay, S. B. Low-dose techniques in CT-guided interventions. Radiographics 32(4), 1109–1119 (2012). 33 P ttk K J t l L d lti d t t t d t h f i di l i filt ti t th i l d l b y q g g p 3. Paprottka, K. J. et al. Low-dose multi-detector computed tomography for periradicular infiltrations at the cervical and lumbar spine. Sci. Rep. 12(1), 4324 (2022).f p p 4. Teeuwisse, W. M. et al. Patient and staff dose during CT guided biopsy, drainage and coagulation. Br. J. Radiol. 74(884), 720–726 (2001).l 5. Paulson, E. K. et al. CT fluoroscopy–guided interventional procedures: Techniques and radiation dose to radiologists. Radiology 220(1), 161–167 (2001).fi 220(1), 161–167 (2001). 36. Corcuera-Solano, I. et al. Repeated head CT in the neurosurgical intensive care unit: Feasibility of sinogram-affirmed iterative i b d l l d CT f ill AJNR A J N di l 35(7) 1281 1287 (2014) 6. Corcuera-Solano, I. et al. Repeated head CT in the neurosurgical intensive care unit: Feasibility of sinogram-affirmed iterative reconstruction-based ultra-low-dose CT for surveillance. AJNR Am. J. Neuroradiol. 35(7), 1281–1287 (2014). 7. Mathieu, K. B. et al. Radiation dose reduction for CT lung cancer screening using ASIR and MBIR: A phantom study. J. Appl. Clin Med. Phys. 15(2), 4515 (2014). 8. Flicek, K. T. References References 1. Takenaka, D. et al. Detection of bone metastases in non-small cell lung cancer patients: Comparison of whole-body diffusion- weighted imaging (DWI), whole-body MR imaging without and with DWI, whole-body FDG-PET/CT, and bone scintigraphy. J. Magn. Reson. Imaging 30(2), 298–308 (2009). 1. Takenaka, D. et al. Detection of bone metastases in non-small cell lung cancer patients: Comparison of whole-body diffusion- weighted imaging (DWI), whole-body MR imaging without and with DWI, whole-body FDG-PET/CT, and bone scintigraphy. J. Magn. Reson. Imaging 30(2), 298–308 (2009). g g g ( ), ( ) 2. Gallucci, P. M. & D’Orazio, F. Image guided interventions in spinal infections. Neuroimaging Clin. N. Am. 25(2), 281–294 (2015). 3 Rimondi E et al Percutaneous CT guided biopsy of the musculoskeletal system: Results of 2027 cases Eur J Radiol 77(1) 34 42 g g g ( ), ( ) 2. Gallucci, P. M. & D’Orazio, F. Image guided interventions in spinal infections. Neuroimaging Clin. N. Am. 25(2), 281–294 (2015). g g g ( ) ( ) 2. Gallucci, P. M. & D’Orazio, F. Image guided interventions in spinal infections. Neuroimaging Clin. N. Am. 25(2), 281–294 (2015). 3. Rimondi, E. et al. Percutaneous CT-guided biopsy of the musculoskeletal system: Results of 2027 cases. Eur. J. Radiol. 77(1), 34–42 (2011).l g g g allucci, P. M. & D’Orazio, F. Image guided interventions in spinal in d l d d b f h l k ( ) 4. Rimondi, E. et al. Computerized tomography guided biopsy in the diagnosis of neoplastic and inflammatory lesions of the pelvis Radiol. Med. 101(1–2), 60–65 (2001). ( ) ( ) 5. Rimondi, E. et al. Percutaneous CT-guided biopsy of the spine: results of 430 biopsies. Eur. Spine J. 17(7), 975–981 (2008). 6. Menon, V. K., Kumar, K. M. & Al Ghafri, K. One-stage biopsy, debridement, reconstruction, and stabilization of pyogenic vertebra osteomyelitis. Glob. Spine J 4(2), 93–100 (2014).hfi 7. Sertic, M. et al. The efficacy of computed tomography-guided percutaneous spine biopsies in determining a causative organism in cases of suspected infection: A systematic review. Can. Assoc. Radiol. J. 70(1), 96–103 (2019).h p y 8. Saifuddin, A. et al. Review article: The current status of CT-guided needle biopsy of the spine. Skelet. Radiol. 50(2), 281–299 (2021) h 9. Yang, S. Y. et al. Percutaneous image-guided spinal lesion biopsies: factors affecting higher diagnostic yield. AJR Am. J. Roentgenol 211(5), 1068–1074 (2018). 0. Rehm, J. et al. References CT-guided percutaneous spine biopsy in suspected infection or malignancy: A study of 214 patients. Rofo 188(12) 1156–1162 (2016). 11. Singh, D. K. et al. Approach-based techniques of CT-guided percutaneous vertebral biopsy. Diagn. Interv. Radiol. 26(2), 143 (2020). CT-guided percutaneous biopsy of spinal lesions. Biomed. Imaging . Peh, W. CT-guided percutaneous biopsy of spinal lesions. Biome 3. McNamara, A. L. et al. Yield of image-guided needle biopsy for infectious discitis: A systematic review and meta-analysis. AJNR Am. J. Neuroradiol. 38(10), 2021–2027 (2017). 4. Nagayama, Y. et al. Deep learning-based reconstruction for lower-dose pediatric CT: Technical principles, image characteristics and clinical implementations. Radiographics 41(7), 1936–1953 (2021). p g p 5. Kertesz, H. et al. Reducing radiation exposure to paediatric patients undergoing [18F]FDG-PET/CT imaging. Mol. Imaging Biol 23(5), 775–786 (2021). 6. Sollmann, N. et al. Low-dose MDCT of patients with spinal instrumentation using sparse sampling: Impact on metal artifacts AJR Am. J. Roentgenol. 216(5), 1308–1317 (2021). g 7. Sollmann, N. et al. Low-dose MDCT: Evaluation of the impact of systematic tube current reduction and sparse sampling on the detection of degenerative spine diseases. Eur. Radiol. 31(4), 2590–2600 (2021). g p 18. Becce, F. et al. Computed tomography of the cervical spine: comparison of image quality between a standard-dose and a low-dose protocol using filtered back-projection and iterative reconstruction. Skelet. Radiol. 42(7), 937–945 (2013). p gi p j 9. Lee, S. H. et al. Diagnostic usefulness of low-dose lumbar multi-detector CT with iterative reconstruction in trauma patients Acomparison with standard-dose CT. Br. J. Radiol. 90(1077), 20170181 (2017). p 0. Wiesner, E. L. et al. Percutaneous CT-Guided biopsies of the cervical spine: Technique, histopathologic and microbiologic yield and safety at a single academic institution. AJNR Am. J. Neuroradiol. 39(5), 981–985 (2018). y g 1. Sollmann, N. et al. Multi-detector CT imaging: Impact of virtual tube current reduction and sparse sampling on detection of vertebral fractures. Eur. Radiol. 29(7), 3606–3616 (2019).i 22. Sollmann, N. et al. Systematic evaluation of low-dose MDCT for planning purposes of lumbosacral periradicular infiltrations. Clin. Neuroradiol. 30, 749–759 (2019).l ( ) 3. Amrhein, T. J. et al. Reducing patient radiation exposure from CT fluoroscopy-guided lumbar spine pain injections by targeting the planning CT. AJR Am. J. Roentgenol. 206(2), 390–394 (2016). 3. Amrhein, T. J. et al. Reducing patient radiation exposure from CT fluoroscopy-guided lumbar spine pain injections by targeting the planning CT. AJR Am. J. Roentgenol. 206(2), 390–394 (2016).f 24. Schindera, S. www.nature.com/scientificreports/ www.nature.com/scientificreports/ 4. Chen, B. et al. First multislit collimator prototype for SparseCT: Design, manufacturing and initial validation. In The Fifth Inter- national Conference on Image Formation in X-Ray Computed Tomography, Salt Lake City. (2018). 44. Chen, B. et al. First multislit collimator prototype for SparseCT: Design, manufacturing and initial validation. In The Fifth Inter- national Conference on Image Formation in X-Ray Computed Tomography, Salt Lake City. (2018). 44. Chen, B. et al. First multislit collimator prototype for SparseCT: Design, manufacturing and initial validation. In The Fifth Inter- national Conference on Image Formation in X-Ray Computed Tomography, Salt Lake City. (2018). f g y p g p y y 5. Koesters, T. et al. SparseCT: Interrupted-beam acquisition and sparse reconstruction for radiation dose reduction. In SPIE Medica Imaging. SPIE (2017). g g ( ) 46. Brady, S. L. et al. Improving image quality and reducing radiation dose for pediatric CT by using deep learning reconstruction. Radiology 298(1), 180–188 (2021). g g 46. Brady, S. L. et al. Improving image quality and reducing radiation dose for pediatric CT by using deep learning reconstruction. Radiology 298(1), 180–188 (2021). g g 6. Brady, S. L. et al. Improving image quality and reducing radiation dose for pediatric CT by using deep learning reconstruction Radiology 298(1), 180–188 (2021). gy 7. Yeoh, H. et al. Deep learning algorithm for simultaneous noise reduction and edge sharpening in low-dose CT images: A pilot study using lumbar spine CT. Korean J. Radiol. 22(11), 1850–1857 (2021). gy 47. Yeoh, H. et al. Deep learning algorithm for simultaneous noise reduction and edge sharpening in low-dose CT images: A pilot study using lumbar spine CT. Korean J. Radiol. 22(11), 1850–1857 (2021). 47. Yeoh, H. et al. Deep learning algorithm for simultaneous noise reduction and edge sharpening in low-dose CT images: A pilot study using lumbar spine CT. Korean J. Radiol. 22(11), 1850–1857 (2021). Competing interests h p g The authors declare no competing interests. Author contributions K.P., N.S.: Substantial contributions to the conception and design of the work; K.P., N.S.: Drafting the work and revising it critically for important intellectual content. All authors: Final approval of the version to be published. K.P., N.S.: Substantial contributions to the conception and design of the work; K.P., N.S.: Drafting the work and revising it critically for important intellectual content. All authors: Final approval of the version to be published. Funding g Open Access funding enabled and organized by Projekt DEAL. We acknowledge support by the B. Braun Founda- tion (project No BBST-D-19-00106), Dr.-Ing. Leonhard Lorenz Foundation, and the Joachim Herz Foundation (awarded to N.S.). References et al. Reducing the radiation dose for CT colonography using adaptive statistical iterative reconstruction: A pilot study AJR Am. J. Roentgenol. 195(1), 126–131 (2010). g 39. Willemink, M. J. et al. Iterative reconstruction techniques for computed tomography Part 1: Technical principles. Eur. Radiol. 23(6), 1623–1631 (2013). ( ) ( ) 40. Willemink, M. J. et al. Iterative reconstruction techniques for computed tomography part 2: Initial results in dose reduction and image quality. Eur. Radiol. 23(6), 1632–1642 (2013).hii g q y 1. Willemink, M. J. & Noel, P. B. The evolution of image reconstruction for CT-from filtered back projection to artificial intelligence Eur. Radiol. 29(5), 2185–2195 (2019). g q y 41. Willemink, M. J. & Noel, P. B. The evolution of image reconstruction for CT-from filtered back projection to artificial int Eur. Radiol. 29(5), 2185–2195 (2019). 42. Mei, K. et al. Is multidetector CT-based bone mineral density and quantitative bone microstructure assessment at the spine still feasible using ultra-low tube current and sparse sampling?. Eur. Radiol. 27(12), 5261–5271 (2017). g p p g ( ) ( ) 43. Dieckmeyer, M. et al. Computed tomography of the spine: Systematic review on acquisition and reconstruction techniques to reduce radiation dose. Clin. Neuroradiol. (2022). https://doi.org/10.1038/s41598-023-32102-9 Scientific Reports \| (2023) 13:5054 \| Additional information Correspondence and requests for materials should be addressed to K.J.P. Correspondence and requests for materials should be addressed to K.J.P. Reprints and permissions information is available at www.nature.com/reprints. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2023 https://doi.org/10.1038/s41598-023-32102-9 Scientific Reports \| (2023) 13:5054 \|
https://openalex.org/W4237661037	https://ousar.lib.okayama-u.ac.jp/files/public/5/58104/20200325142310871880/fulltext.pdf	English	null	Single-session esophagogastroduodenoscopy and endoscopic ultrasound using a forward-viewing radial scan ultrasonic endoscope	Research Square (Research Square)	2,019	cc-by	6,327	Uchida et al. BMC Gastroenterology (2019) 19:220 https://doi.org/10.1186/s12876-019-1141-7 Uchida et al. BMC Gastroenterology (2019) 19:220 https://doi.org/10.1186/s12876-019-1141-7 Open Access © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Abstract Background: Endoscopic ultrasound is useful for obtaining high-resolution images of pancreaticobiliary diseases, but is not readily available for physical checkups. In this study, we evaluated the safety and efficacy of single-session esophagogastroduodenoscopy and endoscopic ultrasound in the detection of upper-gastrointestinal and pancreaticobiliary diseases using a forward-viewing radial scan ultrasonic endoscope. Methods: A total of 148 patients who were scheduled for upper-gastrointestinal screening using an endoscope were prospectively included. All patients were examined by EUS in combination with EGD using a forward-viewing radial scan ultrasonic endoscope. The primary endpoint was the safety of the procedures. The secondary endpoints were the prevalence of diseases, the basal imaging capability of EUS, the procedure time, total dose of propofol, and the correlation between background factors and the prevalence of pancreatic disease. The imaging capability at each region was scored as 0 (invisible) to 2 (sufficient visualization to evaluate the organs). Results: Intraoperative hypotension occurred as an adverse event of intravenous anesthesia in one patient. There were 82 pancreaticobiliary findings and 165 upper-gastrointestinal findings (malignancy not included). Follicular lymphoma of the intra-abdominal lymph nodes was detected in one patient. The mean imaging scores of each section were 1.95 (pancreatic head and papilla), 2.0 (pancreatic body), 1.99 (pancreatic tail), and 1.89 (common bile duct and gallbladder). Age, history of diabetes mellitus, and smoking history were significantly associated with the prevalence of pancreatic diseases. Conclusion: The simultaneous performance of EGD and EUS using a new ultrasonic endoscope is tolerable and safe for upper-gastrointestinal and pancreaticobiliary screening. Keywords: Endoscopic ultrasound, Diagnostic screening program, Pancreatic diseases, Biliary tract diseases Single-session esophagogastroduodenoscopy and endoscopic ultrasound using a forward- viewing radial scan ultrasonic endoscope Daisuke Uchida1,2* , Hironari Kato1, Kazuyuki Matsumoto1, Yuki Ishihara1, Akihiro Matsumi1, Yosuke Saragai1, Saimon Takada1, Shuntaro Yabe1, Shinichiro Muro1, Takeshi Tomoda1, Shigeru Horiguchi1 and Hiroyuki Okada1 Background with this condition increases each year. Recently, the diagnosis and treatment of PC have advanced, but less than 10% of patients are expected to survive for 5 years after the diagnosis [2]. Patients with early-stage PC often have no subjective symptoms, and most are diagnosed incidentally. Thus, a screening test is required for pa- tients with risk factors for PC; however, the optimal ap- proach is not known. It is often difficult to diagnose pancreaticobiliary dis- eases, including malignant tumors such as pancreatic cancer (PC) and biliary cancer (BC), which exhibit a poor prognosis [1]. In particular, the early diagnosis of PC is an urgent issue because the number of patients * Correspondence: pt77172s@okayama-u.ac.jp 1Department of Gastroenterology, Okayama University Hospital, 2-5-1 Shikata-cho, Kita-ku, Okayama 700-8558, Japan 2Center for Innovative Clinical Medicine, Okayama University Hospital, 2-5-1 Shikata-cho, Kita-ku, Okayama 700-8558, Japan * Correspondence: pt77172s@okayama-u.ac.jp 1Department of Gastroenterology, Okayama University Hospital, 2-5-1 Shikata-cho, Kita-ku, Okayama 700-8558, Japan 2Center for Innovative Clinical Medicine, Okayama University Hospital, 2-5-1 Shikata-cho, Kita-ku, Okayama 700-8558, Japan * Correspondence: pt77172s@okayama-u.ac.jp 1Department of Gastroenterology, Okayama University Hospital, 2-5-1 Shikata-cho, Kita-ku, Okayama 700-8558, Japan 2Center for Innovative Clinical Medicine, Okayama University Hospital, 2-5-1 Shikata-cho, Kita-ku, Okayama 700-8558, Japan The detection of premalignant lesions is one solution for improving the survival of PC patients. Transabdom- inal ultrasound is a non-invasive technique; however, * Correspondence: pt77172s@okayama-u.ac.jp 1Department of Gastroenterology, Okayama University Hospital, 2-5-1 Shikata-cho, Kita-ku, Okayama 700-8558, Japan 2Center for Innovative Clinical Medicine, Okayama University Hospital, 2-5-1 Shikata-cho, Kita-ku, Okayama 700-8558, Japan Study design and sites This was a prospective, interventional study conducted at Okayama University Hospital from May 2017 to De- cember 2018. This study was registered in the UMIN protocol registration system (identification number: UMIN000026627). Definitions and outcome measures After receiving approval for the study from the institu- tional review board, we recruited consecutive patients who visited Okayama University Hospital to undergo esophagogastroduodenoscopy (EGD). The primary endpoint was the safety of the procedures. The secondary endpoints were the prevalence of pan- creaticobiliary diseases that were detected by EUS, the prevalence of upper-gastrointestinal diseases, the basal imaging capability in the evaluation of the pancreatico- biliary region, the procedure time, the total dose of pro- pofol, and the correlation between the prevalence of pancreatic diseases and patient background. To evaluate the basal imaging capability, the pancreaticobiliary re- gion was divided into four areas; pancreas head (includ- ing papilla), pancreas body, pancreas tail, and biliary tract including gallbladder. They were assessed and assigned one of three scores: 0–2 points. “0” was assigned when imaging was completely invisible for all areas; “1” was assigned when the image was partially de- lineated; and “2” was assigned when the visualization of the target image was sufficient for the evaluation of the organs (Table 3). The inclusion criteria were as follows: 1. Scheduled for screening EGD. 2. Over fifty years of age (because the prevalence of PC increases with age, especially at ≥50 years of age [9]). 3. The provision of voluntary written consent for participation in this study. Page 2 of 9 Page 2 of 9 Page 2 of 9 Uchida et al. BMC Gastroenterology (2019) 19:220 Uchida et al. BMC Gastroenterology (2019) 19:220 Experimental methods All procedures were carried out by six experienced endosonographers (DU, HK, KM, YS, ST, and SM), who met all of the following criteria: ≥8 years of endoscopy experience, and ≥200 EUS examinations performed > 3 years. Procedures were performed under conscious sed- ation with propofol (0.8 mg/kg induction dose and a 3– 5 mg/kg/h maintenance dose of 1% propofol using an in- fusion pump, 0.5 mg/kg additional proper boluses). At first, EGD was performed. After EGD screening, EUS screening was performed to evaluate the pancreas, biliary tract, and gallbladder. If abnormal findings were de- tected, appropriate examinations or treatments were provided. Instruments screening of the pancreas is sometimes difficult [3]. Endoscopic ultrasound (EUS) is a novel technique for screening of the pancreas, and is suitable for screening high-risk individuals, and various studies have reported the diagnostic performance of EUS [4–7]. However, it is not a routine test, because it requires dedicated endo- scopes, (e.g., endoscopes for radial and linear EUS) [8]. Forward-viewing radial scan ultrasonic endoscope (EG- 580UR; FUJIFILM, Tokyo, Japan) instruments and an endoscopic ultrasonic processor (SU-1; FUJIFILM, Tokyo, Japan) were used. The EG-580UR has an outer diameter of 11.4 mm, a unique capability of bending to 190°, a 2.8 mm working channel, and the super-CCD image quality (Fig. 1). It also has a flexible imaging color enhancement (FICE) mode that can maximize color dif- ferences, such as vascular and mucosal patterns, without the need for tissue staining [10]. EUS images were delin- eated in standard B-mode, and the frequency was set at 7.5 MHz. Recently, a new endoscopic ultrasonic processor (SU- 1; FUJIFILM, Tokyo, Japan) and radial scan ultrasonic endoscope (EG-580UR; FUJIFILM, Tokyo, Japan) equipped with a direct forward view, which enables the simultaneous performance of esophagogastroduodeno- scopy (EGD) and EUS. This endoscope enables the per- formance pancreaticobiliary screening for patients who undergo EGD. We conducted a prospective intervention study evaluating the efficacy of the new ultrasonic endo- scope in upper-gastrointestinal and pancreaticobiliary screening. The exclusion criteria were as follows: The exclusion criteria were as follows: 1. ECOG performance status 3 or 4. 2. Suspected gastrointestinal bleeding. 3. A history of upper gastrointestinal surgery. Partially visible (score “1”) was defined as any visible delineation that did not meet the criteria for score “2”. For example, if the junction of the superior mesenteric vein (SMV), portal vein (PV), and splenic vein (SPV) was unclear despite the pancreatic body parenchyma being visible, the imaging score of the pancreatic body was “1”. 4. Allergy to propofol. 5. Received abdominal computed tomography (CT) or magnetic resonance imaging (MRI) within the past six months. 6. Judged by the attending physician to be ineligible for inclusion in this study). Page 3 of 9 Uchida et al. BMC Gastroenterology (2019) 19:220 Fig. 1 a Shape of the EG-580UR. b: The diameter of the distal end of the 580UR is 11.4 mm, This device has a the unique ability of bending to 190°, and has a 2.8 mm working channel. c: A FICE image of esophagogastric junction obtained using the EG-580UR. d: A white light image of the gastric body obtained using the EG-580UR. e: An EUS image of the pancreatic body obtained using the EG-580UR. Orange arrow indicates a pancreatic cyst. f: An EUS image of the common bile duct obtained using the EG-580UR. Blue arrow indicates a bile duct stone Fig. 1 a Shape of the EG-580UR. b: The diameter of the distal end of the 580UR is 11.4 mm, This device has a the unique ability of bending to 190°, and has a 2.8 mm working channel. c: A FICE image of esophagogastric junction obtained using the EG-580UR. d: A white light image of the gastric body obtained using the EG-580UR. e: An EUS image of the pancreatic body obtained using the EG-580UR. Orange arrow indicates a pancreatic cyst. f: An EUS image of the common bile duct obtained using the EG-580UR. Blue arrow indicates a bile duct stone As another example, if the region adjacent to the splenic hilum and left kidney was unclear, despite the pancreatic tail parenchyma being visible, the imaging score of the pancreatic tail was “1”. The exclusion criteria were as follows: Table 1 Patient characteristics (n = 148) Table 1 Patient characteristics (n = 148) Characteristics Value Gender Female 106 (71.6%) Age, median (range) 69 (51–76) Family history of PC 11 (7.4%) History of diabetes mellitus 28 (23.3%) Smoking status Never 118 Former 21 Daily 9 Alcohol status Daily drinker 44 (29.7%) Daily alcohol consumption (g), median of drinker 7 Reason for EGD screening Follow-up of chronic gastritis due to Helicobacter pylori 94 Follow-up of gastric polyp 16 Check for varices for patients with hepatitis 23 Check for reflux esophagitis 9 Abdominal discomfort 6 Post-cholecystectomy 6 PC pancreatic cancer; EGD esophagogastroduodenoscopy Statistical analysis All statistical analyses were performed using the JMP Pro 13.0 software program. Categorical data were evalu- ated with the chi-squared test. P values of < 0.05 were considered to indicate statistical significance. A multi- variate logistic regression analysis was performed to analyze associations between patient characteristics (items that were associated with prevalence in a univari- ate analysis; P < 0.05) and the prevalence of pancreatic diseases. Adverse events One patient had hypotension (systolic blood pressure < 70 mmHg) during the endoscopic procedure. This might have been due to oversedation with propofol. Hypotension was rapidly reversed following the adminis- tration of ephedrine (4 mg). She recovered from anesthesia without any further adverse events. Forty-four patients had undergone AUS in the past year. Sixteen of these patients were diagnosed with pan- creatic disease (pancreatic cyst, n = 8; early chronic pan- creatitis, n = 8) which was not detected by AUS in the past year. Twelve of these patients were diagnosed gall- bladder disease (adenomyomatosis, n = 7; gallbladder stones, n = 5), that had already been diagnosed by AUS. To date, 88 (59.5%) have already undergone follow-up CT or MRI examinations. None of the patients showed other pancreaticobiliary findings. Patient characteristics 6033 patients underwent EGD screening within the time frame, and 5885 patients were excluded because they were < 50 years of age (n = 722), refused to participate (n = 836), had a history of upper-gastrointestinal surgery (n = 331), or had a history of abdominal CT or MRI within the past six months (n = 3996). Finally, a total of 148 patients were included. The patient characteristics are summarized in Table 1. One hundred six of the par- ticipants were female (71.6%). Eleven participants (7.4%) had one first- or second-degree blood relative with PC. Page 4 of 9 Uchida et al. BMC Gastroenterology (2019) 19:220 Uchida et al. BMC Gastroenterology (2019) 19:220 Page 4 of 9 Uchida et al. BMC Gastroenterology (2019) 19:220 MRI at six months. Twenty-five patients were diagnosed with early chronic pancreatitis, which had 3 to 4 minor features of chronic pancreatitis, according to the Rose- mont classification [11]. No patients had symptoms of chronic pancreatitis. Fourteen patients had gallbladder polyp (suspected cholesterol polyp). Nine patients had adenomyomatosis and thirteen had asymptomatic gall- bladder stones. These gallbladder lesions were scheduled for follow-up AUS at six months. One patient had asymptomatic common bile duct stone, and was treated with endoscopic removal of the common bile duct stone by endoscopic sphincterotomy. Other patients were followed up with AUS, MRI or CT at one year at the pa- tient’s request. MRI at six months. Twenty-five patients were diagnosed with early chronic pancreatitis, which had 3 to 4 minor features of chronic pancreatitis, according to the Rose- mont classification [11]. No patients had symptoms of chronic pancreatitis. Fourteen patients had gallbladder polyp (suspected cholesterol polyp). Nine patients had adenomyomatosis and thirteen had asymptomatic gall- bladder stones. These gallbladder lesions were scheduled for follow-up AUS at six months. One patient had asymptomatic common bile duct stone, and was treated with endoscopic removal of the common bile duct stone by endoscopic sphincterotomy. Other patients were followed up with AUS, MRI or CT at one year at the pa- tient’s request. Twenty-eight (23.3%) participants had a history of dia- betes mellitus. The most common reason for EGD screening was follow-up of chronic gastritis due to Heli- cobacter pylori. There were six post-cholecystectomy pa- tients. Forty-four patients had undergone transabdominal ultrasonography (AUS) in the past year. Prevalence of pancreaticobiliary disease The prevalence of disease is shown in Table 2. There were thirty-two patients with pancreatic cyst. The me- dian cyst size was 5 mm. The largest was a 20-mm mul- tilocular cyst, which was diagnosed as a branch duct intraductal papillary mucinous neoplasm (IPMN) with- out high-risk stigmata or worrisome features. All pa- tients with pancreatic cysts were scheduled for follow-up Table 2 Prevalence of diseases Pancreaticobiliary diseases Pancreatic cysts 32 (21.6%) Size of cyst, median, range (mm) 5 (2–20) Early chronic pancreatitis* 25 (16.9%) Gallbladder polyp 14 (9.5%) Size of polyps, median range (mm) 4 (1–5) Adenomyomatosis 9 (6.1%) Gallbladder stone 13 (8.8%) Common bile duct stone 1 (0.7%) Upper-gastrointestinal diseases Barrett’s esophagus 9 (6%) Reflux esophagitis 41 (27.7%) Candida esophagitis 1 (0.7%) Esophageal papilloma 1 (0.7%) Esophageal submucosal tumor 2 (1.4%) Chronic gastritis 69 (46.6%) Hyperplastic polyp 9 (6%) Fundic gland polyp 14 (9.5%) Gastric ulcer scar 1 (0.7%) Gastric submucosal tumor 8 (5.4%) Duodenal ulcer scar 8 (5.4%) Duodenal submucosal tumor 2 (1.4%) Others detected by EUS Follicular lymphoma of intra-abdominal lymph nodes 1 (0.7%) EUS endoscopic ultrasound Other diseases detected by EUS Other diseases detected by EUS There was one patient with follicular lymphoma of the intra-abdominal lymph nodes. A swollen abdominal lymph node (> 25 mm) was pointed out, and EUS-guided fine needle aspiration (EUS-FNA) was performed. EUS- FNA revealed follicular lymphoma. She was followed closely by hematologists. EUS endoscopic ultrasound Prevalence of upper-gastrointestinal disease As shown in Table 2, various findings were detected by EGD screening. No patients had malignant features. In 56 patients (37.8%) cases, it was difficult to delineate the lesser curvature of the gastric angle when pulling back the scope because of the long distal end of the scope. In these cases, the gastric angle could be evaluated in the short scope position. All patients were scheduled for follow-up EGD at one year. To date, 103 patients (69.6%) have already undergone follow-up EGD using a conventional EGD scope. Fortunately, no patients had other findings on follow-up EGD. Procedure times and total dose of propofol become the scope that is predominantly used for EUS examination. Meanwhile, radial scan EUS is used for screening examinations, and can obtain transverse im- ages of the pancreas, which provides objective evalua- tions. However, these scopes are dedicated EUS devices, and are unsuitable for medical check-ups. The median procedure time was 22 min (EGD, 10 min; EUS, 11 min). The median total dose of propofol was 145 mg. A breakdown of these results is shown in Table 4. Correlation between the prevalence of pancreatic dis- ease and patient background factors. On the other hand, EGD using a forward-viewing endoscope and transabdominal ultrasound are com- monly used in medical check-ups in Japan [19]. A forward-viewing radial scan ultrasonic endoscope and ultrasound processor (SU-1 and EG-580UR, respectively; FUJIFILM, Tokyo, Japan) enable single-session screening of the upper gastrointestinal tract and pancreaticobiliary organ. EUS provides high-resolution images, and is an- ticipated to be highly useful in the early diagnosis of pancreaticobiliary tumors [16]. We conducted this study to evaluate the efficacy of this device for the early diag- nosis of pancreaticobiliary diseases, especially pancreatic malignancies. Age and a history of diabetes mellitus were extracted as factors significantly associated with the prevalence of pancreatic cysts in a univariate analysis. In the multivari- ate logistic regression analysis, age was the only signifi- cant factor. Sex, history of diabetes mellitus, smoking history, and daily alcohol were extracted as factors sig- nificantly associated with the prevalence of early chronic pancreatitis. In the multivariate logistic regression ana- lysis, history of diabetes mellitus and smoking history were significant factors. Detailed data are shown in Table 5. In the present study, we recruited 148 patients scheduled for EGD screening (Table 1). We defined the minimum age as 50 years, because the prevalence of PC increases with age, especially in patients of ≥50 years of age [9]. As a result, a comparatively large number of findings are detected by procedures (Table 2). EGD screening revealed various findings, including reflux esophagitis, chronic gastritis, and gastrointes- tinal benign tumors, but not malignancies. The image quality of the CCD of the EG-580UR device is not equivalent to that of the latest endoscopes. In fact, to delineate the lesser curvature of the gastric angle when pulling back the scope was difficult in 56 pa- tients (37.8%). This constructional disadvantage should be improved. Basal imaging capability of EUS The mean imaging scores of each section were 1.95 (pancreatic head and papilla), 2.0 (pancreatic body), 1.99 (pancreatic tail), and 1.89 (common bile duct and gall- bladder) (Table 3). The biliary imaging score of post- cholecystectomy patients was without gallbladder im- aging. There were no patients with score “0” in any section. Uchida et al. BMC Gastroenterology (2019) 19:220 Page 5 of 9 Table 3 Basal imaging capability of EUS Score definition Mean score 2 1 0 Pancreatic head and papilla Duodenal papilla (the region of confluence of the pancreatic duct and bile ducts on the duodenal muscularis), and the pancreas head (the region surrounded by the SMA and the scope) are clearly visible Partially visible Completely invisible 1.95 Pancreatic body The pancreatic body (the region from the proximal parenchyma of the junction of SMV, PV and SPV up to pancreatic tail) is clearly visible Partially visible Completely invisible 2 Pancreatic tail Pancreatic tail (the region adjacent to the splenic hilum and left kidney) is clearly visible Partially visible Completely invisible 1.99 Common bile duct and gallbladder Common bile duct (from confluence of hepatic duct to pancreatic bile duct including junction of cystic duct), and gallbladder (from neck to fundus) are serially visible Partially visible Completely invisible 1.89 EUS endoscopic ultrasound; SMA superior mesenteric artery; SMV superior mesenteric vein; PV portal vein; SPV splenic vein Discussion Small pancreaticobiliary tumors, such as PC and BC, are difficult to diagnose because they have minimal symp- toms in the early stages. However, advanced pancreati- cobiliary cancers exhibit a poor prognosis. The 5-year survival rate of PC patients is approximately 8%, and PC is the fourth leading cause of cancer death worldwide [12, 13]. The early diagnosis of PC is one solution to im- proving the prognosis of PC, and various approaches have been developed such as genetic screening of high- risk patients [14, 15]. EUS is considered to be a sensitive device for the diag- nosis of pancreaticobiliary diseases [16–18]. In particu- lar, curved linear array EUS can obtain tissue by EUS- guided fine needle aspiration (EUS-FNA), and has Table 4 Procedure time, and total dose of propofol Procedure time (min); median and range Total dose of propofol (mg); median and range EGD EUS total 145 (30–350) 10 (5–25) 11 (10–29) 22 (15–45) EGD esophagogastroduodenoscopy; EUS endoscopic ultrasound Total dose of propofol (mg); median and range 145 (30–350) 145 (30–350) Uchida et al. Discussion BMC Gastroenterology (2019) 19:220 Page 6 of 9 Table 5 The correlation between the prevalence of pancreatic diseases and patient background Pancreatic cyst (n = 32) Early chronic pancreatitis (n = 25) Number Univariate Multivariate Number Univariate Multivariate Odds ratio (95%CI) P value Odds ratio (95%CI) P value Odds ratio (95%CI) P value Odds ratio (95%CI) P value Age ≥70 (n = 68) 22 (32.4%) 3.35 (1.45–7.71) 0.0035 3.18 (1.36–7.38) 0.0073 12 (17.7%) 1.10 (0.47–2.61) 0.8212 < 70 (n = 80) 10 (12.5%) 13 (16.3%) Sex Male (n = 42) 9 (21.4%) 0.98 (0.41–2.35) 0.9714 14 (33.3%) 4.32 (1.76–10.6) 0.0008 2.08 (0.65–6.66) 0.2147 Female (n = 106) 23 (21.7%) 11 (10.4%) Family history of PC Yes (n = 11) 4 (36.4%) 2.22 (0.61–8.14) 0.217 2 (18.2%) 1.10 (0.22–5.44) 0.9055 No (n = 137) 28 (20.4%) 23 (16.8%) History of diabetes mellitus Yes (n = 28) 10 (35.7%) 2.47 (1.01–6.09) 0.0442 2.23 (0.88–5.65) 0.0921 10 (35.7%) 3.89 (1.51–9.99) 0.0032 3.78 (1.33–10.8) 0.0128 No (n = 120) 22 (18.3%) 15 (12.5%) Smoking history Yes (n = 30) 6 (20%) 0.88 (0.33–2.39) 0.8091 12 (40%) 5.38 (2.12–13.7) 0.0002 3.54 (1.04–12) 0.043 No (n = 118) 26 (22%) 13 (11%) Alcohol status Daily (n = 44) 11 (25%) 1.32 (0.57–3.03) 0.5161 13 (29.6%) 3.22 (1.33–7.78) 0.0075 1.06 (0.34–3.3) 0.9223 Never or occasional (n = 104) 21 (20.2%) 12 (11.5%) PC pancreatic cancer Table 5 The correlation between the prevalence of pancreatic diseases and patient background EUS is undoubtedly a superior tool for the diagnosis of pancreatic diseases, including pancreatic cysts. Kamata et al. reported that the EUS was superior to other imaging modalities (e.g., CT or MRI) for the early detection of PC in patients with IPMN [27]. Pausawasdi et al. reported that EUS offered some benefits in the evaluation of pan- creatic cyst [28]. They referred to the possibility of the molecular and biomarker analysis of cyst fluid obtained by EUS-FNA. Barresi et al. reported the efficacy of EUS- through-the-needle biopsy in pancreatic cystic lesions; however, this procedure is still in the investigational stage, because it is associated with a risk of needle tract seeding [29]. Additionally, contrast-enhanced EUS is reported to be an effective tool for the diagnosis of pancreatic cysts [30, 31]. We consider that EUS with high-resolution im- aging can be a tool for identifying and qualitatively diag- nosing pancreaticobiliary diseases. Various findings were detected by EUS screening. Discussion As shown in Table 2, pancreatic cysts, which were detected in 32 patients, were the most common finding. Incidental pancreatic cysts are reported to be associated with increased mortality, and follow-up is recommended for patients in whom they are identified [20, 21]. IPMNs are pancreatic cysts that are associated a risk of malig- nancy; however, pancreatic cysts other than IPMNs may cause pancreatic ductal carcinoma [22]. Laffan et al. re- ported that the prevalence of unsuspected pancreatic cysts detected by MDCT in an outpatient population was 2.6%, which was correlated with increasing age and Asian race [23]. There are also some reports on the prevalence of pancreatic cysts on MRI imaging. De Jong reported that the prevalence was 2.4%, while Lee et al. and Zhang et al. reported that the prevalence was 13.5 and 19.6%, respect- ively [24–26]. Page 7 of 9 Page 7 of 9 Uchida et al. BMC Gastroenterology (2019) 19:220 Uchida et al. BMC Gastroenterology (2019) 19:220 generally good, and the procedure time was tolerable (Tables 3 and 4). Recently, curved linear array EUS scope is widely used. Radial scan EUS scope provides understandable organ image with 360°scanning range, however it cannot collect tissue samples. Kaneko et al. reported that there was not a significant difference be- tween the imaging capability of radial scan scope and curved linear array scope, however, both scope have pros and cons [39]. They reported radial scan scope was su- perior in the delineation of the major duodenal papilla and gallbladder. Front-viewing radial scan scope may be able to expand options for scope choice. Additionally, this scope and single-session EGD and EUS method may widen the training opportunities for novices. Increasing experts of EUS procedure using forward-viewing radial scan EUS scope may be conductive to improve pancrea- ticobiliary diseases. Our result showed that the prevalence of pancreatic cysts was 21.6%, which was clearly higher in comparison to previous reports. This contradictory finding have been due to the fact that our cohort consisted of patients of ≥50 years of age. Actually, an autopsy study revealed that 24.3% of patients (most patients were ≥65 years of age) had pancreatic cysts [32]. Fortunately, there were no ma- lignant findings, such as high-risk stigmata or IPMNs with worrisome features, and the patients with cysts of > 5 mm in size were scheduled for follow-up MRI at six months. Discussion In our data, age and a history of diabetes melli- tus were significantly associated with the prevalence of pancreatic cysts (Table 4). Chronic pancreatitis is a risk factor for PC [33]. Re- cently, the early diagnosis of chronic pancreatitis with EUS has gained attention; however, the extent to which early chronic pancreatitis is associated with pancreatic carcinogenesis is still unclear [11, 34]. In our study, the prevalence of early chronic pancreatitis was 16.9%; this was related to a history of diabetes mellitus and smoking history (Tables 2 and 4). This result is in line with previ- ous reports, and serves as a useful reference for identify- ing high-risk patients [24, 34–36]. A family history of PC is a known risk factor for PC [37, 38]. There was no sig- nificant association between a family history of PC and the prevalence of pancreatic disease in this study. There may be some reasons for this controversial result. Canto et al. reported that individuals with three or more blood relatives with PC, including at least one first-degree rela- tive, should be considered for screening [38]. In our study, there were 11 patients with only one affected first or second-degree blood relative. Additionally, the rela- tively small number of patients may have affected the result. The present study was associated with some limita- tions. The accuracy of the EUS findings was unclear be- cause some participants did not undergo screening with other modalities (e.g., CT or MRI). Most patients were also scheduled for follow-up AUS, MRI or CT at six months; however, some patients who had no pancreati- cobiliary findings did not wish to undergo the examin- ation because of the high cost. Fortunately, there were no other pancreaticobiliary findings in 88 patients (59.5%) who—at the time of writing—have already undergone follow-up using AUS, CT or MRI. Addition- ally, the accuracy of EGD was still unclear. To date, 103 patients (69.6%) have already undergone follow-up EGD using a conventional EGD scope, and no other gastro- intestinal findings were identified in follow-up EGD. It was unclear whether EUS was superior to other modal- ities; however, some patients were diagnosed with pan- creatic diseases that could not be detected by post-AUS. This result proved the high sensitivity of EUS in the diagnosis of pancreatic lesions. These subjects should be evaluated in a future analysis with comparative design including large number of participants. Discussion One patient had enlarged intra-abdominal lymph nodes, and was diagnosed with follicular lymphoma by EUS-FNA and 18-fluoro-2-deoxy-D-glucose positron emission tomography (FDG-PET). She was referred to a hematology specialist, and followed closely without chemotherapy. In this study, there was only one adverse event, which might have been caused by oversedation. We were con- cerned about the risk of increasing the dose of propofol; thus, we excluded patients with an ECOG performance status of 3 or 4. However, most procedures were com- pleted within 30 min as a result. This might have been due to the normal EUS findings in most patients. The imaging capability of EUS was generally good (Table 3). However, these procedures were performed by experienced endosonographers. The technique of EUS imaging is sometimes difficult, and the diagnostic per- formance of EUS is operator-dependent. Increasing the number of endosonographers is also necessary to facili- tate the early diagnosis of pancreaticobiliary diseases. Acknowledgments The authors thank all participants for accomplishing this study. 15. DaVee T, Coronel E, Papafragkakis C, et al. Pancreatic cancer screening in high-risk individuals with germline genetic mutations. Gastrointest Endosc. 2018;87:1443–50. Ethics approval and consent to participate pp p p This study was conducted in compliance with the principles of th This study was conducted in compliance with th This study was conducted in compliance with the principles of the Declaration of Helsinki, and the protocol was approved by the ethics committee of Okayama University Hospital (No.1705–002). All the eligible patients provided written informed consent before participating in this study. 21. Munigala S, Gelrud A, Agarwal B. Risk of pancreatic cancer in patients with pancreatic cyst. Gastrointest Endosc. 2016;84:81–6. 22. Yamaguchi K, Kanemitsu S, Hatori T, et al. Pancreatic ductal adenocarcinoma derived from IPMN and pancreatic ductal adenocarcinoma concomitant with IPMN. Pancreas. 2011;40:571–80. 23. Laffan TA, Horton KM, Klein AP, et al. Prevalence of unsuspected pancreatic cysts on MDCT. AJR Am J Roentgenol. 2008;191:802–7. References 27. Kamata K, Kitano M, Kudo M, et al. Value of EUS in early detection of pancreatic ductal adenocarcinomas in patients with intraductal papillary mucinous neoplasms. Endosc. 2014;46:22–9. 1. Torre LA, Bray F, Siegel RL, et al. Global cancer statistics, 2012. CA Cancer J Clin. 2015;65:87–108. 1. Torre LA, Bray F, Siegel RL, et al. Global cancer statistics, 2012. CA Cancer J Clin. 2015;65:87–108. 2. Poruk KE, Firpo MA, Adler DG, et al. Screening for pancreatic cancer: why, how, and who? Ann Surg. 2013;257:17–26. 28. Pausawasdi N, Ratanachu-Ek T. Endoscopic ultrasonography evaluation for pancreatic cysts: necessity or overkill? Dig Endosc. 2017;29:444–54. 3. Jeon JH, Kim JH, Joo I, et al. Transabdominal ultrasound detection of pancreatic cysts incidentally detected at CT, MRI, or endoscopic ultrasound. AJR Am J Roentgenol. 2018:1–8. 29. Barresi L, Crino SF, Fabbri C, et al. Endoscopic ultrasound-through-the- needle biopsy in pancreatic cystic lesions: a multicenter study. Dig Endosc. 2018;30:760–70. 4. DeWitt J, Devereaux B, Chriswell M, et al. Comparison of endoscopic ultrasonography and multidetector computed tomography for detecting and staging pancreatic cancer. Ann Intern Med. 2004;141:753–63. 30. Kamata K, Kitano M, Omoto S, et al. Contrast-enhanced harmonic endoscopic ultrasonography for differential diagnosis of pancreatic cysts. Endosc. 2015;48:35–41. 5. Jemaa Y, Houissa F, Trabelsi S, et al. Endoscopic ultrasonography versus helical CT in diagnosis and staging of pancreatic cancer. Tunis Med. 2008; 86:346–9. 5. Jemaa Y, Houissa F, Trabelsi S, et al. Endoscopic ultrasonography versus helical CT in diagnosis and staging of pancreatic cancer. Tunis Med. 2008; 86:346–9. 31. Yamamoto N, Kato H, Tomoda T, et al. Contrast-enhanced harmonic endoscopic ultrasonography with time-intensity curve analysis for intraductal papillary mucinous neoplasms of the pancreas. Endosc. 2015;48: 26–34. 6. Krishna SG, Rao BB, Ugbarugba E, et al. Diagnostic performance of endoscopic ultrasound for detection of pancreatic malignancy following an indeterminate multidetector CT scan: a systemic review and meta-analysis. Surg Endosc. 2017;31:4558–67. 6. Krishna SG, Rao BB, Ugbarugba E, et al. Diagnostic performance of endoscopic ultrasound for detection of pancreatic malignancy following an indeterminate multidetector CT scan: a systemic review and meta-analysis. Surg Endosc. 2017;31:4558–67. 32. Kimura W, Nagai H, Kuroda A, et al. Analysis of small cystic lesions of the pancreas. Int J Pancreatol. 1995;18:197–206. 33. Kirkegard J, Mortensen FV, Cronin-Fenton D. Chronic pancreatitis and pancreatic Cancer risk: a systematic review and meta-analysis. Am J Gastroenterol. 2017;112:1366–72. g 7. Yamashita Y, Kitano M, Ashida R. Consent for publication Not applicable. cysts on MDCT. AJR Am J Roentgenol. 2008;191:802–7. 24. de Jong K, Nio CY, Hermans JJ, et al. High prevalence of pancreatic cysts detected by screening magnetic resonance imaging examinations. Clin Gastroenterol Hepatol. 2010;8:806–11. Availability of date and material 16. Maguchi H. The roles of endoscopic ultrasonography in the diagnosis of pancreatic tumors. J Hepato-Biliary-Pancreat Surg. 2004;11:1–3. 16. Maguchi H. The roles of endoscopic ultrasonography in the diagnosis of pancreatic tumors. J Hepato-Biliary-Pancreat Surg. 2004;11:1–3. The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. 17. Canto MI, Goggins M, Yeo CJ, et al. Screening for pancreatic neoplasia in high-risk individuals: an EUS-based approach. Clin Gastroenterol Hepatol. 2004;2:606–21. Funding g This research has received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. This research has received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. 20. Chernyak V, Flusberg M, Haramati LB, et al. Incidental pancreatic cystic lesions: is there a relationship with the development of pancreatic adenocarcinoma and all-cause mortality? Radiol. 2015;274:161–9. Conclusion The screening method using the EG-580UR device is a little different from the traditional method using an or- dinary EUS scope, especially with regard to duodenal manipulation, because this new scope has a slim distal end (11.4 mm) and a small bending radius. At first, this difference perplexed us; however, all operators soon got used to it. Actually, the imaging capability of EUS was In conclusion, a forward-viewing radial scan ultrasonic endoscope, the EG-580UR, was a novel tool for screen- ing of gastrointestinal and pancreaticobiliary diseases. Further developments in manipulation and image quality are expected, which may help to improve the prognosis of pancreaticobiliary diseases. Page 8 of 9 Page 8 of 9 Page 8 of 9 Uchida et al. BMC Gastroenterology (2019) 19:220 Uchida et al. BMC Gastroenterology (2019) 19:220 Abbreviations 11. Catalano MF, Sahai A, Levy M, et al. EUS-based criteria for the diagnosis of chronic pancreatitis: the Rosemont classification. Gastrointest Endosc. 2009; 69:1251–61. AUS: Transabdominal ultrasonography; BC: Biliary cancer; CT: Computed tomography; EGD: Esophagogastroduodenoscopy; EUS: Endoscopic ultrasound; EUS-FNA: Eus-guided fine needle aspiration; FDG-PET: 18-fluoro- 2-deoxy-D-glucose positron emission tomography; FICE: Flexible imaging color enhancement; IPMN: Intraductal papillary mucinous neoplasm; MRI: Magnetic resonance imaging; PC: Pancreatic cancer; PV: Portal vein; SMV: Superior mesenteric vein; SPV: Splenic vein 12. Aier I, Semwal R, Sharma A, et al. A systematic assessment of statistics, risk factors, and underlying features involved in pancreatic cancer. Cancer Epidemiol. 2018;58:104–10. 13. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2018. CA Cancer J Clin. 2018; 68:7–30. 14. Klein AP, Wolpin BM, Risch HA, et al. Genome-wide meta-analysis identifies five new susceptibility loci for pancreatic cancer. Nat Commun. 2018;9:556. Authors’ contributions l d d DU analyzed and interpreted the patient data. HK, KM, YI, AM, YS, ST, SY, SM, TT, SH, and HO performed the endoscopic procedures and the patient care. All authors approved the final version of the manuscript. DU analyzed and interpreted the patient data. HK, KM, YI, AM, YS, ST, SY, SM, TT, SH, and HO performed the endoscopic procedures and the patient care. All authors approved the final version of the manuscript. 18. Liu CL, Lo CM, Chan JK, et al. EUS for detection of occult cholelithiasis in patients with idiopathic pancreatitis. Gastrointest Endosc. 2000;51:28–32. 18. Liu CL, Lo CM, Chan JK, et al. EUS for detection of occult cholelithiasis in patients with idiopathic pancreatitis. Gastrointest Endosc. 2000;51:28–32. 19. Hamashima C, Fukao A. And working group for the quality assurance of endoscopic screening for gastric cancer. Quality assurance manual of endoscopic screening for gastric cancer in Japanese communities. Jpn J Clin Oncol. 2016;46:1053–61. Competing interests The authors declare that they have no competing interests. The authors declare that they have no competing interests. 25. Lee KS, Sekhar A, Rofsky NM, et al. Prevalence of incidental pancreatic cysts in the adult population on MR imaging. Am J Gastroenterol. 2010;105:2079– 84. Received: 20 August 2019 Accepted: 10 December 2019 Received: 20 August 2019 Accepted: 10 December 2019 Received: 20 August 2019 Accepted: 10 December 2019 26. Zhang XM, Mitchell DG, Dohke M, et al. Pancreatic cysts: depiction on single-shot fast spin-echo MR images. Radiol. 2002;223:547–53. References Value of endoscopy for early diagnosis of pancreatic carcinoma "in press". Dig Endosc. 2019. 8. Canto MI, Goggins M, Yeo CJ, et al. Screening for pancreatic neoplasia in high-risk individuals: an EUS-based approach. Clin Gastroenterol Hepatol. 2004;2:606–21. 34. Sato A, Irisawa A, Bhutani MS, et al. Significance of normal appearance on endoscopic ultrasonography in the diagnosis of early chronic pancreatitis. Endosc Ultrasound. 2018;7:110–8. 9. Hori M, Matsuda T, Shibata A, et al. Cancer incidence and incidence rates in Japan in 2009: a study of 32 population-based cancer registries for the monitoring of Cancer incidence in Japan (MCIJ) project. Jpn J Clin Oncol. 2015;45:884–91. 35. Mizuno S, Isayama H, Nakai Y, et al. Prevalence of pancreatic cystic lesions is associated with diabetes mellitus and obesity: an analysis of 5296 individuals who underwent a preventive medical examination. Pancreas. 2017;46:801–5. 10. Negreanu L, Preda CM, Ionescu D, et al. Progress in digestive endoscopy: flexible spectral imaging colour enhancement (FICE)-technical review. J Med Life. 2015;8:416–22. 36. Lew D, Afghani E, Pandol S. Chronic pancreatitis: current status and challenges for prevention and treatment. Dig Dis Sci. 2017;62:1702–12. Page 9 of 9 Uchida et al. BMC Gastroenterology (2019) 19:220 Uchida et al. BMC Gastroenterology (2019) 19:220 39. Kaneko M, Katanuma A, Maguchi H, et al. Prospective randomized, comparative study of delineation capability of radial scanning and curved linear array endoscopic ultrasound for the pancreaticobiliary region. Endosc Int Open. 2014;2:E160–70. Uchida et al. BMC Gastroenterology (2019) 19:220 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
https://openalex.org/W2995958582	https://europepmc.org/articles/pmc6982832?pdf=render	English	null	Lipid Profiling and Stable Isotopic Data Analysis for Differentiation of Extra Virgin Olive Oils Based on Their Origin	Molecules/Molecules online/Molecules annual	2,019	cc-by	13,963	Lipid Proﬁling and Stable Isotopic Data Analysis for Diﬀerentiation of Extra Virgin Olive Oils Based on Their Origin Igor Luki´c 1,† , Alessio Da Ros 2,† , Graziano Guella 3 , Federica Camin 2, Domenico Masuero 2 , Nadia Mulinacci 4, Urska Vrhovsek 2,* and Fulvio Mattivi 2,3,* Igor Luki´c 1,† , Alessio Da Ros 2,† , Graziano Guella 3 , Federica Camin 2, Domenico Masuero 2 , Nadia Mulinacci 4, Urska Vrhovsek 2,* and Fulvio Mattivi 2,3,* 2 Fondazione Edmund Mach, Research and Innovation Centre, Via E. Mach 1, 38010 San Michele all’Adige, Italy; darosalessio92@gmail.com (A.D.R.); federica.camin@unitn.it (F.C.); domenico.masuero@fmach.it (D.M.) 3 Department of Physics, University of Trento, Via Sommarive 14, 38123 Povo Trento, Italy; graziano.guella@unitn.it 4 NEUROFARBA, Pharmaceutical and Nutraceutical Division, University of Florence, Via Ugo Schiﬀ6, 50019 Sesto Fiorentino, Italy; nadia.mulinacci@uniﬁ.it Correspondence: urska.vrhovsek@fmach.it (U.V.); fu * Correspondence: urska.vrhovsek@fmach.it (U.V.); fulvio.mattivi@unitn.it (F.M.) Academic Editor: Roberto Fabiani Received: 14 November 2019; Accepted: 11 December 2019; Published: 18 December 2019 Abstract: To diﬀerentiate extra virgin olive oils (EVOO) according to the origin of purchase, such as monocultivar Italian EVOO with protected denomination of origin (PDO) and commercially-blended EVOO purchased in supermarkets, a number of samples was subjected to the analysis of various lipid species by liquid chromatography/mass spectrometry (LC-ESI-MS/MS, LC-ESI-IT-MS) and proton nuclear magnetic resonance analysis (1H-NMR). Many putative chemical markers were extracted as diﬀerentiators by uni- and multivariate statistical analysis. Commercially-blended EVOO contained higher concentrations of the majority of minor lipids, including free fatty acids, their alkyl (methyl and ethyl) esters, monoglycerides, and diglycerides, which may be indicative of a higher degree of triglyceride lipolysis in these than in monocultivar PDO EVOO. Triterpenoids and particular TAG species were also found in higher proportions in the samples from the commercially-blended EVOO class, suggesting a possible inﬂuence of factors such as the cultivar and geographical origin. The largest diﬀerences between the classes were determined for the concentrations of uvaol and oleanolic acid. The results of the analysis by isotopic ratio mass spectrometry (IRMS) were reasonably consistent with the information about the geographical origin declared on the labels of the investigated EVOOs, showing considerable variability, which possibly also contributed to the diﬀerences in lipid composition observed between the two investigated classes of EVOO. Keywords: extra virgin olive oil; lipids; LC-MS/MS; NMR; IRMS; PDO molecules molecules Molecules 2020, 25, 4; doi:10.3390/molecules25010004 1. Introduction Extra virgin olive oil (EVOO) is appreciated among consumers because of its speciﬁc ﬂavor and nutritional properties. Due to its economic importance, EVOO is among the most common commodities subject to fraud and mislabeling, and for this reason it is protected by regulation. The international trade standard issued by the International Olive Council (IOC) [1] and the corresponding umbrella regulation in the European Union (EU) [2] include a set of analytical and sensory methods to test and conﬁrm the quality grade and authenticity of olive oil. In EU, EVOO can be additionally protected Molecules 2020, 25, 4; doi:10.3390/molecules25010004 www.mdpi.com/journal/molecules Molecules 2020, 25, 4 Molecules 2020, 25, 4 2 of 20 by protected denomination of origin (PDO) [3]. PDO EVOOs are produced according to a set of speciﬁc rules set by the holder of a designation in a speciﬁcation document, governing aspects such as olive cultivars used, cultivation, harvest and processing conditions, and oil physico-chemical and sensory characteristics. In our recent case study conducted on the Italian market it has been shown that oils labelled by the highest quality category grade (EVOO), despite meeting basic regulatory requirements, can diﬀer signiﬁcantly in qualitative terms [4]. Monocultivar PDO EVOOs purchased on family farms were found to be superior to those oﬀered at the same time in supermarkets with respect to their volatile proﬁles and sensory quality. Such large heterogeneity within the EVOO category can certainly inﬂuence and distort consumers’ perception of EVOO quality and in a way discredit the reputation of EVOO in general. Having in mind that the heterogeneity of oils within the EVOO category with respect to origin of purchase is certainly among the less studied topics in the ever-growing scientiﬁc area of olive oil traceability and quality, more studies are needed to ﬁnd reliable chemical markers able to discriminate EVOO based on this criterion. Such ﬁndings would signiﬁcantly contribute in clarifying the interrelationship between EVOO origin, overall quality and price, and would as well provide a basis for designing additional measures of protection of the PDO EVOO class in general, which is most likely target of fraud by mislabeling due to its large economic importance [5]. y g y g g p Triglycerides (triacylglicerols—TAGs), which are basically esters of glycerol and fatty acids (FAs), are the main neutral lipid component of olive oil (ca. 98%) [6]. 1. Introduction Olive oil TAGs contain primarily oleic (C18:1), palmitic (C16:0), linoleic (C18:2n−6), stearic (C18:0), palmitoleic (C16:1), and linolenic (C18:3n−3) acids, while others occur in minor amounts. Monounsaturated FAs (MUFAs) and essential polyunsaturated FAs (PUFAs) are among the most important nutritional elements of EVOO. The consumption of MUFAs has been associated with decrease of several cardiovascular risk factors [7], while EVOO linoleic and linolenic acids are an important source of essential FAs in human nutrition [8]. The percentage of TAGs with equivalent carbon number 42 (ECN 42) in total TAGs, where ECN is the sum of the number of carbon atoms in three constituent FAs in TAG molecule subtracted by 2 × total of double bonds, can be used as a marker for the detection of the presence of seed oils in olive oil [2]. In combination with other olive oil constituents, TAGs have been successfully utilized as markers of varietal [9–11] and geographical origin of EVOO [11]. In virgin olive oils, diglycerides (DAGs) are present in a range of 1% to 3% in the form of 1,2- and 1,3-isomers, whose ratio has been used as a marker of olive oil “freshness” [12]. The percentage of total free FAs (FFAs or acidity) is one of the parameters which is evaluated for the purposes of olive oil quality categorization [1,2]. The composition of olive oil total FA (the sum of those bound with glycerol in TAGs and the FFA forms) is usually determined by gas chromatography with ﬂame-ionization or mass spectral detection after TAG hydrolysis and methylation [2] and may provide important information about olive oil nutritional quality (level of FA unsaturation) and purity. In the last years, proton nuclear magnetic resonance (1H-NMR) spectral analysis of olive oil is often used to determine not only the % molar ratio of fatty acyl chains in TAG but also the relative amount of DAGs and, eventually, to detect the presence of peroxidized acyl chains in TAGs [13]. Several results have also proved the usefulness of FA composition and distribution on the glycerol moiety for the establishment of EVOO cultivar or geographical origin [14–16]. The concentration of alkyl esters of free FAs (FAAE), speciﬁcally ethyl esters, is included among the criteria for olive oil quality grade evaluation [1,2]. 1. Introduction These markers originate mostly from inappropriate handling of olive fruit and oil [17,18] so their concentrations may be used to detect fraudulent mixtures of EVOO with lower quality oils, including deodorized ones. It is general opinion that the standard proﬁling of TAGs, DAGs, and total FFAs has limited discriminative power to diﬀerentiate olive oils according to various criteria [19]. However, it was assumed that recent analytical developments and novel sensitive methods could be able to provide new, more speciﬁc data with more information on olive oil lipids that could be useful for EVOO diﬀerentiation. y p p more speciﬁc data with more information on olive oil lipids that could be useful for EVOO diﬀerentiation. In this work, we applied a multi-methodological approach based on several potent analytical techniques in order to ﬁnd new reliable markers able to discriminate EVOO based on the origin p p In this work, we applied a multi-methodological approach based on several potent analytical techniques in order to ﬁnd new reliable markers able to discriminate EVOO based on the origin Molecules 2020, 25, 4 3 of 20 of purchase. A method was developed based on liquid chromatography with triple quadrupole mass spectrometric detection (LC-ESI-MS/MS) for the simultaneous quantiﬁcation of minor lipids, including the proﬁling of FFAs, which, to our knowledge, has been studied rather scarcely. As well, the method provided a more detailed composition of FA alkyl esters occurring in olive oil in relation to previous studies, supplemented by particular MAGs and triterpenoids. On the other hand, liquid chromatography with quadrupole ion-trap mass spectrometry (LC-ESI-IT-MS) was utilized for the proﬁling of major TAGs, but also aimed to the targeted detection of particular trace TAG species, such as C50:4, C50:3, and C56:3, as well as those containing FAs with odd number of carbon atoms, which, to our knowledge, have not been studied extensively until now. In order to provide more detailed and yet complementary data to those obtained by the other techniques mentioned above, all the samples were subjected to 1H-NMR quantitative analysis to obtain both the distribution among lipid species (TAG/DAG/FFA) and a reliable distribution of acyl chains unsaturation (saturated FA (SFA)/MUFA/PUFA). The latter parameter also allowed establishing the unsaturation index (UI) and the iodine value (IV), two useful chemical parameters in EVOO quality control. 1. Introduction p q y Besides quality in general, the most important aspect that deﬁnes a given PDO and drives the consumers’ preferences towards this class of EVOO is in fact the authenticity of its geographical origin. One of the most useful techniques to prove and authenticate the geographical origin of EVOO is stable isotope ratio analysis by isotope ratio mass spectrometry (IRMS) [20]. For example, the stable isotope ratio of C (13C/12C, expressed as δ13C) of palmitic, oleic, and linoleic FAs was successful in diﬀerentiating olive oils according to the country or region of provenience [20,21]. Both the δ13C and 18O/16O (δ18O) ratios determined for bulk olive oils from various countries were found to change according to the latitude, the distance from the sea and the environmental conditions during growing of the plants [22,23]. The δ13C and δ2H (2H/1H) of n-C29 alkanes were signiﬁcantly more positive in olive oils from the southern compared with northern Mediterranean countries [24]. As for Italian EVOOs, works carried on δ13C and δ18O [25,26], also in combination with δ2H [27,28], proved that it wass possible to distinguish samples from diﬀerent Italian macro areas, as well as Italian from other Mediterranean olive oils. The three isotopic ratios, in particular δ2H and δ18O, were found to be correlated to the climatic (mainly temperature) and geographical (mainly latitude and distance from the coast) characteristics and to the δ18O and δ2H of the surface waters as well to the year of production [25–28]. The δ2H values signiﬁcantly distinguished olive oils produced on the Adriatic from those from the Tyrrhenian coast of Italy in each year [29]. The combination of isotopic analysis with 1H-NMR proﬁling achieved optimal discrimination between Greece, Spain, Italy, Turkey, Crete, France, and between Italy and Tunisia, the country from which the largest amount of olive oil is imported in Europe [30]. In this study, it was assumed that IRMS would be able to provide relevant information on the three isotopic ratios in the two investigated EVOO classes that could be useful for the conﬁrmation of geographical origin declared on their labels, and possibly for their diﬀerentiation. The aim of this study was dual. In the first part, sensitive analytical methods based on complementary LC-MS and NMR techniques were applied to detect less known chemical markers among various lipid species able to differentiate Italian monocultivar PDO EVOOs obtained on family farms from those purchased in supermarkets. 1. Introduction The second goal was to verify the declared geographical origin of the EVOOs from the both classes by IRMS analysis. It was considered that such findings would significantly contribute to EVOO diversification on the market, and would help to clarify the interrelationship between EVOO origin, quality, and price, and in this way support the growth of the niche in the market segment of consumers informed and interested in healthy, quality products with remarkable diversity and clear identity. 2.1. LC-ESI-MS/MS Analysis of Free Fatty Acids, Fatty Acid Methyl and Ethyl Esters, Monoglicerides and Triterpenoids 2.1. LC-ESI-MS/MS Analysis of Free Fatty Acids, Fatty Acid Methyl and Ethyl Esters, Monoglicerides and Triterpenoids Validation parameters for the method of determination of minor lipid compounds in EVOO by LC-ESI-MS/MS are shown in Tables S3 and S4. All the calibration curves exhibited good linearity 4 of 20 Molecules 2020, 25, 4 (r2 values from 0.95 to 1.00). Limits of quantiﬁcation ranged from 0.2 to 40 µg/L depending on the compound. The linearity data were used to assess the percentage of matrix eﬀect (% ME), which was reported in Table S3. The matrix eﬀect was found to be insigniﬁcant because the obtained variability was close to %RSD repeatability values [31]. Thus, curves prepared in solvent were selected for the quantiﬁcation. The coeﬃcients of variation (CV%) did not exceed 15% for intra-day assay and 20% for inter-day assay. The average recovery was in the range from 65% to 125% with %RSD less or equal 20%, which was considered satisfactory. The recovery values obtained surpassed 90% for 14 compounds, were between 80% and 90% for 4 compounds and between 70% and 80% for 3 compounds, with %RSD values between 1% and 11%. This indicated good accuracy, recovery, and precision of the method. It is worth mentioning that several lipids for which the method was also validated (Table S1), such as carnitines, glycerophospholipids, and sphingolipids, were not detected in the samples of this study, since their concentrations were below the determined limits of detection (LOD). In all the investigated EVOO samples 20 minor lipids were identiﬁed, including 10 free fatty acids (FFAs), six FFA methyl and ethyl esters, two monoglycerides (MAGs), and two triterpenoids (Table 1). The most abundant among FAAs was oleic (C18:1) followed by palmitic (C16:0), linoleic (C18:2), stearic (C18:0), and palmitoleic (C16:1) acids, which corresponded to the natural distribution of total FAs (esteriﬁed in TAGs + FFA) in EVOO in general [6]. The methyl and ethyl esters of the most abundant FFAs, that is oleates, palmitates, and linoleates, dominated the alkyl ester composition, while oleic and linoleic acids were a structural part of the only two identiﬁed MAGs (Table 1). Table 1. Concentrations (µg/g) of minor lipids in monocultivar protected designation of origin (PDO) and commercially-blended extra virgin olive oils obtained by LC-ESI-MS/MS analysis. An asterisk () in a row represents signiﬁcant diﬀerences between mean values at p < 0.05 obtained by ANOVA and least signiﬁcant diﬀerence (LSD) test. 2.1. LC-ESI-MS/MS Analysis of Free Fatty Acids, Fatty Acid Methyl and Ethyl Esters, Monoglicerides and Triterpenoids Table 1. Concentrations (µg/g) of minor lipids in monocultivar protected designation of origin (PDO) and commercially-blended extra virgin olive oils obtained by LC-ESI-MS/MS analysis. An asterisk () in a row represents signiﬁcant diﬀerences between mean values at p < 0.05 obtained by ANOVA and least signiﬁcant diﬀerence (LSD) test. Minor Lipids Origin/Class Monocultivar PDO Commercially-Blended Free fatty acids Palmitic acid (C16:0) 285.41 362.95 Palmitoleic acid (C16:1) 10.58 16.57 Stearic acid (C18:0) 27.41 43.39 Oleic acid (C18:1) 809.53 1045.52 Linoleic acid (C18:2) 70.52 93.10 Linolenic acid (C18:3) 5.67 9.76 * Arachidic acid (C20:0) 8.53 11.42 Behenic acid (C22:0) 8.52 9.45 Erucic acid (C22:1) 0.06 0.05 Lignoceric acid (C24:0) 21.85 27.04 Total free fatty acids 1248.09 1619.27 Free fatty acid esters Methyl oleate 4.02 7.09 * Methyl linoleate 0.05 0.24 * Ethyl palmitate 0.10 1.01 * Ethyl stearate 0.00 0.32 * Ethyl oleate 2.29 8.18 * Ethyl linoleate 0.38 1.04 * Total free fatty acid esters 6.83 17.87 * Monoglycerides 1-Oleoyl-rac-glycerol 30.11 53.29 * 1-Linoleoyl-rac-glycerol 8.77 12.84 * Total monoglycerides 38.88 66.13 * Triterpenoids Oleanolic acid 26.71 50.13 * Uvaol 4.58 14.04 * Total triterpenoids 31.28 64.17 * Minor Lipids Origin/Class Monocultivar PDO Commercially-Blended 5 of 20 Molecules 2020, 25, 4 FFAs in olive oil derive from the breakdown of TAGs by lipolysis. There are many factors which can aﬀect the degree of TAG lipolysis, including anomalies during biosynthesis, microbial activity, and environmental factors. Infestation by the olive ﬂy (Bactrocera oleae) is a major cause of high FFA content in olives. Damaged olive fruits, delayed fruit processing, and storage in inappropriate conditions result in increased lipolysis rates, while olive oil extraction which is not properly conducted (e.g., prolonged contact between oil and vegetation water) may also result in high FFA values [6,32]. Therefore, the content of FFA is directly related to the quality of olive oil and reﬂects the care taken from blossoming and fruit set to the eventual sale and consumption of the oil [33]. The FFA content, also known as acidity, is one of the main criteria used to establish diﬀerent categories of olive oil: according to the European community, EVOO, as the highest quality category, must have FFA content below or equal to 0.8% (as oleic acid, w/w), as obtained by the standard titration method [2]. 2.1. LC-ESI-MS/MS Analysis of Free Fatty Acids, Fatty Acid Methyl and Ethyl Esters, Monoglicerides and Triterpenoids In this work, the average total FFA concentration obtained by LC-ESI-MS/MS analysis did not exceed 0.2% in neither of the two investigated EVOO classes (Table 1). Although a tendency towards higher concentrations in commercially-blended than in PDO EVOO was noted for the majority of FFAs, statistically signiﬁcant diﬀerences were found only for linolenic acid. Such a result was, to some extent, in accordance with a previous study in which low-priced EVOO samples were found to contain more FFAs than EVOOs of higher price [34]. FFA content was previously shown to increase during olive oil storage and aging [35,36]. In this work, in contrast to the monocultivar PDO EVOOs which were analyzed relatively fresh, the age of commercially-blended EVOOs was not declared by the producers/sellers and it was practically unknown. It was possible that the samples from the latter class were fully or partially composed from oils obtained in harvests prior to 2016, and that the increased concentrations of particular FFAs partially resulted from TAG chemical hydrolysis during aging. Despite the possible diﬀerences with respect to the EVOO age, it must be kept in mind that all the EVOOs included in this study were carefully selected and sampled at the same time, and were therefore valid and authentic representatives of the both classes of EVOOs oﬀered on the market at that given moment. The content of fatty acid alkyl esters (FAAEs) was principally introduced among the chemical parameters controlled in olive oil quality evaluation [2] to detect blends including low quality olive oils with weak organoleptic defects [17,18]. FAAEs are formed by esteriﬁcation of short-chain alcohols methanol and ethanol with FAAs yielding methyl and ethyl esters, respectively, although transesteriﬁcation with triglycerides or partial glycerides may also be a source [37]. They are generally considered indicators of lower olive oil quality and their high concentration often indicate the use of olive fruits with fermentative alterations [17,38–40]. In fact, it was demonstrated that FAAE formation was not limited mainly by the content of FAA, but it appeared to be strongly related to the concentration of free alcohols in oil, among which ethanol can be produced exclusively by fermentation [37]. However, evidence exists that the content of FAAE can be relatively high even in high quality EVOO, and vice versa [41]. 2.1. LC-ESI-MS/MS Analysis of Free Fatty Acids, Fatty Acid Methyl and Ethyl Esters, Monoglicerides and Triterpenoids In this study, the average total concentration of fatty acid ethyl esters (FAEEs) and the total FAAE concentration were below the maximum limit of 35 mg/kg prescribed by the European Commission regulation for EVOO [2]. Higher concentrations of all the identiﬁed FAAE/FAEEs were found in commercially-blended EVOOs (Table 1), which indicates the possibility that the olives used for the production of particular samples from this class were overripe or of lower quality suﬀering from fermentative alterations. Such results are in agreement with our previous report generated from a study with the same sample set, where commercially-blended olive oils were characterized by lower sensory quality on the average, with a number of samples having a sensory defect, including fusty/muddy sediment, vinegary/winey, or musty [4] which could have originated from undesirable fermentative processes, as reported earlier [38,42]. FAAE concentration was previously shown to increase during storage [41,43], which is another possible cause of the higher concentration found in commercially-blended EVOOs, which were possibly not fresh at the moment of sampling. y p y p g Similar to diglycerides (DAGs), the presence of MAGs in olive oil is a result of either incomplete biosynthesis of TAGs or their later hydrolysis during processing and storage [12,44]. In virgin olive oil, DAGs are present in the range of 1–2.8%, while MAGs are found in amounts lower than 0.25%, Molecules 2020, 25, 4 Molecules 2020, 25, 4 6 of 20 with a much higher proportion of those containing a single fatty acid on position 1 than on position 2 of glycerol moiety [45,46]. The identiﬁcation of only two 1-MAG species in this work conﬁrmed this phenomenon (Table 1). Both compounds, 1-oleoyl- and 1-linoleoyl-rac-glycerol were found in higher concentration in commercial–blended EVOO, conﬁrming the possibility that a higher degree of TAG hydrolysis occurred in these than in PDO EVOOs. The largest portion of triterpenoids in olive fruit is located in its epicarp. Triterpenoid concentration in olive oil can be increased by longer malaxation durations and higher malaxation temperatures during olive processing, depending on the compound and olive cultivar [47]. High levels of particular triterpenoids may indicate the presence of olive pomace oil in EVOO, and the percentage of the sum of triterpene diols uvaol and erythrodiol with respect to total sterols is in fact included among the criteria which are evaluated in testing EVOO authenticity in EU [2]. 2.1. LC-ESI-MS/MS Analysis of Free Fatty Acids, Fatty Acid Methyl and Ethyl Esters, Monoglicerides and Triterpenoids Higher concentrations of the triterpenoids identiﬁed in this study, oleanolic acid and uvaol, were found in commercially-blended EVOO (Table 1). Although a possibility should not be neglected that the malaxation and processing parameters were the cause, it must be kept in mind that the content of triterpenoids strongly depends on cultivar origin [47,48], which is therefore another possible source of the observed diﬀerence. 2.2. LC-ESI-IT-MS Analysis of Triglycerides Relative proportions (%) of triglycerides (TAGs) obtained by LC-ESI-IT-MS analysis in monocultivar PDO and commercially-blended EVOO are reported in Table 2. As expected, TAGs consisting of the most abundant naturally occurring FAs in olive oil, the species 54:3, 52:2, 54:4, and 52:3, dominated the proﬁles in both classes of EVOO. The obtained proﬁles did not fully coincide with those obtainable by the oﬃcial EU method [2], with some TAGs not reported and some additionally identiﬁed, which suggests the method applied in this study could be used as a complementary approach to the standard one for obtaining additional information. Table 2. Relative proportions (%) of triglycerides (TAGs) in monocultivar protected designation of origin (PDO) and commercially-blended extra virgin olive oils obtained by liquid chromatography with quadrupole ion-trap mass spectrometry (LC-ESI-IT-MS) analysis. An asterisk () in a row represents signiﬁcant diﬀerences between mean values at p < 0.05 obtained by ANOVA and least signiﬁcant diﬀerence (LSD) test. TAG Species † TAG Chains ‡ Origin/Class Monocultivar PDO Commercially-Blended TAG 50:1 18:1 16:0 16:0 5.83 5.44 TAG 50:2 18:1 16:1 16:0 3.06 2.84 TAG 50:3 - - - 0.63 0.70 TAG 50:4 - - - 0.07 0.08 TAG 52:1 18:1 18:0 16:0 2.32 2.47 TAG 52:2 18:1 18:1 16:0 19.10 19.06 TAG 52:3 18:2 18:1 16:0 10.36 9.64 TAG 52:4 18:3 18:1 16:0 3.38 3.13 TAG 52:5 - - - 0.46 0.42 TAG 52:6 - - - 0.05 * 0.03 TAG 53:2 18:1 18:1 17:0 0.42 0.44 TAG 53:3 18:1 18:1 17:1 0.68 0.70 TAG 53:4 - - - 0.12 0.14 TAG 54:1 - - - 0.68 0.80 * TAG 54:2 18:1 18:1 18:0 6.37 7.34 * TAG 54:3 18:1 18:1 18:1 25.40 26.59 TAG 54:4 18:2 18:1 18:1 11.86 11.56 TAG 54:5 18:3 18:1 18:1 4.55 4.47 TAG 54:6 - - - 0.74 0.70 TAG 56:1 - - - 0.16 0.18 TAG 56:2 20:0 18:1 18:1 1.47 1.51 TAG 56:3 20:1 18:1 18:1 1.63 1.53 7 of 20 Molecules 2020, 25, 4 Table 2. Cont. 2.2. LC-ESI-IT-MS Analysis of Triglycerides TAG Species † TAG Chains ‡ Origin/Class Monocultivar PDO Commercially-Blended TAG 56:4 - - - 0.43 0.38 TAG 58:1 - - - 0.06 0.06 TAG 58:2 22:0 18:1 18:1 0.38 0.39 TAG 58:3 - - - 0.11 0.10 TAG peroxides § - - - 0.77 0.72 † Number of carbon atoms: double bonds in the structure of the corresponding TAG; ‡ number of carbon atoms: double bonds in each of the three fatty acids in the structure of the corresponding TAG; § proportion (%) of the sum of 52:2, 52:3, 54:3, 54:4, and 54:5 TAG peroxides in total TAGs. † Number of carbon atoms: double bonds in the structure of the corresponding TAG; ‡ number of carbon atoms: double bonds in each of the three fatty acids in the structure of the corresponding TAG; § proportion (%) of the sum of 52:2, 52:3, 54:3, 54:4, and 54:5 TAG peroxides in total TAGs. The diﬀerences between the two EVOO classes with respect to average TAG composition were not large (Table 2). TAG composition is not evaluated among the parameters related to EVOO sensory quality, while a part of EVOO nutritional value linked to TAGs, related mainly to the FA unsaturation level, depends mostly on cultivar, geographical origin and fruit ripening degree [9–11,34]. The diﬀerences between the two classes of EVOO, observed for a relatively small number of TAG species, could be primarily ascribed to the abovementioned factors. TAGs 52:3 and 52:6 occurred in a higher average percentage in monocultivar PDO EVOO, while TAGs 54:1 and 54:2 stood out with higher values in commercially-blended EVOO (Table 2). The proportions of the TAG species that could possibly be associated with particular alterations in production with possible repercussions on olive oil quality, such as species that include FAs with odd number of carbon atoms (53:2 and 53:3) and TAG peroxides, did not diﬀer between the two classes. The proportions of particular minor TAG isomers, not included in the oﬃcial methods [2] and investigated rather scarcely in olive oil up to date, are reported in Table 3. Signiﬁcant diﬀerences between the two EVOO classes were found only for the relative proportions of TAG 50:4 I and II species. † Number of carbon atoms: double bonds in the structure of the corresponding TAG; ‡ Number of carbon atom double bonds in each of the three fatty acids in the structure of the corresponding TAG. Number of carbon atoms: double bonds in the structure of the corresponding TAG; ‡ N double bonds in each of the three fatty acids in the structure of the corresponding TAG. † Number of carbon atoms: double bonds in the structure of the corresponding TAG; ‡ Number of carbon atoms: double bonds in each of the three fatty acids in the structure of the corresponding TAG 2.3. H-NMR Analysis of Lipids The lipid (essentially TAG) composition of EVOO samples was established by measurement and analysis of the corresponding 1H-NMR spectra as described in Section 3.4. The main features of the spectra are outlined in Figure 1 reporting also one of the main TAG species present in olive oils (TAG 54:4, 18:1/18:1/18:2) as a model. The data reported in Table 4 were obtained by area peak integration and simple equations relating them to the number of protons at a given position. Figure 1. 400 MHz proton nuclear magnetic resonance (1H-NMR) spectrum of a monocultivar protected designation of origin (PDO) extra virgin olive oil sample in CDCl3 at 300 K. Figure 1. 400 MHz proton nuclear magnetic resonance (1H-NMR) spectrum of a monocultivar protected designation of origin (PDO) extra virgin olive oil sample in CDCl3 at 300 K. Table 4. Lipids in monocultivar protected designation of origin (PDO) and commercially-blended extra virgin olive oils obtained by NMR analysis. An asterisk () in a row represents signiﬁcant diﬀerences between mean values at p < 0.05 obtained by ANOVA and least signiﬁcant diﬀerence (LSD) test. Table 4. Lipids in monocultivar protected designation of origin (PDO) and commercially-blended extra virgin olive oils obtained by NMR analysis. An asterisk () in a row represents signiﬁcant diﬀerences between mean values at p < 0.05 obtained by ANOVA and least signiﬁcant diﬀerence (LSD) test. Table 4. Lipids in monocultivar protected designation of origin (PDO) and commercially-blended extra virgin olive oils obtained by NMR analysis. An asterisk () in a row represents signiﬁcant diﬀerences between mean values at p < 0.05 obtained by ANOVA and least signiﬁcant diﬀerence (LSD) test. Lipids Origin/Class Monocultivar PDO Commercially-Blended Total triglycerides (TAGs) 99.38 99.05 Total saturated fatty acids (SFA) 15.83 15.52 Total monounsaturated fatty acids (MUFA) 74.95 75.77 Linoleic acid in TAG (C18:2) 7.92 7.61 Linolenic acid in TAG (C18:3) 1.30 1.10 Unsaturation index (UI) 0.95 0.94 Iodine value (IV) 82.33 84.10 * Total 1,2-diacylglycerols (1,2-DAGs) 0.56 0.72 * Estimated free fatty acids and minor lipids 0.07 0.26 * In particular, the ratio of the peaks area F/H represents the best way to validate the approximations used in the approach for the analysis of edible oils relying on the large dominance of TAG species [49]. 2.2. LC-ESI-IT-MS Analysis of Triglycerides Although the source of the observed diﬀerences remained unexplained at this stage, and it could only be assumed that factors such as cultivar and geographical origin, respectively, could have had a signiﬁcant eﬀect, the results obtained are certainly intriguing and imply the need to further investigate the signiﬁcance of the trace TAG species in olive oil. Table 3. Relative proportions (%) of minor triglycerides (TAGs) and particular TAG isomers in monocultivar protected designation of origin (PDO) and commercially-blended extra virgin olive oils obtained by HPLC-HRMS analysis. An asterisk () in a row represents signiﬁcant diﬀerences between mean values at p < 0.05 obtained by ANOVA and least signiﬁcant diﬀerence (LSD) test. TAG Species † TAG Chains ‡ Origin/Class Monocultivar PDO Commercially-Blended TAG 50:4 isomers TAG 50:4 I 18:2 16:1 16:1 24.54 31.20 TAG 50:4 II 18:3 16:1 16:0 75.46 * 68.80 total - - - 100.00 100.00 TAG 50:3 isomers TAG 50:3 II 18:2 16:1 16:0 66.56 67.55 TAG 50:3 III 18:3 16:0 16:0 33.44 32.45 total 18:1 18:1 16:0 100.00 100.00 TAG 56:3 isomers TAG 56:3 I 20:1 18:1 18:1 71.63 71.96 TAG 56:3 II 20:0 18:2 18:1 28.37 28.04 total - - - 100.00 100.00 TAG 53 isomers TAG 53:4 - - - 10.12 10.41 TAG 53:3 18:1 18:1 17:1 56.17 54.79 TAG 53:2 18:1 18:1 17:0 33.71 34.80 total 18:1 18:1 18:0 100.00 100.00 † Number of carbon atoms: double bonds in the structure of the corresponding TAG; ‡ Number of carbon atoms: double bonds in each of the three fatty acids in the structure of the corresponding TAG. 8 of 20 Molecules 2020, 25, 4 2.3. H-NMR Analysis of Lipids Although the averaged UI (0.941 in the ﬁrst set versus 0.950 in the second set) and iodine value (84.0 versus 84.5, respectively) of the two mentioned sets of commercially-blended EVOO were quite similar, the second set was characterized by signiﬁcantly higher relative amounts of SFA and PUFA and lower amount of MUFA (oleic chain, essentially). The diﬀerences between the two classes of the investigated EVOO with respect to the 1H-NMR data can be seen in Table 4. Both classes were characterized by relatively similar major lipid parameters. Monocultivar EVOOs were distinguished by a higher level of linolenic acid and lower iodine value (IV), while a signiﬁcant diﬀerence for unsaturation index (UI) was not found, meaning an unambiguous general conclusion about the diﬀerence between the two classes of EVOO with respect to the level of unsaturation could not be made at this point. Level of saturation of acyl chains in olive oil TAGs may depend on various factors, including geographical position and climate, as well as varietal origin [50]. It is possible that these were among the main sources of both intra- and inter-class variability of lipid composition observed in this study. Stereospeciﬁc distribution of FAs in DAG is known to be aﬀected by several factors. 1,2-DAG isomers are commonly attributed to the incomplete biosynthesis of TAGs in olive fruit, whereas 1,3-DAGs are considered to derive mainly from enzymatic or chemical hydrolysis of TAGs before or during oil extraction [12]. It was shown that 1,2-, 1,3-, and total DAG concentrations in olive oil signiﬁcantly increase as a result of alterations during processing, including, for example, prolonged storage of piled olives before processing [12,51]. During storage 1,2- species isomerize to more stable 1,3-DAGs, making the ratio of 1,3-/1,2-DAG a useful criterion indicative of olive oil age [12,44,51]. In this study, a higher proportion of 1,2-DAG fraction was found in commercially-blended than in PDO EVOO (Table 4), while 1,3-DAG isomers were not identiﬁed. Considering the contents of the other tentative indicators of olive oil age evaluated in this study, such as FFAs and FAAEs, it was tentatively assumed that commercially-blended EVOO were at least partially composed of olive oils obtained during harvests prior to 2016, meaning their age was older, on the average, than that of the PDO ones. 2.3. H-NMR Analysis of Lipids Its value should be exactly 1.500 in an oil containing only TAG species, since the signal F (δH ≈2.30 brt, -CH2 in α-position in the acyl chains) represents six protons and the signal H (δH ≈4.29 dd and δH ≈4.14 dd, -CH2 from sn-1,3 TAG) represents four protons. F/H values higher than 1.500 can be explained by the presence of DAG, MAG, and/or FFA which give their contribution to F but do not contribute to the H peak area. As much as the F/H ratio diverges from 1.500 the contribution of 9 of 20 Molecules 2020, 25, 4 DAG, MAG, and FFA becomes higher, but within the range 1.450 ≤F/H ≤1.550 the approach is still considered reliable. Thus, the F/H ratio indicates whether the approximations are correct and therefore produce trustworthy analytical data for the % molar fraction of SFA, MUFA, and PUFA fatty acyl chains in a targeted olive oil. In this work, the average F/H value of all the analyzed EVOO was 1.512 with a very small relative standard deviation of 0.5%, conﬁrming the calculations were valid and reliable. y g A certain degree of intra-class heterogeneity was observed within both classes of investigated EVOO with respect to the relative proportions of total SFA, MUFA, and PUFA. Concerning the relative amount of the lipid acyl chains, a bimodal distribution of PDO EVOOs was noted, with a major set (14 of 20 PDO samples) showing a classical distribution of SFA/MUFA/PUFA = 15.1 ± 1.5/76.3 ± 1.4/8.6 ± 0.9, whilst the remaining lead to the averaged distribution values lower in MUFA, such as SFA/MUFA/PUFA = 16.1 ± 0.8/72.6 ± 0.8/10.4 ± 0.5. Worth of note, particularly unexpected distributions were observed in particular samples, such as in Ottobratico cultivar PDO EVOO from Reggio Calabria with SFA/MUFA/PUFA = 19.0 ± 0.3/70.8 ± 0.3/10.2 ± 0.2 (richer in SFA and PUFA), and in Taggiasca cultivar PDO EVOO from Imperia with SFA/MUFA/PUFA = 12.5 ± 0.1/80.0 ± 0.2/7.5 ± 0.1 (richer in MUFA). For commercially-blended EVOO a similar intra-class diﬀerentiation was observed, with the most populated set (19 of 25 samples) centered at average molar fractions SFA/MUFA/PUFA = 15.0 ± 0.9/76.9 ± 1.9/8.1 ± 15 and a minor set (6 of 25) centered at average molar fractions SFA/MUFA/PUFA = 17.1 ± 0.5/72.1 ± 1.2/10.8 ± 1.5. 2.4. Multivariate Statistical Analysis Principal component analysis (PCA) separated the samples belonging to the two classes of investigated EVOO according to the origin of purchase relatively successfully (Figure 2). Although monocultivar PDO EVOOs were produced from diﬀerent olive cultivars grown in diﬀerent geographical areas in Italy, they were grouped much closer to each other than the commercially-blended ones, suggesting a greater level of intra-class homogeneity with respect to the proﬁle of lipids. It is possible that the presumed diﬀerences between the average age of the EVOOs from the two classes, and still a relatively homogenous geographical origin of Italian PDO in comparison to possibly heterogeneous provenience of the commercially-blended EVOOs (Italy and other EU countries), were among the causes. Several markers were found to be related to the commercially-blended EVOOs, including all the minor lipid species, such as FFAs, FAAEs, MAGs, and DAGs, and also particular TAG species and iodine value, which corresponded completely to the one-way ANOVA results. As mentioned above, the positions of the commercially-blended EVOO samples on Cartesian plane were rather dispersed, pointing to the very heterogeneous composition of minor lipids and potential quality of these samples. Molecules 2019, 10 of 20 Figure 2. (a) and (c) Separation of olive oils sold as extra virgin olive oil (EVOO) in Italy according to the origin of purchase in two-dimensional space defined by the first three principal components, PC1, PC2, and PC3. Green cycles represent monocultivar protected designation of origin (PDO)EVOO purchased on family farms, while red cycles represent commercially-blended EVOO purchased in supermarkets (b) and (d) Factor loadings of selected variables, i.e., concentrations or proportions of various lipid species, obtained by LC-ESI-MS/MS, LC-ESI-IT-MS, and 1H-NMR, respectively, on PC1, PC2, and PC3. Relati ely ood e a atio obtai ed by hie a hi al lu te i a aly i o fi ed that the t o Figure 2. (a) and (c) Separation of olive oils sold as extra virgin olive oil (EVOO) in Italy according to the origin of purchase in two-dimensional space deﬁned by the ﬁrst three principal components, PC1, PC2, and PC3. Green cycles represent monocultivar protected designation of origin (PDO)EVOO purchased on family farms, while red cycles represent commercially-blended EVOO purchased in supermarkets (b) and (d) Factor loadings of selected variables, i.e., concentrations or proportions of various lipid species, obtained by LC-ESI-MS/MS, LC-ESI-IT-MS, and 1H-NMR, respectively, on PC1, PC2, and PC3. Figure 2. 2.3. H-NMR Analysis of Lipids Knowing that the concentration of 1,2-DAGs decreases during storage, it was expected that the presumably older commercially-blended EVOO would be characterized by lower amounts in relation to PDO EVOO, but this was not the case. Nevertheless, the possibility cannot be rejected that the commercially-blended samples contained higher average concentration of 1,2-DAGs already at the moment of production and/or release on the market, which later decreased but were still higher than that found in monocultivar PDO EVOOs. The diﬀerences observed between the average estimated levels of FAA and other minor lipids found in the two classes of EVOO (Table 4) correspond well to those determined for particular FAAs, 10 of 20 Molecules 2020, 25, 4 FAAEs, and MAGs by the LC-ESI-MS/MS method (Table 1). Commercially-blended EVOOs were characterized by higher levels, which was possibly mainly a result of an increased degree of TAG lipolysis, although the contribution of other factors, such as diﬀerent cultivars and geographical origin, as well as EVOO age and storage conditions, should not be completely excluded. 2.4. Multivariate Statistical Analysis (a) and (c) Separation of olive oils sold as extra virgin olive oil (EVOO) in Italy according to the origin of purchase in two-dimensional space defined by the first three principal components, PC1, PC2, and PC3. Green cycles represent monocultivar protected designation of origin (PDO)EVOO purchased on family farms, while red cycles represent commercially-blended EVOO purchased in supermarkets (b) and (d) Factor loadings of selected variables, i.e., concentrations or proportions of various lipid species, obtained by LC-ESI-MS/MS, LC-ESI-IT-MS, and 1H-NMR, respectively, on PC1, PC2, and PC3. Figure 2. (a) and (c) Separation of olive oils sold as extra virgin olive oil (EVOO) in Italy according to the origin of purchase in two-dimensional space deﬁned by the ﬁrst three principal components, PC1, PC2, and PC3. Green cycles represent monocultivar protected designation of origin (PDO)EVOO purchased on family farms, while red cycles represent commercially-blended EVOO purchased in supermarkets (b) and (d) Factor loadings of selected variables, i.e., concentrations or proportions of various lipid species, obtained by LC-ESI-MS/MS, LC-ESI-IT-MS, and 1H-NMR, respectively, on PC1, PC2, and PC3. 11 of 20 Molecules 2020, 25, 4 Molecules 2020, 25, 4 Relatively good separation obtained by hierarchical clustering analysis conﬁrmed that the two investigated classes of EVOO diﬀered notably with respect to the composition of lipids (Figure 3). Most of the conclusions were similar to those obtained by the PCA analysis: PDO EVOOs formed a more heterogeneous class, characterized by a smaller number of markers, while the commercially-blended EVOOs exhibited rather diverse lipid composition. Molecules 2019, 11 of 20 Figure 3. Hierarchical clustering analysis performed using lipid profiles found in Italian monocultivar protected designation of origin (PDO) extra virgin olive oils (EVOO) purchased on family farms and commercially-blended (CB) EVOO purchased in supermarkets in Italy. The heatmap was generated using 21 most significant compounds (the highest Fisher ratios). The rows in the heatmap represent lipids and the columns indicate samples. The colors of the heatmap cells indicate the abundance of lipids across different samples. The color gradient, ranging from dark blue through white to dark red, represents low middle and high abundance of lipid species Figure 3. Hierarchical clustering analysis performed using lipid proﬁles found in Italian monocultivar protected designation of origin (PDO) extra virgin olive oils (EVOO) purchased on family farms and commercially-blended (CB) EVOO purchased in supermarkets in Italy. The heatmap was generated using 21 most signiﬁcant compounds (the highest Fisher ratios). 2.4. Multivariate Statistical Analysis The rows in the heatmap represent lipids and the columns indicate samples. The colors of the heatmap cells indicate the abundance of lipids across diﬀerent samples. The color gradient, ranging from dark blue through white to dark red, represents low, middle, and high abundance of lipid species. Figure 3. Hierarchical clustering analysis performed using lipid profiles found in Italian monocultivar protected designation of origin (PDO) extra virgin olive oils (EVOO) purchased on family farms and commercially-blended (CB) EVOO purchased in supermarkets in Italy. The heatmap was generated using 21 most significant compounds (the highest Fisher ratios). The rows in the heatmap represent lipids and the columns indicate samples. The colors of the heatmap cells indicate the abundance of lipids across different samples. The color gradient, ranging from dark blue through white to dark red, e e e t lo iddle a d hi h abu da e of li id e ie Figure 3. Hierarchical clustering analysis performed using lipid proﬁles found in Italian monocultivar protected designation of origin (PDO) extra virgin olive oils (EVOO) purchased on family farms and commercially-blended (CB) EVOO purchased in supermarkets in Italy. The heatmap was generated using 21 most signiﬁcant compounds (the highest Fisher ratios). The rows in the heatmap represent lipids and the columns indicate samples. The colors of the heatmap cells indicate the abundance of lipids across diﬀerent samples. The color gradient, ranging from dark blue through white to dark red, represents low, middle, and high abundance of lipid species. PLSDA allowed a rather good differentiation of the two classes of investigated EVOO according to the origin of purchase (Figure 4). Interestingly, the highest variable importance in projection (VIP) scores were attributed to the triterpenoids, such as uvaol and oleanolic acid, which turned out to be the most important differentiators. Such a result confirmed once again the potential of the compounds from the olive oil unsaponifiable fraction to serve as markers according to various criteria. Besides triterpenoids, the 15 most important lipids according to PLSDA included particular FAAE and other minor lipid species abundant in commercially-blended EVOO, while certain TAGs fi d l d l i PDO EVOO PLSDA allowed a rather good diﬀerentiation of the two classes of investigated EVOO according to the origin of purchase (Figure 4). 2.4. Multivariate Statistical Analysis Interestingly, the highest variable importance in projection (VIP) scores were attributed to the triterpenoids, such as uvaol and oleanolic acid, which turned out to be the most important diﬀerentiators. Such a result conﬁrmed once again the potential of the compounds from the olive oil unsaponiﬁable fraction to serve as markers according to various criteria. Besides triterpenoids, the 15 most important lipids according to PLSDA included particular FAAE and other minor lipid species abundant in commercially-blended EVOO, while certain TAGs were conﬁrmed as related to monocultivar PDO EVOO. Molecules 2020, 25, 4 M l l 2019 12 of 20 f 12 of 20 12 f 20 Figure 4. (a) Separation of olive oils sold as extra virgin olive oil (EVOO) in Italy according to the origin of purchase in two-dimensional space by partial least squares discriminant analysis. Green cycles represent monocultivar protected designation of origin (PDO) EVOO purchased on family farms, while red cycles represent commercially-blended (CB) EVOO purchased in supermarkets (b) variable importance in projection (VIP) scores of the variables (lipids) most useful for the differentiation of monocultivar PDO EVOO and commercially-blended EVOO. Figure 4. (a) Separation of olive oils sold as extra virgin olive oil (EVOO) in Italy according to the origin of purchase in two-dimensional space by partial least squares discriminant analysis. Green cycles represent monocultivar protected designation of origin (PDO) EVOO purchased on family farms, while red cycles represent commercially-blended (CB) EVOO purchased in supermarkets (b) variable importance in projection (VIP) scores of the variables (lipids) most useful for the diﬀerentiation of monocultivar PDO EVOO and commercially-blended EVOO. ﬁ f h l b ( ) Figure 4. (a) Separation of olive oils sold as extra virgin olive oil (EVOO) in Italy according to the origin of purchase in two-dimensional space by partial least squares discriminant analysis. Green cycles represent monocultivar protected designation of origin (PDO) EVOO purchased on family farms, while red cycles represent commercially-blended (CB) EVOO purchased in supermarkets (b) variable importance in projection (VIP) scores of the variables (lipids) most useful for the differentiation of monocultivar PDO EVOO and commercially-blended EVOO. Figure 4. (a) Separation of olive oils sold as extra virgin olive oil (EVOO) in Italy according to the origin of purchase in two-dimensional space by partial least squares discriminant analysis. 2.5. Confirmation of Geographical Origin by Isotope Ratio Mass Spectrometry (IRMS) 2.5. Conﬁrmation of Geographical Origin by Isotope Ratio Mass Spectrometry (IRMS) f f g p g y p p y The average values of δ13C, δ2H, and δ18O in monocultivar PDO and commercially-blended EVOO are reported in Table 5. Statistically significant differences were determined for δ2H and δ18O, with lower values found in monocultivar PDO EVOO. According to the literature [30], the isotopic values of olive oil increase with decreasing latitude. It is possible that the contribution of Italian monocultivar PDO EVOO originating from the orchards located further from the sea and at higher latitudes with colder climate (e.g., Brescia, Verona, and Garda PDOs) prevailed and significantly decreased the average δ2H and δ18O isotopic values in monocultivar PDO EVOO, the same as non- Italian commercial EVOO originating from lower latitudes in warmer EU Mediterranean countries possibly had a notable influence on increasing the average δ2H and δ18O values determined in The average values of δ13C, δ2H, and δ18O in monocultivar PDO and commercially-blended EVOO are reported in Table 5. Statistically signiﬁcant diﬀerences were determined for δ2H and δ18O, with lower values found in monocultivar PDO EVOO. According to the literature [30], the isotopic values of olive oil increase with decreasing latitude. It is possible that the contribution of Italian monocultivar PDO EVOO originating from the orchards located further from the sea and at higher latitudes with colder climate (e.g., Brescia, Verona, and Garda PDOs) prevailed and signiﬁcantly decreased the average δ2H and δ18O isotopic values in monocultivar PDO EVOO, the same as non-Italian commercial EVOO originating from lower latitudes in warmer EU Mediterranean countries possibly had a notable inﬂuence on increasing the average δ2H and δ18O values determined in commercially-blended EVOO. mmercially-blended EVOO. Table 5. Average values of stable isotopic ratios obtained for monocultivar protected designation of origin (PDO) and commercially-blended extra virgin olive oils obtained by isotope ratio mass Table 5. Average values of stable isotopic ratios obtained for monocultivar protected designation of origin (PDO) and commercially-blended extra virgin olive oils obtained by isotope ratio mass spectrometry (IRMS). An asterisk () in a row represents signiﬁcant diﬀerences between mean values at p < 0.05 obtained by ANOVA and least signiﬁcant diﬀerence (LSD) test. ometry (IRMS). An asterisk ( ) in a row represents significant differences between me .05 obtained by ANOVA and least significant difference (LSD) test. 2.4. Multivariate Statistical Analysis Green cycles represent monocultivar protected designation of origin (PDO) EVOO purchased on family farms, while red cycles represent commercially-blended (CB) EVOO purchased in supermarkets (b) variable importance in projection (VIP) scores of the variables (lipids) most useful for the diﬀerentiation of monocultivar PDO EVOO and commercially-blended EVOO. 2.5. Confirmation of Geographical Origin by Isotope Ratio Mass Spectrometry (IRMS) 2.5. Conﬁrmation of Geographical Origin by Isotope Ratio Mass Spectrometry (IRMS) Stable Isotopic Ratios Origin/Class Monocultivar PDO Commercially-Blended δ13C −29.65 −29.44 δ2H −151.28 −144.96 δ18O 23 89 26 10 * Stable Isotopic Ratios Origin/Class Monocultivar PDO Commercially-Blended δ13C −29.65 −29.44 δ2H −151.28 −144.96 * δ18O 23.89 26.10 * By correlating the two parameters more linked to geographical origin, i.e., δ2H and δ18O, it was possible to visualize different groupings (Figure 5). As already observed [27], Garda PDO EVOO (Casaliva cultivar) was characterized by the lowest δ2H and δ18O values, most probably due to the production area (far from the sea, higher latitude) and climate (colder than the Mediterranean one). The commercial non-Italian EU EVOO showed the highest δ2H and δ18O values, overlapping only with those of Ragusa PDO (Tonda Iblea cultivar), because the geographical and climatic characteristics of Sicily are similar to those of other European Mediterranean countries, as observed previously [30]. The commercially-blended EVOOs of Italian origin had δ2H and δ18O values By correlating the two parameters more linked to geographical origin, i.e., δ2H and δ18O, it was possible to visualize diﬀerent groupings (Figure 5). As already observed [27], Garda PDO EVOO (Casaliva cultivar) was characterized by the lowest δ2H and δ18O values, most probably due to the production area (far from the sea, higher latitude) and climate (colder than the Mediterranean one). The commercial non-Italian EU EVOO showed the highest δ2H and δ18O values, overlapping only with those of Ragusa PDO (Tonda Iblea cultivar), because the geographical and climatic characteristics of Sicily are similar to those of other European Mediterranean countries, as observed previously [30]. The commercially-blended EVOOs of Italian origin had δ2H and δ18O values overlapping with those 13 of 20 13 of 20 Molecules 2020, 25, 4 Molecules 2019, of Italian monocultivar PDO with the exception of Garda PDO EVOO, which was as expected because the areas of production overlapped as well. overlapping with those of Italian monocultivar PDO with the exception of Garda PDO EVOO, which was as expected because the areas of production overlapped as well. Figure 5. Differentiation of Italian monocultivar protected designation of origin (PDO) extra virgin olive oils purchased on family farms and commercially-blended extra virgin olive oils (EU and Italian origin) purchased in supermarkets in Italy according to δ2H and δ18O stable isotopic ratios. Figure 5. 2.5. Confirmation of Geographical Origin by Isotope Ratio Mass Spectrometry (IRMS) 2.5. Conﬁrmation of Geographical Origin by Isotope Ratio Mass Spectrometry (IRMS) Diﬀerentiation of Italian monocultivar protected designation of origin (PDO) extra virgin olive oils purchased on family farms and commercially-blended extra virgin olive oils (EU and Italian origin) purchased in supermarkets in Italy according to δ2H and δ18O stable isotopic ratios. 3 Materials and Methods Figure 5. Differentiation of Italian monocultivar protected designation of origin (PDO) extra virgin olive oils purchased on family farms and commercially-blended extra virgin olive oils (EU and Italian origin) purchased in supermarkets in Italy according to δ2H and δ18O stable isotopic ratios. Figure 5. Diﬀerentiation of Italian monocultivar protected designation of origin (PDO) extra virgin olive oils purchased on family farms and commercially-blended extra virgin olive oils (EU and Italian origin) purchased in supermarkets in Italy according to δ2H and δ18O stable isotopic ratios. 3. Materials and Me 3.1. EVOO Samples 3.1. EVOO Samples After preliminary selection from a larger group of high quality monocultivar EVOOs with PDO, samples that were produced from olives of Italian cultivars harvested in 2016 were collected from different geographical areas in Italy (price range from 20 to 30 €/L), including Reggio Calabria (cultivar: Ottobratica; n = 3), Perugia (cultivar: Moraiolo; n = 3), Ragusa (cultivar: Tonda Iblea; n = 3), Grosseto (cultivar: Frantoio; n = 3), Imperia (cultivar: Taggiasca; n = 1), Brescia (Garda Bresciano PDO, cultivar: Moraiolo; n = 1), Verona (Garda Orientale PDO, cultivar: Leccino; n = 1), and Riva del Garda (Garda Trentino PDO, cultivar: Casaliva; n = 5). Furthermore, 25 commercially-blended EVOOs were selected according to Nielsen data (New York, NY, USA 2016) as among the most consumed during 2016 in Italy (price range from 3 to 12 €/L) and were purchased from Italian grocery stores (supermarkets), consisting of seven samples with Italian and 18 samples with EU origin declared on their labels. All the samples were stored in dark glass bottles at a controlled temperature of 15 °C After preliminary selection from a larger group of high quality monocultivar EVOOs with PDO, samples that were produced from olives of Italian cultivars harvested in 2016 were collected from diﬀerent geographical areas in Italy (price range from 20 to 30 €/L), including Reggio Calabria (cultivar: Ottobratica; n = 3), Perugia (cultivar: Moraiolo; n = 3), Ragusa (cultivar: Tonda Iblea; n = 3), Grosseto (cultivar: Frantoio; n = 3), Imperia (cultivar: Taggiasca; n = 1), Brescia (Garda Bresciano PDO, cultivar: Moraiolo; n = 1), Verona (Garda Orientale PDO, cultivar: Leccino; n = 1), and Riva del Garda (Garda Trentino PDO, cultivar: Casaliva; n = 5). Furthermore, 25 commercially-blended EVOOs were selected according to Nielsen data (New York, NY, USA 2016) as among the most consumed during 2016 in Italy (price range from 3 to 12 €/L) and were purchased from Italian grocery stores (supermarkets), consisting of seven samples with Italian and 18 samples with EU origin declared on their labels. All the samples were stored in dark glass bottles at a controlled temperature of 15 ◦C before analysis, and gaseous N2 was added in the headspace to prevent oxidation each time the bottles were opened. 3.3. LC-ESI-MS/MS Analysis The samples were prepared by weighing 500 mg of oil in a 10 mL ﬂask, brought to volume with a 2-propanol solution and internal standard (stearic acid d3 at 1 mg/L). The ﬁnal solutions were ﬁltered through 0.22 µm ﬁlters and transferred into 2 mL vials [52]. LC-ESI-MS/MS analysis of FFA and other lipids was carried out using a UHPLC Dionex 3000 (Thermo Fisher Scientiﬁc, Dreieich, Germany), coupled to an API 5500 triple-quadrupole mass spectrometer (Applied Biosystems/MDS Sciex, Toronto, ON, Canada) equipped with an electrospray source. Five microliters of sample were injected into the LC-ESI-MS/MS system using an autosampler (Dionex Thermo Fisher Scientiﬁc, Germany) kept at 10 ◦C. A reversed phase column Ascentis Express C18 (150 mm × 2.1 mm, 2.7 µm; Sigma, Milan, Italy) set at 55 ◦C was used for the compound separation. Flow-rate was 0.26 mL/min and the composition of mobile phases was: solvent A (CH3CN 40% in water, NH4COOH 10 mM and HCOOH 0.1%) and solvent B (CH3CH(OH)CH3 90%, CH3CN 10%, NH4COOH 10 mM and HCOOH 0.1%). Separation was carried out following a 30 min multistep linear gradient, according to the method reported by Della Corte et al. [53]. Selected chemical standards were used to construct calibration curves and data were expressed as mg/kg after normalization on the basis of the internal standard stearic acid d3. The targeted lipids were detected under multiple reaction monitoring (MRM) mode and the compounds were identiﬁed based on their reference standards, retention times, and qualiﬁer and quantiﬁer ions (Table S2). The chromatographic system and data acquisition were managed by Analyst™software version 1.6.1 (Applera Corporation, Norwalk, CT, USA). pp p For the method validation, the US Food and Drug Administration (FDA) recommendations for a bioanalytical method were followed [54]. The validation included the evaluation of linearity, sensitivity, variability, recovery, accuracy, and precision based on calibration standards and quality control (QC). Calibration curves were made in 2-propanol and lipid matrix [53,55] in order to evaluate the percentage of matrix eﬀect (%ME) for each compound. The values were determined by comparing the equality in the slope ratio between the curves in solvent and matrix, using the following formula: %ME = 100% × (1 −slope solvent/slope matrix) [56]. The regression line was created with the least square ﬁt, and the determination coeﬃcient (r2) was also calculated. before analysis, and gaseo bottles were opened. 3.2. Standards and Solvents 3.2. Standards and Solvents The solvents used for the analysis of lipids in EVOO were LC-MS grade methanol, hexane, isopropanol and formic acid, purchased from Honeywell Riedel-de Haën (Seelze, Germany) and all aqueous solutions, including the HPLC mobile phase, were prepared with water purified using a The solvents used for the analysis of lipids in EVOO were LC-MS grade methanol, hexane, isopropanol and formic acid, purchased from Honeywell Riedel-de Haën (Seelze, Germany) and all aqueous solutions, including the HPLC mobile phase, were prepared with water puriﬁed using a Milli-Q system (Millipore, Vimodrone, Milan, Italy). All the analytical standards used for identiﬁcation and calibration are listed in Table S1. 14 of 20 Molecules 2020, 25, 4 14 of 20 3.3. LC-ESI-MS/MS Analysis The linearity was evaluated by preparing diﬀerent levels of independent calibration and adding increasing concentrations of each lipid in diﬀerent concentration ranges. The sensitivity of the method was evaluated with limits of detection (LOD) and limits of quantiﬁcation (LOQ) at the concentration in which the quantizer transitions showed a signal to noise ratio (S/N) of >3 and >10, respectively. To estimate the analytical variability, intra-day and inter-day parameters were calculated by injecting 10 times a middle concentration level QC sample on the same day and re-injecting it for six consecutive days. Intra-day and inter-day variability were evaluated by the coeﬃcients of variation (CV%). The recovery test was carried out to verify the applicability of the LC-ESI-MS/MS technique, and it was determined as the average of the “measured value”/“the expected value” ratio (%). The precision values were calculated as relative standard deviation (%RSD) among the measures replicated in the QC sample. They were obtained by analyzing the same sample 10 times. Since the precision can vary with the concentration, it was appropriated to analyze at least three samples at diﬀerent concentrations (low, medium, and high) with respect to the calibration range for each analyte. The accuracy was calculated as the diﬀerence between the calculated value and the theoretical value divided by the theoretical value, reported as the relative error percentage (%RE). 3.4. H-NMR Analysis of Lipids All the EVOO samples were prepared by addition and suitable mixing of 700 µL of deuterated solvent (CDCl3) to 200 µL of oil (solution about 200 mM) in a 5 mm NMR tube. All the 1H-NMR spectra were acquired at 300 K on a Bruker-Avance 400 MHz NMR spectrometer (Bruker, Bremen, Germany) by using a 5 mm BBI probe with 60◦hard pulse length of 6.6 µs at a transmission power of 0 db. For the acquisition, 32 K complex points were recorded, the spectral width was set to 10 ppm, the frequency oﬀset was set to 4.8 ppm, the relaxation delay was set to 15 s, the acquisition time was 8.2 s, the number 15 of 20 Molecules 2020, 25, 4 of scans was set to 32, and the number of dummy scans was equal to 2. The total experimental time was 13 min. All spectra were acquired without spinning. The chemical shift scale was calibrated by using the residual proton signal of the deuterated solvent (CHCl3 signal at 7.260 ppm). The data were acquired using the software Topspin 2.1 (Bruker Biospin, Rheinstetten, Germany). The resulting spectra were processed manually and automatically with the software MestreNova 12.0.0 (Mestrelab Research SL, Santiago de Compostela, Spain) taking care to achieve good symmetry on all peaks. The baseline was corrected using a polynomial function. The integral data extracted from the spectra were analyzed using standard software (Microsoft Oﬃce Excel 2016). The ratio of the peaks area F/H (Figure 1) was tested to validate the approximations used in this study for calcuation of various lipid species, as suggested earlier [49]. The average F/H value of all the EVOO analyzed in this work was 1.512 with a very small relative standard deviation of 0.5%, conﬁrming the approximations of the calculations were fulﬁlled. The data reported were obtained by the following procedures. The % molar fraction of α-linolenic acid (18:3) was obtained by the ratio of the peak area of ω-3 Me (δH = 0.97 t) with respect to the peak area of F (×2/3), the % molar fraction of linoleic acid (18:2) by the ratio of the peak area of bis-allylic protons (δH = 2.77 brt) with respect to the peak area of F minus the contribution of the previously evaluated % molar fraction of α-linolenic acid (i.e., %18:2 = %PUFA −%18:3). 3.5. LC-ESI-IT-MS Analysis of Triglycerides LC-ESI-IT-MS analysis of the olive oils was performed on a Hewlett-Packard Model 1100 Series liquid chromatograph coupled both to an Agilent 1100 Series DAD (Photo Diode-Array Detector) (Hewlett–Packard Development Company, L.P., Palo Alto, CA, USA), and to a Bruker Esquire-LC quadrupole ion-trap mass spectrometer equipped with atmospheric pressure ESI+ interface (ESI-IT-MS). Isocratic elution was applied with a ﬂow of 0.3 mL/min by using a Kinetex C18 column (2.6 µm 100A 100 × 2.1 mm) (Phenomenex, Sydney, Australia) as stationary phase, and isopropanol:methanol 10/90 (v/v), 10 mM in ammonium acetate, as mobile phase. Samples were prepared by diluting 1:200 (5 mM) each EVOO with a solution of MeOH:CHCl3 8/2 (v/v), and 4 µL were injected. Relative quantitation of TAG species was established by peak area integration of the MS extracted ion currents of the corresponding major ions produced by ESI ionization by assuming the same response time for all the TAGs species. 3.4. H-NMR Analysis of Lipids Finally, MUFA (16:1 + 18:1, essentially) was evaluated by the peak area of allylic protons (δH = 2.01 brt) with respect to the peak area of F minus the contribution of the previously evaluated % molar fraction of PUFA, whilst SFA (16:0 + 18:0 essentially) was evaluated from total peak area of methyl protons (δH = 0.97 + δH = 0.89) with respect to the peak area of F (×2/3) minus the contribution of PUFA and MUFA. The relative contribution of 1,2-DAG was given by the integration of the peak at δH = 3.71 brd attributable to the -CH2 from sn-1,2-DAG, always present in minor amount in olive oils. From these data the average unsaturation index (UI) and the iodine value (IV) of the acyl chains in TAG were evaluated. Worth of note, the standard deviations of all the mentioned measurements, calculated from three technical replicates (ex novo acquisition and data analysis) of a given EVOO sample, were quite low (<1%) for SFA, MUFA, linoleic and linolenic fatty acyl chains, giving good reliability to the approach applied in this study. 3.6. IRMS Analysis The analysis of the stable isotope ratios of H, C and O was performed on the bulk olive oil. 13C/12C (δ13C) was measured (around 0.5 mg of oil) using an isotope ratio mass spectrometer IsoPrime (Isoprime Limited, Manchester, UK) following total combustion in an elemental analyzer (VARIO CUBE, Elementar, Hanau, Germany). 18O/16O (δ18O) and 2H/1H (δ2H) were measured (around 0.3 mg of oil) using an IRMS (Finnigan DELTA XP, Thermo Scientiﬁc, Bremen, Germany) coupled with a pyrolyzer (Finnigan TC/EA, high temperature conversion elemental analyzer, Thermo Scientiﬁc). For δ2H and δ18O analysis, the weighed samples were stored in a desiccator above P2O5 for at least four days before analysis, then put into an auto-sampler equipped with a suitable cover. During Molecules 2020, 25, 4 16 of 20 measurement, dryness was guaranteed by ﬂushing nitrogen continuously over the samples. Before determining the δ2H values, the H3+ factor was veriﬁed to be lower than 8, as suggested in the instrumental manual. The values were denoted in delta in relation to the international V-PDB (Vienna-Pee Dee Belemnite) for δ13C and Vienna-standard mean ocean water (V-SMOW) for δ18O and δ2H, according to the following general equation: δi E = (i RSA −i RREF), where i is the mass number of the heavier isotope of element E, RSA is the respective isotope ratio of the sample and RREF is the relevant internationally recognized reference material [57]. The delta values were multiplied by 1000 and expressed in units “per mil” (%). The δ13C and δ2H values were calculated against two international reference materials (Icosanoic Acid Methyl Esters USGS70, δ13C value: −30.53% and δ2H value: −183.9% and USGS71, δ13C value: −10.5% and δ2H value: −4.9%), through the creation of a linear equation. δ18O was calculated against IAEA 601 (benzoic acid δ18O = +23.3%) and 602 (benzoic acid δ18O = +71.4%), through the creation of a linear equation. Data were therefore reported relative to V-PDB on a scale normalized to LSVEC-NBS19 for δ13C, and relative to the V-SMOW-SLAP scale for δ2H and δ18O. The uncertainty (2 s) of measurements, calculated following the Nordtest approach, which combines within-laboratory reproducibility standard deviation and laboratory bias using PT data [58], was < 0.3% for δ13C analysis, < 0.5% for δ18O and < 3% for δ2H. 3.7. Statistical Data Elaboration Data obtained by the LC-ESI-MS/MS analysis of minor lipids, LC-ESI-IT-MS proﬁling of TAGs, 1H-NMR analysis of major lipid parameters, and IRMS analysis in the investigated EVOO were subjected to one-way analysis of variance (ANOVA) and the average values were compared by least signiﬁcant diﬀerence (LSD) test at the level of p < 0.05. The data were further processed by principal component analysis (PCA) in order to better visualize the diﬀerences between the two classes of EVOO and explain them on the basis of the content and composition of various lipid species. Prior to PCA, the original datasets were reduced to include only the lipids and parameters for which statistically signiﬁcant diﬀerence between the two classes was determined by one-way ANOVA. Partial least squares discriminant analysis (PLSDA) was performed to extract the most useful variables among lipids for the diﬀerentiation of the two investigated classes of EVOO: variable importance in projection (VIP) scores for lipids were determined as the weighted sums of the squares of the weight in the PLSDA. Hierarchical clustering was conducted and a heatmap was generated by Ward algorithm and Euclidean distance analysis. Multivariate statistical elaboration was performed on mean-centered data. ANOVA and PCA data elaboration was performed using Statistica v. 13.2 software (StatSoft Inc., Tulsa, OK, USA), while PLSDA and cluster analysis was conducted using MetaboAnalyst v. 4.0 (http://www.metaboanalyst.ca) created at the University of Alberta, Canada [59]. 4. Conclusions LC-MS and NMR techniques were found to be potent tools to study the variability of various lipid species in EVOO. Their outputs were shown to be relatively complementary and their combined use successfully extracted several chemical markers useful for the diﬀerentiation of the two classes of EVOO with respect to the origin of purchase: monocultivar PDO EVOO from family farms vs. commercially-blended EVOO from supermarkets. Considering that the EVOO samples from both classes were characterized by known (for PDO) and declared/presumed (for commercially-blended) geographical and pedoclimatic heterogeneity and large variations in olive growing and oil producing parameters, the extracted markers could be considered relatively robust. Commercially-blended EVOO contained higher concentrations of the majority of minor lipids, including FAAs, FAAEs, MAGs, and DAGs, which may be indicative of a higher degree of TAG lipolysis in these than in monocultivar PDO EVOO. However, triterpenoids and particular TAG species were also found in higher concentrations/proportions in the samples from the commercially-blended EVOO class, suggesting a possible inﬂuence of other factors, including diverse cultivar and geographical origin, 17 of 20 Molecules 2020, 25, 4 17 of 20 respectively. The results of IRMS analysis were reasonably consistent with the information about the geographical origin declared on the labels of the investigated EVOOs, which showed considerable variability. The results of this study undoubtedly conﬁrmed the heterogeneity of oils which are sold declared as EVOO in Italy in terms of their lipid composition and geographical origin. In that sense, the obtained ﬁndings could signiﬁcantly contribute to EVOO diversiﬁcation on the market, and could help to clarify the interrelationship between EVOO origin, quality, and price, and in this way support the growth of the niche in the market segment of consumers informed and interested in healthy, quality products with remarkable diversity and clear identity. Supplementary Materials: The following are available online, Table S1: standards of minor lipids used in the LC-ESI-MS/MS method development. Table S2: mass transitions (MRM) and instrumental parameters optimized for each metabolite for the analyses by LC-ESI-MS/MS. Table S3: method validation parameters for each metabolite for the analyses by LC-ESI-MS/MS - part I. Table S4: method validation parameters for each metabolite for the analyses by LC-ESI-MS/MS - part II. Author Contributions: Conceptualization, F.M., I.L., N.M., and U.V.; funding acquisition, F.M., U.V., and F.C., data curation, A.D.R., D.M., G.G., F.C., and I.L.; formal analysis, A.D.R., D.M., G.G., and F.C.; methodology, F.M. References 1. International Olive Council. COI/T.15/NC. Nº 3/Rev.13/2019. Trade Standard Applying to Olive Oils and Olive-Pomace Oils 2018. Available online: https://www.internationaloliveoil.org/wp-content/uploads/2019/ 11/COI-T.15-NC.-No-3-Rev.-13-2019-Eng.pdf (accessed on 12 December 2019). 2. European Economic Community. Commission Regulation (EEC) No 2568/91 of 11 July 1991 (and later modiﬁcations) on the characteristics of olive oil and olive-residue oil and the relevant methods of analysis. Oﬀ. J. Eur. Community 1991, L248, 1–83. 2. European Economic Community. Commission Regulation (EEC) No 2568/91 of 11 July 1991 (and later modiﬁcations) on the characteristics of olive oil and olive-residue oil and the relevant methods of analysis. Oﬀ. J. Eur. Community 1991, L248, 1–83. 3. European Union Regulation (EU) No 1151/2012 of the European parliament and of the Council of 21 November 2012 (and later modiﬁcations) on quality schemes for agricultural products and foodstuﬀs. Oﬀ. J. Eur. Union 2012, L343, 1–29. 3. European Union Regulation (EU) No 1151/2012 of the European parliament and of the Council of 21 November 2012 (and later modiﬁcations) on quality schemes for agricultural products and foodstuﬀs. Oﬀ. J. Eur. Union 2012, L343, 1–29. 4. Da Ros, A.; Masuero, D.; Riccadonna, S.; Brki´c Bubola, K.; Mulinacci, N.; Mattivi, F.; Luki´c, I.; Vrhovsek, U. Complementary Untargeted and Targeted Metabolomics for Diﬀerentiation of Extra Virgin Olive Oils of Diﬀerent Origin of Purchase Based on Volatile and Phenolic Composition and Sensory Quality. Molecules 2019, 24, 2896. [CrossRef] [PubMed] 4. Da Ros, A.; Masuero, D.; Riccadonna, S.; Brki´c Bubola, K.; Mulinacci, N.; Mattivi, F.; Luki´c, I.; Vrhovsek, U. Complementary Untargeted and Targeted Metabolomics for Diﬀerentiation of Extra Virgin Olive Oils of Diﬀerent Origin of Purchase Based on Volatile and Phenolic Composition and Sensory Quality. Molecules 2019, 24, 2896. [CrossRef] [PubMed] 5. Alonso-Salces, R.M.; Moreno-Rojas, J.M.; Holland, M.V.; Reniero, F.; Guillou, C.; Héberger, K. Virgin Olive Oil Authentication by Multivariate Analyses of 1H NMR Fingerprints and δ13C and δ2H Data. J. Agric. Food Chem. 2010, 58, 5586–5596. [CrossRef] [PubMed] 5. Alonso-Salces, R.M.; Moreno-Rojas, J.M.; Holland, M.V.; Reniero, F.; Guillou, C.; Héberger, K. Virgin Olive Oil Authentication by Multivariate Analyses of 1H NMR Fingerprints and δ13C and δ2H Data. J. Agric. Food Chem. 2010, 58, 5586–5596. [CrossRef] [PubMed] 6. Angerosa, F.; Campestre, C.; Giansante, L. Analysis and Authentication. In Olive Oil: Chemistry and Technology; Boskou, D., Ed.; AOCS Press: Urbana, IL, USA, 1996; pp. 113–172, ISBN 9781893997882. 7. Schwingshackl, L.; Hoﬀmann, G. 4. Conclusions and U.V.; writing—original draft, A.D.R., F.C., G.G., and I.L.; writing—review and editing, I.L.; supervision: U.V., N.M., and F.M. All the authors read and approved the manuscript. All authors have read and agreed to the published version of the manuscript. Funding: The research was founded by the Autonomous Province of Trento (Italy). Acknowledgments: We thank Agraria Riva del Garda (TN) for supporting the sampling of the Italian monocultivar samples and technical support. Conﬂicts of Interest: The authors declare no conﬂict of interest. Conﬂicts of Interest: The authors declare no conﬂict of interest. References Alkyl Esters of Fatty Acids a Useful Tool to Detect Soft Deodorized Olive Oils. J. Agric. Food Chem. 2008, 56, 6740–6744. [CrossRef] 19. Alves, E.; Domingues, M.R.M.; Domingues, P. Polar Lipids from Olives and Olive Oil: A Review on Their Identiﬁcation, Signiﬁcance and Potential Biotechnological Applications. Foods 2018, 7, 109. [CrossRef] 20. Royer, A.; Gerard, C.; Naulet, N.; Lees, M.; Martin, G.J. Stable isotope characterization of olive oils. I—Compositional and carbon-13 proﬁles of fatty acids. J. Am. Oil Chem. Soc. 1999, 76, 357–363. [CrossRef] 21. Faberi, A.; Marianella, R.M.; Fuselli, F.; La Mantia, A.; Ciardiello, F.; Montesano, C.; Mascini, M.; Sergi, M.; Compagnone, D. Fatty acid composition and δ13 C of bulk and individual fatty acids as marker for authenticating Italian PDO/PGI extra virgin olive oils by means of isotopic ratio mass spectrometry. J. Mass Spectrom. 2014, 49, 840–849. [CrossRef] 22. Bréas, O.; Guillou, C.; Reniero, F.; Sada, E.; Angerosa, F. Oxygen-18 measurement by continuous ﬂow pyrolysis/isotope ratio mass spectrometry of vegetable oils. Rapid Commun. Mass Spectrom. 1998, 12, 188–192. [CrossRef] 23. Angerosa, F.; Bréas, O.; Contento, S.; Guillou, C.; Reniero, F.; Sada, E. Application of stable isotope ratio analysis to the characterization of the geographical origin of olive oils. J. Agric. Food Chem. 1999, 47, 1013–1017. [CrossRef] 24. Mihailova, A.; Abbado, D.; Kelly, S.D.; Pedentchouk, N. The impact of environmental factors on molecular and stable isotope compositions of n-alkanes in Mediterranean extra virgin olive oils. Food Chem. 2015, 173, 114–121. [CrossRef] [PubMed] 25. Iacumin, P.; Bernini, L.; Boschetti, T. Climatic factors inﬂuencing the isotope composition of Italian olive oils and geographic characterisation. Rapid Commun. Mass Spectrom. RCM 2009, 23, 448–454. [CrossRef] [PubMed] 26. Portarena, S.; Gavrichkova, O.; Lauteri, M.; Brugnoli, E. Authentication and traceability of Italian extra-virgin olive oils by means of stable isotopes techniques. Food Chem. 2014, 164, 12–16. [CrossRef] [PubMed] 27. Camin, F.; Larcher, R.; Perini, M.; Bontempo, L.; Bertoldi, D.; Gagliano, G.; Nicolini, G.; Versini, G. Characterisation of authentic Italian extra-virgin olive oils by stable isotope ratios of C, O and H and mineral composition. Food Chem. 2010, 118, 901–909. [CrossRef] 28. Camin, F.; Larcher, R.; Nicolini, G.; Bontempo, L.; Bertoldi, D.; Perini, M.; Schlicht, C.; Schellenberg, A.; Thomas, F.; Heinrich, K.; et al. Isotopic and Elemental Data for Tracing the Origin of European Olive Oils. J. Agric. Food Chem. 2010, 58, 570–577. [CrossRef] [PubMed] 29. References Monounsaturated Fatty Acids and Risk of Cardiovascular Disease: Synopsis of the Evidence Available from Systematic Reviews and Meta-Analyses. Nutrients 2012, 4, 1989–2007. [CrossRef] 8. Alves, E.; Melo, T.; Barros, M.P.; Domingues, M.R.M.; Domingues, P. Lipidomic Proﬁling of the Olive (Olea europaea L.) Fruit towards Its Valorisation as a Functional Food: In-Depth Identiﬁcation of Triacylglycerols and Polar Lipids in Portuguese Olives. Molecules 2019, 24, 2555. [CrossRef] . Blasi, F.; Pollini, L.; Cossignani, L. Varietal Authentication of Extra Virgin Olive Oils by Triacylglycerols Volatiles Analysis. Foods 2019, 8, 58. [CrossRef] 10. Montealegre, C.; Marina Alegre, M.L.; García-Ruiz, C. Traceability Markers to the Botanical Origin in Olive Oils. J. Agric. Food Chem. 2010, 58, 28–38. [CrossRef] Molecules 2020, 25, 4 18 of 20 18 of 20 11. Ollivier, D.; Artaud, J.; Pinatel, C.; Durbec, J.P.; Guérère, M. Triacylglycerol and fatty acid compositions of French virgin olive oils. Characterization by chemometrics. J. Agric. Food Chem. 2003, 51, 5723–5731. [CrossRef] 12. Pérez-Camino, M.C.; Moreda, W.; Cert, A. Eﬀects of olive fruit quality and oil storage practices diacylglycerol content of virgin olive oils. J. Agric. Food Chem. 2001, 49, 699–704. [CrossRef] 13. Mannina, L.; Sobolev, A.P. High resolution NMR characterization of olive oils in terms of quality, authenticity and geographical origin. Magn. Reson. Chem. MRC 2011, 49 (Suppl. 1), S3–S11. [CrossRef] 14. Amaral, J.S.; Mafra, I.; Oliveira, M.B.P.P. Chapter 63 - Characterization of Three Portuguese Varietal Olive Oils Based on Fatty Acids, Triacylglycerols, Phytosterols and Vitamin E Proﬁles: Application of Chemometrics. In Olives and Olive Oil in Health and Disease Prevention; Preedy, V.R., Watson, R.R., Eds.; Academic Press: San Diego, CA, USA, 2010; pp. 581–589, ISBN 978-0-12-374420-3. 15. Peršuri´c, Ž.; Safti´c, L.; Mašek, T.; Kraljevi´c Paveli´c, S. Comparison of triacylglycerol analysis by MALDI-TOF/MS, fatty acid analysis by GC-MS and non-selective analysis by NIRS in combination with chemometrics for determination of extra virgin olive oil geographical origin. A case study. LWT Food Sci. Technol. 2018, 95, 326–332. [CrossRef] 16. Ollivier, D.; Artaud, J.; Pinatel, C.; Durbec, J.; Guerere, M. Diﬀerentiation of French virgin olive oil RDOs by sensory characteristics, fatty acid and triacylglycerol compositions and chemometrics. Food Chem. 2006, 97, 382–393. [CrossRef] 17. Bendini, A.; Cerretani, L.; Valli, E.; Lercker, G.; Mazzini, C. Application of analytical methods to determine mildly deodorized olive oils in commercial extra virgin olive oils. Ind. Aliment. 2009, 48, 46–51. 18. Pérez-Camino, M.D.; Cert, A.; Romero-Segura, A.; Cert-Trujillo, R.; Moreda, W. References Bontempo, L.; Camin, F.; Larcher, R.; Nicolini, G.; Perini, M.; Rossmann, A. Coast and year eﬀect on H, O and C stable isotope ratios of Tyrrhenian and Adriatic italian olive oils. Rapid Commun. Mass Spectrom. 2009, 23, 1043–1048. [CrossRef] Molecules 2020, 25, 4 19 of 20 19 of 20 30. Camin, F.; Pavone, A.; Bontempo, L.; Wehrens, R.; Paolini, M.; Faberi, A.; Marianella, R.M.; Capitani, D.; Vista, S.; Mannina, L. The use of IRMS, 1H NMR and chemical analysis to characterise Italian and imported Tunisian olive oils. Food Chem. 2016, 196, 98–105. [CrossRef] 31. Walorczyk, S. Validation and use of a QuEChERS-based gas chromatographic-tandem mass spectrometric method for multiresidue pesticide analysis in blackcurrants including studies of matrix eﬀects and estimation of measurement uncertainty. Talanta 2014, 120, 106–113. [CrossRef] 32. Muik, B.; Lendl, B.; Molina-Díaz, A.; Ayora-Cañada, M.J. Direct, reagent-free determination of free fatty acid content in olive oil and olives by Fourier transform Raman spectrometry. Anal. Chim. Acta 2003, 487, 211–220. [CrossRef] 33. Tarhan, ˙I.; Ismail, A.A.; Kara, H. Quantitative determination of free fatty acids in extra virgin olive oils by multivariate methods and Fourier transform infrared spectroscopy considering diﬀerent absorption modes. Int. J. Food Prop. 2017, 20, S790–S797. [CrossRef] 34. Fiorini, D.; Boarelli, M.C.; Conti, P.; Alfei, B.; Caprioli, G.; Ricciutelli, M.; Sagratini, G.; Fedeli, D.; Gabbianelli, R.; Pacetti, D. Chemical and sensory diﬀerences between high price and low price extra virgin olive oils. Food Res. Int. 2018, 105, 65–75. [CrossRef] [PubMed] 35. Tawﬁk, M.S.; Huyghebaert, A. Interaction of packaging materials and vegetable oils: Oil stability. Food Chem. 1999, 64, 451–459. [CrossRef] 36. Méndez, A.I.; Falqué, E. Eﬀect of storage time and container type on the quality of extra-virgin olive oil. Food Control 2007, 18, 521–529. [CrossRef] 37. Conte, L.; Mariani, C.; Gallina Toschi, T.; Tagliabue, S. Alkyl esters and related compounds in virgin olive oils: Their evolution over time. Riv. Ital. Sostanze Gr. 2014, 91, 21–29. 38. Di Serio, M.G.; Giansante, L.; Di Loreto, G.; Faberi, A.; Ricchetti, L.; Di Giacinto, L. Ethyl esters versus fermentative organoleptic defects in virgin olive oil. Food Chem. 2017, 219, 33–39. [CrossRef] 39. Gómez-Coca, R.B.; Moreda, W.; Pérez-Camino, M.C. Fatty acid alkyl esters presence in olive oil vs. organoleptic assessment. Food Chem. 2012, 135, 1205–1209. [CrossRef] 40. Pérez-Camino, M.C.; Moreda, W.; Mateos, R.; Cert, A. Determination of Esters of Fatty Acids with Low Molecular Weight Alcohols in Olive Oils. J. Agric. References Food Chem. 2002, 50, 4721–4725. [CrossRef] 41. Gómez-Coca, R.B.; Fernandes, G.D.; Pérez-Camino, M. del C.; Moreda, W. Fatty acid ethyl esters (FAEE) in extra virgin olive oil: A case study of a quality parameter. LWT - Food Sci. Technol. 2016, 66, 378–383. [CrossRef] 42. Morales, M.T.; Luna, G.; Aparicio, R. Comparative study of virgin olive oil sensory defects. Food Chem. 2005, 91, 293–301. [CrossRef] 43. Costa, R.; Bartolomeo, G.; Saija, E.; Rando, R.; Albergamo, A.; Dugo, G. Determination of Alkyl Esters Content in PDO Extra Virgin Olive Oils from Sicily. J. Food Qual. 2017, 2017, 3078105. [CrossRef] 44. Amelotti, G.; Daghetta, A.; Ferrario, A. Content and structure of partial glycerides in virgin olive oils. Their evolution by diﬀerent working process and preservation form. Riv. Ital. Sostanze Gr. 1989, 66, 681–692. 45. Paganuzzi, V. Monoglycerides in vegetable oils. Note I. Riv. Ital. Sostanze Gr. 1987, 64, 411–414. 46. Paganuzzi, V. Monoglycerides in vegetable oils. Note IV: Raw Oils of Law Unsaturation. Riv. Ital. Sostanze Gr. 1999, 76, 457–471. 47. Allouche, Y.; Jiménez, A.; Uceda, M.; Paz Aguilera, M.; Gaforio, J.J.; Beltrán, G. Inﬂuence of olive paste preparation conditions on virgin olive oil triterpenic compounds at laboratory-scale. Food Chem. 2010, 119, 765–769. [CrossRef] 48. Allouche, Y.; Jiménez, A.; Uceda, M.; Aguilera, M.P.; Gaforio, J.J.; Beltrán, G. Triterpenic content and chemometric analysis of virgin olive oils from forty olive cultivars. J. Agric. Food Chem. 2009, 57, 3604–3610. [CrossRef] [PubMed] 49. Castejón, D.; Mateos-Aparicio, I.; Molero, M.D.; Cambero, M.I.; Herrera, A. Evaluation and Optimization of the Analysis of Fatty Acid Types in Edible Oils by 1H-NMR. Food Anal. Methods 2014, 7, 1285–1297. [CrossRef] 50. Inglese, P.; Famiani, F.; Galvano, F.; Servili, M.; Esposto, S.; Urbani, S. Factors Aﬀecting Extra-Virgin Olive Oil Composition. In Horticultural Reviews; Janick, J., Ed.; John Wiley & Sons, Ltd: Chichester, West Sussex, UK, 2011; Volume 38, pp. 83–147, ISBN 978-0-470-87237-6. 20 of 20 Molecules 2020, 25, 4 20 of 20 51. Circi, S.; Ingallina, C.; Vista, S.; Capitani, D.; Di Vecchia, A.; Leonardi, G.; D’Achille, G.; Centauri, L.; Camin, F.; Mannina, L. A Multi-Methodological Protocol to Characterize PDO Olive Oils. Metabolites 2018, 8, 43. [CrossRef] 2. Wabaidur, S.M.; AlAmmari, A.; Aqel, A.; AL-Tamrah, S.A.; Alothman, Z.A.; Ahmed, A.Y.B.H. Determina of free fatty acids in olive oils by UPHLC–MS. J. Chromatogr. B 2016, 1031, 109–115. [CrossRef] 53. References Della Corte, A.; Chitarrini, G.; Di Gangi, I.M.; Masuero, D.; Soini, E.; Mattivi, F.; Vrhovsek, U. A rapid LC–MS/MS method for quantitative proﬁling of fatty acids, sterols, glycerolipids, glycerophospholipids and sphingolipids in grapes. Talanta 2015, 140, 52–61. [CrossRef] 54. U.S. Department of Health and Human Services; Food and Drug Administration; Center for Drug Evaluation and Research; Center for Veterinary Medicine. Bioanalytical Method Validation: Guidance for Industry. 2018. Available online: https://www.fda.gov/ﬁles/drugs/published/Bioanalytical-Method-Validation-Guidance- for-Industry.pdf (accessed on 12 December 2019). 55. Pérez-Navarro, J.; Da Ros, A.; Masuero, D.; Izquierdo-Cañas, P.M.; Hermosín-Gutiérrez, I.; Gómez-Alonso, S.; Mattivi, F.; Vrhovsek, U. LC-MS/MS analysis of free fatty acid composition and other lipids in skins and seeds of Vitis vinifera grape cultivars. Food Res. Int. 2019, 125, 108556. [CrossRef] 56. Kwon, H.; Lehotay, S.J.; Geis-Asteggiante, L. Variability of matrix eﬀects in liquid and gas chromatography- mass spectrometry analysis of pesticide residues after QuEChERS sample preparation of diﬀerent food crops. J. Chromatogr. A 2012, 1270, 235–245. [CrossRef] [PubMed] 57. Coplen, T.B. Guidelines and recommended terms for expression of stable-isotope-ratio and gas-ratio measurement results. Rapid Commun. Mass Spectrom. 2011, 25, 2538–2560. [CrossRef] [PubMed] 57. Coplen, T.B. Guidelines and recommended terms for expression of stable-isotope-ratio and gas-ratio measurement results. Rapid Commun. Mass Spectrom. 2011, 25, 2538–2560. [CrossRef] [PubMed] 58. Lindholm, G. The role of Nordtest in testing and quality assurance. Accreditation Qual. Assur. 1998, 3, 296–297. [CrossRef] 58. Lindholm, G. The role of Nordtest in testing and quality assurance. Accreditation Qual. Assur. 1998, 3, 296–297. [CrossRef] 59. Xia, J.; Sinelnikov, I.V.; Han, B.; Wishart, D.S. MetaboAnalyst 3.0—Making metabolomics more meaningful. Nucleic Acids Res. 2015, 43, W251–W257. [CrossRef] [PubMed] 59. Xia, J.; Sinelnikov, I.V.; Han, B.; Wishart, D.S. MetaboAnalyst 3.0—Making metabolomics more meaningful. Nucleic Acids Res. 2015, 43, W251–W257. [CrossRef] [PubMed] Sample Availability: Not available. © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). Sample Availability: Not available. © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
https://openalex.org/W2329658694	https://periodicos.ufv.br/ojs/reveng/article/download/166/135	Portuguese	null	UTILIZAÇÃO DO BIOSSÓLIDO DE CERVEJARIA COMO SUBSTRATO PARA PRODUÇÃO DE MUDAS DE EUCALIPTO: ANÁLISE DO DESENVOLVIMENTO INICIAL DAS MUDAS E ASPECTOS AMBIENTAIS	Engenharia na Agricultura	2,010	cc-by	2,481	RESUMO Este trabalho foi conduzido com o objetivo de avaliar o desenvolvimento de mudas de Eucalyptus grandis em substrato de lodo da estação de tratamento de efluente de indústria cervejeira. Os tratamentos consistiram de substratos proveniente de misturas de biossólidos com casca de arroz carbonizada, correspondentes a M1 - 80% de biossólido e 20% de casca de arroz carbonizada, M2 - 60% de biossólido e 40% de casca de arroz carbonizada, M3 - 50% de biossólido e 50% de casca de arroz carbonizada, M4 - 40% de biossólido e 60% de casca de arroz carbonizada e Testemunha composta de solo sem adição de lodo. A mistura que apresentou melhores resultados estatitiscamente foi a mistura M2 (60% lodo de esgoto + 40% casca de arroz carbonizada). Palavras-chave: resíduos sólidos, reciclagem agrícola, meio ambiente MATERIAL E MÉTODOS Este trabalho foi conduzido, nos meses de outubro-novembro de 2009, no Setor de Fruticultura da Universidade Federal de Viçosa, campus de Florestal, em Florestal-MG. Os substratos utilizados foram misturas de lodo após compostagem (biossólidos) das estações de tratamento de esgoto e água de indústria cervejeira misturada à casca de arroz carbonizadas. Além das formas convencionais de utilização do lodo de esgoto por incineração, disposição em aterro sanitário, uso como matéria-prima em alguns produtos cerâmicos e cimento, conversão em óleo combustível, e aplicação na agricultura como condicionador de solo (TSUTIYA, 2001), o uso como substrato para produção de mudas florestais destaca-se como uma alternativa econômica e ambientalmente correta. Os tratamentos consistiram de substratos proveniente de misturas de biossólidos com casca de arroz carbonizada, correspondentes a M1 - 80% de biossólido e 20% de casca de arroz carbonizada, M2 - 60% de biossólido e 40% de casca de arroz carbonizada, M3 - 50% de biossólido e 50% de casca de arroz carbonizada, M4 - 40% de biossólido e 60% de casca de arroz carbonizada e Testemunha composta de solo sem adição de lodo. O termo biossólido proposto pela WEF- Water Environmental Federation é utilizado para designar o resíduo proveniente das estações de tratamento de esgoto (ETE). Segundo Silva et al. (2004), os biossólidos são sinônimos de lodo de esgoto e contêm matéria orgânica, macro e micronutrientes que exercem um papel fundamental na produção agrícola e na manutenção da fertilidade do solo. A mistura dos substratos foi realizada por meio da pesagem e da determinação dos volumes de acordo com cada seqüência de mistura. Em seguida, foram transferidas para recipientes de polietileno. Cada recipiente recebeu em torno de três sementes de E. grandis que foram cobertas por uma fina camada de casca de arroz carbonizada. O uso adequado dos biossólidos constitui- se excelente fertilizante orgânico promovendo o crescimento dos organismos, melhorando o nível de fertilidade e aumentando a capacidade de troca de cátions do solo (VALIM et al., 2001). O uso agrícola é uma forma mundialmente aceita para destinação final dos biossólidos, por sua constituição apresentar altos teores de matéria orgânica, macro e micronutrientes para as plantas (TSUTIYA, 2001). Foram realizadas duas regas diárias e não se utilizou adubações. Quando as mudas atingiram 8 cm de alturas procedeu-se o desbaste, deixando-se a muda central mais vigorosa. INTRODUÇÃO demanda de aproximadamente 11 bilhões mudas a serem produzidas anualmente no país. Considerando ainda a utilização de tubetes para a confecção das mudas e que cada tubete de 50 cm³ é preenchido com 30 gramas de substrato a base de biossólidos, mais 20 gramas de casca de arroz carbonizadas, o somatório dessa mistura representa aproximadamente 330 mil toneladas de biossólidos que deixam de ser incinerados, aterrados, ou dispostos no mar poluindo o meio ambiente. O Brasil é o quinto maior fabricante de cerveja do mundo com produção anual de 8,5 bilhões de litros, mas um tímido consumidor da bebida, com cerca de 47 litros ano-1 por habitante, bem distante do maior consumidor a Republica Tcheca (120 litros ano-1 por habitante), setor este, que vem sendo impulsionado por uma taxa anual de crescimento da ordem de 2,2% ao ano (ROSA et al., 2006). Com a expansão do mercado cervejeiro brasileiro vem a preocupação com a geração de resíduos sólidos, principalmente, nas etapas de filtragem, envase e tratamento de água e efluentes líquidos, sendo o lodo um dos mais representativos da ordem de 0,8 kg de lodo para cada hectolitro de cerveja produzida (SANTOS, 2005). Considerando a produção brasileira de cerveja de 8,5 bilhões de litros por ano, tem-se, a geração de aproximadamente 68.000 toneladas de lodo por ano. Este trabalho foi conduzido com o objetivo de avaliar o desenvolvimento de mudas de Eucalyptus grandis em substrato de lodo da estação de tratamento de efluente de indústria cervejeira. USE OF BREWING BIOSOLIDS AS SUBSTRATE FOR PRODUCTION OF EUCALYPTUS SEEDLINGS: ANALYSIS OF EARLY SEEDLING GROWTH AND ENVIRONMENTAL ASPECTS This study was conducted to evaluate the growth development of Eucalyptus grandis seedlings raised in the sludge obtained from effluent treatment plant of a brewing industry. The treatments consisted of substrate originating from biosolids mixtures with charred, corresponding peel of rice M1 - 80% of biosolid and 20% of charred peel of rice, M2 - 60% of biosolid and 40% of charred peel of rice, M3 - 50% of biosolid and 50% of charred peel of rice, M4 - 40% of biosolid and 60% of charred peel of rice and witness composed of soil without mud addition. The best results were obtained in 60% sludge and 40% carbonized rice hull mixture. Keywords: solid waste, recycling, agriculture, environment 305 Engenharia na agricultura, viçosa - mg, V.18 N.4, julho / agosto 2010 ANTÔNIO, A. C. et al. INTRODUÇÃO UTILIZAÇÃO DO BIOSSÓLIDO DE CERVEJARIA COMO SUBSTRATO... UTILIZAÇÃO DO BIOSSÓLIDO DE CERVEJARIA COMO SUBSTRATO... Quadro 1. Análise química, matéria seca, do composto do biossólido Parâmetro Valor pH 6 Nitrogênio 1,4 dag kg-1 Fósforo 0,6 dag kg-1 C/N 10/1 Soma de NPK 2 dag kg-1 Cádmio 1,43 mg kg-1 Chumbo 18,04 mg kg-1 Cromo 59,02 mg kg-1 Níquel 28,72 mg kg-1 Selênio 1,94 mg kg-1 Coliformes termotolerantes Ausente Salmonella Ausente Quadro 2. Análise química do solo Parâmetro Valor pH 5,4 Fósforo 2,5 mg dm-³ Potássio 84 mg dm-³ Cálcio 2,1 cmolc dm-³ Magnésio 0,3 cmolc dm-³ Alumínio 0,1 cmolc dm-³ Hidrogênio + Alumínio 3,47 cmolc dm-³ Soma de Bases 2,61 cmolc dm-³ Matéria Orgânica 1,5 dag kg-1 Fósforo Remanescente 14,5 mg L-1 Zinco 1,7 mg dm-³ Ferro 63,3 mg dm-³ Manganês 36,3 mg dm-³ Cobre 1,9 mg dm-³ Quadro 1. Análise química, matéria seca, do composto do biossólido Parâmetro Valor pH 6 Nitrogênio 1,4 dag kg-1 Fósforo 0,6 dag kg-1 C/N 10/1 Soma de NPK 2 dag kg-1 Cádmio 1,43 mg kg-1 Chumbo 18,04 mg kg-1 Cromo 59,02 mg kg-1 Níquel 28,72 mg kg-1 Selênio 1,94 mg kg-1 Coliformes termotolerantes Ausente Salmonella Ausente Quadro 2. Análise química do solo Parâmetro Valor pH 5,4 Fósforo 2,5 mg dm-³ Potássio 84 mg dm-³ Cálcio 2,1 cmolc dm-³ Magnésio 0,3 cmolc dm-³ Alumínio 0,1 cmolc dm-³ Hidrogênio + Alumínio 3,47 cmolc dm-³ Soma de Bases 2,61 cmolc dm-³ Matéria Orgânica 1,5 dag kg-1 Fósforo Remanescente 14,5 mg L-1 Zinco 1,7 mg dm-³ Ferro 63,3 mg dm-³ Manganês 36,3 mg dm-³ Cobre 1,9 mg dm-³ MATERIAL E MÉTODOS O composto de biossólidos e o solo foram submetidos à análise química, conforme resultados apresentados nos Quadros 1 e 2. No Brasil foram plantados em 2008, segundo Fonseca (2010), 6,6 milhões de hectares de florestas compostas de áreas plantadas com eucalipto, pinus e outras espécies. Considerando-se espaçamento convencional de 3 x 2 metros isto representa uma Foram analisados nas mudas de eucalipto os seguintes parâmetros: altura das mudas, comprimento de raiz, diâmetro do colo, biomassa REVENG 305-309 p. REVENG 305-309 p. REVENG 305-309 p. 306 Engenharia na agricultura, viçosa - mg, V.18 N.4, julho / agosto 2010 RESULTADOS E DISCUSSÃO ra, comprimento de raiz, diâmetro do colo, biomassa seca e relação altura/diâmetro do colo em as de eucalipto Os tratamentos com biossólidos foram superiores à testemunha quanto ao diâmetro do colo, sendo M1, M2 e M4 os mais relevantes. Segundo Carneiro (1995), o diâmetro do colo é um parâmetro utilizado para demonstrar a capacidade de sobrevivência da muda no campo, podendo-se definir doses de fertilizantes a serem aplicados. • O uso de lodo de esgoto de estação de tratamento de indústria cervejeira após compostagem como componente de substratos para produção de mudas é uma alternativa viável para sua disposição final e constitui uma ferramenta a ser utilizada para produção de mudas de espécies florestais, substratos para jardinagem e recuperação de áreas degradadas; Os tratamentos com biossólidos proporcionaram maior produção de biomassa seca do que a testemunha e os tratamentos M1, M2 e M4 foram superiores em relação ao M3. A produção de biomassa seca nos tratamentos com o biossólido foi superior a 10 (dez) vezes a produção de biomassa seca da testemunha. Faustino et al. (2005) trabalhando com mudas de eucalipto e substrato à base de biossólidos obteve diferença estatística para biomassa seca em tratamento com 75% de biossólido abaixo deste valor os dados foram semelhantes à testemunha (solo sem adição de lodo). • A utilização de biossólidos provenientes da compostagem do lodo de esgoto de estações de tratamento de indústria cervejeira como insumo na produção de mudas de Eucalyptus grandis mostrou-se uma opção exeqüível para a produção de mudas, devido ao baixo custo e ao aporte significativo de nutrientes e matéria orgânica. A relação altura/ diâmetro do colo foi superior no tratamento M3 e na testemunha. Segundo Trigueiro & Guerrini (2003), em geral, as mudas de eucalipto apresentam maiores incrementos no desenvolvimento em altura do que em diâmetro de colo e, conseqüentemente, os valores apresentados para a relação A/D em todos os tratamentos, inclusive a testemunha, foram acima da faixa considerada ideal por Carneiro (1995). Segundo esse autor, a relação A/D, parâmetro que exprime qualidade em qualquer fase do período de produção de mudas, deve situar- se entre os limites de 5,4 a 8,1. Considerando a faixa proposta por Carneiro (1995) os tratamentos que mais aproximaram desta foram M1, M2 e M4. REFERÊNCIAS BIBLIOGRÁFICAS FONSECA, F.H. da. Apresentação do anuário estatístico da ABRAF 2010 – Ano Base 2009. http://www.abraflor.org.br/estatisticas.asp. 27 abr. 2010. CARNEIRO, J.G.A. Produção e controle de qualidade de mudas florestais. Viçosa: Editora Folha de Viçosa, 1995. 451p. FAUSTINO, R.; KATO, M.T.; FLORÊNCIO, L.; GAVAZZA, S. Lodo de esgoto como substrato para produção de mudas de Senna siamea Lam. Revista Brasileira de Engenharia Agrícola e Ambiental, Campina Grande v.9, (Suplemento), p.278 - 282, 2005. RESULTADOS E DISCUSSÃO seca e relação altura/diâmetro do colo. A altura das mudas e comprimento de raiz foram determinadas por medição direta feitas com réguas milimetradas. O diâmetro do colo foi analisado por meio de paquímetro. Para determinação da biomassa seca, amostras foram levadas para secar em estufa a 65 0C, até atingir massa constante, sendo, a massa quantificada em balança de precisão. A relação altura/ diâmetro do colo foi determinada pela relação direta entre ambos. Os resultados de altura, comprimento de raiz, diâmetro do colo, biomassa seca e relação altura/ diâmetro do colo estão apresentados no Quadro 3. Os tratamentos com biossólido no substrato apresentaram altura das mudas estatisticamente superiores à testemunha. Isso pode estar relacionado com a maior concentração de nitrogênio (N) e fósforo (P) presentes nos substratos. Segundo Novais et al. (1980) e Novais et al. (1982), N e P são nutrientes altamente requeridos nos estádios iniciais de desenvolvimento das mudas de eucalipto. O delineamento experimental foi o inteiramente casualizados (DIC) com 5 (cinco) tratamentos (4 proporções de biossólido mais a testemunha) com 50 (cinqüenta) repetições sendo cada parcela constituída por uma única muda de eucalipto. A concentração de fósforo foi menor que a de nitrogênio no biossólido, todavia, as mudas necessitam, para seu desenvolvimento, de quantidades menores de fósforo em relação ao nitrogênio. Segundo Faustino et al. (2005) como a concentração de potássio, geralmente, é baixa nos biossólidos, poderá ocorrer deficiência ao longo do tempo em que as mudas permanecerem em viveiro. Os dados foram submetidos à análise de variância e as médias comparadas pelo teste Tukey a 5% de probabilidade. REVENG 305-309 p. 307 Engenharia na agricultura, viçosa - mg, V.18 N.4, julho / agosto 2010 ANTÔNIO, A. C. et al. Quadro 3. Altura, comprimento de raiz, diâmetro do colo, biomassa seca e relação altura/diâmetro do colo em mudas de eucalipto Tratamento Altura (cm) Comprimento de raiz (cm) Diâmetro do colo (mm) Biomassa Seca (g) Relação A/D M1 9,98 A 15,01 BC 1,12 A 0,17 A 9,49 B M2 10,20 A 17,42 B 1,19 A 0,21 A 8,98 B M3 7,55 B 15,37 BC 0,66 B 0,11 B 13,53 A M4 10,50 A 15,37 A 1,10 A 0,20 A 9,74 B Testemunha 2,90 C 12,36 C 0,24 C 0,01 C 13,97 A *As médias seguidas pela mesma letra na coluna não diferem entre si pelo teste de Tukey a 5% de probabilidade. CONCLUSÕES • O melhor desenvolvimento das mudas de Eucalyptus grandis foi obtido com substrato contendo 60% de biossólido e 40% de casca de arroz carbonizada, 40% de biossólido e 60% de casca de arroz carbonizada e 80% de NOVAIS, R. F.; BARROS, N. F.; NEVES, J. C.; COUTO, C. Níveis críticos de fósforo no solo para o eucalipto. Revista Árvore, v.6, n.1, p.29-37, 1982. 308 REVENG 305-309 p. 308 Engenharia na agricultura, viçosa - mg, V.18 N.4, julho / agosto 2010 UTILIZAÇÃO DO BIOSSÓLIDO DE CERVEJARIA COMO SUBSTRATO... UTILIZAÇÃO DO BIOSSÓLIDO DE CERVEJARIA COMO SUBSTRATO... NOVAIS, R.F.; RÊGO, A.K.; GOMES, J.M. Nível crítico de potássio no solo e na planta para o crescimento de mudas de Eucalyptus grandis W.Hill ex Maiden e de Eucalyptus cloeziana F.Muell. Revista Árvore, v.4, n.1, p.14-23, 1980. biodigestão anaeróbica. São Carlos, SP: Embrapa Instrumentação Agropecuária, Documentos n.13, 50p., 2004. TRIGUEIRO, R.M. & GUERRINI, I.A. Uso de biossólido como substrato para produção de mudas de eucalipto. Scientia Forestalis, n.64, p.150-162, 2003. ROSA, S.E.S. da; COSENZA, J.P.; LEÃO, L.T.S. Panorama do setor de bebidas no Brasil. BNDES Setorial, Rio de Janeiro, n.23, p.101-150, mar. 2006. TSUTIYA, M. T. Qualidade de biossólidos produzidos em estações de tratamento de esgotos na região metropolitana de São Paulo. In: Congresso Brasileiro de Engenharia Sanitária e Ambiental, 21, 2001, Rio de Janeiro. Anais... Rio de Janeiro, RJ, ABES, 2001, CD Rom. SANTOS, M.S. & RIBEIRO, M.F. Cervejas e refrigerantes. São Paulo: CETESB, Série P+L, 2005. 58p. SILVA, W.T.L. da; NOVAES, A. P. de; MARTIN- NETO, L.; MILORI, D. M. B. P.; SIMÕES, M. L.; HANEDA, R. N.; FIALHO, L. L.; LEONELLI, F. C. V. Método de aproveitamento biossólido proveniente de lodo de esgoto residencial através de processo de compostagem seguindo de VALIM, M.C.A. et al. Compostagem de esgoto com resíduos agrícolas através da aeração forçada positiva. In: Congresso Brasileiro de Engenharia Sanitária e Ambiental, 19, 2001, Vitória. Anais... Rio de Janeiro, RJ, ABES, 2001, CD Rom. REVENG 305-309 p. 309 Engenharia na agricultura, viçosa - mg, V.18 N.4, julho / agosto 2010 309 Engenharia na agricultura, viçosa - mg, V.18 N.4, julho / agosto 2010 305-309 p.
https://openalex.org/W3208847801	https://link.springer.com/content/pdf/10.1140/epjc/s10052-021-09700-w.pdf	English	null	The energy spectrum of cosmic rays beyond the turn-down around $$\varvec{10^{17}}$$ eV as measured with the surface detector of the Pierre Auger Observatory	European physical journal. C, Particles and fields	2,021	cc-by	21,885	The energy spectrum of cosmic rays beyond the turn-down around 1017 eV as measured with the surface detector of the Pierre Auger Observatory Pierre Auger Collaboration⋆ The Pierre Auger Observatory, Av. San Martín Norte 306, 5613 Malargüe, Mendoza, Argentina; URL: http://www.auger.org Pierre Auger Collaboration⋆ The Pierre Auger Observatory, Av. San Martín Norte 306, 5613 Malargüe, Mendoza, Argentina; URL: http://www.auger.org Received: 10 May 2021 / Accepted: 29 September 2021 / Published online: 2 November 2021 © The Author(s) 2021 Abstract We present a measurement of the cosmic-ray spectrum above 100PeV using the part of the surface detector of the Pierre Auger Observatory that has a spacing of 750 m. An inﬂection of the spectrum is observed, conﬁrming the presence of the so-called second-knee feature. The spectrum is then combined with that of the 1500m array to produce a single measurement of the ﬂux, linking this spectral feature with the three additional breaks at the highest energies. The combined spectrum, with an energy scale set calorimetrically via ﬂuorescence telescopes and using a single detector type, results in the most statistically and systematically precise measurement of spectral breaks yet obtained. These mea- surements are critical for furthering our understanding of the highest energy cosmic rays. ated by diffusing in the moving magnetic heterogeneities in shocks accordingly to their rigidity. That the CR composi- tion gets heavier for two decades in energy above the knee energy could thus reﬂect that heavier elements, although sub- dominant below the knee, are accelerated to higher energies, until the iron component falls off steeply at a point of turn- down around ≃1016.9 eV. Such a bending has been observed in several experiments at a similar energy, referred to as the “second knee” or “iron knee” [8–11]. The recent observa- tions of gamma rays of a few 1014 eV from decaying neutral pions, both from a direction coincident with a giant molec- ular cloud [12] and from the Galactic plane [13], provide evidence for CRs indeed accelerated to energies of several 1015 eV, and above, in the Galaxy. A dozen of sources emit- ting gamma rays up to 1015 eV have even been reported [14], and the production could be of hadronic origin in at least one of them [15]. However, the nature of the sources and the mechanisms by which they accelerate CRs remain in gen- eral undecided. In particular, that particles can be effectively accelerated to the rigidity of the second knee in supernova remnants is still under debate, see e.g. [16]. Eur. Phys. J. C (2021) 81:966 https://doi.org/10.1140/epjc/s10052-021-09700-w Regular Article - Theoretical Physics ⋆e-mail: spokespersons@auger.org 1 Introduction The 1600 SD locations which make up the SD-1500 are shown in black while the stations which belong only to the SD-750 and the boarder of this sub-array are highlighted in cyan The FD consists of ﬁve telescopes at four sites which look out over the surface array, see Fig. 1. Four of the tele- scopes (shown in blue) cover an elevation range from 0◦ to 30◦while the ﬁfth, the High Elevation Auger Telescopes (HEAT), covers an elevation range from 30◦to 58◦(shown in red). Each telescope is used to collect the light emitted from air molecules excited by charged particles. After ﬁrst selecting the UV band with appropriate ﬁlters (310–390nm), the light is reﬂected off a spherical mirror onto a camera of 22×20 hexagonal, 45.6mm, photo-multiplier tubes (PMTs). In this way, the longitudinal development of the particle cas- cades can be studied and the energy contained within the electromagnetic sub-showers can be measured in a calori- metric way. Thus the FD can be used to set an energy scale for the Observatory that is calorimetric and so is independent of simulations of shower development. p The SD, the data of which are the focus of this paper, consists of two nested hexagonal arrays of water Cherenkov detectors (WCDs). The layout, shown in Fig. 1, includes the SD-1500, with detectors spread apart by 1500m and total- ing approximately 3000km2 of effective area. The detectors of the SD-750 are instead spread out by 750m, yielding an effective area of 24km2. SD-750 and SD-1500 include iden- tical WCDs, cylindrical tanks of pure water with a 10m2 base and a height of 1.2m. Three 9” PMTs are mounted to the top of each tank and view the water volume. When relativistic secondaries enter the water, Cherenkov radiation is emitted, reﬂectedviaaTyvekliningintothePMTs,anddigitizedusing 40 MHz 10-bit Flash Analog to Digital Converters (FADCs). Each WCD along with its digitizing electronics, communi- cation hardware, GPS, etc., is referred to as a station. Fig. 1 The layout of the SD and FD of the Pierre Auger Observatory are shown above. The respective ﬁelds of view of the ﬁve FD sites are shown in blue and orange. The 1600 SD locations which make up the SD-1500 are shown in black while the stations which belong only to the SD-750 and the boarder of this sub-array are highlighted in cyan well-accredited,afullunderstandingofhowitoccursishence lacking. 1 Introduction The steepening of the energy spectrum of cosmic rays (CRs) at around 1015.5 eV, ﬁrst reported in [1], is referred to as the “knee” feature. A widespread view for the origin of this bending is that it corresponds to the energy beyond which the efﬁciency of the accelerators of the bulk of Galactic CRs is steadily exhausted. The contribution of light elements to the all-particle spectrum, largely dominant at GeV energies, remains important up to the knee energy after which the heav- ier elements gradually take over up to a few 1017 eV [2–6]. This ﬁts with the long-standing model that the outer shock boundaries of expanding supernova remnants are the Galac- tic CR accelerators, see e.g. [7] for a review. Hydrogen is indeed the most abundant element in the interstellar medium that the shock waves sweep out, and particles are acceler- Above 1017 eV, the spectrum steepens in the interval lead- ing up to the “ankle” energy, ∼5×1018 eV, at which point it hardens once again. The inﬂection in this energy range is not as sharp as suggested by the energy limits reached in the Galactic sources to accelerate iron nuclei beyond the iron- knee energy [17]. Questions arise, then, on how to make up the all-particle spectrum until the ankle energy. The harden- ing around 1017.3 eV in the light-particle spectrum reported in [18] is suggestive of an extragalactic contribution to the all-particle spectrum steadily increasing. It has even been argued that an additional component is necessary to account for the extended gradual fall-off of the spectrum and for the mass composition in the iron-knee-to-ankle region, be it of Galactic [17] or extragalactic origin [19]. Supplementary Information The online version contains supplementary material available at https://doi.org/10.1140/epjc/ s10052-021-09700-w. Supplementary Information The online version contains supplementary material available at https://doi.org/10.1140/epjc/ s10052-021-09700-w. WhiletheconceptthattheGalactic-to-extragalactictransi- tion occurs somewhere between 1017 eV and a few 1018 eV is ⋆e-mail: spokespersons@auger.org 12 3 966 Page 2 of 25 Eur. Phys. J. C (2021) 81 :966 Fig. 1 The layout of the SD and FD of the Pierre Auger Observatory are shown above. The respective ﬁelds of view of the ﬁve FD sites are shown in blue and orange. 1 Introduction The approximately power-law shape of the spectrum in this energy range may mask a complex superposition of different components and phenomena, the disentanglement of which rests on the measurements of the all-particle energy spectrum, and of the abundances of the different elements as a function of energy, both of them challenging from an exper- imental point of view. On the one hand, the energy range of interest is accessible only through indirect measurements of CRs via the extensive air showers that they produce in the atmosphere. Therefore, the determination of the properties of the CRs, especially their mass and energy, is prone to systematic effects. On the other hand, different experiments, different instruments and different techniques of analysis are used to cover this energy range, so that a unique view of the CRs is only possible by combining measurements the matching of which inevitably implies additional systematic effects. Using data collected over 15 years with the SD-1500, we recently reported the measurement of the CR energy spec- trum in the range covering the region of the ankle up to the highest energies [20,21]. In this paper we extend these mea- surements down to 1017 eV using data from the SD-750: not only is the detection technique consistent but the same meth- ods are used to treat the data and build he spectrum. The paper is organized as follows: we ﬁrst explain how, with the SD-750 array, the surface array is sensitive to primaries down to 1017 eV in Sect. 2; in Sect. 3, we describe how we recon- struct the showers up to determining the energy; we illustrate in Sect. 4 the approach used to derive the energy spectrum from SD-750; ﬁnally, after combining the spectra measured by SD-750 and SD-1500, we present the spectrum measured using the Auger Observatory from 1017 eV upwards in Sect. 5 and discuss it in the context of other measurements in Sect. 6. TheaimofthispaperistopresentameasurementoftheCR spectrum from 1017 eV up to the highest observed energies, based on the data collected with the surface-detector array of the Pierre Auger Observatory. The Observatory is located in the Mendoza Province of Argentina at an altitude of 1400m above sea level at a latitude of 35.2◦S, so that the mean atmospheric overburden is 875g/cm2. Extensive air showers induced by CR-interactions in the atmosphere are observed via a hybrid detection using a ﬂuorescence detector (FD) and a surface detector (SD). 2 Identiﬁcation of showers with the SD-750: from the trigger to the data set Originally, two triggers were implemented into the sta- tion ﬁrmware, called threshold (TH), more adept to detect muons, and time-over-threshold (ToT), more suited to iden- tify the electromagnetic component. Both of these have set- tings which require the signal to be higher in amplitude or longer than what is observed for a muon traveling vertically through the water volume. As such, they have the inherent limitation of being insensitive to signals which are smaller than (or equal to) that of a single muon, thus prohibiting the measurement of pure electromagnetic signals, which are generally smaller. The implementation of an additional set of station-level trig- ger algorithms in mid-2013 is particularly relevant for the operation of the SD-750. Their inclusion in this work extends the energy range over which the SD-750 triggers with > 98% probability from 1017.2 eV down to 1017 eV. To identify showers, a hierarchical set of triggers is used which range in scope from the individual station-level up to the selection of events and the rejection of random coin- cidences. The trigger chain, extensively described in [22], has been used since the start of the data taking of the SD- 1500, and was successively adopted for the SD-750. In short, station-level triggers are ﬁrst formed at each WCD. They are then combined with those from other detectors and examined for spatial and temporal correlations, leading to an array trig- ger, which initiates data acquisition. After that, a similar hier- archical selection of physics events out of the combinatorial background is ultimately made. To bolster the sensitivity of the array to such small sig- nals, two additional triggers were designed. The ﬁrst, time- over-threshold-deconvolved (ToTd), ﬁrst removes the typi- cal exponential decay created by Cherenkov light inside the water volume, after which the ToT algorithm is applied. The second, multiplicity-of-positive-steps (MoPS), is designed to select small, non-smooth signals, a result of many electro- magnetic particles entering the water over a longer period of time than a typical muon pulse. This is done by counting the number of instances in the waveform where consecutive bins are increasing in amplitude. Both of the trigger algorithms are described in detail in Appendix A. We describe in this section the design of the triggers (Sect. 2.1). We then illustrate their effect on the data, at the level of the amplitude of detected signals (Sect. 123 123 Eur. Phys. J. C (2021) 81 :966 Page 3 of 25 966 2 Identiﬁcation of showers with the SD-750: from the trigger to the data set 1 The response of an individual WCD to secondary particles has been studied using unbiased FADC waveforms and dedicated studies of sig- nals from muons [23]. 2 Identiﬁcation of showers with the SD-750: from the trigger to the data set 2.2) and on the timing of detected signals in connection with the event selection (Sect. 2.3). Finally we describe the energy at which acceptance is 100% (Sect. 2.4). A more detailed description of the trigger algorithms can be found in Appendix A. The implementation of the ToTd and MoPS (the rate of which is around 0.3Hz, compared to 0.6Hz of ToT and 20Hz of TH) did not require any modiﬁcation in the logic of the array trigger, which calls for a coincidence of three or more SD stations that pass any combination of the triggers described above with compact spacing, spatially and tempo- rally [22]. We note that in spite of the low rate of the ToTd and MoPS relative to TH and ToT, the array rate more than doubled after their implementation. This, as will be shown in the following, is due to the extension of measurements to the more abundant, smaller signals. 2.1 The electromagnetic triggers Using the station-level triggers, the digitized waveforms are constantly monitored in each detector for patterns consistent with what would be expected as a result of air-shower sec- ondary particles (primarily electrons and photons of 10 MeV on average, and GeV muons) entering the water volume.1 The typical morphologies include large signals, not neces- sarily spread in time, such as those close to the shower core, or sequences of small signals spread in time, such as those nearby the core in low-energy showers, or far from the core in high-energy ones. Atmospheric muons, hitting the WCDs at a rate of 3kHz, are the primary background. The output from the PMTs has only a small dependence on the muon energy. The electromagnetic and hadronic background, while also present, yields a total signal that is usually less than that of a muon. Consequently, the atmospheric muons are the pri- mary impediment to developing a station-level trigger for small signal sizes without contaminating the sampling of an air shower with spurious muons. 2.3 Effects of ToTd and MoPS on signal timing The increased responsiveness of the ToTd and MoPS algo- rithms tosmaller signals, speciﬁcally due to the electromag- netic component, has an effect also on the observed timing of the signals. In general, the electromagnetic signals are expected to be delayed with respect to the earliest part of the shower which is muon-rich, the delay increasing with axial distance. Further, in large events, stations that pass these trig- gers tend to be on the edge of the showers, where the front is thicker, thus increasing the variance of the arrival times. Such effects can be seen through the distribution of the start times for stations that pass the ToTd and MoPS triggers. Fig. 3 The increase in station multiplicity when including the ToTd and MoPS triggers versus the original multiplicity with only ToT and TH. The black circles show the median increase in that multiplicity bin is the convolution of three effects, the uniformity of the array, the decreasing density of particles as a function of perpen- dicular distance to the shower axis (henceforth referred to as the axial distance), and the shape of the CR spectrum result- ing in the negative slope above ≃7VEM. Furthermore there is a decreasing efﬁciency of the ToT and TH at small signal sizes. The range of additional signals that are now detectable via the ToTd and MoPS triggers are shown in dashed red. As expected, ToTd and MoPS triggers increase the probability of the SD to detect small amplitude signals, namely between 0.3 and 5VEM. That the high-signal tail of this distribution ends near 10VEM is consistent with a previous study [24] that estimated that the ToT+TH triggers were fully efﬁcient above this value. The residuals of the pulse start times with respect to a plane front ﬁt of the three stations with the largest signals in the event are shown in Fig. 4 for different trigger types. The entries shown in blue correspond to stations that passed the ToT algorithm, the ones in green to stations that pass the TH trigger (but not the ToT trigger), and those in red to stations that pass the ToTd and/or MoPS triggers, only. For each of the trigger types, there is a clear peak near zero, which reﬂects the approximately planar shower front close to the core. 2.2 Effect of ToTd and MoPS on signals amplitudes 3 The increase in station multiplicity when including the ToTd and MoPS triggers versus the original multiplicity with only ToT and TH. The black circles show the median increase in that multiplicity bin increase is characterized in Fig. 3, which shows the number of additional triggered stations per event as a function of the number of stations that pass the TH and ToT triggers, after removing spuriously triggered stations. The median increase of multiplicity in each horizontal bin is shown by the black circles and indicates a typical increase of one station per event. 2.2 Effect of ToTd and MoPS on signals amplitudes The ToTd and MoPS triggers extend the range over which signals can be observed at individual stations into the region whichis dominatedbythebackgroundmuons that arecreated inrelativelylowenergyairshowers.Byremaininginsensitive to muon-like signals, these two triggers increase the sensi- tivity of the SD to the low-energy parts of the showers that have previously been below the trigger threshold. The effects of the additional triggers can be seen in the distribution of the observed signal sizes. An example of such a distribution, based on one month of air-shower data, is shown in Fig. 2. The signal sizes are shown in the calibration unit of one vertical equivalent muon (VEM), the total deposited charge of a muon traversing vertically through the water volume [22]. For the stations passing only the ToT and TH triggers (shown in solid black), the distribution of deposited signals 12 3 Eur. Phys. J. C (2021) 81 :966 966 Page 4 of 25 Fig. 2 Distribution of the signal sizes at individual stations which pass the TH and ToT triggers (solid black) and signals which pass only the ToTd and/or MoPS triggers (dashed red) Fig. 4 Distributions of start times with respect to a plane front for stations that pass the ToT and TH algorithms, in blue and in green, respectively. The signals due to ToTd and MoPS are shown in red. P i i id l d d l i h h l Fig. 4 Distributions of start times with respect to a plane front for stations that pass the ToT and TH algorithms, in blue and in green, respectively. The signals due to ToTd and MoPS are shown in red. Positive residuals correspond to a delay with respect to the plane wave expectation Fig. 2 Distribution of the signal sizes at individual stations which pass the TH and ToT triggers (solid black) and signals which pass only the ToTd and/or MoPS triggers (dashed red) Fig. 2 Distribution of the signal sizes at individual stations which pass the TH and ToT triggers (solid black) and signals which pass only the ToTd and/or MoPS triggers (dashed red) Fig. 4 Distributions of start times with respect to a plane front for stations that pass the ToT and TH algorithms, in blue and in green, respectively. The signals due to ToTd and MoPS are shown in red. Positive residuals correspond to a delay with respect to the plane wave expectation Fig. 2 The energy-cut corresponding to the full-efﬁciency threshold increases with zenith angle, due to the increasing attenuation of the electromagnetic component with slant depth. The zenith angle 40◦was chosen as a balance to have good statistical precision and a low energy threshold. 2.3 Effects of ToTd and MoPS on signal timing Stations that pass the TH condition, but not the ToT one, tend to capture isolated muons, including background The additional sensitivity to small air-shower signals also increases the multiplicity of triggered stations per event. This 123 123 Page 5 of 25 966 Eur. Phys. J. C (2021) 81 :966 Fig. 5 The detection efﬁciency of the SD-750 for air showers with θ < 40◦is shown for the original (dashed red) and expanded (solid blue) station-level trigger sets with bands indicating the systematic uncertain- ties. The trigger efﬁciency was determined using data above 1016.8 eV and is extrapolated below this energy (shown in gray) Table 1 Temporal window limits tlow and thigh used to remove stations from an event, for each station-level trigger algorithm Trigger type tlow (ns) thigh (ns) ToT −397 1454 ToTd −468 2285 MoPS −477 2883 TH −485 1379 Table 1 Temporal window limits tlow and thigh used to remove stations from an event, for each station-level trigger algorithm muons arriving randomly in time. This explains the vertical offset, ﬂat and constant, in the green curve. In turn, the lack of such a baseline shift in the blue and red distributions gives evidence that the ToT, TOTd and MoPS algorithms reject background muons effectively. This is particularly success- ful for the ToTd and MoPS that accept very small signals, of approximately 1VEM in size. One can see that these distri- butions have different shapes and that, in particular, the start time distributions of signals that pass the ToTd and MoPS have much longer tails than those of the TOT triggers, includ- ing a second distribution beginning around 1.5µs possibly due to heavily delayed electromagnetic particles. Fig. 5 The detection efﬁciency of the SD-750 for air showers with θ < 40◦is shown for the original (dashed red) and expanded (solid blue) station-level trigger sets with bands indicating the systematic uncertain- ties. The trigger efﬁciency was determined using data above 1016.8 eV and is extrapolated below this energy (shown in gray) the chance that a shower will produce a given SD trigger as a function of axial radius, was calculated for all trigger types. The LTP was then parameterized as a function of the observed air-shower zenith angle and energy. 2.3 Effects of ToTd and MoPS on signal timing It is important to note that because the LTP is derived using observed air showers as a function of energy, this calculation reﬂects the efﬁciency as a function of energy based on the true underly- ing mass distribution of primary particles. Further details of this method can be found in [25]. The extended time portion of showers accessed by the ToTd and MoPS triggers has implications on the procedure used to select physical events from the triggered ones [22]. In this process, non-accidental events, as well as non-accidental stations, are disentangled on the basis of their timing. First, we identify the combination of three stations where they form atriangle,inwhichatleasttwolegsare750mlong,andwhere they have the largest summed signal among all such possible conﬁgurations. These stations make up the event seed and the arrival times of the signals are ﬁt to a plane front. Addi- tional stations are then kept if their temporal residual, t, is within a ﬁxed window, tlow < t < thigh. Motivated by the differing time distributions, updated tlow and thigh values were calculated based on which trigger algorithm was sat- isﬁed. Using the distributions of timing residuals, shown in Fig. 4, the baseline was ﬁrst subtracted. Then the limits of the window, tlow and thigh, were chosen such that the middle 99% of the distribution was kept. The trigger-wise limits are summarized in Table 1. The SD-750 trigger efﬁciency was then determined via a study in which isotropic arrival directions and random core positions were simulated for ﬁxed energies between 1016.5 and1018 eV.Eachstationonthearraywasrandomlytriggered using the probability given by the LTP. The set of stations that triggered were then checked against the compactness criteria of the array-level triggers, as described in [22]. The resulting detection probability for showers with zenith angles < 40◦is shown as a solid blue line in Fig. 5 as a function of energy. The detection efﬁciency becomes almost unity (> 98%) at around 1017 eV.2 For comparison, we show in the same ﬁgure, in dashed red, the detection efﬁciency curve for the original set of station-triggers, TH and ToT, in which the full efﬁciency is attained at a larger energy, i.e., around 1017.2 eV. 2.4 Effect of the ToTd and MoPS on the energy above which acceptance is fully-efﬁcient 2.4 Effect of the ToTd and MoPS on the energy above which acceptance is fully-efﬁcient A description for the detection efﬁciency, ϵ(E), below 1017 eV, will be important for unfolding the detector effects close to the threshold energy (see Sect. 4). This quantity was ﬁt using the results of the LTP simulations with θ < 40◦and Mostrelevanttothemeasurementofthespectrumisthedeter- mination of the energy threshold above which the SD-750 becomes fully efﬁcient. To derive this, events observed by the FD were used to characterize this quantity as a function of energy and zenith angle. The FD reconstruction requires only a single station be triggered to yield a robust determina- tion of the shower trajectory. Using the FD events with ener- gies above 1016.8 eV, the lateral trigger probability (LTP), 12 3 966 Page 6 of 25 Eur. Phys. J. C (2021) 81 :966 is well-parameterized by is well-parameterized by is well-parameterized by The lateral fall-off of the signal, S(r), with increasing distance, r, to the shower axis in the shower plane is mod- eled with a lateral distribution function (LDF). The stochastic variations in the location and character of the leading inter- action in the atmosphere result in shower-to-shower ﬂuctu- ations of the longitudinal development that propagate onto ﬂuctuations of the lateral proﬁle, sampled at a ﬁxed depth. Showers induced by identical primaries at the same energy and at the same incoming angle can thus be sampled at the ground level at a different stage of development. The LDF is consequently a quantity that varies on an event-by-event basis. However, the limited degrees of freedom, as well as the sparse sampling of the air-shower particles reaching the ground, prevent the reconstruction of all the parameters of the LDF for individual events. Instead, an average LDF, ⟨S(r)⟩, is used in the reconstruction to infer the expected signal, S(ropt), that would be detected by a station located at a ref- erence distance from the shower axis, ropt [27,28]. This ref- erence distance is chosen so as to minimize the ﬂuctuations of the shower size, down to ≃7% in our case. The observed distribution of signals is then adjusted to ⟨S(r)⟩by scaling the normalization, S(ropt), in the ﬁtting procedure. ϵ(E) = 1 2 1 + erf lg(E/eV) −μ σ , (1) (1) where erf(x) is the error function, μ = 16.4 ± 0.1 and σ = 0.261 ± 0.007. 3 This is primarily due to the instabilities in the wireless communica- tions systems as well as periods where large fractions of the array were not functioning. 3.1 Estimation of the shower size where ρ = ln(r/(450 m)), and β and γ are two structure parameters. The overall steepness of the fall-off of the signal from the core is governed by β, while the concave devia- tion from a power-law function is given by γ . The values of β and γ have been obtained in a data-driven manner, by using a set of air-shower events with more than three stations, none of which have a saturated signal. The zenith angle and the shower size are used to trace the age dependence of the structure parameters based on the following parameteriza- tion in terms of the reduced variables t = sec θ −1.27 and u = ln S(450) −5: where ρ = ln(r/(450 m)), and β and γ are two structure parameters. The overall steepness of the fall-off of the signal from the core is governed by β, while the concave devia- tion from a power-law function is given by γ . The values of β and γ have been obtained in a data-driven manner, by using a set of air-shower events with more than three stations, none of which have a saturated signal. The zenith angle and the shower size are used to trace the age dependence of the structure parameters based on the following parameteriza- tion in terms of the reduced variables t = sec θ −1.27 and u = ln S(450) −5: The general strategy for the reconstruction of air showers using the SD-750 array is similar to that used for the SD- 1500 array which is detailed extensively in [26]. In this pro- cess, the arrival direction is obtained using the start times of signals, assuming either a plane or a curved shower front, as the degrees of freedom allow. The lateral distribution of the signal is then ﬁtted to an empirically-chosen function to infer the size of the air shower, which is used as a surrogate for the primary energy. The reconstruction algorithm thus pro- duces an estimate of the arrival direction and the size of the air shower via a log-likelihood minimization. β = (β0 + β1t + β2t2)(1 + β3u), (3) γ = γ0 + γ1u. (4) (3) (3) (4) (4) 3 This is primarily due to the instabilities in the wireless communica- tions systems as well as periods where large fractions of the array were not functioning. 2.4 Effect of the ToTd and MoPS on the energy above which acceptance is fully-efﬁcient For events used in this analysis, there is an additional requirement regarding the containment of the core within the array: only events in which the detector with the high- est signal is surrounded by a hexagon of six stations that are fully operational are used. This criterion not only ensures adequate sampling of the shower but also allows the aper- ture of the SD-750 to be evaluated in a purely geometrical manner [22]. With these requirements, the SD-750 data set used below consists of about 560,000 events with θ < 40◦ and E > 1017 eV recorded between 1 January 2014 and 31 August 2018. The minimum energy cut is motivated by the lowest energy to which we can cross-calibrate with adequate statistics the energy scale of the SD with that of the FD (see Sect. 3.3). The corresponding exposure, E, after removal of time periods when the array was unstable3 (<2% of the total) is E = (105 ± 4)km2 sryr. The reference distance, or optimal distance, ropt, has been determined on an event-by-event basis by ﬁtting the mea- sured signals to different hypotheses for the fall-off of the LDF with distance to the core as in [28]. Via a ﬁt of many power-law-like functions, the dispersion of signal expecta- tions has been observed to be minimal at ropt ≃450m, which is primarily constrained by the geometry of the array. The expected signal at 450m from the core, S(450), has thus been chosen to deﬁne the shower-size estimate. 3 Energy measurements with the SD-750 Inthissection,themethodfortheestimationoftheair-shower energy is detailed together with the resulting energy resolu- tion of the SD-750 array. The measurement of the actual shower size is ﬁrst described in Sect. 3.1 after which the corrections for attenuation effects are presented in Sect. 3.2. The energy calibration of the shower size after correction for attenuation is presented in Sect. 3.3. The energy resolution function is ﬁnally derived in Sect. 3.4. The functional shape chosen for the average LDF is a parabola in a log-log representation of ⟨S(r)⟩as a function of the distance to the shower core, ln⟨S(r)⟩= ln S(450) + β ρ + γ ρ2, (2) (2) 3.2 Correction of attenuation effects There are two signiﬁcant observational effects that impact the precision of the estimation of the shower size. Both of these effects are primarily a result of the variable slant depth that a shower must traverse before being detected with the SD. Since the mean atmospheric overburden is 875g/cm2 at the location of the Observatory, nearly all observed showers in the energy range considered in this analysis have already reached their maximum size and have started to attenuate [29]. Thus, an increase in the slant depth of a shower results inamoreattenuatedcascadeattheground,directlyimpacting the observed shower size. We empirically chose a functional form which describes the relative amount of attenuation of the air shower, fCIC(θ) = 1 + ax + bx2. (6) (6) The scaling of this function is normalized to the attenuation of a shower arriving at 35◦by choosing x = sin2 35◦−sin2 θ. For a given air shower, the observed shower size can be scaled using Eq. (6) to get the equivalent signal of a shower arriv- ing with the reference zenith angle, S35, via the relationship S(450) = S35 fCIC(θ). The scaling of this function is normalized to the attenuation of a shower arriving at 35◦by choosing x = sin2 35◦−sin2 θ. For a given air shower, the observed shower size can be scaled using Eq. (6) to get the equivalent signal of a shower arriv- ing with the reference zenith angle, S35, via the relationship S(450) = S35 fCIC(θ). Isotropy implies that dN/d sin2 θ is constant. Thus, the shape of fCIC(θ) is determined by ﬁnding the parameters a and b for which the CDF of events above S(450) > Scut fCIC(θ) is linear in sin2 θ using an Anderson-Darling test [33]. The parameter Scut deﬁnes the size of a shower with θ = 35◦at which the CIC tuning is performed, the choice of which is described below. The ﬁrst observational effect is related to the changing weather at the Observatory. Fluctuations in the air pressure equate to changes in the local overburden and thus showers observed during periods of relatively high pressure result in anunderestimatedshowersize.Similarly,thevariationsinthe air density directly change the Molière radius which directly affects the spread of the shower particles. The increased lat- eral spread of the secondaries, or equivalently, the decrease in the density of particles on the ground, also leads to a sys- tematically underestimated shower size. 3.1 Estimation of the shower size For any speciﬁc set of values p = {βi, γi}, the reconstruction is then applied to calculate the following χ2-like quantity, 3 Page 7 of 25 966 Page 7 of 25 966 Eur. Phys. J. C (2021) 81 :966 Table 2 Best-ﬁt {βi, γi} values deﬁning the structure parameters of the LDF Parameter Value β0 2.95 ± 0.02 β1 −1.0 ± 0.2 β2 0.7 ± 0.2 β3 0.02 ± 0.01 γ0 0.26 ± 0.09 γ1 −0.02 ± 0.01 Table 2 Best-ﬁt {βi, γi} values deﬁning the structure parameters of the LDF Parameter Value β0 2.95 ± 0.02 β1 −1.0 ± 0.2 β2 0.7 ± 0.2 β3 0.02 ± 0.01 γ0 0.26 ± 0.09 γ1 −0.02 ± 0.01 events detected with the SD [30]. From this relationship, a model was constructed to scale the observed value of S(450) to what would have been measured had the shower been instead observed at a time with the daily and yearly aver- age atmosphere. When applying this correction to individ- ual air showers, the measurements from the weather stations located at the FD sites are used. The values of S(450) are scaled up or down according to these measurements, result- ing in a shift of at most a few percent. The shower size is eventually the proxy of the air-shower energy, which is cali- brated with events detected with the FD (see Sect. 3.3). Since the FD operates only at night when, in particular, the air den- sity is relatively low, the scaling of S(450) to a daily and yearly average atmosphere corrects for a ≃0.5% shift in the assigned energies. globally to all events: Q2(p) = 1 Ntot Nevents k=1 Nk j=1 (Sk, j −⟨S(r j, p)⟩)2 σ 2 k, j . (5) (5) The second observational effect is geometric, wherein showers arriving at larger zenith angles have to go through more atmosphere before reaching the SD. To correct for this effect, the Constant Intensity Cut (CIC) method [31] is used. The CIC method relies on the assumption that cosmic rays arrive isotropically, which is consistent with observations in the energy range considered [32]. The intensity is thus expected to be independent of arrival direction after correct- ing for the attenuation. Deviations from a constant behavior can thus be interpreted as being due to attenuation alone. 3.1 Estimation of the shower size Based on this property, the CIC method allows us to deter- mine the attenuation curve as function of the zenith angle and therefore to infer a zenith-independent shower-size esti- mator. The sum over Nk stations is restricted to those with observed signals larger than 5VEM to minimize the impact of upward The sum over Nk stations is restricted to those with observed signals larger than 5VEM to minimize the impact of upward ﬂuctuations of the station signals far from the core and hence to avoid biases from trigger effects, and to stations more than 150m away from the core. The uncertainty σk, j is propor- tional to Sk, j [26]. Ntot is the total number of stations in all such events. The best-ﬁt {βi, γi} values are collected in Table 2. The sum over Nk stations is restricted to those with observed signals larger than 5VEM to minimize the impact of upward ﬂuctuations of the station signals far from the core and hence to avoid biases from trigger effects, and to stations more than 150m away from the core. The uncertainty σk, j is propor- tional to Sk, j [26]. Ntot is the total number of stations in all such events. The best-ﬁt {βi, γi} values are collected in Table 2. 3.3 Energy calibration of the shower size The conversion of the shower size, corrected for attenuation, is based on a special set of showers, called golden hybrid events, which can be reconstructed independently by the FD and by the SD. The FD allows for a calorimetric estimate of the primary energy except for the contribution carried away by particles that reach the ground. The amount of this so-called invisible energy, ≃20% at 1017 eV and ≃15% at 1018 eV, has been evaluated using simulations [34] tuned to measurementsat1018.3 eVsoastocorrectforthediscrepancy in the muon content of simulated and observed showers [35]. The empirical relationship between the FD energy measure- ments, EFD, and the corrected SD shower size, S35, allows for the propagation of the FD energy scale to the SD events. Fig. 6 Top: histogram of reconstructed shower sizes and zenith angles. The solid black line represents the shape of fCIC at 10VEM. Bottom: same distribution but as a function of corrected shower size, S35, and zenith angle. The dashed black line indicates the mapping of the solid black line in the top ﬁgure after inverting the effects of the CIC correc- tion Fig. 6 Top: histogram of reconstructed shower sizes and zenith angles. The solid black line represents the shape of fCIC at 10VEM. Bottom: same distribution but as a function of corrected shower size, S35, and zenith angle. The dashed black line indicates the mapping of the solid black line in the top ﬁgure after inverting the effects of the CIC correc- tion Table 3 The energy dependence of the CIC parameters (Eq. (6)) are given below k0 k1 k2 a 2.42 −0.886 0.268 b −4.56 5.61 −2.47 Table 3 The energy dependence of the CIC parameters (Eq. (6)) are given below k0 k1 k2 a 2.42 −0.886 0.268 b −4.56 5.61 −2.47 FD events were selected based on quality and ﬁducial cri- teria aimed at guaranteeing a precise estimation of EFD as well as at minimizing any acceptance biases towards light or heavy mass primaries introduced by the ﬁeld of view of the FD telescopes. The cuts used for the energy calibra- tion are similar to those described in [29,36]. They include the selection of data when the detectors are properly oper- ational and the atmosphere properties like clouds coverage and the vertical aerosol depth are suitable for a good deter- mination of the air-shower proﬁle. 3.2 Correction of attenuation effects Both the air-density and pressure have typical daily and yearly cycles that imprint similar cycles upon the estimation of the shower size. Sincetheattenuationthat ashower undergoes beforebeing detected is related to the depth of shower maximum and the particle content, the shape of fCIC(θ) is dependent on both the energy and the average mass of the primary particles at that energy. Further, this implies that a single choice of Scut could introduce a mass and/or energy bias. Thus, Eq. (6) was extended to allow the polynomial coefﬁcients, k ∈{a, b}, The relationship between these two atmospheric param- eters and the estimated shower sizes has been studied using 123 12 3 3 Eur. Phys. J. C (2021) 81 :966 966 Page 8 of 25 Fig. 6 Top: histogram of reconstructed shower sizes and zenith angles. The solid black line represents the shape of fCIC at 10VEM. Bottom: same distribution but as a function of corrected shower size, S35, and zenith angle. The dashed black line indicates the mapping of the solid black line in the top ﬁgure after inverting the effects of the CIC correc- tion Fig. 7 Correlation between the SD shower-size estimator, S35, and the reconstructed FD energy, EFD, for the selected hybrid events Fig. 7 Correlation between the SD shower-size estimator, S35, and the reconstructed FD energy, EFD, for the selected hybrid events Fig. 7 Correlation between the SD shower-size estimator, S35, and the reconstructed FD energy, EFD, for the selected hybrid events 3.3 Energy calibration of the shower size So we choose a power-law type relationship, Table 4 The systematic uncertainties on the FD energy scale are given below. Lines with multiple entries represent the values at the low and high end of the considered energy range (≃1017 and ≃1019 eV, respec- tively) Systematic Uncertainty (%) Absolute ﬂuorescence yield 3.6 Atmosphere and scattering 2–6 FD Calibration 10 Longitudinal proﬁle reconstruction 7–5.5 Invisible energy 3–1.5 ESD = ASB 35, (7) ESD = ASB 35, (7) educated guess for σsh(S35), as expected from Monte-Carlo simulationsforamass-compositionscenariocompatiblewith data [29]. The total resolution σSD(S35)/S35 is then extracted from data as explained next in Sect. 3.4 and used in a second iteration. which is expected from Monte-Carlo simulations in the case of a single logarithmic dependence of Xmax with energy. The energy of an event with S35 = 1VEM arriving at the refer- ence angle, A, and the logarithmic slope, B, are ﬁtted to the data by means of a maximum likelihood method which mod- els the distribution of golden-hybrid events in the plane of energies and shower sizes. The use of these events allows us to infer A and B while accounting for the clustering of events in the range 1017.4 to 1017.7 eV observed in Fig. 7 due to the fall-off of the energy spectrum combined with the restrictive golden-hybrid acceptance for low-energy, dim showers. A comprehensive derivation of the likelihood function can be found in [37]. which is expected from Monte-Carlo simulations in the case of a single logarithmic dependence of Xmax with energy. The energy of an event with S35 = 1VEM arriving at the refer- ence angle, A, and the logarithmic slope, B, are ﬁtted to the data by means of a maximum likelihood method which mod- els the distribution of golden-hybrid events in the plane of energies and shower sizes. The use of these events allows us to infer A and B while accounting for the clustering of events in the range 1017.4 to 1017.7 eV observed in Fig. 7 due to the fall-off of the energy spectrum combined with the restrictive golden-hybrid acceptance for low-energy, dim showers. A comprehensive derivation of the likelihood function can be found in [37]. The resulting relationship is shown as the red line in Fig. 7 with best-ﬁt parameters such that A = (13.2 ± 0.3)PeV and B = 1.002 ± 0.006. 3.3 Energy calibration of the shower size The goodness of the ﬁt is supported by the χ2/NDOF = 2120/1978 (p = 0.013). We use these val- ues of A and B to calibrate the shower sizes in terms of ener- gies by deﬁning the SD estimator of energies, ESD, according to Eq. (7). The SD energy scale is set by the calibration proce- dure and thus it inherits the A and B calibration-parameters uncertainties and the FD energy-scale uncertainties, listed in Table 4. The systematic uncertainty, after addition in quadra- ture, of the energy scale is about 14% and is almost energy independent. The energy independence is a consequence of the 10% uncertainty of the FD calibration, which is the dom- inant contribution. The probability density function entering the likelihood procedure, detailed in [37], is built by folding the cosmic-ray intensity, as observed through the effective aperture of the FD, with the resolution functions of the FD and of the SD. Note that to avoid the need to model accurately the cosmic- ray intensity observed through the effective aperture of the telescopes (and thus to reduce reliance on mass assumptions), the observed distribution of events passing the cuts described above is used. The FD energy resolution, σFD(E)/EFD, is typically between 6% and 8% [38]. It results from the sta- tistical uncertainty arising from the ﬁt to the longitudinal proﬁle, the uncertainties in the detector response, the uncer- tainties in the models of the state of the atmosphere, and the uncertainties in the expected ﬂuctuations from the invisible energy. The SD shower-size resolution, σSD(S35)/S35, is, on the other hand, comprised of two terms, the detector sam- pling ﬂuctuations, σdet(S35), and the shower-to-shower ﬂuc- tuations, σsh(S35). The former is obtained from the sum of the squares of the uncertainties from the reconstructed shower size and zenith angle, and from the attenuation-correction terms that make up the S35 assignment. The latter stem from the stochastic nature of both the depth of ﬁrst interaction of the primary and the subsequent development of the parti- cle cascade. This contribution thus depends on the CR mass composition and on the hadronic interactions in air showers. For this reason, the derivation of A and B follows a two-step procedure. A ﬁrst iteration of the ﬁt is carried out by using an 3.3 Energy calibration of the shower size A further quality selec- tion includes requirements on the uncertainties of the energy assignment (less than 12%) and of the reconstruction of the depth at the maximum of the air-shower development (less than 40 g cm−2). A possible bias due to a selection depen- dency on the primary mass is avoided by using an energy dependent ﬁducial volume determined from data as in [29]. R i i h d i h 1017 V to be functions of S(450) via k(S(450)) = k0 + k1y + k2y2 where y = lg(S(450)/VEM). The function fCIC(θ, S(450)) was tuned using an unbinned likelihood. The ﬁt was performed so as to guarantee equal intensity of the integral spectra using eight threshold values of Scut between 10 and 70VEM, evenly spaced in log-scale. These values were chosen to avoid triggering biases on the low end and the dwindling statistics on the high end. The best ﬁt parameters are given in Table 3. The resulting 2D distribution of the number of events, in equal bins of sin2 θ and lg S35, is shown in Fig. 6, bottom panel. It is apparent that the number of events above any sin2 θ value is equalized for any constant line for lg S35 ≳0.7. The magnitude of the CIC correction is (−27 ± 4)% for vertical showers (depending on S(450)) and +15% for a zenith angle of 40◦. Restricting the data set to events with EFD ≥1017 eV, (to ensure that the SD is operating in the regime of full efﬁciency) there are 1980 golden-hybrid events available to establish the relationship between S35 and EFD. Fourty-ﬁve 3 123 Eur. Phys. J. C (2021) 81 :966 Page 9 of 25 966 Page 9 of 25 966 events in the energy range between 1016.5 eV and 1017 eV are included in the likelihood as described in [37]. As S35 depends on the mass composition of the primary particles, the relation between S35 and EFD, shown in Fig. 7, accounts for the trend of the composition change with energy inher- ently as the underlying mass distribution is directly sampled by the FD. Measurements of ⟨Xmax⟩suggest that this compo- sition trend follows a logarithmic evolution up to an energy of 1018.3 eV, beyond which the number of events available for this analysis is too small to affect the results in any way [36]. 3.4 Resolution function of the SD-750 array The events were weighted based on the primary mass accord- ing to the Global Spline Fit (GSF) model [41] to account for the changing mass-evolution near the second knee and ankle. The reconstructed values of S(450) were corrected by applying the energy-dependent CIC method to obtain val- ues for S35 and these values were then calibrated against the Monte-Carlo energies. During the calibration, a further weighting was performed based on the energy distribution of golden hybrid events to account for the hybrid detection efﬁciency. Following the calibration procedure, each MC event was assigned an energy in the FD energy scale (i.e. EMC →S35 →EFD). Fig. 8 An example of the ratio of the energy assignments for the SD and FD is shown with black crosses for the energy bin indicated in the plot. The best ﬁt ratio distribution for this bin is shown by the black line Fig. 9 The total SD energy resolution, as calculated using the golden hybrid events (red circles) is shown in bins with equal statistics. The parameterization of the resolution is shown by the solid blue line and the corresponding 68% conﬁdence interval in dashed lines. The energy res- olution, calculated using mass-weighted MC air showers (gray squares), is shown as a veriﬁcation of the method The SD energy resolution was calculated using the mass- weighted simulations and is shown in gray squares in Fig. 9. Indeed, the simulated and measured SD resolutions show a similar trend and agree to within the uncertainties, supporting the golden hybrid method. Fig. 9 The total SD energy resolution, as calculated using the golden hybrid events (red circles) is shown in bins with equal statistics. The parameterization of the resolution is shown by the solid blue line and the corresponding 68% conﬁdence interval in dashed lines. The energy res- olution, calculated using mass-weighted MC air showers (gray squares), is shown as a veriﬁcation of the method In the energy region at-and-below 1017 eV, systematic effects also enter into play on the energy estimate. An energy- dependent offset, a bias, is thus expected in the resolution function for several reasons: For two independent, Gaussian-distributed random vari- ables, X and Y, their ratio, z = X/Y, produces a ratio distribution that depends on the means (μX, μY ) and standard deviations (σX, σY ) of the two variables, PDF(z; μX, μY , σX, σY ). 3.4 Resolution function of the SD-750 array The SD resolution as a function of energy is needed in sev- eral steps of the analysis. In the regime of full efﬁciency, it can be considered as a Gaussian function centered on the true energy, the width of which reﬂects the statistical uncertainty associated with the detection and reconstruction processes on one hand, and the stochastic development of the parti- cle cascade on the other hand. The combination of the two can be estimated for the golden hybrid events, thus allowing us to account for the contribution of the shower-to-shower ﬂuctuations in a data-driven way. Each event observed by the SD and FD results in two independent measurements of the air-shower energy, ESD and EFD, respectively. Unlike for the SD, the FD directly provides a view of the shower development so a total energy resolution, σFD(E), can be estimated for each of the golden hybrid events. Using the known σFD(E), the resolution of SD can be determined by studying the distribution of the ratio of the two energy measurements. 12 3 966 Page 10 of 25 Eur. Phys. J. C (2021) 81 :966 Fig. 8 An example of the ratio of the energy assignments for the SD and FD is shown with black crosses for the energy bin indicated in the plot. The best ﬁt ratio distribution for this bin is shown by the black line An example of the measured energy-ratio distributions is shown in Fig. 8 with the ﬁtted curve overlaid on the data points. Carrying out the ﬁt in different energy bins, the SD resolution, shown by the red points in Fig. 9, is represented by, σSD(E) E = (0.06 ± 0.02) + (0.05 ± 0.01) 1 EeV E . (8) (8) The corresponding curve is overlaid in blue, bracketed by the 68% conﬁdence region. 17 To measure the spectrum above the 1017 eV threshold, the knowledge of the resolution function, which induces bin-to- bin migration of events, and of the detection efﬁciency are also required for energies below this threshold. As a ver- iﬁcation, particularly in the energy region where Eq. (8) is extrapolated, a Monte-Carlo analysis was performed. A set of 325,000 CORSIKA [39] air showers were used, con- sisting of proton, helium, oxygen, and iron primaries with energies above 1016 eV. EPOS-LHC [40] was used as the hadronicinteractionmodel.Theairshowerswererunthrough the full SD simulation and reconstruction algorithms. 3.4 Resolution function of the SD-750 array The ratio of the assigned and expected values as a function of energy are shown (red circles) along with the parameterization (blue line) given in Eq. (9) Fig. 10 The bias of the energy assignment for the SD-750 was studied using Monte Carlo simulations, weighted according to the GSF model [41]. The ratio of the assigned and expected values as a function of energy are shown (red circles) along with the parameterization (blue line) given in Eq. (9) Fig. 11 Residuals of the SD-750 raw spectrum with respect to the power-law function J ref(E). Data points from the SD-1500 spectrum measurement are superimposed Table 5 Best-ﬁt parameters for the relative energy bias of the SD-750, bSD(E), given in Eq. (9) Parameter Value Uncertainty b0 −3 1 b1 26 4 b2 0.35 0.02 b3 12.7 0.1 b4 −0.0039 0.0008 venting a fair sampling of S35 values over the underlying mass distribution. Table 5 Best-ﬁt parameters for the relative energy bias of the SD-750, bSD(E), given in Eq. (9) 3.4 Resolution function of the SD-750 array Likewise, the ratio of the two energy measurements, z = ESD/EFD, follows such a dis- tribution to ﬁrst order. Because the FD sets the energy scale of the Observatory, there is inherently no bias in the energy measurements with respect to its own scale and thus, on aver- age, μFD(E) = 1. Using the golden hybrid data set, the ratio distribution was ﬁt in an unbinned likelihood analysis, PDF(z; μSD(E), 1, σSD(E), σFD(E)). 1. The application of the trigger below threshold, combined with the ﬁnite energy resolution, cause an overestimate of the shower size, on average, which is then propagated to the energy assignment. 2. The linear relationship assumed in Eq. (7) cannot account for a possible sudden change in the evolution of the mass- composition with energy. Such a change would require a broken power law for the energy calibration relationship. 3. In the energy range where the SD is not fully efﬁcient, the SD efﬁciency is larger for light primary nuclei, thus pre- 123 Eur. Phys. J. C (2021) 81 :966 Page 11 of 25 966 5 966 Fig. 10 The bias of the energy assignment for the SD-750 was studied using Monte Carlo simulations, weighted according to the GSF model [41]. The ratio of the assigned and expected values as a function of energy are shown (red circles) along with the parameterization (blue Fig. 11 Residuals of the SD-750 raw spectrum with respect to the power-law function J ref(E). Data points from the SD-1500 spectrum measurement are superimposed Fig. 10 The bias of the energy assignment for the SD-750 was studied using Monte Carlo simulations, weighted according to the GSF model [41]. The ratio of the assigned and expected values as a function of energy are shown (red circles) along with the parameterization (blue line) given in Eq. (9) Fig. 11 Residuals of the SD-750 raw spectrum with respect to the power-law function J ref(E). Data points from the SD-1500 spectrum measurement are superimposed Fig. 11 Residuals of the SD-750 raw spectrum with respect to the power-law function J ref(E). Data points from the SD-1500 spectrum measurement are superimposed Fig. 10 The bias of the energy assignment for the SD-750 was studied using Monte Carlo simulations, weighted according to the GSF model [41]. 4 Measurement of the energy spectrum To build the energy spectrum from the reconstructed energy distribution, we need to correct the raw spectrum, obtained as J raw i = Ni/(EEi), for the bin-to-bin migrations of events due to the ﬁnite accuracy with which the energies are assigned. The energy bins are chosen to be regularly sized in decimal logarithm, lg Ei = 0.1, commensurate with the energy resolution. The level of migration is driven by the resolution function, the detection efﬁciency in the energy range just below the threshold energy, and the steepness of the spectrum. To correct for these effects, we use the bin- by-bin correction approach presented in [21]. It consists of folding the detector effects into a proposed spectrum func- tion, J(E, k), with free parameters, k, such that the result describes the set of the observed number of events Ni. The set of expectations, νi, is obtained as νi(k) = j Ri jμ j(k), where the Ri j coefﬁcients (reported in a matrix format in the Supplementary material) describe the bin-to-bin migra- tions, and where μ j are the expectations in the case of an ideal detector obtained by integrating the proposed spectrum over E j and E j + E j scaled by E. The optimal set of free parameters, ˆk, is inferred by minimizing a log-likelihood function built from the Poisson probabilities to observe Ni events when νi(ˆk) are expected. venting a fair sampling of S35 values over the underlying mass distribution. This is expected from a com- bination of primarily the resolution effects to be unfolded and of a possible mismatch, within the energy-dependent bud- get of uncorrelated uncertainties, of the SD-1500 and SD- 750 ESD energy scales. Below ≃1017.4 eV, a slight roll-off begins. Overall, these residuals are suggestive of a power-law function to describe the data leading up to the ankle energy where the spectrum hardens, with a gradually changing spec- tral index over the lowest energies studied. Consequently, the proposed function is chosen as three power laws with transi- tions occurring over adjustable energy ranges Table 6 Best-ﬁt values of the spectral parameters (Eq. (11)). The parameter ω12 is ﬁxed to the value constrained in [21]. Note that the parameters γ0 and E01 correspond to features below the measured energy region and are treated only as aspects of the unfolding ﬁxed to their best-ﬁt values to infer the uncertainties of the measured spectral parameters Parameter Value ±σstat ± σsyst J0/(km2 yrsreV) (1.09 ± 0.04 ± 0.28) ×10−13 ω01 0.49 ± 0.07 ± 0.34 γ1 3.34 ± 0.02 ± 0.09 E12/eV (3.9 ± 0.8 ± 1.1) ×1018 γ2 2.6 ± 0.2 ± 0.1 γ0 2.64 – ﬁxed E01/eV 1.24×1017 – ﬁxed ω12 0.05 – ﬁxed Fig. 13 Unfolded energy spectrum of the SD-750, scaled by E2.6 of the response of the detector and shown in Fig. 12. It is obtained as Ji = μi νi J raw i = ci J raw , (12) where the μi and νi coefﬁcients are estimated using the best- ﬁt parameters ˆk Their ratios deﬁne the bin by bin correc Fig. 12 Unfoldedenergyspectrumderivedusingdata fromthe SD-750 array Fi 12 U f ld d t d i d i d t f th SD 750 Table 6 Best-ﬁt values of the spectral parameters (Eq. (11)). The parameter ω12 is ﬁxed to the value constrained in [21]. Note that the parameters γ0 and E01 correspond to features below the measured energy region and are treated only as aspects of the unfolding ﬁxed to their best-ﬁt values to infer the uncertainties of the measured spectral parameters Parameter Value ±σstat ± σsyst J0/(km2 yrsreV) (1.09 ± 0.04 ± 0.28) ×10−13 ω01 0.49 ± 0.07 ± 0.34 γ1 3.34 ± 0.02 ± 0.09 E12/eV (3.9 ± 0.8 ± 1.1) ×1018 γ2 2.6 ± 0.2 ± 0.1 γ0 2.64 – ﬁxed E01/eV 1.24×1017 – ﬁxed ω12 0.05 – ﬁxed Fig. venting a fair sampling of S35 values over the underlying mass distribution. 12 Unfoldedenergyspectrumderivedusingdata fromthe SD-750 array Fig. 13 Unfolded energy spectrum of the SD-750, scaled by E2.6 The reference function in this energy range, as reported in [21], is J ref(E) = J ref 0 E 1018.5 eV −γ ref 1 , (10) (10) with J ref 0 = 1.315×10−18 km−2 yr−1 sr−1 eV−1 and γ ref 1 = 3.29. The residuals of the SD-1500 unfolded spectrum with respect to J ref(E) are also shown as open squares in Fig. 11. The sharp transition at ≃1018.7 eV to a different power law corresponds to the spectral feature known as the ankle. Such a transition is also observed, with much lower sensitivity, using data from the SD-750 array. Below ≃1018.7 eV and down to ≃1017.4 eV, one can see a shift of the raw SD-750 spectrum compared to J ref(E). This is expected from a com- bination of primarily the resolution effects to be unfolded and of a possible mismatch, within the energy-dependent bud- get of uncorrelated uncertainties, of the SD-1500 and SD- 750 ESD energy scales. Below ≃1017.4 eV, a slight roll-off begins. Overall, these residuals are suggestive of a power-law function to describe the data leading up to the ankle energy where the spectrum hardens, with a gradually changing spec- tral index over the lowest energies studied. Consequently, the proposed function is chosen as three power laws with transi- tions occurring over adjustable energy ranges, Fig. 13 Unfolded energy spectrum of the SD-750, scaled by E2.6 of the response of the detector and shown in Fig. 12. It is obtained as Ji = μi νi J raw i = ci J raw , (12) (12) where the μi and νi coefﬁcients are estimated using the best- ﬁt parameters ˆk. Their ratios deﬁne the bin-by-bin correc- tions used to produce the unfolded spectrum. The correc- tion applied extends from 0.84 at 1017 eV to 0.99 around the ankle (see Appendix B). The best-ﬁt spectral parameters are reported in Table 6, while the statistical correlations between the parameters are detailed in Appendix B (Table 9). The goodness-of-ﬁt of the forward-folding procedure is attested by the deviance of 15.9, which, if considered to follow the C statistics [42], can be compared4 to the expectation of 16.2 ± 5.6 to yield a p-value of 0.50. where the μi and νi coefﬁcients are estimated using the best- ﬁt parameters ˆk. 4 Note that the p-value for a proposed function which does not include a transition from γ0 to γ1 can be rejected with more than 20σ conﬁdence. venting a fair sampling of S35 values over the underlying mass distribution. Because there is an insufﬁcient number of FD events which pass the ﬁducial cuts below 1017 eV, the bias was char- acterized, using the same air-shower simulations as used for the resolution cross-check. The remaining relative energy bias is shown in Fig. 10. The ratio between the reconstructed and expected values are shown as the red points as a function of EFD. A larger bias of ≃20% is seen at low energies, where upward ﬂuctuations are necessarily selected by the triggering conditions. In the range considered for the energy spectrum, E > 1017 eV, the bias is 3% or less. To complete the description of the SD resolution function, the relative bias was ﬁt to an empirical function, To choose the proposed function, we plot in Fig. 11 the residuals (red dots) of the SD-750 raw spectrum with respect to a reference function, J ref(E), that ﬁts the SD-1500 spec- trum below the ankle energy down to the SD-1500 thresh- old energy, 1018.4 eV. A re-binning was applied at and above 1019 eV to avoid too large statistical ﬂuctuations. bSD(E) = b0(lg E eV −b1) exp −b2(lg E eV −b3)2 + b4. (9) (9) The corresponding best ﬁt parameters (blue line in Fig. 10) are given in Table 5. The corresponding best ﬁt parameters (blue line in Fig. 10) are given in Table 5. 12 3 966 Page 12 of 25 Eur. Phys. J. C (2021) 81 :966 Fig. 12 Unfoldedenergyspectrumderivedusingdata fromthe SD-750 array The reference function in this energy range, as reported in [21], is J ref(E) = J ref 0 E 1018.5 eV −γ ref 1 , (10) with J ref 0 = 1.315×10−18 km−2 yr−1 sr−1 eV−1 and γ ref 1 = 3.29. The residuals of the SD-1500 unfolded spectrum with respect to J ref(E) are also shown as open squares in Fig. 11. The sharp transition at ≃1018.7 eV to a different power law corresponds to the spectral feature known as the ankle. Such a transition is also observed, with much lower sensitivity, using data from the SD-750 array. Below ≃1018.7 eV and down to ≃1017.4 eV, one can see a shift of the raw SD-750 spectrum compared to J ref(E). venting a fair sampling of S35 values over the underlying mass distribution. Their ratios deﬁne the bin-by-bin correc- tions used to produce the unfolded spectrum. The correc- tion applied extends from 0.84 at 1017 eV to 0.99 around the ankle (see Appendix B). The best-ﬁt spectral parameters are reported in Table 6, while the statistical correlations between the parameters are detailed in Appendix B (Table 9). The goodness-of-ﬁt of the forward-folding procedure is attested by the deviance of 15.9, which, if considered to follow the C statistics [42], can be compared4 to the expectation of 16.2 ± 5.6 to yield a p-value of 0.50. J(E, k)=J0 E 1017 eV −γ0 1 i=0 1+ E Ei j 1 ωi j (γi−γ j)ωi j , (11) (11) with j = i + 1. The normalization factor J0, the three spectral indices γi, and the transition parameter ω01 consti- tute the free parameters in k. The transition parameter ω12, constrained with much more sensitivity using data from the SD-1500, is ﬁxed at ω12 = 0.05 [21]. with j = i + 1. The normalization factor J0, the three spectral indices γi, and the transition parameter ω01 consti- tute the free parameters in k. The transition parameter ω12, constrained with much more sensitivity using data from the SD-1500, is ﬁxed at ω12 = 0.05 [21]. Combining all the ingredients at our disposal, we obtain the ﬁnal estimate of the spectrum, Ji, unfolded for the effects 4 Note that the p-value for a proposed function which does not include a transition from γ0 to γ1 can be rejected with more than 20σ conﬁdence. 123 Eur. Phys. J. C (2021) 81 :966 Page 13 of 25 966 Page 13 of 25 966 966 Fig. 14 Evolution of the spectral index with energy. The measured spectral points were ﬁt to power laws within a sliding window of lg E = 0.3. The values of γ1 and γ2 are represented by the dashed and dash-dotted lines, for reference The ﬁtting function is shown in Fig. 13, superimposed to the spectrum scaled by E2.6, allowing one to better appreci- ate its characteristics, from the turn-over at around 1017 eV up to a few 1019 eV, thus including the ankle. The turn-over is observed with a very large exposure, unprecedented at such energies. venting a fair sampling of S35 values over the underlying mass distribution. However, as indicated by the magnitude of the transition parameter, ω01 ≃0.49, the change of the spec- tral index occurs over an extended lg E ≃0.5 energy range, so that the spectral index γ0 cannot be observed but only indirectly inferred. Also, the value of the energy break, E01 ≃1.24×1017 eV, turns out to be close to the threshold energy. These two facts thus imply that, while a spectral break is found beyond any doubt, it cannot wholly be characterised, as only the higher energy portion is actually observed. Con- sequently, the ﬁt values describing E01 and γ0 are not to be considered as true measurements but as necessary parame- ters in the ﬁt function, the statistical resolutions of which are on the order of 35%. Once we infer their best-ﬁt values, we use these values as “external parameters” to estimate the uncertainties of the other spectral parameters. This proce- dure gives rise to an increase of the systematic uncertainties, but is necessary as E01 and γ0 are not directly observed. Beyond the smooth turn-over around E01, the intensity can be described by a power-law shape as J(E) ∝E−γ1, up to E12 = (3.9 ± 0.8) ×1018 eV, the ankle energy, the value of which is within 1.4σ of that found with the much larger expo- sure of the SD-1500 measurement of the spectrum, namely (5.0 ± 0.1)×1018 eV. Also the value of γ1 = 3.34 ± 0.02 is within 1.8σ of that obtained with the SD-1500 between 1018.4 and 1018.7 eV (3.29 ± 0.02). Fig. 14 Evolution of the spectral index with energy. The measured spectral points were ﬁt to power laws within a sliding window of lg E = 0.3. The values of γ1 and γ2 are represented by the dashed and dash-dotted lines, for reference Fig. 15 Systematic uncertainties in the ﬂux measurement as a function of energy. The main contributions are shown separately The characteristics of the measured spectrum can also be studied by looking at the evolution of the spectral index as a function of energy, γ (E). Rather than relying on the empiri- cally chosen unfolding function, this slope parameter can be directlyﬁtusingthevaluescalculatedin J(E).Power-lawﬁts were performed for a sliding window of width lg E = 0.3. The resulting estimations of the so obtained spectral indexes are shown in Fig. 14. Fig. 5 The combined SD-750 and SD-1500 energy spectrum The spectrum obtained in Sect. 4 extends down to 1017 eV and at the high-energy end overlaps with the one recently reported in [21] using the SD-1500 array. The two spectra are superimposed in Fig. 16. Beyond the overall consistency observed between the two measurements, a combination of them is desirable to gather the information in a single energy spectrum above 1017 eV obtained with data from both the SD-750 and the SD-1500 of the Pierre Auger Observatory. We present below such a combination considering adjustable re-scaling factors in exposures, δE, and ESD energy scales, δESD, within uncorrelated uncertainties. Fig. 17 SD energy spectrum after combining the individual measure- ments by the SD-750 and the SD-1500 scaled by E2.6. The ﬁt using the proposed function (Eq. (13)) is overlaid in red along with the one sigma error band in gray The combination is carried out using the same bin-by- bin correction approach as in Sect. 4. The joint likelihood function, L(s, δE, δESD), is built from the product of the individual Poissonian likelihoods pertaining to the two SD measurements, L750 and L1500. These two individual likeli- hoods share the same proposed function, arrays. For each array, a change of the associated exposure E →E + δE impacts the νi coefﬁcients accordingly, while a change in energy scale ESD →ESD +δESD impacts as well the observed number of events in each bin. Additional likeli- hood factors, LδE and LδESD, are thus required to control the changes of the exposure and of the energy-scale within their uncorrelated uncertainties. The likelihood factors described below account for δE and δESD changes associated with the SD-750 only. We have checked that allowing additional free parameters, such as the δE corresponding to the SD-1500, does not improve the deviance of the best ﬁt by more than one unit, and thus their introduction is not supported by the data. J(E, s) = J0 E E0 −γ0 3 i=0 1 + E Ei j 1 ωi j (γi−γ j)ωi j 3 i=0 1 + E0 Ei j 1 ωi j (γi−γ j)ωi j , (13) (13) with j = i + 1 and E0 = 1018.5 eV. As in [21], the transition parameters ω12, ω23 and ω34 are ﬁxed to 0.05. venting a fair sampling of S35 values over the underlying mass distribution. 15 Systematic uncertainties in the ﬂux measurement as a function of energy. The main contributions are shown separately energies of the individual events were shifted by ±14% and the unfolding procedure was repeated. The result is shown as dashed red lines in Fig. 15. The values of the spectral index ﬁts present a consistent picture of the evolution. Beginning at the lowest energies shown, γ (E) increases ﬁrst quite rapidly, ﬁnally approach- ing a value of 3.3 leading up to the ankle asymptotically. Unsurprisingly, this is the value found for γ1 in the unfold- ing of both the SD-750 and SD-1500 spectra [21]. Beyond that of the energy scale, the additional uncer- tainties are subdominant but are important to understand as they have energy dependence and some are uncorrelated with other ﬂux measurements made at the Observatory. Such knowledge is particularly important for the combination of the two SD spectra presented later in Sect. 5. The most rel- evant of these energy-dependent uncertainties is associated with the procedure of the forward-folding itself. The uncer- tainties in the resolution function and in the detection efﬁ- ciency all contribute a component to the overall unfolding uncertainty. The forward-folding process was hence repeated by shifting, within the statistical uncertainties, the parame- terizations of the energy resolution (Eq. (8)) and efﬁciency The systematic uncertainties that affect the measurement of the spectrum are dominated by the overall uncertainty of the energy scale, detailed in [43], and is, itself, dominated by the absolute calibration of the ﬂuorescence telescopes (10%). The total uncertainty in the energy scale is σE/E = 14%. Once propagated, the steepness of the spectrum as a function ofenergyampliﬁesthisuncertainty,roughlyasσJ /J = (γ1− 1)σE/E, resulting in a total ﬂux uncertainty of σJ/J ≃35%. However, for a more exact calculation of the uncertainty, the 12 3 3 966 Page 14 of 25 Eur. Phys. J. C (2021) 81 :966 Fig. 16 Superimposed SD spectra to be combined scaled by E2.6, the SD-750 (red circles) and the SD-1500 (black squares) Fig. 17 SD energy spectrum after combining the individual measure- ments by the SD-750 and the SD-1500 scaled by E2.6. The ﬁt using the proposed function (Eq. (13)) is overlaid in red along with the one sigma error band in gray Fig. venting a fair sampling of S35 values over the underlying mass distribution. 16 Superimposed SD spectra to be combined scaled by E2.6, the SD-750 (red circles) and the SD-1500 (black squares) parameterization, and by bracketing the bias with the pure proton/iron mass primaries below full efﬁciency. The impact of the resolution uncertainties on the unfolding procedure is the larger, in particular at the highest energies. On the other hand, the energy bias and reduced efﬁciency below 1017 eV only impacts the ﬁrst few bins. These various components are summed in quadrature and are shown by the dotted blue line in Fig. 15. These inﬂuences are clearly seen to impact the spectrum by <4%. The last signiﬁcant uncertainty in the ﬂux is related to the calculation of the geometric exposure of the array. This quantity has been previously studied and is 4% for the SD- 750 which directly translates to a 4% energy-independent shift in the ﬂux [24]. The resulting systematic uncertainties of the spectral parameters are given in Table 6. For completeness, beyond the summary information provided by the spectrum param- eterization, the correlation matrix of the energy spectrum is given in the Supplementary material. It is obtained by repeat- ing the analysis on a large number of data sets, sampling randomly the systematic uncertainties listed above. Fig. 16 Superimposed SD spectra to be combined scaled by E2.6, the SD-750 (red circles) and the SD-1500 (black squares) Fig. 17 SD energy spectrum after combining the individual measure- ments by the SD-750 and the SD-1500 scaled by E2.6. The ﬁt using the proposed function (Eq. (13)) is overlaid in red along with the one sigma error band in gray 5 The combined SD-750 and SD-1500 energy spectrum In this way, the same parameters s are used during the minimisation process to calculate the set of expectations νi(s, δE, δESD) of the two Both likelihood factors are described by Gaussian distri- butions with a spread given by the uncertainty pertaining to the exposure and to the energy-scale. The joint likelihood 123 Page 15 of 25 966 Eur. Phys. J. C (2021) 81 :966 966 Fig. 18 SD-750spectrum(solidredcircles)nearthesecondkneealong with the measurements from Akeno [44], GAMMA [45], IceTop [9], KASCADE-Grande [46], TALE [10], Tien Shan [47], Tibet-III [48], Tunka-133 [11], Yakutsk [49]. The experiments that set their energy scale using calorimetric observations are indicated by solid colored markers while those with an energy scale based entirely on simulations are shown by gray markers Fig. 18 SD-750spectrum(solidredcircles)nearthesecondkneealong with the measurements from Akeno [44], GAMMA [45], IceTop [9], KASCADE-Grande [46], TALE [10], Tien Shan [47], Tibet-III [48], Tunka-133 [11], Yakutsk [49]. The experiments that set their energy scale using calorimetric observations are indicated by solid colored markers while those with an energy scale based entirely on simulations are shown by gray markers Table 7 Best-ﬁt values of the combined spectral parameters (Eq. (13)). The parameter ω12, ω23 and ω34 are ﬁxed to the value constrained in [21]. Note that the parameters γ0 and E01 correspond to features below the measured energy region and should be treated only as aspects of the combination Parameter Value ±σstat ± σsyst J0 /(km2 yrsreV) (1.309 ± 0.003 ± 0.400) ×10−18 ω01 0.43 ± 0.04 ± 0.34 γ1 3.298 ± 0.005 ± 0.10 E12/eV (4.9 ± 0.1 ± 0.8) ×1018 γ2 2.52 ± 0.03 ± 0.05 E23/eV (1.4 ± 0.1 ± 0.2) ×1019 γ3 3.08 ± 0.05 ± 0.10 E34/eV (4.7 ± 0.3 ± 0.6) ×1019 γ4 5.2 ± 0.2 ± 0.1 γ0 2.64 – ﬁxed E01/eV 1.24×1017 – ﬁxed ω12 0.05 – ﬁxed ω23 0.05 – ﬁxed ω34 0.05 – ﬁxed Fig. 18 SD-750spectrum(solidredcircles)nearthesecondkneealong with the measurements from Akeno [44], GAMMA [45], IceTop [9], KASCADE-Grande [46], TALE [10], Tien Shan [47], Tibet-III [48], Tunka-133 [11], Yakutsk [49]. 5 The combined SD-750 and SD-1500 energy spectrum The experiments that set their energy scale using calorimetric observations are indicated by solid colored markers while those with an energy scale based entirely on simulations are shown by gray markers The outcome of the forward-folding ﬁt is the set of param- eters s, δE, δA and δB that allow us to calculate the expec- tation values μi and νi, and thus the correction factors ci, for both arrays separately. The resulting combined spectrum, obtained as function reads then as L(s, δE, δESD) = L750 × L1500 × LδE × LδESD. (14) (14) The allowed change of exposure, δE, is guided by the sys- tematic uncertainties in the SD-750 exposure, σE/E = 4%. Hence, the constraining term for any change in the SD-750 exposure reads, dropping constant terms, as The allowed change of exposure, δE, is guided by the sys- tematic uncertainties in the SD-750 exposure, σE/E = 4%. Hence, the constraining term for any change in the SD-750 exposure reads, dropping constant terms, as J comb i = ci,750 Ni,750 + ci,1500 Ni,1500 Eeff i Ei , (17) J comb i = ci,750 Ni,750 + ci,1500 Ni,1500 Eeff i Ei , (17) (17) −2 ln LδE(δE) = δE σE 2 . (15) (15) is shown in Fig. 17. Here, the observed number of events N 750 i in each bin is calculated at the re-scaled energies, while the effective exposure, Eeff i , is the shifted one of the SD-750 in the energy range where Ni,1500 = 0, the one of the SD- 1500 in the energy range where Ni,750 = 0, and the sum E750 + δE + E1500 in the overlapping energy range. The set of spectral parameters are collected in Table 7, while the corresponding correlation matrix is reported in Appendix B (Table 11) for δE, δA and δB ﬁxed to their best-ﬁt values. The change in exposure is δE/E = +1.4%, while the one in energy scale follows from δA/A = −2.5% and δB/B = +0.8%. The goodness-of-ﬁt is evidenced by a deviance of 37.2 for an expected value of 32 ± 8. We also note that the parameters describing the spectral shape are in agreement with those of the two individual spectra from the SD arrays. is shown in Fig. 17. 6 Discussion previously reported measurement using the SD-1500, a uni- ﬁed SD spectrum was calculated by combining the respec- tive observed ﬂuxes, energy resolutions, and exposures. The result has partial coverage of the second knee and full cover- age of the ankle, an additional inﬂection at ≃1.4×1019 eV, and the suppression. This procedure is applied to spectra inferred from a single detector type (i.e. water-Cherenkov detectors), but can be used for the combination of any spec- tral measurements for which the uncorrelated uncertainties can be estimated. We have presented here a measurement of the CR spectrum in the energy range between the second knee and the ankle, which is covered with high statistics by the SD-750, includ- ing 560,000 events with zenith angles up to 40◦and energies above 1017 eV. The measurement includes a total exposure of 105km2 sryr and an energy scale set by calorimetric obser- vations from the FD telescopes. We note a signiﬁcant change in the spectral index and with a width that is much broader than that of the ankle feature. The impressive regularity of the all-particle spectrum observed in the energy region between the second knee and the ankle can hide an underlying intertwining of different astrophysical phenomena, which might be exposed by look- ing at the spectrum of different primary elements. In the future, further measurements will allow separation of the intensities due to the different components. On the one hand, Xmax values will be determined down to 1017 eV using the three HEAT telescopes. On the other hand, the determina- tion of the muon component of EAS above 1017 eV will be possible using the new array of underground muon detectors [35], co-located with the SD-750. This will help us in study- ing whether the origin of the second knee stems from, for instance, the steep fall-off of an iron component, as expected for Galactic CRs characterized by a rigidity-dependent maxi- mum acceleration energy for particles with charge Z, namely Emax(Z) ≃Z Eproton max . In addition, we will be able to extend the measurement of the energy spectrum below 1017 eV with a denser array of 433m-spaced detectors and with the anal- ysis of the Cherenkov light in FD events [55]. The extension will allow us to lower the threshold and to further explore the second-knee region in more detail. 5 The combined SD-750 and SD-1500 energy spectrum Here, the observed number of events N 750 i in each bin is calculated at the re-scaled energies, while the effective exposure, Eeff i , is the shifted one of the SD-750 in the energy range where Ni,1500 = 0, the one of the SD- 1500 in the energy range where Ni,750 = 0, and the sum E750 + δE + E1500 in the overlapping energy range. The set of spectral parameters are collected in Table 7, while the corresponding correlation matrix is reported in Appendix B (Table 11) for δE, δA and δB ﬁxed to their best-ﬁt values. The change in exposure is δE/E = +1.4%, while the one in energy scale follows from δA/A = −2.5% and δB/B = +0.8%. The goodness-of-ﬁt is evidenced by a deviance of 37.2 for an expected value of 32 ± 8. We also note that the parameters describing the spectral shape are in agreement with those of the two individual spectra from the SD arrays. Likewise, uncertainties in A and B, δA and δB, translate into uncertainties in the SD-750 energy scale. Statistical contri- butions stem from the energy calibration of S35, which are by essence uncorrelated to those of the SD-1500. Other uncor- related contributions of the systematic uncertainties from the FD energy scales propagated to the SD-1500 and SD-750 could enter into play. The magnitude of such systematics, σsyst, is difﬁcult to quantify. By testing several values for σsyst, we have checked, however, that such contributions have a negligible impact on the combined spectrum. Hence, the constraining term for any change in energy scale can be con- sidered to stem from statistical uncertainties only and reads as −2 ln LESD(δA, δB) = [σ −1]AA(δA)2 + [σ −1]BB(δB)2 + 2[σ −1]AB(δA)(δB), (16) The impact of the systematic uncertainties, dominated by those in the energy scale, on the spectral parameters are reported in Table 7. For completeness, beyond the summary information provided by the spectrum parameterization, the correlation matrix of the energy spectrum itself is also given in the Supplementary material. (16) where the notation [σ]i j stands for the coefﬁcients of the variance-covariance matrix of the A and B best-ﬁt estimates and [σ −1] is the inverse of this matrix. 12 3 3 3 966 Page 16 of 25 Eur. Phys. J. C (2021) 81 :966 6 Discussion The consistency of all these observations strongly supports a sce- nario of Galactic CRs characterised by a rigidity-dependent maximum acceleration energy for particles with charge Z, namely Emax(Z) ≃Z Eproton max , to explain the knee structures. The measurements of the all-particle ﬂux from various experiments [9–11,44–49] in the energy region surrounding the second knee are shown in Fig. 18. Experiments which set their energy scale using calorimetric measurements are plotted using colored markers (Auger SD-750, TA TALE, TUNKA-133, Yakutsk) while the measurements shown in gray markers represent MC-based energy assignments. The spread between various experiments is statistically signiﬁ- cant. However, all these measurements are consistent with the SD-750 spectrum within the 14% energy scale system- atic uncertainty. Understanding the nature of the off-sets in the energy scales is beyond the scope of this paper. However, we note that the TALE spectrum agrees rather well with the SD-750 spectrum, offset by 5 to 6% in energy. The agree- ment is notable given that at-and-above the ankle, an energy scale off-set of around 11% is required to bring the spectral measurements with SD-1500 of the Auger Observatory and the SD of the Telescope Array into agreement [54]. Acknowledgements The successful installation, commissioning, and operation of the Pierre Auger Observatory would not have been pos- sible without the strong commitment and effort from the technical and administrative staff in Malargüe. We are very grateful to the fol- lowing agencies and organizations for ﬁnancial support: Argentina – Comisión Nacional de Energía Atómica; Agencia Nacional de Promo- ción Cientíﬁca y Tecnológica (ANPCyT); Consejo Nacional de Investi- gaciones Cientíﬁcas y Técnicas (CONICET); Gobierno de la Provincia de Mendoza; Municipalidad de Malargüe; NDM Holdings and Valle Las Leñas; in gratitude for their continuing cooperation over land access; Australia – the Australian Research Council; Belgium – Fonds de la Recherche Scientiﬁque (FNRS); Research Foundation Flanders (FWO); Brazil – Conselho Nacional de Desenvolvimento Cientíﬁco e Tecnológico(CNPq);FinanciadoradeEstudoseProjetos(FINEP);Fun- dação de Amparo à Pesquisa do Estado de Rio de Janeiro (FAPERJ); São Paulo Research Foundation (FAPESP) Grants No. 2019/10151- 2, No. 2010/07359-6 and No. 1999/05404-3; Ministério da Ciên- cia, Tecnologia, Inovações e Comunicações (MCTIC); Czech Repub- lic – Grant No. 6 Discussion Such a change has been observed by a number of other experiments, and via various detection methods. Most notably, the nature of this feature was linked to a soften- ing of the heavy-mass primaries beginning at 1016.9 eV by the KASCADE-Grande experiment, leading to the moniker iron knee [8]. Additional analyses by the Tunka-133 [50] and IceCube [9] collaborations have given further evidence that high-mass particles are dominant near 1017 eV and thus that it is their decline that largely deﬁnes the shape of the all-particle spectrum. The hypothesis is also supported by a preliminary study of the distributions of the depths of the shower maximum, Xmax, measured at the Auger Observa- tory [36,51]. These have been parametrized according to the hadronic models EPOS-LHC [40], QGSJetII-04 [52] and Sibyll2.3 [53]. From these parametrizations, the evolution over energy of the fractions of different mass groups, from protons to Fe-nuclei, has been derived. From all three models, a fall-off of the Fe component above 1017 eV is inferred. The consistency of all these observations strongly supports a sce- nario of Galactic CRs characterised by a rigidity-dependent maximum acceleration energy for particles with charge Z, namely Emax(Z) ≃Z Eproton max , to explain the knee structures. Such a change has been observed by a number of other experiments, and via various detection methods. Most notably, the nature of this feature was linked to a soften- ing of the heavy-mass primaries beginning at 1016.9 eV by the KASCADE-Grande experiment, leading to the moniker iron knee [8]. Additional analyses by the Tunka-133 [50] and IceCube [9] collaborations have given further evidence that high-mass particles are dominant near 1017 eV and thus that it is their decline that largely deﬁnes the shape of the all-particle spectrum. The hypothesis is also supported by a preliminary study of the distributions of the depths of the shower maximum, Xmax, measured at the Auger Observa- tory [36,51]. These have been parametrized according to the hadronic models EPOS-LHC [40], QGSJetII-04 [52] and Sibyll2.3 [53]. From these parametrizations, the evolution over energy of the fractions of different mass groups, from protons to Fe-nuclei, has been derived. From all three models, a fall-off of the Fe component above 1017 eV is inferred. 123 Appendix A: The electromagnetic trigger algorithms The two additional triggers build upon the ToT trigger in two ways, applying more sophisticated analyses to the signal waveform. They are aimed at further suppressing the muon background so as to enhance the sensitivity to pure electromagnetic signals, which are generally smaller. The ToTd trigger uses the typical decay time of Cherenkov light inside the water volume, τ = 67ns, to deconvolve the exponential tail of the pulses before applying the ToT condi- tion. This has the effect of reducing the inﬂuence of muons in the trigger, since the typical signal from a muon, with fast rise time and ≈60ns decay constant, is compressed into one or two time bins. The exponential tail of the signal is decon- volved using Di = Si −Si−1e−t/τ 1 −e−t/τ , (A.1) (A.1) where Si is the signal in the i-th time-bin and t = 25ns is the ADC bin-width. For an exponential decay with the mean decay time, the deconvolved values, Di, would be zero. How- ever for an exponential decay with statistical noise that is pro- portional to √Si, the set {Di} would exponentially decrease with an increased decay length τ ′ = 2τ. After performing the deconvolutioninEq.(A.1),thetriggerissatisﬁedif≥13ADC bins (≥325ns) are above 0.2 IVEM, in coincidence between two of the three PMTs, within a sliding 3µs (120bin) time window. An example of a waveform which passes the ToTd trigger and its deconvolution are shown in the top two plots of Fig. 19. Only 11 bins are above 0.2 IVEM in the original waveform such that it cannot pass the traditional TOT algo- rithm. However the deconvolution has the 13 bins required to be above the threshold. Data Availability Statement This manuscript has data included as electronic supplementary material. The online version of this article contains supplementary material, which is available to authorized users. Open Access This article is licensed under a Creative Commons Attri- bution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, pro- vide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indi- cated otherwise in a credit line to the material. Appendix A: The electromagnetic trigger algorithms (ILP) Grant No. LABEX ANR-10-LABX-63 within the Investisse- ments d’Avenir Programme Grant No. ANR-11-IDEX-0004-02; Ger- many – Bundesministerium für Bildung und Forschung (BMBF); Deutsche Forschungsgemeinschaft (DFG); Finanzministerium Baden- Württemberg; Helmholtz Alliance for Astroparticle Physics (HAP); Helmholtz-Gemeinschaft Deutscher Forschungszentren (HGF); Min- isterium für Innovation, Wissenschaft und Forschung des Landes Nordrhein-Westfalen; Ministerium für Wissenschaft, Forschung und Kunst des Landes Baden-Württemberg; Italy – Istituto Nazionale di Fisica Nucleare (INFN); Istituto Nazionale di Astroﬁsica (INAF); Ministero dell’Istruzione, dell’Universitá e della Ricerca (MIUR); CETEMPS Center of Excellence; Ministero degli Affari Esteri (MAE); México – Consejo Nacional de Ciencia y Tecnología (CONACYT) No. 167733; Universidad Nacional Autónoma de México (UNAM); PAPIIT DGAPA-UNAM; The Netherlands – Ministry of Educa- tion, Culture and Science; Netherlands Organisation for Scientiﬁc Research (NWO); Dutch national e-infrastructure with the support of SURF Cooperative; Poland -Ministry of Science and Higher Educa- tion, grant No. DIR/WK/2018/11; National Science Centre, Grants No. 2013/08/M/ST9/00322, No. 2016/23/B/ST9/01635 and No. HAR- MONIA 5–2013/10/M/ST9/00062, UMO-2016/22/M/ST9/00198; Por- tugal – Portuguese national funds and FEDER funds within Programa Operacional Factores de Competitividade through Fundação para a Ciência e a Tecnologia (COMPETE); Romania – Romanian Min- istry of Education and Research, the Program Nucleu within MCI (PN19150201/16N/2019 and PN19060102) and project PN-III-P1- 1.2-PCCDI-2017-0839/19PCCDI/2018 within PNCDI III; Slovenia – Slovenian Research Agency, grants P1-0031, P1-0385, I0-0033, N1- 0111; Spain – Ministerio de Economía, Industria y Competitividad (FPA2017-85114-P and PID2019-104676GB-C32), Xunta de Gali- cia (ED431C 2017/07), Junta de Andalucía (SOMM17/6104/UGR, P18-FR-4314) Feder Funds, RENATA Red Nacional Temática de Astropartículas (FPA2015-68783-REDT) and María de Maeztu Unit of Excellence (MDM-2016-0692); USA – Department of Energy, Con- tracts No. DE-AC02-07CH11359, No. DE-FR02-04ER41300, No. DE- FG02-99ER41107 and No. DE-SC0011689; National Science Foun- dation, Grant No. 0450696; The Grainger Foundation; Marie Curie- IRSES/EPLANET; European Particle Physics Latin American Net- work; University of Delaware Research Foundation (UDRF) – 2019; and UNESCO. The ToTd and MoPS triggers were designed to be insensitive to atmospheric muons such that they enable the detection of small electromagnetic signals from air showers. The typical morphology of a waveform from a ∼GeV muon is a ≃150ns (≃6 ADC bins) pulse with an amplitude of ≃1 IVEM, where IVEM is the maximum amplitude of a signal created by a muon that traverses the water volume vertically [23]. Thus, the ToTd and MoPS algorithms are used to look for signals that do not ﬁt this criteria. 5 For example, four bins with Si ≤Si+1 ≤Si+2 ≤Si+3 is considered one positive step, not three positive steps. 6 Discussion MSMT CR LTT18004, LM2015038, LM2018102, CZ.02.1.01/0.0/0.0/16_013/0001402, CZ.02.1.01/0.0/0.0/18_046/ 0016010 and CZ.02.1.01/0.0/0.0/17_049/0008422; France – Centre de Calcul IN2P3/CNRS; Centre National de la Recherche Scien- tiﬁque(CNRS);ConseilRégionalIle-de-France;DépartementPhysique Nucléaire et Corpusculaire (PNC-IN2P3/CNRS); Département Sci- ences de l’Univers (SDU-INSU/CNRS); Institut Lagrange de Paris y ( ) p The measurements of the all-particle ﬂux from various experiments [9–11,44–49] in the energy region surrounding the second knee are shown in Fig. 18. Experiments which set their energy scale using calorimetric measurements are plotted using colored markers (Auger SD-750, TA TALE, TUNKA-133, Yakutsk) while the measurements shown in gray markers represent MC-based energy assignments. The spread between various experiments is statistically signiﬁ- cant. However, all these measurements are consistent with the SD-750 spectrum within the 14% energy scale system- atic uncertainty. Understanding the nature of the off-sets in the energy scales is beyond the scope of this paper. However, we note that the TALE spectrum agrees rather well with the SD-750 spectrum, offset by 5 to 6% in energy. The agree- ment is notable given that at-and-above the ankle, an energy scale off-set of around 11% is required to bring the spectral measurements with SD-1500 of the Auger Observatory and the SD of the Telescope Array into agreement [54]. Additionally, we have presented a robust method to com- bine energy spectra. Using the result from the SD-750 and a 123 123 Page 17 of 25 966 Eur. Phys. J. C (2021) 81 :966 Appendix A: The electromagnetic trigger algorithms The correlations between systematic uncertainties are provided in the Supplementary material lg(E/eV) N J±σstat±σsyst km2 yr sr eV 17.05 217094 6.568 +0.015 +2.0 −0.015 −1.8 ×10−14 17.15 132828 3.302 +0.010 +1.0 −0.010 −0.9 ×10−14 17.25 79931 1.625 +0.006 +0.5 −0.006 −0.5 ×10−14 17.35 47509 7.860 +0.038 +2.5 −0.038 −2.3 ×10−15 17.45 27889 3.738 +0.023 +1.2 −0.023 −1.1 ×10−15 17.55 16407 1.775 +0.014 +0.6 −0.014 −0.5 ×10−15 17.65 9695 8.44 +0.09 +2.9 −0.09 −2.5 ×10−16 17.75 5653 3.95 +0.05 +1.5 −0.05 −1.1 ×10−16 17.85 3317 1.86 +0.03 +0.7 −0.03 −0.5 ×10−16 17.95 1990 8.91 +0.20 +3.6 −0.20 −2.6 ×10−17 18.05 1158 4.14 +0.12 +1.8 −0.12 −1.2 ×10−17 18.15 651 1.85 +0.07 +0.9 −0.07 −0.5 ×10−17 18.25 367 8.35 +0.43 +4.1 −0.45 −2.4 ×10−18 18.35 235 4.26 +0.27 +2.2 −0.28 −1.2 ×10−18 18.45 139 2.01 +0.17 +1.1 −0.17 −0.6 ×10−18 18.55 79 9.0 +1.0 +5.2 −1.0 −2.5 ×10−19 18.65 45 4.1 +0.7 +2.4 −0.6 −1.2 ×10−19 18.75 31 2.3 +0.4 +1.3 −0.4 −0.6 ×10−19 18.85 29 1.7 +0.3 +0.9 −0.3 −0.5 ×10−19 19.10 36 2.8 +0.5 +1.6 −0.5 −0.8 ×10−20 19.40 7 5.7 +2.4 +3.2 −2.4 −1.6 ×10−21 reported in Table 10 and the correlation matrix of the spectral parameters at the nominal energy scale in Table 11 (statistical uncertainties). Fig. 20 The scaling factor that has been applied to the raw spectrum to produce the unfolded spectrum (see Eq. (12)) and the statistical uncer- tainty Fig. 20 The scaling factor that has been applied to the raw spectrum to d h f ld d ( E (12)) d h i i l Fig. 20 The scaling factor that has been applied to the raw spectrum to produce the unfolded spectrum (see Eq. (12)) and the statistical uncer- tainty Table 8 SD-750 spectrum data. The correlations between systematic uncertainties are provided in the Supplementary material Table 8 SD-750 spectrum data. The correlations between systematic uncertainties are provided in the Supplementary material Fig. 19 Top: Example waveform which passes the ToTd algorithm. Middle: The deconvolution (Eq. (A.1)) of the ﬁrst waveform along with the threshold to pass the algorithm (dashed red line). Bottom: Example waveform which passes the MoPS algorithm fall within this range, the trigger condition is satisﬁed. An example of a waveform which passes the MoPS trigger is shown in the bottom plot of Fig. 19. Appendix A: The electromagnetic trigger algorithms If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permit- ted use, you will need to obtain permission directly from the copy- right holder. To view a copy of this licence, visit http://creativecomm ons.org/licenses/by/4.0/. The second, MoPS, counts the number of instances, in a sliding 3µs window, in which there is a monotonic increase of the signal amplitude. Each such instance of successive increases in the digitized waveform is what we deﬁne as a positive step.5 For each positive step, the total vertical increase, j, must be above that of typical noise, and below the characteristic amplitude of a vertical muon, namely 3 < j ≤31. If more than four of the positive-step instances g y Funded by SCOAP3. 12 3 Eur. Phys. J. C (2021) 81 :966 966 Page 18 of 25 g y ( ) Fig. 19 Top: Example waveform which passes the ToTd algorithm. Middle: The deconvolution (Eq. (A.1)) of the ﬁrst waveform along with the threshold to pass the algorithm (dashed red line). Bottom: Example waveform which passes the MoPS algorithm fall within this range, the trigger condition is satisﬁed. An example of a waveform which passes the MoPS trigger is shown in the bottom plot of Fig. 19. Appendix B: Spectrum data We report in this appendix several data of interest. Note that more can be found in the Supplemental Material in electronic format. The bin migration is corrected to produce the unfolded spectrum.Themagnitudeofthecorrectionfactor,asdescribed by Eq. (12), is shown in Fig. 20 along with the statistical uncertainty band. The energy spectrum of the SD-750 array is reportedinTable8andthecorrelationmatrixof thespectral parameters at the nominal energy scale in Table 9 (statisti- cal uncertainties). Finally, the combined energy spectrum is Fig. 20 The scaling factor that has been applied to the raw spectrum to produce the unfolded spectrum (see Eq. (12)) and the statistical uncer- tainty Table 8 SD-750 spectrum data. References galactic cosmic rays beyond PeV energies. Phys. Rev. Lett. 126, 141101 (2021). https://doi.org/10.1103/PhysRevLett.126.141101. arXiv:2104.05181 1. G.V. Kulikov, G.B. Khristiansen, On the size spectrum of extensive air showers. J. Exp. Theor. Phys. 35, 635 (1958) 14. LHAASO Collaboration, Ultrahigh-energy photons up to 1.4 peta- electronvolts from 12 gamma-ray galactic sources. Nature 594, 33–36 (2021). https://doi.org/10.1038/s41586-021-03498-z 14. LHAASO Collaboration, Ultrahigh-energy photons up to 1.4 peta- electronvolts from 12 gamma-ray galactic sources. Nature 594, 33–36 (2021). https://doi.org/10.1038/s41586-021-03498-z 2. HEGRA Collaboration, Energy spectrum and chemical composi- tionofcosmicraysbetween0.3PeVand10 eVdeterminedfromthe Cherenkov light and charged particle distributions in air showers. Astron. Astrophys. 359, 682 (2000). arXiv:astro-ph/9908202 2. HEGRA Collaboration, Energy spectrum and chemical composi- tionofcosmicraysbetween0.3PeVand10 eVdeterminedfromthe Cherenkov light and charged particle distributions in air showers. Astron. Astrophys. 359, 682 (2000). arXiv:astro-ph/9908202 15. LHAASO Collaboration, Discovery of the Ultra-high energy gamma-ray source LHAASO J2108+5157. arXiv:2106.09865 16. P. Cristofari, P. Blasi, E. Amato, The low rate of Galactic pevatrons. Astropart. Phys. 123, 102492 (2020). https://doi.org/10.1016/j. astropartphys.2020.102492. arXiv:2007.04294 3. J.W. Fowler, L.F. Fortson, C.C.H. Jui, D.B. Kieda, R.A. Ong, C.L. Pryke et al., A measurement of the cosmic ray spectrum and com- position at the knee. Astropart. Phys. 15, 49 (2001). https://doi.org/ 10.1016/S0927-6505(00)00139-0. arXiv:astro-ph/0003190 3. J.W. Fowler, L.F. Fortson, C.C.H. Jui, D.B. Kieda, R.A. Ong, C.L. Pryke et al., A measurement of the cosmic ray spectrum and com- position at the knee. Astropart. Phys. 15, 49 (2001). https://doi.org/ 10.1016/S0927-6505(00)00139-0. arXiv:astro-ph/0003190 17. A.M. Hillas, Can diffusive shock acceleration in supernova rem- nants account for high-energy galactic cosmic rays? J. Phys. G 31, R95 (2005). https://doi.org/10.1088/0954-3899/31/5/R02 4. EAS-TOP Collaboration, The cosmic ray primary composition in the “knee” region through the EAS electromagnetic and muon mea- surements at EAS-TOP. Astropart. Phys. 21, 583 (2004). https:// doi.org/10.1016/j.astropartphys.2004.04.005 18. KASCADE-Grande Collaboration, KASCADE-Grande measure- ments of energy spectra for elemental groups of cosmic rays. Astropart. Phys. 47 (2013) 54. https://doi.org/10.1016/j. astropartphys.2013.06.004. arXiv:1306.6283 5. MACRO, EAS-TOP Collaboration, The primary cosmic ray com- position between 1015 and 1016 eV from extensive air show- ers electromagnetic and TeV muon data. Astropart. Phys. 20, 641 (2004). https://doi.org/10.1016/j.astropartphys.2003.10.004. arXiv:astro-ph/0305325 19. R. Aloisio, V. Berezinsky, P. Blasi, Ultra high energy cosmic rays: implications of Auger data for source spectra and chemical compo- sition. JCAP 10, 020 (2014). https://doi.org/10.1088/1475-7516/ 2014/10/020. arXiv:1312.7459 6. A.P. Garyaka, R.M. Martirosov, S.V. Ter-Antonyan, N. Nikolskaya, Y.A. Gallant, L. Jones et al., Rigidity-dependent cosmic ray energy spectra in the knee region obtained with the GAMMA experi- ment. Astropart. Phys. Appendix B: Spectrum data The bin migration is corrected to produce the unfolded spectrum.Themagnitudeofthecorrectionfactor,asdescribed by Eq. (12), is shown in Fig. 20 along with the statistical uncertainty band. The energy spectrum of the SD-750 array is reportedinTable8andthecorrelationmatrixof thespectral parameters at the nominal energy scale in Table 9 (statisti- cal uncertainties). Finally, the combined energy spectrum is 123 123 Page 19 of 25 966 Eur. Phys. J. C (2021) 81 :966 Table 9 Elements of the correlation matrix (statistical uncertainties) of the spectral parameters describing the SD-750 energy spectrum at the nominal energy scale J0 γ1 E12 γ2 ω01 J0 1 0.978 −0.067 −0.120 0.998 γ1 1 −0.094 −0.109 0.967 E12 1 −0.814 −0.059 γ2 1 −0.123 ω01 1 Table 10 Combined SD spectrum data. The correlations between sys- tematic uncertainties are provided in the Supplementary material lg(E/eV) J±σstat±σsyst km2 yr sr eV 17.05 6.341 +0.015 +2.1 −0.015 −1.9 ×10−14 17.15 3.191 +0.010 +1.1 −0.010 −0.9 ×10−14 17.25 1.577 +0.006 +0.5 −0.006 −0.5 ×10−14 17.35 7.643 +0.039 +2.6 −0.039 −2.3 ×10−15 17.45 3.650 +0.024 +1.3 −0.024 −1.1 ×10−15 17.55 1.739 +0.015 0.6 −0.015 −0.5 ×10−15 17.65 8.32 +0.09 +3.0 −0.09 −2.4 ×10−16 17.75 3.90 +0.05 +1.4 −0.05 −1.1 ×10−16 17.85 1.85 +0.03 +0.7 −0.03 −0.5 ×10−16 17.95 8.87 +0.20 +3.3 −0.20 −2.5 ×10−17 18.05 4.14 +0.12 +1.6 −0.12 −1.2 ×10−17 18.15 1.90 +0.07 +0.7 −0.07 −0.5 ×10−17 Table 10 continued lg(E/eV) J±σstat±σsyst km2 yr sr eV 18.25 8.47 +0.43 +3.3 −0.44 −2.4 ×10−18 18.35 4.17 +0.28 +1.7 −0.27 −1.2 ×10−18 18.45 1.929 +0.007 +0.7 −0.007 −0.5 ×10−18 18.55 9.041 +0.042 +2.9 −0.042 −2.0 ×10−19 18.65 4.294 +0.026 +1.3 −0.026 −0.9 ×10−19 18.75 2.167 +0.016 +0.6 −0.016 −0.4 ×10−19 18.85 1.226 +0.011 +0.3 −0.011 −0.2 ×10−19 18.95 6.82 +0.08 +1.6 −0.08 −1.3 ×10−20 19.05 3.79 +0.05 +0.9 −0.05 −0.7 ×10−20 19.15 2.07 +0.03 +0.5 −0.03 −0.4 ×10−20 19.25 1.04 +0.02 +0.2 −0.02 −0.2 ×10−20 19.35 0.53 +0.01 +1.6 −0.01 −1.3 ×10−20 19.45 2.49 +0.08 +0.9 −0.08 −0.7 ×10−21 19.55 1.25 +0.05 +0.5 −0.05 −0.3 ×10−21 19.65 5.99 +0.32 +2.4 −0.32 −1.8 ×10−22 19.75 1.95 +0.17 +0.9 −0.17 −0.7 ×10−22 19.85 8.1 +1.0 +4.0 −0.9 −2.8 ×10−23 19.95 1.8 +0.5 +1.0 −0.4 −0.7 ×10−23 20.05 5.5 +2.5 +3.3 −1.8 −2.2 ×10−24 20.15 2.9 +1.7 +1.9 −1.2 −1.2 ×10−24 Table 9 Elements of the correlation matrix (statistical uncertainties) of the spectral parameters describing the SD-750 energy spectrum at the nominal energy scale 123 123 123 Eur. Phys. J. 966 Page 20 of 25 Table 11 Elements of the correlation matrix (statistical uncertainties) of the spectral parameters describing the combined SD energy spectrum J0 γ1 E12 γ2 E23 γ3 E34 γ4 ω01 J0 1 −0.470 0.552 −0.357 −0.383 −0.095 −0.033 0.035 −0.258 γ1 1 −0.585 0.524 0.877 0.358 0.075 −0.085 0.966 E12 1 −0.896 −0.425 0.192 0.119 0.110 −0.493 γ2 1 0.455 −0.385 −0.217 −0.154 0.475 E23 1 0.252 −0.063 −0.174 0.858 γ3 1 0.474 0.136 0.366 E34 1 0.805 0.075 γ4 1 −0.078 ω01 1 Appendix B: Spectrum data C (2021) 81 :966 966 Page 20 of 25 966 Page 20 of 25 References Dawson (Pierre Auger Collaboration), The Energy Scale of the Pierre Auger Observatory. PoS ICRC2019, 231 (2019). https:// doi.org/10.22323/1.358.0231 28. D.W. Newton, J. Knapp, A.A. Watson, The optimum distance at which to determine the size of a giant air shower. Astropart. Phys. 26, 414 (2007). https://doi.org/10.1016/j.astropartphys.2006.08. 003. arXiv:astro-ph/0608118 44. M. Nagano, M. Teshima, Y. Matsubara, H. Dai, T. Hara, N. Hayashida et al., Energy spectrum of primary cosmic rays above 1017 eV determined from the extensive air shower experiment at Akeno. J. Phys. G 18, 423 (1992). https://doi.org/10.1088/ 0954-3899/18/2/022 29. J. Bellido (Pierre Auger Collaboration), Depth of maximum of air- shower proﬁles at the Pierre Auger Observatory: measurements above 1017.2 eV and composition implications. PoS ICRC2017, 506 (2017). https://doi.org/10.22323/1.301.0506 45. S. Ter-Antonyan, Sharp knee phenomenon of primary cosmic ray energy spectrum. Phys. Rev. D 89, 123003 (2014). https://doi.org/ 10.1103/PhysRevD.89.123003. arXiv:1405.5472 30. Pierre Auger Collaboration, Impact of atmospheric effects on the energy reconstruction of air showers observed by the surface detectors of the Pierre Auger Observatory. JINST 12, P02006 (2017). https://doi.org/10.1088/1748-0221/12/02/ p02006. arXiv:1702.02835 46. KASCADE-Grande Collaboration, KASCADE-Grande energy spectrum of cosmic rays interpreted with post-LHC hadronic inter- action models. PoS ICRC2015, 359 (2016). https://doi.org/10. 22323/1.236.0359 31. J. Hersil, I. Escobar, D. Scott, G. Clark, S. Olbert, Observations of extensive air showers near the maximum of their longitudi- nal development. Phys. Rev. Lett. 6, 22 (1961). https://doi.org/ 10.1103/PhysRevLett.6.22 47. E.N. Gudkova, N.M. Nesterova, Results of the further analysis of data from the Tien Shan array in the energy spectrum of primary cosmic rays in the energy Range of 2 × 1013 - 3 × 1017 eV. Phys. Atomic Nuclei 83, 629 (2020). arXiv:2010.04236 48. TIBET III Collaboration, The All-particle spectrum of primary cos- mic rays in the wide energy range from 1014 eV to 1017 eV observed with the Tibet-III air-shower array. Astrophys. J. 678, 1165 (2008). https://doi.org/10.1086/529514. arXiv:0801.1803 32. Pierre Auger Collaboration, Cosmic-ray anisotropies in right ascension measured by the Pierre Auger Observatory. Astrophys. J. 891, 142 (2020). https://doi.org/10.3847/1538-4357/ab7236. arXiv:2002.06172 49. S.P. Knurenko, Z.E. Petrov, R. Sidorov, I.Y. Sleptsov, S.K. Starostin, G.G. Struchkov, Cosmic ray spectrum in the energy range 1015–1018 eV and the second knee according to the small Cherenkov setup at the Yakutsk EAS array, Proc. of 33rd ICRC (2013). arXiv:1310.1978 33. T.W. Anderson, D.A. Darling, A test of goodness of ﬁt. J. Am. Stat. Assoc. 49, 765 (1954) 34. References Pierre Auger Collaboration, Data-driven estimation of the invis- ible energy of cosmic ray showers with the Pierre Auger Obser- vatory. Phys. Rev. D 100, 082003 (2019). https://doi.org/10.1103/ PhysRevD.100.082003. arXiv:1901.08040 50. O.A. Gress, T.I. Gress, E.E. Korosteleva, L.A. Kuzmichev, B.K. Lubsandorzhiev, L.V. Pan’kov et al., The study of primary cosmic rays energy spectrum and mass composition in the energy range 0.5–50 PeV with TUNKA Eas Cherenkov array. Nucl. Phys. B- Proc. Suppl. 75, 299 (1999) 35. Pierre Auger Collaboration, Direct measurement of the muonic content of extensive air showers between 2 × 1017 and 2 × 1018 eV at the Pierre Auger Observatory. Eur. Phys. J. C 80, 751 (2020). https://doi.org/10.1140/epjc/s10052-020-8055-y 51. Pierre Auger Collaboration, Depth of maximum of air-shower proﬁles at the Pierre Auger Observatory: measurements above 1017.2 eV and composition implications. PoS ICRC2017, 506 (2018). https://doi.org/10.22323/1.301.0506 36. Pierre Auger Collaboration, Depth of maximum of air-shower pro- ﬁles at the Pierre Auger Observatory. II. Composition implica- tions. Phys. Rev. D 90, 122006 (2014). https://doi.org/10.1103/ PhysRevD.90.122006. arXiv:1409.5083 52. S. Ostapchenko, QGSJET-II: physics, recent improvements, and results for air showers, in EPJ Web of Conferences, vol. 52, p. 02001 (EDP Sciences, 2013) 37. H.P. Dembinski, B. Kégl, I.C. Mari¸s, M. Roth, D. Veberiˇc, A like- lihood method to cross-calibrate air-shower detectors. Astropart. Phys. 73, 44 (2016). https://doi.org/10.1016/j.astropartphys.2015. 08.001. arXiv:1503.09027 53. F. Riehn, H.P. Dembinski, R. Engel, A. Fedynitch, T. Gaisser, T. Stanev, The hadronic interaction model Sibyll 2.3c and Feynman scaling. PoS ICRC2017, 301 (2017). https://doi.org/10.22323/1. 301.0301. arXiv:1709.07227 38. Pierre Auger Collaboration, The energy scale of the Pierre Auger Observatory. PoS ICRC2019, 231 (2020). https://doi.org/ 10.22323/1.358.0231 39. D.Hecketal.,CORSIKA:aMonteCarlocodetosimulateextensive air showers. Report fzka 6019 (1998) 54. O. Deligny, (Pierre Auger and Telescope Array Collaborations), The energy spectrum of ultra-high energy cosmic rays measured at the Pierre Auger Observatory and at the Telescope Array. PoS ICRC2019, 234 (2019). https://doi.org/10.22323/1.358.0234 40. T. Pierog, I. Karpenko, J.M. Katzy, E. Yatsenko, K. Werner, EPOS LHC: test of collective hadronization with data measured at the CERN Large Hadron Collider. Phys. Rev. C 92, 034906 (2015). https://doi.org/10.1103/PhysRevC.92.034906. arXiv:1306.0121 55. V. Novotny (Pierre Auger Collaboration), Measurement of the spectrum of cosmic rays above 1016.5 eV with Cherenkov- dominated events at the Pierre Auger Observatory. PoS ICRC2019, 374 (2019). https://doi.org/10.22323/1.358.0374 41. H.P. Dembinski, R. Engel, A. Fedynitch, T. Gaisser, F. Riehn, T. Stanev, Data-driven model of the cosmic-ray ﬂux and mass com- position from 10 GeV to 1011 GeV. References 28, 169 (2007). https://doi.org/10.1016/j. astropartphys.2007.04.004. arXiv:0704.3200 20. Pierre Auger Collaboration, Features of the energy spectrum of cosmic rays above 2.5×1018 eV using the Pierre Auger Observa- tory. Phys. Rev. Lett. 125, 121106 (2020). https://doi.org/10.1103/ PhysRevLett.125.121106. arXiv:2008.06488 21. PierreAugerCollaboration,Measurementofthecosmic-rayenergy spectrum above 2.5×1018 eV using the Pierre Auger Observa- tory. Phys. Rev. D 102, 062005 (2020). https://doi.org/10.1103/ PhysRevD.102.062005. arXiv:2008.06486 p p y 7. P. Blasi, The origin of galactic cosmic rays. Astron. Astrophys. Rev. 21, 70 (2013). https://doi.org/10.1007/s00159-013-0070-7. arXiv:1311.7346 8. KASCADE-Grande Collaboration, The spectrum of high-energy cosmic rays measured with KASCADE-Grande. Astropart. Phys. 36, 183 (2012). https://doi.org/10.1016/j.astropartphys.2012.05. 023. arXiv:1206.3834 22. Pierre Auger Collaboration, Trigger and Aperture of the Surface Detector Array of the Pierre Auger Observatory. Nucl. Instrum. Meth. A 613, 29 (2010). https://doi.org/10.1016/j.nima.2009.11. 018. arXiv:1111.6764 9. IceCube Collaboration, Cosmic ray spectrum and composition from PeV to EeV using 3 years of data from IceTop and Ice- Cube. Phys. Rev. D 100, 082002 (2019). https://doi.org/10.1103/ PhysRevD.100.082002. arXiv:1906.04317 23. Pierre Auger Collaboration, Calibration of the surface array of the Pierre Auger Observatory. Nucl. Instrum. Meth. A 568, 839 (2006). https://doi.org/10.1016/j.nima.2006.07.066. arXiv:2102.01656 p g j 24. Pierre Auger Collaboration, The Pierre Auger Cosmic Ray Obser- vatory. Nucl. Instrum. Meth. A 798, 172 (2015). https://doi.org/10. 1016/j.nima.2015.06.058. arXiv:1502.01323 10. Telescope Array Collaboration, The cosmic-ray energy spectrum between 2 PeV and 2 EeV Observed with the TALE detector in monocular mode. Astrophys. J. 865, 74 (2018). https://doi.org/10. 3847/1538-4357/aada05. arXiv:1803.01288 25. Pierre Auger Collaboration, The lateral trigger probability function for the ultra-high energy cosmic ray showers detected by the Pierre Auger Observatory. Astropart. Phys. 35, 266 (2011). https://doi. org/10.1016/j.astropartphys.2011.08.001. arXiv:1111.6645 11. N.M. Budnev et al., The primary cosmic-ray energy spectrum mea- sured with the Tunka-133 array. Astropart. Phys. 117, 102406 (2020). https://doi.org/10.1016/j.astropartphys.2019.102406 26. Pierre Auger Collaboration, Reconstruction of events recorded with the surface detector of the pierre auger observatory. JINST 15, P10021 (2020). https://doi.org/10.1088/1748-0221/15/ 10/P10021. arXiv:2007.09035 12. A. Albert et al., Evidence of 200 TeV photons from HAWC. Astro- phys. J. Lett. 907, L30 (2021). https://doi.org/10.3847/2041-8213/ abd77b. arXiv:2012.15275 13. Tibet ASgamma Collaboration, First detection of sub-PeV dif- fuse gamma rays from the galactic disk: evidence for ubiquitous 13. Tibet ASgamma Collaboration, First detection of sub-PeV dif- fuse gamma rays from the galactic disk: evidence for ubiquitous 3 Eur. Phys. J. C (2021) 81 :966 Page 21 of 25 966 27. A.M. Hillas, Derivation of the EAS spectrum. Acta Phys. Acad. Sci. Hung. 29, 355 (1970) 43. B. References PoS ICRC2017, 533 (2017). https://doi.org/10.22323/1.301.0533 42. M. Bonamente, Distribution of the C statistic with applica- tions to the sample mean of Poisson data. J. Appl. Stat. 47, 2044 (2020). https://doi.org/10.1080/02664763.2019.1704703. arXiv:1912.05444 123 123 Eur. Phys. J. C (2021) 81 :966 966 Page 22 of 25 P. Abreu71, M. Aglietta51,53, J. M. Albury12, I. Allekotte1, A. Almela8,11, J. Alvarez-Muñiz78, R. Alves Batista79, G. A. Anastasi51,62, L. Anchordoqui86, B. Andrada8, S. Andringa71, C. Aramo49, P. R. Araújo Ferreira41, J. C. Arteaga Velázquez66, H. Asorey8, P. Assis71, G. Avila10, A. M. Badescu74, A. Bakalova31, A. Balaceanu72, F. Barbato44,45, R. J. Barreira Luz71, K. H. Becker37, J. A. Bellido12,68, C. Berat35, M. E. Bertaina51,62, X. Bertou1, P. L. Biermann95, P. Billoir34, V. Binet6, K. Bismark8,38, T. Bister41, J. Biteau36, J. Blazek31, C. Bleve35, M. Boháˇcová31, D. Boncioli45,56, C. Bonifazi25, L. Bonneau Arbeletche20, N. Borodai69, A. M. Botti8, J. Brack97, T. Bretz41, P. G. Brichetto Orchera8, F. L. Briechle41, P. Buchholz43, A. Bueno77, S. Buitink14, M. Buscemi46, M. Büsken8,38, K. S. Caballero-Mora65, L. Caccianiga48,58, F. Canfora79,80, I. Caracas37, J. M. Carceller77, R. Caruso46,57, A. Castellina51,53, F. Catalani18, G. Cataldi47, L. Cazon71, M. Cerda9, J. A. Chinellato21, J. Chudoba31, L. Chytka32, R. W. Clay12, A. C. Cobos Cerutti7, R. Colalillo49,59, A. Coleman92 , M. R. Coluccia47, R. Conceição71, A. Condorelli44,45, G. Consolati48,54, F. Contreras10, F. Convenga47,55, D. Correia dos Santos27, C. E. Covault84, S. Dasso3,5, K. Daumiller40, B. R. Dawson12, J. A. Day12, R. M. de Almeida27, J. de Jesús8,40, S. J. de Jong79,80, G. De Mauro79,80, J. R. T. de Mello Neto25,26, I. De Mitri44,45, J. de Oliveira17, D. de Oliveira Franco21, F. de Palma47,55, V. de Souza19, E. De Vito47,55, M. del Río10, O. Deligny33, A. Di Matteo51, C. Dobrigkeit21, J. C. D’Olivo67, L. M. Domingues Mendes71, R. C. dos Anjos24, D. dos Santos27, M. T. Dova4, J. Ebr31, R. Engel38,40, I. Epicoco47,55, M. Erdmann41, C. O. Escobar94, A. Etchegoyen8,11, H. Falcke79,80,81, J. Farmer91, G. Farrar89, A. C. Fauth21, N. Fazzini94, F. Feldbusch39, F. Fenu51,53, B. Fick88, J. M. Figueira8, A. Filipˇciˇc75,76, T. Fitoussi40, T. Fodran79, M. M. Freire6, T. Fujii91,98, A. Fuster8,11, C. Galea79, C. Galelli48,58, B. García7, A. L. Garcia Vegas41, H. Gemmeke39, F. Gesualdi8,40, A. Gherghel-Lascu72, P. L. Ghia33, U. Giaccari79, M. Giammarchi48, J. Glombitza41, F. Gobbi9, F. Gollan8, G. Golup1, M. Gómez Berisso1, P. F. Gómez Vitale10, J. P. Gongora10, J. M. González1, N. González13, I. Goos1,40, D. Góra69, A. Gorgi51,53, M. Gottowik37, T. D. Grubb12, F. Guarino49,59, G. P. Guedes22, E. Guido51,62, S. Hahn8,40, P. Hamal31, M. R. Hampel8, P. Hansen4, D. Harari1, V. M. Harvey12, A. Haungs40, T. Hebbeker41, D. Heck40, G. C. Hill12, C. Hojvat94, J. R. Hörandel79,80, P. Horvath32, M. Hrabovský32, T. Huege14,40, A. Insolia46,57, P. G. Isar73, P. Janecek31, J. A. Johnsen85, J. Jurysek31, A. Kääpä37, K. H. Kampert37, N. Karastathis40, B. Keilhauer40, J. Kemp41, A. Khakurdikar79, V. V. Kizakke Covilakam8,40, H. O. Klages40, M. Kleifges39, J. Kleinfeller9, M. Köpke38, N. Kunka39, B. L. Lago16, R. G. Lang19, N. Langner41, M. A. Leigui de Oliveira23, V. Lenok40, A. Letessier-Selvon34, I. Lhenry-Yvon33, D. Lo Presti46,57, L. Lopes71, R. López63, L. Lu93, Q. Luce38, J. P. Lundquist75, A. Machado Payeras21, G. Mancarella47,55, D. Mandat31, B. C. Manning12, J. Manshanden42, P. Mantsch94, S. Maraﬁco33, A. G. Mariazzi4, I. C. Mari¸s13, G. Marsella46,60, D. Martello47,55, S. Martinelli8,40, H. Martinez19, O. Martínez Bravo63, M. Mastrodicasa45,56, H. J. Mathes40, J. Matthews87, G. Matthiae50,61, E. Mayotte37, P. O. Mazur94, G. Medina-Tanco67, D. Melo8, A. Menshikov39, K.-D. Merenda85, S. Michal32, M. I. Micheletti6, L. Miramonti48,58, D. Mockler13,38, S. Mollerach1, F. Montanet35, C. Morello51,53, M. Mostafá90, A. L. Müller8, M. A. Muller21, K. Mulrey14, R. Mussa51, M. Muzio89, W. M. Namasaka37, A. Nasr-Esfahani37, L. Nellen67, M. Niculescu-Oglinzanu72, M. Niechciol43, D. Nitz88, D. Nosek30, V. Novotny30, L. Nožka32, A. Nucita47,55, L. A. Núñez29, M. Palatka31, J. Pallotta2, P. Papenbreer37, G. Parente78, A. Parra63, J. Pawlowsky37, M. Pech31, F. Pedreira78, J. P¸ekala69, R. Pelayo64, J. Peña-Rodriguez29, E. E. Pereira Martins8,38,J. Perez Armand20,C. Pérez Bertolli8,40,M. Perlin8,40,L. Perrone47,55,S. Petrera44,45,T. Pierog40, M. Pimenta71, V. Pirronello46,57, M. Platino8, B. Pont79, M. Pothast79,80, P. Privitera91, M. Prouza31, A. Puyleart88, S. Querchfeld37, J. Rautenberg37, D. Ravignani8, M. Reininghaus8,40, J. Ridky31, F. Riehn71, M. Risse43, V. Rizi45,56, W. Rodrigues de Carvalho20, J. Rodriguez Rojo10, M. J. Roncoroni8, M. Roth40, E. Roulet1, A. C. Rovero5, P. Ruehl43, S. J. Safﬁ12, A. Saftoiu72, F. Salamida45,56, H. Salazar63, G. Salina50, J. D. Sanabria Gomez29, F. Sánchez8, E. M. Santos20, E. Santos31, F. Sarazin85, R. Sarmento71, C. Sarmiento-Cano8, R. Sato10, P. Savina33,47,55, C. M. Schäfer40, V. Scherini47, H. Schieler40, M. Schimassek8,38, M. Schimp37, F. Schlüter8,40, D. Schmidt38, O. Scholten14,83, P. Schovánek31, F. G. Schröder40,92, S. Schröder37, J. Schulte41, A. Schulz38, S. J. Sciutto4, M. Scornavacche8,40, A. Segreto46,52, S. Sehgal37, R. C. Shellard15, G. Sigl42, G. Silli8,40, O. Sima72,99, R. Šmída91, P. Sommers90, J. F. Soriano86, J. Souchard35, R. Squartini9, M. Stadelmaier8,40, D. Stanca72, S. Staniˇc75, J. Stasielak69, P. Stassi35, A. Streich8,38, M. Suárez-Durán13, T. Sudholz12, T. Suomijärvi36, A. D. Supanitsky8, Z. Szadkowski70, A. Tapia28, C. Taricco51,62, C. Timmermans79,80, O. Tkachenko40, P. Tobiska31, C. J. Todero Peixoto18, B. Tomé71, Z. Torrès35, A. Travaini9, P. Travnicek31, C. Trimarelli45,56, M. Tueros4, R. Ulrich40, M. Unger40, L. Vaclavek32, M. Vacula32, J. F. Valdés Galicia67, L. Valore49,59, E. Varela63, A. Vásquez-Ramírez29, D. Veberiˇc40, C. Ventura26, I. D. Vergara Quispe4, V. Verzi50, J. Vicha31, J. Vink82, S. Vorobiov75, H. Wahlberg4, C. Watanabe25, A. A. Watson96, M. Weber39, A. Weindl40, L. Wiencke85, Pierre Auger Collaboration 123 Eur. Phys. J. C (2021) 81 :966 Page 23 of 25 966 H. Wilczy´nski69, M. Wirtz41, D. Wittkowski37, B. Wundheiler8, A. Yushkov31, O. Zapparrata13, E. Zas78, D. Zavrtanik75,76, M. Zavrtanik75,76, L. Zehrer75 H. Wilczy´nski69, M. Wirtz41, D. Wittkowski37, B. Wundheiler8, A. Yushkov31, O. Zapparrata13, E. Zas78, D. Zavrtanik75,76, M. Zavrtanik75,76, L. Zehrer75 1 Centro Atómico Bariloche and Instituto Balseiro (CNEA-UNCuyo-CONICET), San Carlos de Bariloche, Argentina 2 Centro de Investigaciones en Láseres y Aplicaciones CITEDEF and CONICET Villa Martelli Argentina 1 Centro Atómico Bariloche and Instituto Balseiro (CNEA-UNCuyo-CONICET), San Carlos de Bariloche, Argentina 2 Centro de Investigaciones en Láseres y Aplicaciones, CITEDEF and CONICET, Villa Martelli, A artamento de Física and Departamento de Ciencias de la Atmósfera y los Océanos, FCEyN, Univers s and CONICET Buenos Aires Argentina 3 Departamento de Física and Departamento de Ciencias de la Atmósfera y los Océanos, FCEyN, U 3 Departamento de Física and Departamento de Ciencias de la Atmósfera y los Océanos, FCEyN, Universidad de Buenos 3 Departamento de Física and Departamento de Ciencias de la Atmósfera y los Océanos, FCEyN, Universidad de Buenos Aires and CONICET, Buenos Aires, Argentina Aires and CONICET, Buenos Aires, Argentina 4 IFLP, Universidad Nacional de La Plata and CONICET, La Plata, Argentina 5 Instituto de Astronomía y Física del Espacio (IAFE, CONICET-UBA), Buenos Aires, Argentina 6 5 Instituto de Astronomía y Física del Espacio (IAFE, CONICET-UBA), Buenos Aires, Argentina 6 Instituto de Física de Rosario (IFIR)-CONICET/U.N.R. and Facultad de Ciencias Bioquímicas y Farmacéuticas U.N.R., Rosario, Argentina 6 Instituto de Física de Rosario (IFIR)-CONICET/U.N.R. Pierre Auger Collaboration and Facultad de Ciencias Bioquímicas y Farmacéuticas U.N.R., Rosario, Argentina Rosario, Argentina 7 Instituto de Tecnologías en Detección y Astropartículas (CNEA, CONICET, UNSAM), Universidad Tecnológic 8 Instituto de Tecnologías en Detección y Astropartículas (CNEA, CONICET, UNSAM), Buenos Aires, Argentina 9 Ob t i Pi A M l ü A ti 8 Instituto de Tecnologías en Detección y Astropartículas (CNEA, CONICET, UNSAM), Buenos Aires, Argentina 9 Observatorio Pierre Auger, Malargüe, Argentina Observatorio Pierre Auger, Malargüe, Argentina 10 Observatorio Pierre Auger and Comisión Nacional de Energía Atómica, Malargüe, Argentina 11 Universidad Tecnológica Nacional-Facultad Regional Buenos Aires, Buenos Aires, Argentina 12 University of Adelaide, Adelaide, SA, Australia 13 Université Libre de Bruxelles (ULB), Brussels, Belgium Université Libre de Bruxelles (ULB), Brussels, 14 Vrije Universiteit Brussels, Brussels, Belgium 15 Centro Brasileiro de Pesquisas Fisicas, Rio de Janeiro, RJ, Brazil 5 Centro Brasileiro de Pesquisas Fisicas, Rio de Ja 16 Centro Federal de Educação Tecnológica Celso Suckow da Fonseca, Nova Friburgo, Brazil 17 Instituto Federal de Educação, Ciência e Tecnologia do Rio de Janeiro (IFRJ), Rio de Janei 18 Escola de Engenharia de Lorena, Universidade de São Paulo, Lorena, SP, Brazil 18 Escola de Engenharia de Lorena, Universidade de São Paulo, Lorena, SP, Brazil 19 Instituto de Física de São Carlos, Universidade de São Paulo, São Carlos, SP, Brazil 19 Instituto de Física de São Carlos, Universidade de São Paulo, São Carlos, SP, Brazil 20 Instituto de Física, Universidade de São Paulo, São Paulo, SP, Brazil 20 Instituto de Física, Universidade de São Paulo, São Paulo, SP, Brazil 21 Universidade Estadual de Campinas, IFGW, Campinas, SP, Brazil 21 Universidade Estadual de Campinas, IFGW, Campinas, SP, Brazil 22 Universidade Estadual de Feira de Santana, Feira de Santana, Brazil 22 Universidade Estadual de Feira de Santana, Feira de Santana, Brazil 23 Universidade Federal do ABC, Santo André, SP, Brazil 4 Universidade Federal do Paraná, Setor Palotina, 25 Instituto de Física, Universidade Federal do Rio de Janeiro, Rio de Janeiro, RJ, Brazil 25 Instituto de Física, Universidade Federal do Rio de Janeiro, Rio de Janeiro, RJ, Brazil 26 Observatório do Valongo, Universidade Federal do Rio de Janeiro (UFRJ) 27 Universidade Federal Fluminense, EEIMVR, Volta Redonda, RJ, Brazil 28 27 Universidade Federal Fluminense, EEIMVR, Volta Redonda, RJ, Brazil 28 28 Universidad de Medellín, Medellín, Colombia 29 Universidad Industrial de Santander, Bucaramanga, Colombia 30 29 Universidad Industrial de Santander, Bucaramanga, Colombia 30 30 Institute of Particle and Nuclear Physics, Faculty of Mathematics and Physics, Charles University, Prague, Czech Republic 30 Institute of Particle and Nuclear Physics, Faculty of Mathematics and Physics, Charles University, Prag Czech Republic 31 Institute of Physics of the Czech Academy of Sciences, Prague, Czech Republic 31 Institute of Physics of the Czech Academy of Sciences, Prague, Czech Republic 32 Palacky University, RCPTM, Olomouc, Czech Republic 32 Palacky University, RCPTM, Olomouc, Czech Republic 33 CNRS/IN2P3, IJCLab, Université Paris-Saclay, Orsay, France 33 CNRS/IN2P3, IJCLab, Université Paris-Saclay, Orsay, France 34 Laboratoire de Physique Nucléaire et de Hautes Energies (LPNHE), Sorbonne Université, Université de Paris, CNRS-IN2P3, Paris, France 34 Laboratoire de Physique Nucléaire et de Hautes Energies (LPNHE), Sorbonne Université, Université de Paris, CNRS-IN2P3, Paris, France 35 Univ. Pierre Auger Collaboration Grenoble Alpes, CNRS, Grenoble Institute of Engineering Univ. Grenoble Alpes, LPSC-IN2P3, 38000 Grenoble, France 35 Univ. Grenoble Alpes, CNRS, Grenoble Institute of Engineering Univ. Grenoble Alpes, LPSC-IN2P3, 38000 Grenob France 36 Université Paris-Saclay, CNRS/IN2P3, IJCLab, Orsay, France 36 Université Paris-Saclay, CNRS/IN2P3, IJCLab, Orsay, France 37 Department of Physics, Bergische Universität Wuppertal, Wuppertal, Germany 38 38 Institute for Experimental Particle Physics, Karlsruhe Institute of Technology (KIT), K 39 Institut für Prozessdatenverarbeitung und Elektronik, Karlsruhe Institute of Technology (KIT), Karlsruhe, German 39 Institut für Prozessdatenverarbeitung und Elektronik, Karlsruhe In g gy 40 Institute for Astroparticle Physics, Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany 41 III. Physikalisches Institut A, RWTH Aachen University, Aachen, Germany 41 III. Physikalisches Institut A, RWTH Aachen University, Aachen, Germany 42 II. Institut für Theoretische Physik, Universität Hamburg, Hamburg, Germany 43 Department Physik-Experimentelle Teilchenphysik, Universität Siegen, Siegen, Ger 44 Gran Sasso Science Institute, L’Aquila, Italy 12 3 Eur. Phys. J. C (2021) 81 :966 966 Page 24 of 25 45 INFN Laboratori Nazionali del Gran Sasso, Assergi (L’Aquila), Ital 46 INFN, Sezione di Catania, Catania, Italy 46 INFN, Sezione di Catania, Catania, Italy 47 INFN, Sezione di Lecce, Lecce, Italy 48 INFN, Sezione di Milano, Milan, Italy 49 INFN, Sezione di Napoli, Naples, Italy 50 INFN, Sezione di Roma “Tor Vergata”, Rome, Italy 51 INFN, Sezione di Torino, Turin, Italy 52 Istituto di Astroﬁsica Spaziale e Fisica Cosmica di Palermo (INAF), Palermo, Italy 52 Istituto di Astroﬁsica Spaziale e Fisica Cosmica di Palermo (INAF), Palermo, Italy 52 Istituto di Astroﬁsica Spaziale e Fisica 53 Osservatorio Astroﬁsico di Torino (INAF), Turin, Italy 54 Dipartimento di Scienze e Tecnologie Aerospaziali, Politecnico di Milano, Milan, Italy 54 Dipartimento di Scienze e Tecnologie Aerospaziali, Politecnico di Milano, Milan, Italy 55 Dipartimento di Matematica e Fisica “E. De Giorgi”, Università del Salento, Lecce, Italy 55 Dipartimento di Matematica e Fisica “E. De Giorgi”, Università del Salento, Lecce, Italy 56 Dipartimento di Scienze Fisiche e Chimiche, Università dell’Aquila, L’Aquila, Italy 56 Dipartimento di Scienze Fisiche e Chimiche, Università dell’Aquila, L’Aquila, Italy 57 Dipartimento di Fisica e Astronomia, Università di Catania, Catania, Italy 57 Dipartimento di Fisica e Astronomia, Università di Catania, Catania, Italy 58 Dipartimento di Fisica, Università di Milano, Milan, Italy ipartimento di Fisica “Ettore Pancini”, Università 59 Dipartimento di Fisica Ettore Pancini , Università di Napoli Federico II , Naples, Italy 60 Dipartimento di Fisica e Chimica ”E Segrè” Università di Palermo Palermo Italy 60 Dipartimento di Fisica e Chimica ”E. Segrè”, Università di Palermo 60 Dipartimento di Fisica e Chimica ”E. Segrè”, Università di Palermo, Palermo, Italy Dipartimento di Fisica e Chimica ”E. 36 Université Paris-Saclay, CNRS/IN2P3, IJCLab, Orsay, France Segrè”, Univ 61 Dipartimento di Fisica, Università di Roma “Tor Vergata”, Rome, Italy 62 61 Dipartimento di Fisica, Università di Roma “Tor Vergata”, Rome, Italy 6 62 Dipartimento di Fisica, Università Torino, Turin, Italy 62 Dipartimento di Fisica, Università Torino, Turin, Italy 63 Benemérita Universidad Autónoma de Puebla, Puebla, Mexico 63 Benemérita Universidad Autónoma de Puebla, Puebla, Mexico 64 Unidad Profesional Interdisciplinaria en Ingeniería y Tecnologías Avanzadas del Instituto Politécnico Nacional (UPIITA-IPN), Mexico, D.F., Mexico 64 Unidad Profesional Interdisciplinaria en Ingeniería y Tecnologías Avanzadas del Instituto Poli (UPIITA-IPN), Mexico, D.F., Mexico 64 Unidad Profesional Interdisciplinaria en Ingeniería y Tecnologías Avanzadas del Instituto Politécnico Nacional (UPIITA-IPN), Mexico, D.F., Mexico 95 Now at Max-Planck-Institut für Radioastronomie, Bonn, Germany 96 Now at School of Physics and Astronomy, University of Leeds, Leeds, UK (UPIITA-IPN), Mexico, D.F., Mexico 65 Universidad Autónoma de Chiapas, Tuxtla Gutiérrez, Chiapas, Mexico 65 Universidad Autónoma de Chiapas, Tuxtla Gutiérrez, Chiapas, Mexico 66 Universidad Michoacana de San Nicolás de Hidalgo, Morelia, Michoacán, México 66 Universidad Michoacana de San Nicolás de Hidalgo, Morelia, Michoa 67 Universidad Nacional Autónoma de México, Mexico, D.F., Mexico 67 Universidad Nacional Autónoma de México, Mexico, D.F., Mexico 68 Facultad de Ciencias Naturales y Formales, Universidad Nacional de San Agustin de Arequipa, Arequipa, Peru 69 Institute of Nuclear Physics PAN Krakow Poland iencias Naturales y Formales, Universidad Nacional de San Agustin de Arequipa, Arequipa, Peru 68 Facultad de Ciencias Naturales y Formales, Universidad Nacional de San Agustin de Arequipa, A 69 68 Facultad de Ciencias Naturales y Formales, Universidad Nacional de 69 Institute of Nuclear Physics PAN, Krakow, Poland 69 Institute of Nuclear Physics PAN, Krakow, Poland 70 Faculty of High-Energy Astrophysics, University of Łód´z, Łód´z, Poland 70 Faculty of High-Energy Astrophysics, University of Łód´z, Łód´z, Poland 70 Faculty of High-Energy Astrophysics, University of Łód´z, Łód´z, Poland 71 Laboratório de Instrumentação e Física Experimental de Partículas – LIP and Instituto Superior Técnico-IST, Universidade de Lisboa-UL, Lisbon, Portugal 71 Laboratório de Instrumentação e Física Experimental de Partículas – LIP and Instituto Superior T Universidade de Lisboa-UL, Lisbon, Portugal Universidade de Lisboa-UL, Lisbon, Portugal 72 “Horia Hulubei” National Institute for Physics and Nuclear Engineering, Bucharest-Magurele 73 Institute of Space Science, Bucharest-Magurele, Romania 73 Institute of Space Science, Bucharest-Magurele, Romania 74 University Politehnica of Bucharest, Bucharest, Romania sity Politehnica of Bucharest, Bucharest, Romania 74 University Politehnica of Bucharest, Bucharest, Romania 75 Center for Astrophysics and Cosmology (CAC), University of Nova Gorica, Nova Gorica, Slovenia 77 Universidad de Granada and C.A.F.P.E., Granada, Spain 78 77 Universidad de Granada and C.A.F.P.E., Granada, Spain ersidad de Granada and C.A.F.P.E., Granada, Spai 78 Instituto Galego de Física de Altas Enerxías (IGFAE), Universidade de Santiago de Compostela, Santiago de Compostela, Spain 78 Instituto Galego de Física de Altas Enerxías (IGFAE), Universidade de Santiago de Compostela, Santiago de Compostela, Spain 79 IMAPP, Radboud University Nijmegen, Nijmegen, The Netherlands 80 Nationaal Instituut voor Kernfysica en Hoge Energie Fysica (NIKHEF), Science Park, Amsterdam, The Netherlands 81 80 Nationaal Instituut voor Kernfysica en Hoge Energie Fysica (NIKHEF), Science Park, Amster 81 Stichting Astronomisch Onderzoek in Nederland (ASTRON), Dwingeloo, The Netherlands 82 Faculty of Science, Universiteit van Amsterdam, Amsterdam, The Netherlands 82 Faculty of Science, Universiteit van Amsterdam, Amsterdam, The Netherlands 83 83 Kapteyn Astronomical Institute, University of Groningen, Groningen, The Netherlands 84 84 Case Western Reserve University, Cleveland, OH, USA 84 Case Western Reserve University, Cleveland, OH, USA 85 Colorado School of Mines, Golden, CO, USA 85 Colorado School of Mines, Golden, CO, USA t of Physics and Astronomy, Lehman College, City University of New York, Bronx, NY, USA 86 Department of Physics and Astronomy, Lehman College, City University of New York, Br 87 87 Louisiana State University, Baton Rouge, LA, USA 87 Louisiana State University, Baton Rouge, LA, USA 88 Michigan Technological University, Houghton, MI, USA 88 Michigan Technological University, Houghton, MI, USA 89 New York University, New York, NY, USA 90 Pennsylvania State University, University Park, PA, USA 90 Pennsylvania State University, University Park, PA, USA 91 University of Chicago, Enrico Fermi Institute, Chicago, IL, 91 University of Chicago, Enrico Fermi Institute, Chicago, IL, USA 91 University of Chicago, Enrico Fermi Institute, Chicago, IL, USA p y y y 93 Department of Physics and WIPAC, University of Wisconsin-Madison, Madison, WI, USA 93 Department of Physics and WIPAC, University of Wisconsin-Madison, Madison, WI, U 94 Now at Fermi National Accelerator Laboratory, Fermilab, Batavia, IL, USA 12 3 99 Now at University of Bucharest, Physics Department, Bucharest, Romania 97 Now at Colorado State University, Fort Collins, CO, USA 99 Now at University of Bucharest, Physics Department, Bucharest, Romania Now at School of Physics and Astronomy, University of Leeds, Leeds, UK 97 Now at Colorado State University, Fort Collins, CO, USA 98 Now at Hakubi Center for Advanced Research and Graduate School of Science, Kyoto University, Kyoto, Japan 99 Now at University of Bucharest, Physics Department, Bucharest, Romania 95 Now at Max-Planck-Institut für Radioastronomie, Bonn, Germany 96 97 Now at Colorado State University, Fort Collins, CO, USA 98 Now at Hakubi Center for Advanced Research and Graduate School of Science, Kyoto University, Kyoto, Japan 99 Now at University of Bucharest, Physics Department, Bucharest, Romania 98 Now at Hakubi Center for Advanced Research and Graduate School of Science, Kyoto Univer Eur. Phys. J. C (2021) 81 :966 Page 25 of 25 966 123 123 123
https://openalex.org/W2775292870	https://aip.scitation.org/doi/pdf/10.1063/1.5004512	English	null	The evolution of in-plane magnetic anisotropy in CoFeB/GaAs(001) films annealed at different temperatures	AIP advances	2,017	cc-by	4,058	The evolution of in-plane magnetic anisotropy in CoFeB/GaAs(001) films annealed at different temperatures Special Collection: 62nd Annual Conference on Magnetism and Magnetic Materials Hongqing Tu; Ji Wang; Lujun Wei; Yuan Yuan; W. Zhang; Biao You; Jun Du AIP Advances 8, 056101 (2018) https://doi.org/10.1063/1.5004512 AIP Advances 8, 056101 (2018) https://doi.org/10.1063/1.5004512 RESEARCH ARTICLE \| MAGNETISM AND MAGNETIC MATERIALS \| DECEMBER 01 2017 Articles You May Be Interested In Element specific spin and orbital moments of nanoscale CoFeB amorphous thin films on GaAs(100) AIP Advances (September 2016) Investigations on magnetostatic interactions among CoFeB layers with perpendicular anisotropy in [Ta/CoFeB/MgO]N multilayers J. Appl. Phys. (March 2014) Thermally induced perpendicular magnetic anisotropy in CoFeB/MgO/CoFeB based magnetic tunnel junction AIP Conference Proceedings (May 2016) Element specific spin and orbital moments of nanoscale CoFeB amorphous thin films on GaAs(100) AIP Advances (September 2016) Thermally induced perpendicular magnetic anisotropy in CoFeB/MgO/CoFeB based magnetic tunnel junction 24 October 2024 04:09:10 Hongqing Tu,1,2 Ji Wang,1 Lujun Wei,1 Yuan Yuan,1 W. Zhang,1 Biao You,1,3 and Jun Du1,3,a (Presented 7 November 2017; received 13 September 2017; accepted 9 October 2017; published online 1 December 2017) A considerable in-plane uniaxial magnetic anisotropy (UMA) ﬁeld (Hu ∼300 Oe) could be achieved when the amorphous CoFeB ﬁlm was deposited on the GaAs(001) wafer by magnetron-sputtering after proper etch-annealed procedure. In order to get deep insights into the mechanism of the UMA, the ﬁlm was annealed at different temperatures and the evolution of the in-plane magnetic anisotropy was investigated carefully. With increasing the annealing temperature (TA), the UMA could be main- tained well when TA reached 250◦C, but became very weak at 300◦C. However, when TA was elevated to 400◦C, another UMA (Hu ∼130 Oe) was built accompanied with a fourfold magnetic anisotropy with its strength of about 50 Oe. In terms of the magnetic anisotropy evolution along with TA, the anelastic strain, which is thought to be resulted from the interfacial interaction between CoFeB and GaAs, may play a dominant role in producing the enhanced UMA based on the ‘bond-orientational’ anisotropy (BOA) model. © 2017 Author(s). All article content, except where oth- erwise noted, is licensed under a Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). https://doi.org/10.1063/1.5004512 24 October 2024 04:09:10 aAuthor to whom correspondence should be addressed. Electronic address: jdu@nju.edu.cn AIP ADVANCES 8, 056101 (2018) The evolution of in-plane magnetic anisotropy in CoFeB/GaAs(001) ﬁlms annealed at diﬀerent temperatures Hongqing Tu,1,2 Ji Wang,1 Lujun Wei,1 Yuan Yuan,1 W. Zhang,1 Biao You,1,3 and Jun Du1,3,a INTRODUCTION Ferromagnetic (FM) ﬁlms deposited on semiconductors (SCs) can often exhibit signiﬁcant uni- axial magnetic anisotropy (UMA), which has important applications in spintronic devices such as information storage and magnetic ﬁeld sensors. For example, when a ultrathin Fe,1–6 Co7–9 or FeCo10–12 ﬁlm was epitaxially deposited on GaAs(001) substrate, a sizable in-plane UMA ﬁeld (Hu) of ∼1 kOe could be easily achieved and its value depends strongly on the ﬁlm thickness.1 Although these phenomena have been generally observed by many research groups, the micro- scopic origin of the UMA remains a controversial topic. Two different ideas have emerged about the origin of the UMA: (i) magnetoelastic coupling due to the lattice mismatch between GaAs and the FM layer and resulting anisotropic strain or (ii) electronic hybridization between the FM layer and GaAs.1 Moreover, these FM ﬁlms often display fourfold magnetic anisotropy (FMA) at proper thicknesses in the ﬁlm plane, which is undoubtedly caused by the magneto-crystalline anisotropy (MCA). Because FMA is often superimposed with UMA, it impedes us to well understand the origin of UMA. Recently, to rule out the inﬂuence MCA, amorphous CoFeB material has been ﬁrst employed to deposit on GaAs(001) ﬁlm by A. T. Hindmarch et al13,14 and a signiﬁcant UMA (Hu ∼150 Oe) without FMA could be observed in the ﬁlm plane. In our recent work, an amazingly enhanced in-plane UMA (Hu ∼300 Oe) could be achieved when amorphous CoFeB ﬁlm was deposited on GaAs(100) © Author(s) 2017 2158-3226/2018/8(5)/056101/6 8, 056101-1 056101-2 Tu et al. AIP Advances 8, 056101 (2018) AIP Advances 8, 056101 (2018) AIP Advances 8, 056101 (2018) wafer substrate after proper etch-annealed procedure.15 In these studies, this kind of pure UMA was tentatively attributed to the anelastic strains induced by the CoFeB-GaAs interface according to the ‘bond-orientational’ anisotropy (BOA) model,14,16–18 rather than the Fe-As bonding charge transfer (BCT) model2,13 and Neel-Taniguchi (N-T) directional pair ordering model.19 Although a consensus has been reached that the UMA is interface-induced, the details of how the anelastic strains are seeded at the interface and produce the UMA accordingly are quite difﬁcult to be clariﬁed, even if some exquisite structural characterizations (e.g. transmission electron microscopy) are employed. Therefore, direct experimental evidences supporting the above UMA mechanism are still absent now. INTRODUCTION Fortunately, to our knowledge, the anelastic strain is temporarily nonrecoverable, for example, after removing stress but, in many cases, may be recoverable by annealing.14,18 If this argument is true, the UMA in the CoFeB/GaAs(001) ﬁlm, which is currently considered to result from the anelastic strains, will weaken remarkably or even vanish after proper annealing. In this presented work, in order to get deep insights of the UMA mechanism, the CoFeB/GaAs(001) ﬁlm owning enhanced UMA at the as-deposited state was annealed at different temperatures, and the evolution of the in-plane magnetic anisotropy along with the annealing temperature (TA) was investigated carefully. EXPERIMENTAL RESULTS The commercial GaAs(001) wafers were used with the major-ﬂat direction along [110] and the secondary-ﬂat direction along [1-10]. These wafers were diced into 10 mm × 10 mm pieces as sub- strates. Before deposition of the CoFeB ﬁlm, the substrate surface needs to be etched and cleaned by proper procedures.15,20 At room temperature, a 3.5 nm thick CoFeB ﬁlm was deposited on GaAs(001) substrates by dc magnetron sputtering at normal incidence from a commercial Co56Fe24B20 alloy target. Both cross-sectional transmission electron microscope (TEM) and x-ray diffraction (XRD) results show that the as-deposited CoFeB ﬁlm is amorphous. A Ta ﬁlm of 2 nm was deposited as capping layer to prevent the CoFeB ﬁlm from oxidation. The base pressure was lower than 8.0×10-6 Pa, and the Ar pressure was kept at 0.3 Pa during ﬁlm deposition. Then the sample was cut into several pieces, and each of them was annealed at a speciﬁc temperature for two hours with the ambient pressure lower than 1.0×10-5 Pa. The in-plane magnetic hysteresis (M-H) loop was measured by a SQUID-VSM (Quantum Design). To precisely determine the magnetic anisotropies of these ﬁlms, ferromagnetic resonance (FMR) measurements were carefully carried out with the frequency of the resonance cavity ﬁxed at 9.78 GHz. The in-plane azimuthal FMR spectra were recorded by varying the angle which describes the deviation from the [1-10] direction (0◦) in the ﬁlm plane. The surface morphology and the root-mean-square (rms) roughness of the ﬁlms were obtained by an atomic force microscope (AFM, Veeco Dimension V). It should be emphasized that all the measurements were carried out at room temperature (RT). 24 October 2024 04:09:10 RESULTS AND DISCUSSION Theoretically, the magnetization dynamics are generally studied by the Landau-Lifshitz-Gilbert (LLG) equation as dm dt = −γµ0m × Heff + αm × dm dt (1) (1) where m is the unit magnetization vector, and the Heff is the effective magnetic ﬁeld. Fi 1( ) h th di t t t d ib th FMR t t I th where m is the unit magnetization vector, and the Heff is the effective magnetic ﬁeld. Figure 1(a) shows the coordinate system to describe the FMR measurement geometry. In th case of enhanced in-plane UMA, the total energy per unit volume is modeled as where m is the unit magnetization vector, and the Heff is the effective magnetic ﬁeld. Figure 1(a) shows the coordinate system to describe the FMR measurement geometry. In the case of enhanced in-plane UMA, the total energy per unit volume is modeled as E = −H0Ms[sinθMsinθH cos (ϕH −ϕM) + cos θM cos θH] −(2πM2 S −KP)sin2θM −K2cos2ϕMsin2θM −K4 8 3 + cos 4 (ϕM −ϕu) sin4 (θM) (2) (2) Here,thedensitiesofZeemanenergy,effectivedemagnetizedenergy,in-planeuniaxialanisotropy energy, and fourfold anisotropy energy are described in sequence. Ms denotes the saturate magne- tization, KP, K2 and K4 denote the out-of-plane uniaxial, in-plane uniaxial and in-plane fourfold anisotropy energy constants, respectively. 056101-3 Tu et al. AIP Advances 8, 056101 (2018) FIG. 1. (a) The coordinate system illustrating the FMR measurement conﬁgurations. The AFM images illustrating the surface morphology of the CoFeB/GaAs(001) ﬁlm at as-deposited state (b) and annealed at 400◦C (c). FIG. 1. (a) The coordinate system illustrating the FMR measurement conﬁgurations. The AFM images illustrating the surface morphology of the CoFeB/GaAs(001) ﬁlm at as-deposited state (b) and annealed at 400◦C (c). Derived from Eq. (1) and Eq. (2), the precession frequency f and equilibrium equation can be expressed as21,22 f = γ 2π p HaHb (3) (3) 24 October 2024 04:09:10 H0 sin (ϕH −ϕM) = 1 2H2 sin 2ϕM + 1 4H4 sin 4 (ϕM −ϕu) , (4) (4) where Ha = H0 cos (ϕH −ϕM) + 4πMeff + H2cos2ϕM + 1 2H4 3 + cos 4 (ϕM −ϕu) , Hb = H0 cos (ϕM - ϕH) + H2 cos 2ϕM + H4 cos 4 (ϕM −ϕu), 4πMeff = 4πMS −HP. H2 and H4 denote the in-plane uniaxial and fourfold anisotropy ﬁeld constants, and the gyromagnetic ratio γ = guB/ℏ. RESULTS AND DISCUSSION Since only in-plane FMR measurements were carried out in the present study, θH = θM = 0◦. Therefore, the parameters such as H2, H4 and 4πMeff can be obtained by ﬁtting calculation according to the above equations. q In order to investigate the inﬂuence of annealing on the magnetic anisotropy, the as-deposited CoFeB(3.5 nm)/GaAs(001) ﬁlm was annealed with TA varying from RT to 400◦C. Before mag- netic measurements, the surface morphologies of all the samples were ﬁrst characterized by AFM. The AFM images for the as-deposited and the 400◦C-annealed samples are exhibited respectively in Fig. 1(b) and Fig. 1(c), which indicate that the surface morphologies of these two samples are much similar. Similar results could be also found in other samples with TA varying between RT and 400◦C. The rms roughness values calculated from the AFM results are 0.82 nm, 0.94 nm, 0.84 nm and 0.96 nm for the as-deposited, 250◦C-annealed, 300◦C-annealed and 400◦C-annealed samples, respectively. These results show unambiguously that the annealing with TA increased up to 400◦C does not have too much inﬂuence on the surface morphology of the CoFeB/GaAs ﬁlm. The M-H loops for all the CoFeB ﬁlms annealed at different temperatures were measured along two speciﬁc directions (i.e. [110] and [1-10]), which are shown in Fig. 2. Fig. 2(a) shows a rectangular loop along [110] combined with an obviously inclined but slightly hysteretic loop along [1-10], indicating a fairly well-deﬁned UMA with easy axis (EA) along [110] and hard axis (HA) along [1-10] in the as-deposited sample. This result can be kept well when TA is increased to 250◦C, which is veriﬁed by the loops shown in Fig. 2(b). However, with increasing TA further, the shape of the M-H loops changes signiﬁcantly. For examples, when TA reaches 300◦C, it seems that the EA changes to the HA and vice versa, as shown in Fig. 2(c). Meanwhile, the saturation ﬁeld of the M-H loop obtained at the HA decreases remarkably in comparison with that in Fig. 2(a) or Fig. 2(b), indicating obvious weakening of the UMA. When TA is increased to 400◦C, the EA switches AIP Advances 8, 056101 (2018) 056101-4 Tu et al. FIG. 2. The in-plane M-H loops for the CoFeB/GaAs(001) ﬁlm at as-deposited state (a), and annealed at 250◦C (b), 300◦C (c) and 400◦C (d). FIG. 2. RESULTS AND DISCUSSION The in-plane M-H loops for the CoFeB/GaAs(001) ﬁlm at as-deposited state (a), and annealed at 250◦C (b), 300◦ (c) and 400◦C (d). back to the [110] direction. In addition, double M-H loops can be observed at the [1-10] direction. The obvious variation of the M-H loops shown above indicates that the in-plane magnetic anisotropy has been altered after annealing with TA larger than 250◦C. In order to quantize the variation of the in-plane magnetic anisotropy ﬁelds, the in-plane azimuthal FMR spectra have been investigated carefully. back to the [110] direction. In addition, double M-H loops can be observed at the [1-10] direction. The obvious variation of the M-H loops shown above indicates that the in-plane magnetic anisotropy has been altered after annealing with TA larger than 250◦C. In order to quantize the variation of the in-plane magnetic anisotropy ﬁelds, the in-plane azimuthal FMR spectra have been investigated carefully. 24 October 2024 04:09:10 The experimental angular dependences of resonance ﬁeld (Hr) for all the samples mentioned in Fig. 2 are shown by the square dots in Fig. 3. By eye inspected, the experimental Hr ∼ϕH curve shows slight or serious deviation from standard 180◦-symmetry, suggesting that the in-plane UMA is not perfectly deﬁned for each sample. Therefore, besides the UMA, FMA has to be taken into account for ﬁtting. According to Eq. (3) and Eq. (4), the experimental angular dependence of Hr can be all well ﬁtted, as shown by the red curves in Fig. 3. By ﬁtting calculation, the val- ues of H2, H4, 4πMeff and the directions of EAs for UMA and FMA are obtained and listed in Table I. For the as-deposited sample, H2 and H4 are about 280 Oe and 10 Oe respectively, indicating a signiﬁcant UMA accompanied with an extremely weak FMA. Note that the EA for UMA is along For the as-deposited sample, H2 and H4 are about 280 Oe and 10 Oe respectively, indicating a signiﬁcant UMA accompanied with an extremely weak FMA. Note that the EA for UMA is along FIG. 3. The in-plane angular dependent resonance ﬁeld for the CoFeB/GaAs(001) ﬁlm at as-deposited state (a), and annealed at 250◦C (b), 300◦C (c) and 400◦C (d). The black dots and red lines represent the experimental and ﬁtted results, respectively. FIG. 3. RESULTS AND DISCUSSION the EA of FMA is along <100> direction, which is 45◦deviated from that (<110>) of FeCo alloy ﬁlm, suggesting that CoFeB has not been crystallized to FeCo yet after annealed at 300◦C. Similar results (not shown here) could be also observed when TA is elevated to 350◦C. We consider that the anelastic strain, which may play dominant role in the UMA for the as-deposited sample based on the BOA model,16–18 has been recoverable by annealing at appropriate temperatures. There- fore, the initial UMA with EA along [110] is eliminated accordingly. As for the rebuilt UMA with EA along [1-10] and the accompanied FMA with EA along <100>, they may be originated from the interfacial bonding between As and Fe (Co) after the surface of GaAs(001) is reconstructed upon proper annealing, which needs further studies such as elaborate characterization of interfacial microstructure. RESULTS AND DISCUSSION Therefore, it is reasonable to consider that the amorphous CoFeB ﬁlm has almost completely crystallized to FeCo alloy ﬁlm with Boron atoms separated out after annealed at 400◦C. 24 October 2024 04:09:10 As for the 300◦C-annealed sample, H2 and H4 are 35.6 Oe and 43.3 Oe respectively, indicating comparably stronger UMA and FMA. These results indicate that the UMA has weakened about an order of magnitude (from about 300 Oe to 36 Oe) in comparison with those of the as-deposited or 250◦C-annealed samples. More interestingly, the EA (HA) of UMA has switched from [110] to [1-10] (from [1-10] to [110]), in agreement with the M-H loops shown in Fig. 2(c). According to our literature survey, when Fe, Co or FeCo alloy ﬁlm was grown on GaAs(001) substrate, the EA (HA) of UMA unexceptionally pointed to the [110] ([1-10]) direction.1,3,4,6,9–12 On the other hand, As for the 300◦C-annealed sample, H2 and H4 are 35.6 Oe and 43.3 Oe respectively, indicating comparably stronger UMA and FMA. These results indicate that the UMA has weakened about an order of magnitude (from about 300 Oe to 36 Oe) in comparison with those of the as-deposited or 250◦C-annealed samples. More interestingly, the EA (HA) of UMA has switched from [110] to [1-10] (from [1-10] to [110]), in agreement with the M-H loops shown in Fig. 2(c). According to our literature survey, when Fe, Co or FeCo alloy ﬁlm was grown on GaAs(001) substrate, the EA (HA) of UMA unexceptionally pointed to the [110] ([1-10]) direction.1,3,4,6,9–12 On the other hand, the EA of FMA is along <100> direction, which is 45◦deviated from that (<110>) of FeCo alloy ﬁlm, suggesting that CoFeB has not been crystallized to FeCo yet after annealed at 300◦C. Similar results (not shown here) could be also observed when TA is elevated to 350◦C. We consider that the anelastic strain, which may play dominant role in the UMA for the as-deposited sample based on the BOA model,16–18 has been recoverable by annealing at appropriate temperatures. There- fore, the initial UMA with EA along [110] is eliminated accordingly. As for the rebuilt UMA with EA along [1-10] and the accompanied FMA with EA along <100>, they may be originated from the interfacial bonding between As and Fe (Co) after the surface of GaAs(001) is reconstructed upon proper annealing, which needs further studies such as elaborate characterization of interfacial microstructure. RESULTS AND DISCUSSION The in-plane angular dependent resonance ﬁeld for the CoFeB/GaAs(001) ﬁlm at as-deposited state (a), and annealed at 250◦C (b), 300◦C (c) and 400◦C (d). The black dots and red lines represent the experimental and ﬁtted results, respectively. AIP Advances 8, 056101 (2018) 056101-5 Tu et al. 056101-5 Tu et al. TABLE I. All the parameters obtained by ﬁtting calculations for the CoFeB/GaAs(001) ﬁlm at as-deposited state and annealed at 250◦C, 300◦C and 400◦C. TABLE I. All the parameters obtained by ﬁtting calculations for the CoFeB/GaAs(001) ﬁlm at as-deposited state and annealed at 250◦C, 300◦C and 400◦C. Temperature H2(Oe) H2 (EA) H4 (Oe) H4(EA) 4πMeff(Oe) TABLE I. All the parameters obtained by ﬁtting calculations for the CoFeB/GaAs(001) ﬁlm at as-deposited state and annealed at 250◦C, 300◦C and 400◦C. Temperature H2(Oe) H2 (EA) H4 (Oe) H4(EA) 4πMeff(Oe) as-deposited 278.9 [110] 9.6 <100> 10111 250◦C 311.1 [110] 11.1 <100> 10067 300◦C 35.6 [1-10] 43.3 <100> 8236 400◦C 131.1 [110] 49.7 <110> 8055 [110] while along the <100> directions for FMA. These features are in good accordance with those of the M-H loops shown in Fig. 2(a). The very weak FMA may be resulted from a small quantity of Fe, Co or FeCo nanocrystals resided in the CoFeB layer. When TA is increased up to 250◦C, the magnetic anisotropy of the as-deposited sample can be kept well with a little enhancement of UMA (∼311 Oe) and FMA (H4 ∼11 Oe). However, when TA is increased to be 300◦C or even larger, the in-plane magnetic anisotropy becomes very complicated. Let us ﬁrst discuss the results for the 400◦C-annealed sample. As displayed in Table I, H2 and H4 are about 131 Oe and 50 Oe respectively, indicating that the UMA and FMA are both considerable in this sample. More importantly, although the EA for UMA keeps along the [110] direction, the EA for FMA has switched to the <110> direction, which is 45◦rotated away from that of the as-deposited sample. All these results, including the direction of EA and the strength of UMA/FMA, coincide well with those of Fe34Co66 alloy ﬁlm grown on GaAs(001) substrate when its thickness is larger than six monolayers (∼1.2 nm).12 Similar results can be also found in other FeCo alloy ﬁlms with similar composition (e.g. Fe31Co69)11 grown on the same substrate. CONCLUSIONS In summary, we have demonstrated the evolution of in-plane magnetic anisotropy of the CoFeB/GaAs(001) ﬁlm along with TA. The UMA ﬁeld is close to 300 Oe in the as-deposited sample and kept well at TA = 250◦C. However, it decreases one order of magnitude when TA is elevated to 300◦C. Meantime, its EA is rotated 90◦away from the initial direction. When TA is increased to 400◦C, the in-plane magnetic anisotropy is much similar to that of a CoFe ﬁlm deposited on GaAs(001) substrate, indicating that the amorphous CoFeB ﬁlm has already crystallized to CoFe alloy ﬁlm. These results support the argument that the anelastic strain plays a dominant role in the UMA for as-deposited CoFeB on GaAs(001). 056101-6 Tu et al. AIP Advances 8, 056101 (2018) AIP Advances 8, 056101 (2018) ACKNOWLEDGMENTS This work was supported by National Key Research and Development Program of China (2016YFA0300803), National Basic Research Program of China (2014CB921101), National Natural Science Foundations of China (Nos. 51471085, 51331004, 11174131), Jiangsu NSF (BK20140054). This work was supported by National Key Research and Development Program of China (2016YFA0300803), National Basic Research Program of China (2014CB921101), National Natural Science Foundations of China (Nos. 51471085, 51331004, 11174131), Jiangsu NSF (BK20140054). G. Bayreuther, J. Premper, M. Sperl, and D. Sander, Phys. Rev. B 86, 054418 (2012). 1 G. Bayreuther, J. Premper, M. Sperl, and D. Sander, Phys. Rev. B 86, 054418 (2012). 2 2 J. J. Krebs, B. T. Jonker, and G. A. Prinz, J. Appl. Phys. 61, 2596 (1987). 3 3 D. M. Gillingham, M. Tselepi, A. Ionescu, S. J. Steinmuller, H. E. Beere, D. A. Ritchie, and J. A. C. Bland, Phys. Rev. B 76, 214412 (2007). E. M. Kneedler, B. T. Jonker, P. M. Thibado, R. J. Wagner, B. V. Shanabrook, and L. J. Whitman, Phys. Rev. B 56, 8163 (1997). y , , p , , , y , ( ) A. Ionescu, M. Tselepi, D. M. Gillingham, G. Wastlbauer, S. J. Steinm¨uller, H. E. Beere, D. A. Ritchie, and J. A. C. Bland, Phys. Rev. B 72, 125404 (2005). 7 E. Gu, M. Gester, R. J. Hicken, C. Daboo, M. Tselepi, S. J. Gray, J. A. C. Bland, and L. M. Brown, Phys. Rev. B. 52, 14704 (1995). 8 S. J. Blundell, M. Gester, J. A. C. Bland, C. Daboo, E. Gu, M. J. Baird, and A. J. R. Ives, J. Appl. Phys. 73, 5948 (1993) 9 9 Y. Z. Wu, H. F. Ding, C. Jing, D. Wu, G. L. Liu, V. Gordon, G. S. Dong, and X. F. Jin, Phys. Rev. B 57, 935 (1998). 10 L. C. Chen, J. W. Dong, B. D. Schultz, C. J. Palmstrom, J. Berezovsky, A. Isakovic, P. A. Crowell, and N. Tabat, J. Vac. Sci. Technol. B 18, 2057 (2000). 11 F. Bianco, P. Bouchon, M. Sousa, G. Salis, and S. F. Alvarado, J. Appl. Phys. 104, 083901 (2008). 12 12 M. Dumm, M. Zolﬂ, R. Moosbuhler, M. Brockmann, T. Schmidt, and G. Bayreuther, J. Appl. Phys. 87, 5457 (2000). 13 A. T. Hindmarch, C. J. Kinane, M. MacKenzie, J. N. Chapman, M. Henini, D. Taylor, D. A. Arena, and J. Dvorak, Phys. Rev. Lett. 100, 117201 (2008). A. T. 1 G. Bayreuther, J. Premper, M. Sperl, and D. Sander, Phys. Rev. B 86, 054418 (2012). 2 S. Iihama, S. Mizukami, H. Naganuma, M. Oogane, Y. Ando, and T. Miyazaki, Phys. Rev. B 89, 174416 (2014). 2 J. J. Krebs, B. T. Jonker, and G. A. Prinz, J. Appl. Phys. 61, 2596 (1987). 3 S. Iihama, Q. Ma, T. Kubota, S. Mizukami, Y. Ando, and T. Miyazaki, Appl. Phys. Exp. 5, 083001 (2012). ACKNOWLEDGMENTS Hindmarch, A. W. Rushforth, R. P. Campion, C. H. Marrows, and B. L. Gallagher, Phys. Rev. B 83, 212404 (2011). H. Q. Tu, B. You, Y. Q. Zhang, Y. Gao, Y. B. Xu, and J. Du, IEEE Trans. Magn. 51, 2005104 (2015). Q Q g g ( ) 16 T. Egami, C. D. Graham, W. Dmowski, P. Zhou, and P. J. Flanders, IEEE Trans. Magn. 23, 2269 (1987) 17 17 Y. Suzuki, J. Haimovich, and T. Egami, Phys. Rev. B 35, 2162 (1987). 18 24 October 2024 04:09:10 18 X. Yan, M. Hirscher, and T. Egami, Phys. Rev. B 43, 9300 (1991). 18 X. Yan, M. Hirscher, and T. Egami, Phys. Re 19 L. N´eel, J. Phys. Radium. 15, 225 (1954). 19 L. N´eel, J. Phys. Radium. 15, 225 (1954). y ( ) 20 G. A. Baca, in Fabrication of GaAs Devices, Institution of Engineering and Technology, London, UK, 2009, p. 66. 21 20 G. A. Baca, in Fabrication of GaAs Devices, Institution of 21 21 S. Iihama, Q. Ma, T. Kubota, S. Mizukami, Y. Ando, and T. Miyazaki, Appl. Phys. Exp. 5, 083001 (2012). 22 1 S. Iihama, Q. Ma, T. Kubota, S. Mizukami, Y. Ando, and T. M 2 22 S. Iihama, S. Mizukami, H. Naganuma, M. Oogane, Y. Ando, and T. Miyazaki, Phys. Rev. B 89, 174416 (2014).
W4239651846.txt	https://figshare.com/articles/thesis/From_EOQ_to_JIT_with_Storage_Consideration_Coordinating_a_Two_Level_Supply_Chain/14647002/1/files/28127163.pdf	en		From EOQ to JIT with Storage Consideration: Coordinating a Two Level Supply Chain	null	2,021	cc-by	18,049	NOTE TO USERS This reproduction is the best copy available. ® UMI Reproduced with perm ission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FROM EOQ TO JIT WITH STORAGE CONSIDERATION: COORDINATING A TWO LEVEL SUPPLY CHAIN by Yohan Jeju John Bachelor of Technology in Mechanical Engineering University of Kerala, Thin •" ananthapnram, Kerala, India 1994 Master in Business Administration University of Wales, Cardiff, United Kingdom 1997 A thesis Presented to Ryerson University in partial fulfillment of the requirements for the degree of Master of Applied Science in the Program of Mechanical Engineering Toronto, Ontario, Canada, 2004 © \ ohan Jeju John 2004 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number; EC52940 INFORMATION TO USERS The quality of this reproduction is dependent upon the quaiity of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. UMI UMI Microform EC52940 Copyright 2008 by ProQuest LEG. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. ProQuest LLC 789 E. Eisenhower Parkway PC Box 1346 Ann Arbor, Ml 48106-1346 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. BORROW ER’S PAGE Ryerson University requires the signatures of all persons using or photocopying this thesis. Please sign below, and give address and date. Ill Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT FROM EOQ TO JIT WITH STORAGE CONSIDERATION: COORDINATING A TWO LEVEL SUPPLY CHAIN by Yohan John Many organizations are faced with a decision to choose between two inventory systems namely JIT (Just in Time) and EOQ (Economic Order Quantity). This thesis models the cost drivers into the EOQ model and extends it to the JIT scenario. They include cost savings like space, synergy of coordination, and other cost factors like rework and penalty costs. It looks at the total cost of the supply chain with two players and calculates space in terms of storage spaces of equal capacity. Results sho wed that considering space in EOQ brought savings to the chain. It has brou^it down the order quantity closer to, and many times equal to JIT ordering quantities. Coordination in the chain has brought further savings. Moving to JIT (ordering daily supply of demand) from the point, where space is accounted and there is coordination between the two levels, did not require much reduction in ordering costs. IV Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS Author’s Declaration........................................................................................... ii Borrower’s Page.......................................................................... iii Abstract................................................................................................................iv Table o f Contents..................................................................................................v List of Figures & Tables.................................................................................... vii Acknowledgments............................................................................................ viii Chapter l_Introduction and Literature Review..............................................1 1.1 Research Background............................................................................................ 1 1.2 Overview of Economic Order Quantity................................................................ 1 1.3 Overview of Just In Time (JIT)............................................................................. 2 1.4 EOQ/jrr Model Consideration of Inventory........................................................6 1.5 Coordination in Supply Chain............................................................................. 17 1.6 Profit Sharing....................................................................................................... 19 1.7 Objectives and Scope of Thesis...........................................................................20 1.8 Thesis Layout.......................................................................................................20 Chapter 2_Problem Description and Mathematical Model........................21 2.1 Problem Objectives............................................................................................. 21 2.2 Nomenclature...................................................................................................... 24 2.3 Mathematical Model........................................................................................... 25 2.4 Optimal Solution Procedure............................................................................... 41 Chapter 3_Numerical Results and Statistical Analysis...............................47 V Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3.1 Numerical Example..............................................................................................47 3.2 Statistical Analysis....................................................................................... Chapter 4_Conclusion and Future Research............................................... 60 References........................................................................................................... 63 Appendices.........................................................................................................66 VI Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 55 LIST OF FIGURES & TABLES List of Figures Number Poge Figure 2.1 : Two level supply chain matrix 23 Figure 2.2: Inventory cycle of the retailer 27 Figure 2.3 : Inventory cycle of the supplier 31 Figure 2.4: Optimal solution procedure for no coordination 42 Figure 2.5: Optimal solution procedure for coordination 42 List o f Tables Number Table 3.1 : Value of Q obtained for each X, from equation (2.2.4) Poge 50 Table 3.2: Value of Q obtained for each X from equation (2.4.3) 52 Table 3.3: Summary table of the total cost for all scenarios 55 Table 3.4: Summary averages from computational trial 57 vu Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGMENTS I would like to thank Dr. M. Y. Jaber for his consistent and valuable input to this thesis. His patience, encouragement and guidance throughout the duration of tire thesis were substantial. His comments and suggestions were o f great importance and have helped me in the completion of the thesis. I would further like to sincerely appreciate Dr. Jaber for adjusting his schedule to meet with me out o f the way in order to help me complete this thesis. I would also like to extend my gratitude to Dr. Greg Kawall, the Chair of the examining committee and its members. Dr. M. Y. Jaber, Dr. Ahmed El-Bouri and Dr. Saeed Zolfaghaii for their valuable suggestion to improve the thesis. I would also thank my parents and my in-laws for the moral support, encouragement and valuable prayers for me during this time. The members of my church, “Zion Gospel Assembly”, were particularly helpful when I needed their support and prayers. Many helped by taking care of my son Jonathan when my wife and I could not juggle our schedules. Most of all I would like to thank my Lord and Savior Jesus Christ who helped me in every situation and worked it for my good. Finally, this thesis would not have come to completion but for the consistent and constant encouragement and prayers of my wife and dear friend Serene John and the delight that my son Jonathan brings me. Thank you. vm Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C hapter 1 INTRODUCTION AND LITERATURE REVIEW 1.1 Research Background In a competitive environment, corporations face many challenges to survive. They are constantly looking for ways to bring value to their share holders. Analysts judge the success o f an organization through various measures like earnings per share, net earnings before interest and tax, and a host o f ratios like the acid test or quick ratio, debt to equity, current ratio and the likes. Many of these ratios pertain to inventory like inventory turnover, receivables ratio, days in inventory and so on. Managing inventory and the associated costs properly can produce significant and positive impact on these ratios. When it comes to inventory, a few important decisions need to be made like, how much needs to be ordered and how often it needs to be ordered. A delicate balance needs to be struck between the cost o f carrying this inventory and the cost of ordering or setup (in case o f manufacturing). This quantity is referred to as the Economic Order Quantity (EOQ) or the Economic Manufacturing Quantity (EMQ). For the sake of consistency in this thesis, the term Economic Order Quantity will used. 1.2 Overview of Economic Order Quantity The EOQ model in its basic form has essentially two components, namely, the order costs and the holding costs. The order cost is often referred to as ordering cost when the context is that of retailer or buyer. This component of the total cost is the 1 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cost incurred for a particular order or lot (in case of manufacturing). This is fixed and is incurred for every order, irrespective of the quantity ordered. Thus, if the quantity ordered is less, more orders need to be placed to fill the demand. This causes the order cost to increase. In order to reduce the impact o f setup on the cost, large quantities are ordered on each order and thus the number o f times it is ordered per year is reduced. However, this causes another problem, since large quantities increase the holding cost. The holding cost is cost associated with keeping inventory. This has many components to it like insurance on inventory, property tax, obsolescence, spoilage, shrinkage, utilities, interest, handling, etc. This component increases in proportion with the quantity stored. This component works against the drive to redr ce the ordering fi-equency and order large quantities. Numerous research papers have been published ever since the concept of Economic Order Quantity (EOQ) had been introduced by Ford Harris in 1915. The problem, however, is that EOQ in its simplest form leads to a lot of miscalculations on the optimum lot size problem (Jones 1991). Jones (1991) argued that if EOQ has properly accounted for the costs it would lead to Just In Time (JIT) lot sizes. 1.3 Overview of Just In Time (JIT) The recurring theme for Just In Time is to eliminate waste and improve flow of materials (Fuller 1995). This means that the adoption of JIT requires cutting down or eliminating inventoiy that is not needed. Orders are placed on an immediate need i Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. basis and enough for a day or till next delivery. In a manufacturing environment this means that orders are processed on a pull basis instead o f push.. To make it worth while to order or manufacture in small quantities or batches, the setup costs have to be brought down or holding costs have to be higher. The following two sections discuss these in detail. 1.2.1 Setup/ Ordering Costs One o f the key components to moving the basic EOQ model towards JIT is to reduce the setup cost. This can drive the quantity down as well as bring tire total cost down besides bringing the benefits of operating in JIT. The JIT literature identifies a variety of ways to bring down the setup cost, namely, the need for inspection or setup of inspection station for incoming stock is eliminated, no annual re-bidding or re-tendering, paper work reduced and a more informal ordering process, long term contracts established, etc. One o f the suggested ways in the literature (Pan & Liao 1989) to reduce the order cost component o f the cost is to increase deliveries for a particular order. Thus, the cost is split over a number of deliveries. Pan & Liao (1989) argue that the purchase cost does not necessarily increase when going to JIT, since the total order is the same but it is just that the fi'equency of deliveries increases. One of the issues that are ignored in this assumption is that costs for such deliveries have to be accounted somewhere. If the vendor bears this cost it would be passed on to the retailer as cost increase. The other problem is based on the reaction o f the vendor to the decision of the retailer. There are two scenarios here: one, the vendor also decides to move to Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. JTT and two, the vendor decides to continue to produce in EOQ. In the former scenario, costs are incurred from the purchase of raw materials (Fazel et al 1998), increased inspection to deliver zero defects to the retailer and so on. In the latter scenario, the vendor has to hold the inventory till it is shipped. This situation does not help in zero defect purchases. To ensure zero defects, the vendor has to increase the costs associated with inspection. In either case scenario, there would be a cost increase that would be transmitted to the retailer in terms of price increase. Ramasesh (1990) also approaches the problem in terms o f blanket orders with multiple shipments like Pan & Liao (1989). However, Ramasesh (1990) includes the cost o f multiple shipments in the model, namely, the freight cost associated. In a way this is a cost that needs to be considered. But it is often the case that the purchase cost includes the cost o f landing the product at the premises. This supports the argument that the price increases when moving towards JTT. Another factor that needs to be considered is that many organizations that purchase from a variety of sources optimize on their own transportation network by combining different loads from vendors in the same locality. This helps in bringing the shipment cost considerably lower. In such circumstances, the cost associated would be calculated in terms o f the volume o f the product rather than the number of shipments. Thus it is possible to bring order cost down by increasing the number of shipments per order, but this comes at a cost in terms of increased price for the product. Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission. 1.2.2 Holding Cost The other aspect that needs to be considered in moving towards JTT is the holding cost. Billington (2003) considered the total cost reduction in lot sizing by reducing the holding cost. The author argued that it is possible to reduce the holding cost by investing in automating the factory thereby reducing handling costs. Further, capital investments can reduce the cost of obsolescence and spoilage. Billington (2003) found that reducing holding cost does not necessarily bring the total cost down. The investment needed to reduce the holding cost should be considered. However, Billington (2003) he did not consider the impact of cost of storage for these additional units. Reducing tlie holding cost leads to increase in lot size which in turn would impact the storage cost. It can be argued that cost saved in reducing the holding cost is offset by the cost incurred in increased storage space. This, however, is very dependant on the size of product. Small but expensive products can bring savings if holding cost is reduced. It is also worth mentioning that the handling costs o f small products would be relatively small compared to larger ones so also the savings that can be obtained. Further, in this thesis, the cost associated with obsolescence, spoilage or returns has been considered separately. This is due to the fact that these items, although they constitute a small percentage, have other costs like penalty costs, opportunity costs and storage costs associated with them. Thus it is essential for them to be considered separately. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. On the other hand, some authors (e.g. Jones 1991) have argued that the holding costs used in the standard EOQ equation do not reflect the actual costs incurred for holding the inventory. Costs like facilities leasing costs, depreciation, interest, taxes, insurance, utilities, handling costs, inspection costs, rework, scrap, and administration costs are not considered in the holding cost function. Considering these factors as well as reducing the set up costs can reconcile JIT with EOQ. While it can be accepted that these costs are not included in the holding costs, it can also be argued that these are often not a direct linear ftmction of the lot size quantity and therefore need to be separated from the holding cost function. Functions like facilities costs and rework have been accounted for separately in this thesis. Including these factors in the lot sizing equation has brought the lot size closer to jrr. The other factors that are not a linear function are space cost and rework cost. Rework is often part of the EOQ process. When JIT is implemented, tlie approach is to eliminate rework. This cost is often neglected in the analysis when considering the cost of inventory. The model proposed considers the impact of this quantity. The following section looks in to the literature specifically pertaining to space. 1.4 E O Q /JIT Model Consideration of Inventory Many authors have discussed and modeled space in the classical EOQ equation. In this section, a critical analysis on these papers is done followed by a detailed look into a few articles. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Among the authors that worked on space is Aghazadeh (2001), who illustrated that the best course for a retailer is to opt for a quantity discount model that will result in the lowest cost. The author argues that JTT philosophy is not applicable in a retailer situation. The example of Wal Mart is quoted where they use distribution centers to store inventory rather than at the individual stores. The holding cost of the distribution stores is presumably lower than that of the individual stores. It is also argued that holding inventory on a JIT basis could result in stock outs which would prove very expensive for a retailer. Through the illustration, the author concludes that taking advantage of the price quantity discount is much cheaper that ordering in JIT quantities. The model that was used, however, has one major drawback. It does not consider space as a cost factor. The holding cost only includes the carrying cost. It does not consider the cost o f warehousing these items. The author considers coffee filter as an example which admittedly is an item with low volume, but retailers do not deal with just coffee filters. Many of the products are of high volume and sometimes low profit margin. A typical example of this will be ice salt or water conditioning salts that cost very low but has a high volume. Space needs to be considered in determining the EOQ for these situations. Further, storing these filters for a long period of time (one year in the example quoted) could lead to spoilage or shrinkage or even obsolescence. This need not be the case with coffee filters if packaged properly but many products caimot be stored for this long. A closer look into the work of some of the authors is done in the following sections. 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1.4.1 Joshi In his article Joshi (1990) discusses the case where bulky, inexpensive and low risk items (BIL) are incorporated in the EOQ equation. The author contends that the holding costs look at all items in the same manner irrespective of the volume the item occupies. It is mentioned that the bulky items considered do not require bins or can be stacked in single or multi-story storage rack systems. They are stored in pallets and sometimes can be double stacked. If is the fixed component of the order cost, D is the annual demand, Q is the order quantity, h is the holding cost and SC is the annual space cost, the total cost is given by T C := D — + ^ + SC g 2 (l.l) The storage cost component in this situation is given as SC = p x k x N Where N = ~ mxn n is the number o f units that fit on a pallet, m is the number of pallets that can be stacked one over the other, p is the area occupied by one pallet and k is the storage cost per square foot. The author considers two scenarios. One is where the space is fixed assignment per product and the other is when the assignment is dynamic. Fixed assignment means that space for Q units is allocated for this product and not utilized for any other products. For the dynamic assignment the space utilization is far better. The space 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. that is available is allocated for other products. This is what is done in many warehouses where the space allocation is computerized. The author argues that due to the efficient utilization of space the annual requirement would be that for •y units. The space cost therefore in this scenario would be denoted as SC = p x k x N SC = { p x k ) x f O ^ \2xmxn^ Thus the EOQ for the two scenarios are given as EOQ Space (1.2) ( Fixed ) mn EOQ Space (1.3) ( Dynamic ) mn The article is limited in its application that the author considers only floor space where pallets are placed. This poses a problem since most large warehouses rely on bins and multi-story storage rack system for most of their products. Further space is a very valuable commodity that irrespective of the kind of the product space should be considered in the calculation of the EOQ. The other quesi.u. • that this article raises is the consideration of space as a continuous function. It is a common practice that inventory is not mixed in bins or locations. If the EOQ is calculated as a little over a bin, the space actually occupied would be that of two bins. Thus the term N = — should be an integer and m xn Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. f o ^ should in actuality read as iV = Int ——— t-1 .In the example quoted by the author, ym-x-n y the maximum units that can be accommodated in a pallet (double stacked) is 40 units. But if the EOQ results in recommending 41 units then the space occupied would be that of 80 units. The error is higher for fixed space allocation than for dynamic. Because of the nature of the bins it would be beneficial to use equation 1.2 and 1.3 to get an approximation and then iteratively calculating the total cost for the range of ± wi x n . In the calculation the author did not utilize the model but iteratively calculated all the multiples of the space capacity to obtain the optimum solution. Adopting this method fiirther neglects the impact of holding cost. 1.4.2 Rao & Bahari-Kashani Rao & Bahari-Kashani (1990) took a similar approach to accounting for space in their model. In their model they considered the space fimction to be given as = (1.4) > 1 Where Sj is the fixed cost for storage of the y-th unit per year, P is the number of available storage units and fo r 1 0 J = l Co P, Otherwise, Where C, is the capacity of storage i. The consideration of space in this way would flexible to include multiple storage locations with various size and corresponding costs. 10 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The total cost is given by C ( Q ) = ^ + h ^ + SC (1.6) The authors consider storage space to be fixed and hence the fihst derivative eliminates SC (Storage Cost). Thus EOQ G ' f f ( .7 ) The author concludes that the optimum cost can be obtained by computing the maximum capacity of fewer than the number o f storage areas needed for Q to obtain the optimum. Considering space as a constant in calculating the EOQ brings a lot of inaccuracies. The problem being that storage cost is directly dependant on the value of Q although it is a stepped fimction. The author also suggests that the maximum capacity of the storage areas be considered and the total cost computed and compared for the optimiun. Considering only the maximum capacity of the storage areas negates the influence of the holding cost and ordering cost if any on the EOQ. The optimum quantity could be less than a full storage capacity. 1.4.3 Fazel Fazel (1997) and Fazel et al (1998) compared the cost of EOQ to that of JIT. Fazel (1997) considered EOQ with no price quantity discount and Fazel et al (1998) with price quantity discount. The price quantity discount considered is an all unit quantity discount. The attempt in both the papers was to model the costs to aid decision makers regarding the move to JIT form EOQ. 11 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Since both the papere have a lot of similarities, the latter article will be the subject of study. In this paper the total cost for EOQ is given as T C ,= ^ * ^ * ( c \-n ^ Q ) D fo r TQ = ^ +^ + (C-- )D fo r (1,8) Q<Q, (1.9) Q>Q. Where ce is the cost of one unit of product when ordering in EOQ, Qmax is the order quantity beyond which there would not be any further price quantity discount and tte is the quantity discount rate. The price quantity discount is an all units quantity discount that is considered, thus the price changes as follows with the increase in Q as follows: Ce = 4 ~ ' ^ eQ M Q^Qnuuc Ce=cTM Where, is the price of the product for Q = Q, and is the fixed price when the quantity exceeds a certain level (Qmoi) The optimum quantity for this would be the minimum total cost for either (m o ) or 2** = ^ f o r Q->Q. (l.ll) or that of <2ff 12 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The author assumes in the article that the total cost for operating in JIT is a simple form of the product of JIT price and demand as follows: TCj =Cj D (1.12) Here cj is the JTT cost o f the product. It is argued that the supplier would raise cost due to the frequent deliveries inspection and other costs involved in moving to JTT or JTT deliveries. The paper then narrows down the application of the model to where Q' < and derives the cost difference for JIT and EOQ as 2 = ^ +^ + ( c ° .- it,Q 'p - C jD (1.13) Fazel et al (1998) concluded that for lower demands it would be cheaper to use JIT instead of EOQ and for higher demands it would be worthwhile to use EOQ. The indifference point in demand was also calculated. This is the demand at which it does not make any difference whether the EOQ or JIT is adopted. This means that the value of Z is zero at this point. The model described in both the articles has some drawbacks. Like Schniederjans & Cao (2000) pointed out, it ignores the space saved by the fact that the batch size is very small. The space savings should be considered as a major cost savings. This is especially the case when the products involved are voluminous. This point is further elaborated on the upcoming sections. The other assumption that the author considers is that there is no holding costs and ordering (setup) cost. It is argued that these costs are largely reduced and thus can 13 ^ PROPERTY OF IBRAffr R'tTBSCW UrJiVERSiTY LlBRAfl Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. be neglected. This is flawed since ordering in JIT is many a times characterized by long term contracts. To establish long term contracts more resources are spent and thus this cost is high, although it is spread over many deliveries. Thus to assume that there is no ordering cost would lead to erroneous decisions. The other aspect that is ignored is that of the holding cost. Here again since the quantities are small the holding cost is assumed as negligible. Although, there is some truth to this assumption, the error compounds when a huge retailers warehouse containing 40,000 different products is considered. The holding cost can also be high when the expensive items are considered like refrigerators, washing machines, dryers etc. One of the points that the model neglects is that, the holding cost is constant, irrespective of the price of the product. Since the holding cost includes the cost of CEÇ)ital among other things, presumably the cost should vary with the increase in price. Granted the difference ($0.3/dollar discount/unit/year) is very minor in the example that the author illustrated. If the order quantity is 2500 units the holding cost would be change by negative $375. If the item is more expensive or the demand is higher this could make significant impact. 1.4.4 Schniederjans & Cao Cao & Schniedegans (2004) introduced the concept of considering the savings in space in comparing the JIT model with EOQ and price discounts. This was the latest in a series o f articles by the authors, where model by Fazel (1997) and Fazel et al. (1998) was improved upon with the consideration of space savings. 14 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The nomenclature used here is in continuation to that o f by Fazel (1997) and Fazel et al. (1998)’s model. Cao & Schniedeijans (2004) extended to total cost o f JIT to include (1.14) Where m = — fo r q m>\ q is the order quantity under JIT which is less than Q. C is the annual cost o f a square foot o f facility and F is the square feet saved by initially adopting a JIT system. Thus the difference in cost is given by: Z = — —^ ^ + ^ + (c£ - c^)D + CF Q Q 2 ^^ (1.15) V / or by substituting the value of Q from Q = 1-' ^ Z = V2ADh [ 1 - y j + (c^ - )D + CF see working in Appendix 1.1 (1.16) The authors suggested that the TCj could be presented without the holding costs. This is never true even when q = u where u is the usage rate (D = w * demand periods in a year). That is, if the production facility shifts away from EOQ to JIT, the recommended lot size is ^ = u, which is producing as demanded per demand period (say 1 day). Even though q = u, inventory may be negligible, but surely not zero. This is also the case with cross dock. There is a point where the products enter and /or leave the warehouse. There are handling charges associated with the volume 15 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and facilities costs while in the facility. Thus although the holding charges are small due to the low units it can never be zero. Another aspect is that of the annual facility space reduction termed as CxF. The authors suggest the term as savings to be deducted from Fazel’s equation (1.12) for the total cost for JIT. The space requirements for any given product in either an EOQ or JIT system would be the lot size quantity, which in turn determines the space requirements for finished items, work-in-progress, raw material, other components, etc. If CxF represents the space requirement when using EOQ, then savings when adopting JIT policy would be CxF=CxS (1.17) \ Q. where S is the total space available to store Q .l îq = Q, then annual g facility space reduction is zero. The authors have not considered the magnitude of facility reduction which could directly impact the decision on the value of q. The cost indifference function is another aspect that would be impractical concerning the papers by Cao & Schniedeijans (2004). It suggests that for a particular demand point the EOQ is more cost effective than JIT. Going along with the author’s argument, which suggests that if demand in the next period is such that JIT is more cost effective then shift to JIT now, otherwise retain EOQ system. This will also suggest that if you have a JIT system implemented, then shift to EOQ system if the demand in the coming period favors EOQ to JIT. In a dynamic environment, it would then suggest to follow a flip-flop policy! Firms can only 16 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. move towards a JIT system if continuous improvement programs are implemented, an issue that authore ignored. As a summary, it can be said that all the authors considered the JIT in isolation. A lth o u ^ it was mentioned that there is a need o f cooperation with the vendors this was not modeled. In the next section, papers that address the issue o f cooperation with the other members of the chain are discussed. This is important when modeling the move to JIT since it is very difficult to implement JIT without cooperation with the vendors. 1.5 C oordination in Supply chain As mentioned earlier, while many papers discussed earlier considered JIT and various aspects therein, the problem was approached with the picture of only the organization concerned. It did not model partnership with suppliers, although it was discussed and agreed upon as important to the success of JIT. It has been found that larger firms could exercise more power in implementing JIT than smaller firms that found it hard to obtain cooperation fi-om its suppliers (Munson et al 1999). The impact of coordination within the supply chain was discussed at length by Munson & Rosenblatt (2001). The authors modeled a three level supply chain namely a single supplier, single manufacturer and single retailer and the impact of savings that can be obtained if there is coordination between the three parties. Under circumstances where there is no coordination, the manufacturer sets their orders quantity based on the order (EOQ) of the retailer. This would be a lot size multiplier of the order of the retailer. The supplier would in turn base their order 17 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. quantity based on the quantity ordered by the manufacturer. This would be a lot size multiplier o f the manufacturers order quantity. In a situation where there is coordination the leader in the supply chain would initiate the change by enticing the retailer to increase their order quantity by offering quantity discounts. This would prompt the retailer to increase the order quantity to a certain point that would bring savings to the manufacturer. A similar approach would be done for the supplier. Coordination would not necessarily bring savings to every party concerned, however, Munson & Rosenblatt (2001) argues that when considering the chain as a whole there would be savings and these savings can be shared through quantity discounts, price increases and so on. The savings could be concentrated at the manufacturer’s or the supplier’s or the retailer’s side. The challenge is to ensure that the savings obtained at one point in the chain is distributed evenly or fairly among the players concerned. This responsibility would depend on the leader of the chain. If the leader exercises a great influence in the chain it need not share the savings and can force the others in the chain to take the losses. This issue was addressed by Munson et al (1999), where the authors examined th : dynamics o f the industry concerning the aspect o f coordination. Their article was a practical outlook on the industry and the dynamics of coordination. The authors argued that there are definite advantages in exploiting coordination but if the power o f the supply chain leader is used in an abusive way to maximize profits at the leader’s end, it could lead to negative consequences like boycott or legal implications. In the short term, this approach could prove profitable to the chain 18 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. leader but ii long run it weakens the chain and forces the abused to take retaliatoiy measures. This leads to another question namely, if the members of the chain do decide to cooperate how will the resulting savings be distributed. The following section takes a brief survey o f the literature in this scenario. 1.6 P ro fit S haring The cooperation between the members of the supply chain to improve efficiency and costs raises another issue namely, how the savings generated would be distributed among the players. The brash approach for a leader would be to maximize and hold on to all the savings obtained being the leader. This would lead to a weakened chain as discussed earlier and long term effects. One o f the approaches would be adopting the model developed by Abad & Jaggi (2003). In this model the dynamics o f setting price in a price sensitive environment through the length of credit was highlighted. The influence o f price affects the demand and thereby the order quantity. They examine the difference in cooperative and non-cooperative scenarios. The model balances the choice between offering a small unit price and no credit against high unit price and some trade credit. Although this aspect is not discussed at length in this thesis, it is acknowledged that this issue is imperative to coordinating in a supply chain. Besides Abad & Jaggi’s (2003) model other ways of profit sharing can be based on the investment ratio of the players or a simplistic 50% split through quantity discounts and trade credit and so on. 19 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1.7 Objectives and Scope of Thesis In an environment of continuous improvement and increased competitiveness many organizations are adopting JIT. The benefits of JIT are often understated in many models due to the fact that the synergy of l)space savings, 2)coordination, 3)reduction in defects and 4)reduced setups are not modeled. Through the past sections in the literature review the object was to highlight the research done so far on the EOQ model. The model that is developed here, unlike others that studied the impact of these factors independently, illustrates the impact of each of these factors together on the total cost of a two level supply chain. 1.8 Thesis Layout Chapter 2 o f the thesis contains the problem description and the development of the mathematical model. The solution procedure is dealt with in a separate section. Chapter 3 illustrates the solution through an example. The behavior of the model is then studied through the generation of random scenarios and the results computed. The results are analyzed to see the overall impact coordination has on the total cost. This is follov/ed, in chapter 4, by the conclusion of the analysis and suggestions for further research. The appendices tliat follow includes the results in a tabulated form, and the step in the working of various equations. 20 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chapter 2 PROBLEM DESCRIPTION AND MATHEMATICAL MODEL 2.1 Problem Objectives The problem that is being addressed in this thesis has five aspects that can be described in a figure 2.1. The matrix describes a two level supply chain and the relationship between the two levels. The two can be representative of a supplier and retailer or a manufacturer and retailer or a supplier and a manufacturer. In this thesis, it is assumed that there are only two parties involved, i.e, there is only one player in the upper level that supplies to the lower level that consists of only one player. This kind of relationship is often seen in the marketplace where the retailer requires firom the manufacturer exclusive rights to a product. Under the initial circumstances, the two parties make order quantity decisions independent of each other. The lower level (say retailer) decides independently what the optimum order quantity is based on their parameters. The upper level (say supplier) then bases their optimum manufacturing quantity based on a lot size multiplier X. of the order quantity of the retailer. This situation is described in the first box (No Space', No Coordination). Moving towards the right would be when the two levels of the chain decide to cooperate and decide on an optimum order quantity. This could mean that the lower level moves away fi’om the local optimum but results in a greater savings for the ' I n th is th e sis, th e te rm “ N o Space” is u s e d to m e a n th a t th e c o s t fo r s p a c e is n o t c o n s id e re d a s a fu n c tio n o f Q (o r d e r q u a n tity ). I t is fa c to re d as a c o n s ta n t in th e m a th e m a tic a l m o d e l. 21 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. chain as a whole. This situation is described in the second box (No Space, Coordination). The retailer can be compensated by the supplier for the savings foregone in terms of discounts or lump sum compensation. The distribution of this savings, however, is not dealt within detail in this thesis. Next, the impact of space on these two scenarios is examined. Space is often an expensive commodity and the model integrates the impact of space and bin capacity to develop the optimum order quantity and lot size multiplier. Moving to j r r involves reducing the order quantity and processing orders on a lot for lot basis. This means that the supplier and retailer do not have a lot o f inventory. It would also be fair to say that the supplier has a lot size multiplier of one. It is also worth mentioning here that it is not possible to have JIT without proper coordination between the two players. Thus, this area of the matrix is indicated by the shade. Another consideration that has been included in the model is that when adopting EOQ, there maybe a lot of defects in a shipment. This is resource lost in the chain and needs to be accounted for in the model. Along with the defects brings the need to inspect the products on arrival. This causes the retailer to incur the cost of screening the shipment on arrival. These costs are eliminated when considering JIT since the approach is zero defects. 22 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Savings 1 No Coordination Coordination Case I Case II Case III Case IV JIT % " Ï ÎZ3 Case V Figure 2.1: Two Level supply chain Matrix The model developed below explores each scenario in detail and elucidates the impact that each relationship indicated in the matrix has on the overall cost of a two level supply chain. Thus players in a two-level supply chain can evaluate the options and corresponding savings they can have in their decision to move towards JTT. In this chapter, the nomenclature used is first defined followed by the section on the mathematical model. The section on the mathematical model first establishes the assumptions and then models 1) no coordination no space 2) coordination no space 3) no coordination space 4) coordination space and finally 5) JIT. This is followed by the stepwise description of the optimal solution procedure. 23 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.2 Nomenclature 2.2.1 Input parameters: i=r,s subscript for retailer or supplier Ai = the fixed component of the order cost ($). a ~ The variable component of the order cost. It is assumed that when operating according to the EOQ policy, lots received need to be screened for quality. The variable cost is the screening cost per unit ($/unit). /?= the fi-action of units not conforming to quality that cannot be sold and thus needs to be returned. (0 <yS <1). ru= the cost of reworking ($/unit) c,= the unit purchasing cost for EOQ purchasing (S/unit). hi= the unit holding cost when operating according to EOQ policy (S/unit/yr). Ni = total number of storage areas, where N=l, 2, ... and each storage area has a maximum capacity of V. Ri = the annual cost per a storage area (S/storage area). V= opportunity cost per unit for not being able to sell the defective items ($/unit) p = penalty charge per returned unit to the supplier ($/unit). D = annual demand rate (Units/yr) n=Number of working days per year 2.2.2 Decision Variables: Q e = order quantity when operating under EOQ policy (units) 24 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Qj = order quantity when operating under JIT policy (units) X = lot size multiplier (1,2,3,..... ) 2.3 M athematical Model In this model, each scenario is formulated. The model considers the impact of defects, returns and penalties for returned goods. This is incorporated into the model to see the impact it would have when approaching JIT where it is assumed that defects are zero (i.e., total quality., Fuller 1995 ). 2.3.1 Assumptions The assumptions associated with the classical Economic Order Quantity Model are held in this model, namely: 1) Demand is known, constant and independent. 2) Lead time is known and constant. 3) Receipt of inventory is instant and complete 4) There are no quantity discounts. 5) There is no shelf life for the product 6) Equipment capacity is infinite 7) Unlimited storage capacity is available Following the assumptions mentioned above the first case scenario is considered. 2.3.2 No Coordination, No Space. When a retailer is operating on EOQ, it is assumed that not all items conform to quality and a fraction (fi) of quantity (Q) is assumed to be defective. These defective items are either sold at a discounted price or returned to the supplier at the 25 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. end of each cycle. In this model, it is assumed that the retailer returns the defective products to the supplier and charges the supplier a penalty p per returned item. The penalty cost would include the holding cost of the defective items, the cost of screening and an opportunity cost for not having these items of good quality in the first place. 2.3.2.3 Retailer’s Cost The figure 2.2 below illustrates the inventory cycle of the retailer following EOQ. The maximum and minimum inventory levels in a cycle are Q and PQ. It is assumed that the retailer holds on to the defective items till the next shipment arrives. Once the new shipment of Q units arrive, the defective items are loaded back on the trailer and returned to the supplier. Again, the Q units that are supplied by the supplier has PQ units that are defective and are held by the retailer till the next shipment. Thus, the cycle length T is given by: ^ Qa-P) D 26 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. With a fraction defective, the model can be illustrated as follows Average Inventor}’ Level T im e Figure 2.2 : Inventory cycle of the retailer 27 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The maximum and minimum inventory levels in a cycle are Q and PQ. The per lot cost for the retailer, L C r (Q ) , is the sum of all the costs in a given cycle, which are: 1. Procurement cost, PCXQ) P C X Q ) = A r + aQ + C r Q (2.1.1a) 2. Holding Cost, HCXQ) HCr(2)=Average Inventory Level x ' (2,u i) 2 ' T D V I D / 3. Storage Cost, SCr 5C=Kr (2.1.1c) a constant such that it does not impact the decision variable. In this scenario, it is assumed that there is plenty of storage. However, for the sake of comparison of the costs of the different scenarios in a like manner, the storage cost has been given a value that is the greater, required for Q when there is coordination and no coordination. The assumption here is that space is not included in the decision variable, however that cost is valid and accounted for as overheads or otherwise. It is argued here that this function is significant and produces an impact on overall cost. 4. Penalty Refund, PnCXQ) PnC, ( Q) = K P + (c^+y +a)PQ {2. L i d ) Therefore, the per lot cost for the retailer, 28 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I,C ^0=(2.1.1a)+(2.1.1b)+(2.1.1cH 2.Lld) L C ,(Q ) = A ,+ c Q + c,Q + h, -K P + ^ + a ) f S Q The retailer’s annual cost, The first derivative o f the above equation is given as; dQ Q ‘( \ - P ) 2 SC^ is considered a constant. It is assumed that space is available and not utilized for any other purposes. The second derivative is as follows: Setting the first derivative equal to zero and solving for Q Qe = ^ 4 working in Appendix j 2 or where D ^, = — —— is the adjusted demand. 29 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (2.1.: 23.2.b Supplier’s Cost The supplier delivers Q units every T units o f time. The decision that the supplier has to make is on the lot size multiplier X. The lot size multiplier in this thesis is considered as an integer, since the assumption is that the supplier provides this product only to this retailer. These can be exclusive products provided to a particular retailer hke retail brands. This research adopts the assumption used in Munson & Rosenblatt (2001) in this matter. Figure 2.3 illustrates the inventory cycle of the supplier. The supplier’s inventory follows a stepped model. They manufacture or purchase in lot sizes of XQ and deplete inventory in steps o f Q. The dotted line indicates the actual inventoiy that the supplier is holding. This includes the defects that are to be reworked or being reworked that are returned to the supplier. Each shipment returns with a load o f PQ defective units. It is assumed here that these items can be turned around before the next shipment is delivered. 30 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The following figure illustrates the Inventory level of the supplier i S(^-V T im c XT- Figure 2.3 : Inventory cycle of the supplier 31 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The per lot cost for the supplier, L C s ( k , Q ), is the sum of the following costs in a given cycle, which are: 1. Procurement cost, P C s ( k , P C s(k,Q )-A i^ c ^ Q Q ) (2.1.6a) 2. Holding Cost, iîCj(2, Q) H c ,(k . Q ) = h, +K m = {2.1.6b) The second term h[XQP is representative of the carrying cost o f the returned / defective items, where h[ = — (c^-\-a-\-v + a>). c, 3. Storage Cost, SCs S Q = Y^ (2.1.6c) a constant. The storage cost for the supplier in this scenario follows the same pattern as that of the retailer. It is assumed that there is plenty of storage available and this cost will not impact the decision variable. Again, for the sake of comparison of the costs of the different scenarios in a like manner, the storage cost has been given a value that is the greater, required fo ilQ , when there is coordination and no coordination. 4. Rework Cost, R C s ( X , R C /X ,Q ) = cûXfiQ Q ) (2.1.6d) 32 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5. Penalty Cost PnCs(Q) charged by the retailer. This would be a cost to the supplier for supplying defective items to the retailer. P nC /Q ) = PnC,(Q) = h j + ( c ,+ v + a)pQ (2.1.6e) Therefore, the per lot cost for the supplier, L C sd Q)= (2.1.6a)+(2.1.6b)+(2.1.6c)+(2.1.6d)+(2.1.6e) The Supplier’s annual cost, C /A ,0 = Where XT = D Thus Q (l-^ )X (I-P) ' (\-P) 2 (\-P) K ?Q + ( c , + v + a ) - ^ + (2.1.7) The value of Qe is substituted in the above equation and the optimal value of 1 is computed such that c /2 -i.6 ^ ;> c /A ,6 j< c /2 + i,g f;v k > 2 (2.1.8) and C ,( X .Q J < C /X + l.Q ,) forX-1 (2.1.9) Thus, in this scenario the supplier and the retailer independently evaluate the optimum cost that each can obtain. There is no attempt on the supplier’s or the 33 Reoroduced with Dermission of the coDvripht owner. Further reproduction prohibited without permission. retailer’s side to influence the other sides order quantity. In the next section the scenario o f coordination between the supplier and retailer is examined. 2.3.3 Coordination, No Space. In this scenario the two levels of the chain cooperate with each other to bring the total chain cost down to the optimum level. This means that the supplier or the retailer departs from their optimum order quantity in the interest of the total chain cost. Thus the total chain cost is optimized instead of the retailer’s and the supplier’s costs individually. The total chain cost is therefore given as C /A .0 = + C /0 (2.27) (2.2.2) The partial first derivative is: Setting the above equation to zero the following is obtained (working in Appendix 13 ) Q ex)= I 2d (ia ^ + a ) (2 2 4 ) The partial second derivative is: a"C/ 1 0 ^ _ 2/ 7) 2/ D .... ^ >2 (2.2.5) 34 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. For every value of A, a unique value for Qe(^) is obtained. The value of A and Qf.(X) is substituted C J A ,0 and the optimal value is computed such that > c / A ,g / A ; ; < c / r A + v , g / A 4. 7;; v A^ 2 (z z d ) And C /A ,6 g r A ;;< C /r ^ + i;,G ^ r ^ + 7;;for A=1 (2.2.7) Thus the optimal cost of the total chain is computed. This cost can be less than or equal to that of no coordination. It may be noted that when the cost of the chain is optimized the retailer could be loosing money since they have gone away from their optimum and the supplier saves over and above what the retailer has lost. Thus the savings get clustered at the suppliers end. Based on the dynamics o f the supply chain relationship, as discussed in the literature review, this savings can be distributed. In the next two scenarios, space is considered as an input variable and the impact on the cost of the chain examined, 2.3.4 No Coordination, Space. In this scenario Space is considered as a function of Q. It can be argued, however, that this function is a stepped function for incremental increases in Q. Most retailers/ suppliers store items in pallets, bins, kanbans or some kind of storage area. The storage area used each have a capacity V. The quantity that needs to be stored is given by Q and the number of storage areas required is given as N = Q +1 V (2.3.1) 35 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. where fx"] is a round-down function, fx lis the smallest integer less than or equal to X . Thus the Storage Cost for the retailer per cycle denoted by SCr(Q) is (2.3.2) It is assumed that V = V^=V^ since the volume of the product does not usually change from the supplier to the retailer. It is accepted that there are situations where These are examples where a supplier processes the products and the manufacturer further processes it before it reaches the end consumer. A typical example of this can be that of a printed circuit board making its way through the supplier to the manufacturer who assembles components on them and assembles them in a box that is passed to the end consumer. The space required to accommodate the printed circuit board is a lot smaller that the space required for the fully assembled product. However, in a supplier and retailer context there is rarely a case where the volume of the product changes. The former can be very easily accommodated in to the model with minor modification. Thus, when space is considered the retailers annual cost equation (2.1.3) now becomes: V (“ .3) The first derivative of the above equation is given as: 36 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A,D dC,(Q) The second derivative is as follows: > 0, v g > 1 Setting the first derivative equation (2.3.4) equal to zero and solving for Q Qr = It . . (2.3.5) working is in Appendix 1.4 Since the storage has been taken as a continuous function, the value C r(0 can be iteratively computed for all values between (Q-yV) to (Q+yV) to obtain gg . y is a sufficiently large number that is enough to accommodate the error due to the noncontinuous function. The storage cost for the supplier is given as s c ,( i,Q ) ^ R ,N , = J \ (2.3.6) \ ^ / The above equation ignores the space required for the total shipment of the supplier. This is because Q units get shipped out the moment it arrives. Thus the space required to accommodate that is not considered. Substituting (2.3.6) in (2.2.2) the following can be obtained C,(l.Q)= A.D cofiD c.D , ô fA -i; ■+ h . + h^pQ + (c^+ v + a) h[Pp 4 ---------- — ■ ' ----- h R G(^ + A -l) PD 37 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. +1 (2.3.7) Again, the supplier would try to optimize their cost by adjusting the lot size multiplier by iteratively substituting the value of A= 1,2, 3, 4, .... etc. The value of Qe from equation (2.3.5) is substituted in the equation (2.3.7) and the optimal value of X is computed such that V x> 2 (2.3.8) and + &rX=l (2.3.9) The total chain cost would be the sum of the supplier’s and retailer’s cost function. This would represent the total cost o f the chain if both the supplier and the retailer consider space as an important cost factor, but do not influence each other’s order quantity. 2.3.5 Coordination, Space. In this scenario, the two levels of the chain cooperate with each other to bring the total chain cost down to the optimum level. The difference here is that space cost is considered. Thus substituting the space cost equation (2.3.2) and (2.3.6) in (2.2.2) the total chain cost can be given as: V n -w sn-w -i a-P) 2 n -w (2.4.1) R. The Supplier cost and the retailer cost include storage cost as a function of Q. 38 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The partial first derivative is; (2.4.2) V Setting the above equation to zero the following is obtained Cf,U-l ^ ~ ^X(l-/]){VhXl + fi)+VhX^-i}+2K+2R^{p + À-l)} (243^ The partial second derivative is: a ^ C / X Q , ^ 2A,D ^ 2A,D aÿ Q’ a - P ) Q ' a - P ) x (2.4.4) Again for every value of -A a unique value for Qe(V is obtained. Since the storage has been taken as a continuous function the value Cc(X 0 c a n be iteratively computed for all values between (Q- y V) to (Q+ y to obtain Qgfbr each value of A. y is a sufficiently large arbitrar y number that is enough to accommodate the error due to the non-continuous function. The value of A and Qe(X) firom equation (2.4.3) is substituted in equation (2.4.1) and C^(X,,Q) computed such that c /r A - i ; , g / A - > c / A . 0 rA;; < c / r A+ i ; , G / A + v A> 2 (2.4. j ) And C /A ,G /A ;;< C /r A + U ,G /A + 7;;For A= 1 (2.4.6) Thus, the cost of coordination with space can be calculated. In each scenario, the cost o f the chain either remains the same or reduces, due to various factors modeled. When space is considered, it influences the cost and the quantity ordered. 39 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. This in itself is an approach to JTT. Reducing the order quantity to a day to day basis can be considered as JIT. This means that there is no wastage of space and products do not wait in inventory bins to be processed. The following subsection looks at how far does the chain need to reduce costs from Coordination Space to JIT. 2.3.6 Operating Under JIT The approach in this section is different from the other sections. Here the values are known and what needs to be obtained is the amount of reduction in setup needed to reach that level. When operating under JIT, the lot size multiplier X is equal to one since the supplier would be building inventory on a lot for lot basis. Further, owing to continuous improvement and immediate identification and corresponding rectification of problems it can be assumed that approaches Zero. Substituting these values in (2.4.3), Ideally in a JIT world Qj = — . Where it is assumed here that n = 365 working days n per year. This would be the scenario where products are ordered on a day to day basis. However, bringing Qj down would cause the total cost to increase. For this scenario to be profitable Ar and As should be brought down to a feasible level. Having computed the total cost in each scenario, the optimum cost is considered as 40 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. a benchmark. The cost for operating under JIT is computed and the reduction needed in setup costs is computed. 2.4 O ptim al Solution Procedure The following sections will illustrate the solution procedure to arrive at the optimal order quantity in each of the abovementioned scenarios. It is acknowledged that the model assumes the order quantity and number of storage bins to be continuous for calculation purposes but in reality it is not possible to order 566.6667 units of a product. It should be either 566 or 567 units. This is also the case where number of bins is concerned. If 100 units fit into one bin the number of bins required to accommodate 566 units would be 6 bins. It cannot be 5.66 bin or 5. This necessitates the search for the optimum solution by the iterative search surrounding solution generated by the model. The following two flowcharts illustrates the solution procedure for No Coordination and Coordination 41 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C a lc u la te C r(Q ) & Q , A= 0 , K = Infiniti A= A+1 C a lc u la te C s(A , Q g ) fo r K Qe No Yes ltCs(A, Q e ) in c r e a s in g , Yes No K= C s(A , Q g ) Nirend/ End K = C s(A , Q g ) S to r e A & Q g a s O p tim a l S o lu tio n Figure 2.4: Optimal solution procedure for no coordination (case I and case HI) A= 0 , K = Infiniti No Y «5 Yes No K= C c(A . Q e ) —^ S t o r e A & Q e a s O p tim a l S o lu tio n Figure 2.5: Optimal solution procedure for coordination (case II and case IV) 42 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The following is the solution algorithm adopted: 2.4.1 No Coordination, No Space ( = hit Step 1: Set 2AD 2 K a -fij / Step 2 : For Q = Qe-yVto QE+yF The value of Q for which C^fQJ is minimum is the retailer’s EOQ. y is a sufficiently large arbitrary number that is enough to accommodate the error due to the non-continuous function. Step 3 : For 1 = 1 to sufficiently large integer (say 2max)- ^max is usually a value beyond which the cost function for the supplier is constantly increasing. KJ^Q + (Cr+v + aJa-P) The value of IQ for which C JX .Q ) is minimum is the supplier’s EOQ and the total chain cost is C J l,Q ) = C^(k,Q) + C^(Q) 2.4.2 Coordination, No Space. Step 1 : For X= 1 to sufficiently large number (say Xmax) generate the values of o m - I + Xmax is a sufficiently large arbitrary integer beyond which the minimum total cost C^^X.QgJ is constantly increasing. 43 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Step 2 : For Q — Qe- y V to Qe+ y V for each X SC ,+ - ^ + h J Q + ( c ,+ v + a)- ^ y is again a sufficiently large arbitrary number that is enough to accommodate the error due to a non-continuous fimction. The value of Q and X for which C^(X,Q) is minimum is the retailer’s EOQ and the suppliers lot size multiplier respectively. 2.4.3 No Coordination, Space Step 1: Set Step 2 : For Q = QE-yV to QE+yV — The value of Q for which C / Q ) is minimum is the retailer’s EOQ. n -w y is a sufficiently large number that is enough to accommodate error due to a noncontinuous function. Step 3 : For 1 = 1 to sufficiently large integer (say Xmax) œPD ^ + Cc^+v + a The value of 2 g for which C /X .Q ) is minimum is the supplier’s EOQ and the total chain cost is Ç J X , 0 = C /X ,Q ) + C / 0 44 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Àmax is a sufficiently large integer beyond wliich the supplier cost is constantly increasing. 2.4.4 Coordination, Space. Step 1 ; For 1 = 1 to sufficiently large number (say Xmax) generate the values of O «)= ^ I 2V D (U ,A,) ^jX(l-fi){VhXl + J3)+VhXX-^)+2R, + 2R^{J3 + X~l)} Amoxis a sufficiently large integer beyond which the minimum total cost C^(X,Q^) is constantly increasing. Step 2 : For Q = Qe- y V to Qe+ y V for each X y is a sufficiently large arbitrary number that is enough to accommodate error due to a non-continuous function. The value of Q and X for which C^(X.Q) is minimum is the retailer’s EOQ and the suppliers lot size multiplier. 2.4.5 Operating under JIT Step 1: Compute the value of Q such that Qj = — where n=365 working days per n year Step 2: Substitute the value of Qj in equation (2.4.1) with X =land p=0 and compute the total cost. 45 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Step 3: If the total cost obtained is less than that computed in each of the above scenarios, no further action required. If the total cost obtained is greater than that of any o f the above scenarios, then Ar and As has to be reduced together uniformly to a level such that the total cost is less. In the following chapter, the procedure detailed in section 2.4 is illustrated through an example and the results of the statistical analysis performed using a number of sets on parameters described. 46 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C h a p te r 3 NUMERICAL RESULTS AND STATISTICAL ANALYSIS This chapter presents numerical examples to illustrate the solution procedure to the mathematical models developed in Chapter 2. It also investigates the behavior of the model tested under varying sets o f parameters. 3.1 N um erical Example This section presents numerical examples for the models developed in Chapter 2 to illustrate the savings that could be obtained when the inventory system shifts fi'om tradition EOQ inventory policy to contemporary JIT policy. These models represent the cases when there is no-coordination between the supplier and the retailer and no space considerations (Case I: NC-NS), coordination between the supplier and the retailer and no space considerations (Case II; C-NS), no-coordination between the supplier and the retailer with space considerations (Case III: NC-S), coordination between the supplier and the retailer with space considerations (Case IV: C-S), and the case where the supplier and the retailer operate under just-in-time policy (Case V: JIT). Each case, i.e., NC-NS, C-NS, NC-S, C-S, and JIT, is examined to see the progress in savings as each step towards the implementation of the ideal JIT as described by Cao and Schniedeijans (2004) is considered. The following is a solved example to illustrate the savings that could be obtained. In the example that is illustrated let Ar =420 ($), As = 480 ($) , a =0.105 (S/unit), ^=0.046 {0 < P <1), Cr= 21 ($/unit), Cs = 10 ($/unit), hr= 2.73 ($/imit/yr), A;=1.8 47 Reproduced with permission of the copyr' ih t owner. Further reproduction prohibited without permission. ($/unit/yr), Rr= 1800 ($/storage area), Rs — 1500 ($/storage area), v = 3.675 ($/imit), D = 120,000 (units/yr), V=10 units Case I: No Coordination, No Space From equation (1.3): Qe = 2x420x120,000 = Int{6,369.43) = 6,369 2.73 X( 1 - 0 .0 4 6 / J For Q = Qe-y V to Qe+y V Here y is taken as 2. This is a sufFiciently large arbitrary number from the results generated that cover the minimum value of C^(Q) and the error due to a noncontinuous function. For Q = 6350 to 6389 substituting the values in equation (2.1.3) PD Note that SC^ is kept constant and the value assigned is the cost of space needed to accommodate the greater of the two EOQ’s obtained in case I and case II, namely “no coordination no space” and “coordination no space”. In the sub-section that follows, it can be seen that when the space requirement for “Coordination No space” was calculated, it was found to be 282 storage areas, since this scenario requires 637 storage areas the greater o f the two is taken as SC^ for both the scenarios. This is done to bring consistency for the space cost to the two scenarios so that the two can be compared. Thus SC^ = $1,146,600. 48 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix 1.5 shows the values for Cr(Q) and their corresponding Q and graphically shows the variation. It is obtained that : It is interesting to note that the change in total cost is in decimals as the optirhum is approached. For practical purposes, this gives the decision maker a lot of flexibility to adjust the order quantity to satisfy other decision parameters with limited impact on the cost. To calculate the supplier’s side of the cost, equation (2.1.7) is used for X from 1 to Àmax for the optimal cost. The computational logic used here is that 1 is tested from 1 to a very large number. As the values of cost for each X, C;(X, Q), are computed they are stored and if this displays an increasing trend the program exits. Here X is computed to a o f 9. The software is set this way since the increase in X, C^(X, Q) takes a parabolic shape. This means that if two or three consecutive values show an increasing trend then the minimum has been reached. Appendix 1.6 shows the values for Cs(X, Q) and the corresponding value of X and depicted graphically. The suppliers optimum cost is at; Thus the total chain cost is c / 2 , G) = c , r A, 0 + C /G ) = % 49 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Case I I : Coordination, No Space The approach here is combined where for each X, look for the optimum Q. To save computing time, the range of X for which the total cost is minimum, is first established. This is done by iteratively substituting the value of 1 in equation (2.2.4) and generating the value of Q. This in turn is substituted in equation (2.2.2) to get the total cost of the chain. The values ofX for which the total chain cost is minimum is identified and a search for the optimum value of X and Q is done. Thus the value of X for which the total cost was minimum was 8 and optimum value was searched for from X= 6 to 10 is substituted in equation (2.2.2). The values of Q obtained are tabulated in table 3.1. Q ;;; 6 7 8 9 10 , 3257 3000 2795 2626 2485 C,(X,Q) $8,884,758 $8,883,713 $8,883,401 $8,883,552 $8,884,008 Table 3.1 : Value of Q obtained for each X from equation (2.2.4) For each X and Q, iteratively the surrounding values of Q were searched for the best solution, i.e. if Q(8)=2795, the best solution is searched for in the range of ±30 units. This range is selected to be sufficiently high arbitrary number such that the minimum lies within this range. The results are tabulated in Appendix 1.7. Thus, from this tabulated results, the optimal value is + c.p&y j; = gj.dgo.jgp+ ^j.203,037 = Here again, it is interesting to note that the change in total cost is in decimals as the optimum is approached. However, this is not the case for the supplier cost and the 50 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. retailer cost taken individually, that in turn varies by a few dollars. Also note the difference in the value generated from table 3.1 and the results obtained from appendix 1.7. This is due to the error of a non-continuous function. It is worthwhile also to note that the supplier’s cost reduced ($5,218,309$5,203,031= $15,278), while the retailer cost increased ($3,674,525-$3,680,369=$5,844), however, the total chain cost has reduced ($8,892,834- $8,883,400=$9,434). The retailer can be compensated by the supplier for the change through quantity discounts or can be forced to order in lot sizes of 2,815 units by the supplier and the savings kept by the supplier, if the supplier is the chain leader. Alternately, if cooperation is the aim then the net savings can be shared based on investments or total purchases or so. Case III: No Coordination, Space When No Coordination with space is considered the EOQ is represented by equation (2.3.5) and the total cost of the retailer is represented by (2.3.3). Thus substituting the input variables in (2.3.5). gg=539 units C^(539) = $2,707,253 The best solution is searched for from Q=520 to 559 and tabulated in Appendix 1.8. The graph of the variation in the total costs takes the form of a saw tooth pattern. This is the impact of the storage areas that are taken into account. 51 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Now searching for X (lot size multiplier) that is optimal to the supplier, consider 1 from 1 to Imax where is 6. The value of Xmax is different for each set of parameters since the software stops searching for the optimal solution when the result o f increasing X starts to generate increasing values of supplier cost. Substituting these values and all the input parameters in equation (2.3.7) X= 2 C J2,539) = $1,512,867 Again the supplier cost for X= 1 to 6 is tabulated in Appendix 1.9 The total cost is therefore C/2,539) = C/2,539) + C/539) = $2,707,253 + $1,512,867 = $4,220,120 Thus, the optimal cost for No Coordination, Space is found. Case IV: Coordination, Space In this scenario the approach is combined where for each X the optimum Q is obtained. From equation (2.4.3) X= 1 to X^ax where Xmax is a large number beyond which the values are increasing. The values of Q obtained for each value of X are tabulated in table 3.2. , X Q 1 2 3 4 5 775 494 385 325 286 Table 3.2: Value of Q obtained for each X from equation (2.4.3) For each Xand Q, iteratively the surrounding values of Q were searched for the best solution. The best solution is searched for in the range of ±30 units. The results are not included but follows the pattern of that of Coordination No Space. Only that 52 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. this has a saw tooth pattern to the graph due to the cost o f storage space that is included. Thus, from the table the optimal result is Q (2,478) = $2.708,882 + $1,509,456 CJ2,179)== $4,218,338 Note that the similar situation has occurred with regard to the supplier cost where it is reduced ($1,512,867-$ 1,509,456= $3,411), the retailers cost increased ($2,707,253-52,708,882= -$1,629), and the total chain has cost has a net reduction ($4,220,120-54,218,338=51,782). The sharing of the savings can be done based on the dynamics of the supply chain relationship and the exchange of funds through quantity discounts or methods suggested by Abed & Jaggi (2003). Case (V): Operating under JIT As the supplier and retailer move towards JIT, it is assumed that the retailer ships product on a need to basis. That is the retailer buys only what is required for that day and the supplier buys or manufactures only what would be ordered by the retailer. This means that X=1 and Q = 365 365 = 329 units Moving to JIT means that the supplier and the retailer move towards total quality. This would increase the cost of Cr and Cs, thereby increasing the holding cost 53 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. parameters hr and hs. It would be safe to assume that Cr and Cs increased by a minimum of one times the screening cost. It is also assumed that p is zero and the retailer does not need to do any screening since it is total quality that is being received. Thus Cr and Cs would he $21.11 and $10.1 Iper unit and hr and h^ would be $2.74 and $1.82 per unit per year. Substituting these values in the equations: CXQ)= $2,745,751 C X lQ ) = $1,389,300 C X lQ ) = $4,135,051 Thus in this scenario adopting JIT has already proved cheaper than all the other alternatives. The major factor that contributed to the lower cost is the supply of total quality products (ie p=0). It is worthwhile to note that more savings can be obtained by reducing the set up cost or ordering costs although no reduction is required. Thus, it can be said that for this set of parameters it would be very valuable for the retailer and the supplier to move towards JIT. The summary results on these five scenarios are tabulated in table 3.3. 54 Reproduced with permission of the copyright owner. Further reproduccion prohibited without permission. mm# KB iSùppliérlsV- $3,674,525 $5,218,309 $8,892,834 6,369 5 $3,680,369 $5,203,031 $8,883,400 2,815 8 $2,707,253 $1,512,867 $4,220,120 539 2 $2,708,882 $1,509,456 $4,218,338 478 2 $2,745,751 $1,389,300 $4,135,051 329 1 NC- No Coordination, C- Coordination, NS - No Space, S - Space, JIT- Just in Time Table 3.3: Summary table of the total cost for all scenarios 3.2 Statistical Analysis In the following section the model is tested through varying sets of input parameters to understand the behavior o f the model. The parameters selected are adopted from the Munson & Rosenblatt (2001) example. The demand (D) is varied randomly (following a uniform distribution) from 10,000 to 500,000 units in increments of 10,000. The setup costs (Ar & As) for the supplier and the retailer is also varied randomly (following a uniform distribution) from $50 to $500 in increments of $10. The cost for the storage area (Rr & Rs) for the supplier and the retailer follows the similar randomness pattern from $1000 to $2000 in increments of $100. The fraction of defects (/J) ranges uniformly from 0.01 to 0.05 in increments of 0.001. The cost for the supplier (cs) varies uniformly from $10 to $100 in increments of $1/-. The cost for the retailer (c j is based on the supplier cost such that: 55 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. =ac^+v where a is the markup that varies between 1.3 and 1.8 unifoimly with intervals of 0.1 and v is the value added by the supplier to the product. In the analysis the value of v has been taken as 5. Thus the value of Cr is dependant on Cs but the randomness is introduced in the markup using the above-mentioned relationship. The holding cost {hr & hs) is dependant on the value o f Cr and Cs. This can be expressed as; hj - (pi XC; where (pi is the annual holding cost percentage for firm i. (p-, varies fi'om 12 to 25% of the cost for the supplier or the retailer in increments of 1%. This again is generated randomly following a uniform distribution. The randomly generated values were tabulated and the total cost in each scenario calculated. The base reference cost is the total cost for No Coordination No Space. The summary of the averages of output measures fiom the computational study of 8410 trials are tabulated in Table 3.4. 56 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ■1 2 3 4 5 6 7 8 9 10 11 12 13 m M Ê IÊ à Ê S m m Ê Ê Ê IÊ Ê M M B $48,337 Cc(X,0)(NC-NS) - CefiCO)(C-NS) 0.11% Cc(X,Q)(NC-NS) - Cc(X,0)(C-NS) % $2,944,919 Cc(X,Q)(NC-NS) - Cc(X,0)(NC-S) 10.95% Cc(X,Q)(NC-NS) - Cc(X,0)(NC-S) % $3,005,102 Cc(^0)(NC-NS) - Cc(X,0)(C-S) 11.08% Cc(^Q)(NC-NS) - Cc()sO)(C-S) % 1.99% % Setup reduction on JIT 60.39% % Reduction in Q (NC-NS to C-NS) 35.60% % Reduction in XQ (NC-NS to C-NS) 78.03% % Reduction in Q (NC-NS to NC-S) 90.77% % Reduction in X.Q (NC-NS to NC-S) 86.63% % Reduction in Q (NC-NS to C-S) 92.73% % Reduction in XQ (NC-NS to C-S) Table 3.4: Summary averages from computational trials $44,903 0.05% $1,930,626 9.49% $1,957,834 9.47% 9.22% 10.82% 9.54% 6.23% 3.56% 4.55% 2.58% The output parameters are defined below Cc(^Q)(NC-NS) - Cc(?t,Q)(C-NS)=> This is the difference in the total cost o f the chain between two scenarios. (NC-NS (No Coordination No Space), C-NS (Coordination No Space), NC-S (No Coordination Space), C-S (Coordination Space).) Cc(X,Q)(NC-NS) - Cc(X,Q)(C-NS)%=> This is the percentage difference between the total cost of the two scenarios. % Setup reduction on JIT=> the cost to operate on JIT is compared with the minimum cost of operating in the other scenarios. This parameter is the percentage reduction in setup needed to bring the total cost of operating in JIT to the minimum. It was observed that in all the experiments that coordinating in the chain brought some cost savings. When space was considered there has always been considerable savings that can be seen from the summary table 57 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. It can be said that it is definitely advantageous for organizations to move to JIT but computing the cost involved is often conducted in an erroneous manner since the actual costs and savings are hidden. Following the model developed above can lead to the conclusion that the cost involved in reducing setup costs to move to JIT is far lower than scenario o f no coordination no space. This is illustrated by the point that the percentage reduction in setup required is around 1.99 % on average. Among the other observations firom the trials are the maximum savings computed when moving fi'om no coordination, no space to coordination no space is 0.33% and the minimum is close to zero. Similarly, maximum savings from no coordination no space to no coordination space is 71.42% and the minimum is 1%. The numbers are slightly higher when comparing coordination space with no coordination no space namely 71.55% and 1.07%. This is due to the savings obtained due to coordination. This supports the fact that coordination between the supplier and retailer can bring a lot more savings. Considering the percentage reduction in setup to make JIT feasible, the maximum reduction in setup required is 75.24% and the minimum is 0%. Another surprising observation is that 93.76% of the trials did not require any reduction in setup cost for JIT to be feasible. This means that for these case scenarios, moving to JIT lot sizes require little if no investments to reduce setup cost. Other observations include, that if the supplier and retailer setup costs are equal the average percentage of improvement reduces slightly to below average from 10.95% to 10.10% for No Coordination Space and from 11.08% to 10.22% for Coordination Space. 58 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. This chapter details the solution of an example set of parameters and the overall behavior o f the results of a number of trials conducted. The next chapter concludes this thesis and contains suggestions for further research. 59 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C h a p ter 4 COÎ'TCLUSION AND FUTURE RESEARCH In this research, the classical EOQ model was extended to reflect and therein compare the actual costs associated with adopting the classical EOQ approach and the JIT approach for inventory. The model brings out the advantages of coordination within the supply chain which is often an integral part of an organization that successfully adopts JIT. The results from the random experiments illustrated that if an organization considered coordination (Case II) with its supplier or retailer on its own to begin with, can reduce the total chain cost by an average of $48,337. In a competitive market this saving can be used to the advantage of the organizations concerned. It must however be noted that the percentage reduction in order quantity for the retailer was down by an average of 60% and that of the supplier by 35.60%. The huge reduction in order quantity did not reflect in total cost since space occupied by the inventory was not properly accounted for in this scenario. Considering space alone as a cost factor without coordination (Case III) brings significant impact to the cost. The results indicate that a total chain cost savings of 10.95%, on average, can be obtained in this scenario. The reduction in order quantity and there in the average inventory on hand reduced by 78.03% for the retailer and 90.77% for the supplier. This indicates considerable savings in the chain. One can argue that these quantities approach JIT lot sizes. 60 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. However, the best case is combining both the scenarios, namely considering space and coordination (Case IV), which brings us closest to JIT. In this scenario the total chain cost savings that was obtained was an average o f 11.08%. The reduction in order quantity was the best with an average reduction of 86.63% for the retailer and 92.73% for the supplier. In case III and case IV, it can be observed that order quantities of the supplier are reduced to an average o f 10%. This can be an indication that the supplier is working on a lot for lot basis that would be in accordance with the principles of JIT. Ordering in JIT is characterized by total quality shipment that are manufactured or delivered on a need-to basis. This requires establishing long term contracts (WatersFuller 1995) and coordination between the supplier and the retailer. It can be observed that case TV characterized by coordination and consideration of space in itself is the closest to JIT ordering. Moving form case IV to JIT ordering (Case V), it can be observed that the reduction in setup needed to make JIT profitable is around 1.99%. It was also observed that in 93% cases no reduction in setup was ever required. This result concurs with Jones (1991) argument that in EOQ, if all costs are appropriately accounted for would approximate JIT lot sizes. This is the conclusion that the model alludes to. The example illustrated in chapter 3 (table 3.3) shows the reduction in order quantity to JIT levels. Thus reducing setup costs, although needed, does not need to be of that much magnitude than estimated with no coordination and no space. This opens scope for lot more improvement in costs as setup is further improved. 61 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. There is however a lot more scope for further research. In the analysis that was done in this thesis the space allocation considered was fixed (Joshi 1990). Considering the space to be dynamic can improve the savings further more. Although this would be of benefit to organizations that have numerous products, it is not the scenario when organizations deal with a few products only. Abed & Jaggi’s (2003) extensive research could be extended into the model to illustrate the dynamics of the players and the mode in which the profit sharing can be done. This can be further investigated in future research. The model assumes that the production rate to be infinite. Further research can be done to study the impact of finite production rates on the model. The economics of Price sensitivity to demand and demand sensitivity to price are other interesting areas that can be incorporated into the model. It is also suggested that further research can be done by extending this model to three members o f the supply chain. This would be in line with Munson & Rosenblatt (2001) model extended to JIT with the consideration of space and defects. The model can also be extended to consider single retailer with multiple vendors as well as multiple retailers and single vendor. 62 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. REFERENCES Abad, P.L., Jaggi, C.K., A Joint Approach fo r Setting Unit Price and the Length o f the Credit Periodfo r a Seller when End Demand is Price Sensitive, International. Journal of Production Economics 83 (2003), pp.l 15-122 Aghazadeh, Seyed-Mahmoud, Comparison o f Just-In-Time Inventory and the Quantity Discount Model In Retail Outlets, Logistics Information Management Volume 14, Number 3,(2001), pp.201-207 Billington, Peter J. Holding Cost Reduction in the EOQ Model. Journal of American Academy of Business, Cambridge, September (2003), pp.409-415 Cao, Qing., Schniedeijans, Marc J. A Revised EMQ/JITProduction-Run Model: An examination o f inventory and production costs. International Journal of Production Economics, 87(2004) pp.83-95. Chyr, Fuchiao, Lin, Tsong Ming, Ho, Chin-Fu, Comparison Between Just-In-Time and EOQ System, Engineering Costs and Production Economics, 18 (1990) pp. 233-240 Fazel Farzaneh, Fischer Klaus P., Gilbert Erika W.UITPurchasing Vs. EOQ with a price discount: An Analytical Comparison o f Inventory Costs, Int. J. Production Economics 54 (1998), pp.101-109 Fazel Farzaneh, A Comparative Analysis o f Inventory costs ofJIT and EOQ Purchasing, International Journal of Physical Distribution & Logistics Management, Vol 27 No.8.(1997), pp.496-504 63 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Gupta, Omprakash K., Kini Ranjan B., Is Price-Quantity Discount Dead In a JustIn-Time Environment?, Memational Journal o f Operations & Production Management, Vol. 15 No. 9 (1995), pp.261-270 Jones, Daniel J., JIT & The EOQ Model.Management Accounting, Management Accounting, February (1991),72,8, pp.54-57 Joshi, Kailash, Storage Space Costs and the EOQ Model, Journal of Supply Chain Management; Summer (1990); 26,3; pp.37-41 Munson, Charles L., Rosenblatt, Meir J., Coordinating a Three-level Supply Chain with Quantity Discounts, HE Transactions (2001) 33, pp.371-384 Munson, Charles L., Rosenblatt, Meir, Rosenblatt Zehava, The Use and Abuse o f Power in Supply Chains, Business Horizons/January-February (1999), pp.55-65 Pan, Andrew C.,Liao, Ching-Jong., An Inventory Model Under Just-In-Time Purchasing Agreements., Production and Inventory Management Journal, First Quarter 30,1 (1989) pp.49-52 Ramasesh, Ranga V.Jlecasting The Traditional Inventory Model To Implement Just-In-Time Purchasing, Production and Inventory Management Journal, First Quarter (1990) pp.71-75 Rao, S. Subba, Bahari-Kashani, Hamid, Economic Order Quantity and Storage Size-Some Considerations, Engineering Costs and Production Economics, 19 (1990), pp.201-204 Schniedeijans, Marc J., Cao, Qing., An Alternative Analysis o f Inventory costs o f JIT and EOQ purchasing. International Journal of Physical Distribution & Logistics Management. Vol 31 No.2 (2001) pp. 109-123 64 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Schniedeijans, Marc J., Cao, Qing., A Note on JTTPurchasing Vs. EOQ with Price Discounts: An Expansion o f Inventory Costs. International Journal of Production Economies, 65 (2000) pp.289-294 Fuller, Niall Waters, Just-In-Time Purchasing and Supply: A Review o f the Literature, International Journal of Operations & Production Management, Vol. 15. No.9 (1995), pp.220-236 65 ! Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDICES Appendix 1.1 Working of equation ftom Scheniederians & Cao + Lp ~ c,)d + CF or by substituting the value of Q from Z= j2 j2 2 2v4D. h Z = - jo ^ yiT% 2v4D 2 h I n yfr^ 2A D 2 O 2 + (^c^^ - c )d / / + CF 2 U Z = 4 2 Â D h -^ 4 2 lÂ D h + { c ^ ,-C j )d + CF Z = V ü 5 â \| i - ^ j + (fH )d + CF 66 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix 1.2 Working of Retailer’s Order quantity for no coordination no space From Equation (2.1.3) C M = ^ 5 ^ -" + +K - h M - ( C r +v + a)- (2.1.3) . !.. . derivative of the above equation is given as: xlk(. : ------ éiR — + .Y: 2' Setting the above equal to zero 2 lA^D & = J , (2 1-4) 67 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix 1,3 The following is the working of the retailer’s order quantity for Coordination, No Space. From Equation (2.2.4) aC ,(X .Q )_ SQ A,D Q‘a- P) A,D 2 ^ eVi-WA (2.2.4) 2 Setting equation (2.2.4) equal to zero jg r i - ; ) ; 2 2D{?u4^+A^) ( 1 - ^)X{h^ ( l - ( ^ ) + ( a - 1)) 2 2 68 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix 1.4 Working o f the Retailer’s order quantity for No Coordination, Space From Equation (2.3.4) (2.3.4) Equating to zero and solving for Q 4Z) _ p ,F (l-y g )+ 2 J ? ; G Y i- ) g ; L 2F 2 ^ DF & = 2A.DV (2.3.5) 69 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix 1.5 Case I No Coordination No Space: The table below shows the total retailers cost (Cr(Q)) tabulated against the value of each order quantity (Q). Highlighted in black is the minimum cost for Q=6369 6350 6351 6352 6353 6354 6 355 6356 6357 6358 6359 6360 6361 6362 6363 6364 6365 6366 6367 6368 3 .6 74,524.5973 3,67 4 ,5 2 4 .5 8 9 5 3 ,6 74.524.5822 3 ,6 74,524.5752 3 ,6 7 4,524.5687 3 ,6 7 4,524.5626 3,674,524.5568 3.6 7 4 ,5 2 4 .5 5 1 5 3,6 7 4 ,5 2 4 .5 4 6 6 3 ,6 7 4,524.5422 3,674.524.5381 3.674,524.5344 3,674,524.5312 3,674.524.5284 3.674.524.5259 3,674,524.5239 3,674,524.5223 3,674,524.5211 3,674,524.5203 6370 6371 6372 6373 6374 6375 6376 6377 6378 6379 6380 6381 6382 3 ,6 7 4,524.5200 3,674,524.5204 3.674.524.5212 3,674.524.5225 3.674.524.5242 3,674.524.5262 3,674.524.5287 3,674,524.5316 3,674.524.5349 3,674,524.5386 3,674,524.5427 3,674,524.5472 3,674,524.5521 70 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. N ü a Ë 6383 6384 6385 6386 6387 6388 6389 3 ,6 7 4 ,5 2 4 .5 5 7 5 3 ,6 7 4 ,5 2 4 .5 6 3 2 3,6 7 4 ,5 2 4 .5 6 9 3 3,6 7 4 ,5 2 4 .5 7 5 9 3 ,6 7 4 ,5 2 4 .5 8 2 8 3 ,6 7 4 ,5 2 4 .5 9 0 2 3 ,6 7 4 ,5 2 4 .5 9 7 9 Case I No Coordination No Space: Figure below i lustrâtes the change in the retailers cost with the change in Q Retailer’s C ost Function 3,674,52^.6200 3,674,524.0000 3,674,524.5800 I 3,674,524.5600 -CrtQ)! 3,674,524.5400 3,674,524.5200 3,674,524.5000 3,674,524.4800 O) CM in oo in lO iS 1 0 (0 (0 ( 0 ro ro ro ro ro ( 0 ( fO 0 ( 0ro (O (O (O (O V fs. ro CO ^ ps. ro (O fs. ro (O o ro CO 00 CO ro (O to (O 6 ro (O O) 00 CO (O Q (Units) 71 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix 1.6 Case I No C ordlnation No Space: Table below shows the variation the supplier cost with the change in the lot size multiplier (1) I 2 3 4 5 6 7 8 9 5,318,309.64 5,247,211.32 5,227,333.28 5,220,260.31 5,218,309.37 5,218,919.44 5,220,992.95 5,223,981.11 5,227,579.03 The figure below illustrates the variation the suppliers cost (Cs(?^,Qe)) with the lot size multiplier (X.) Supplier's C o st Function 5 .3 4 0 .0 0 0 .0 0 5.3 2 0 .0 0 0 .0 0 5 .3 0 0 .0 0 0 .0 0 5 .2 8 0 .0 0 0 .0 0 of 5.2 6 0 .0 0 0 .0 0 I 5 .2 4 0 .0 0 0 .0 0 Cs(A, Q) 5.2 2 0 .0 0 0 .0 0 5 .2 0 0 .0 0 0 .0 0 ■??ar 'V - '- ' C 5 .1 8 0 .0 0 0 .0 0 5.1 6 0 .0 0 0 .0 0 5 6 A 72 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix 1.7 Case II: Coordination. No Space The following are the values obtained for each value of 1 for Q=Q-3 V to Q+3 V 3,228 3,229 3,230 3,231 3,232 3,233 3,234 3,235 3,236 3,237 3,238 3,239 3,240 3,241 3,242 3,243 3,244 3,245 3,246 3,247 3,248 3,249 3,250 3,251 3,252 3,253 3,254 3,255 3,256 3,257 3,258 3,259 3,260 3,261 3,262 3,263 3,264 3,678,505.6124 3,678,501.8461 3,678,498.0829 3,678,494.3229 3,678,490.5660 3,678,486.8122 3,678,483.0616 3,678,479.3141 3,678,475.5697 3,678,471.8284 3,678,468.0902 3,678,464.3552 3,678,460.6232 3,678,456.8944 3,678,453.1687 3,678,449.4461 3,678,445.7265 3,678,442.0101 3,678,438.2967 3,678,434.5865 3,678,430.8793 3,678,427.1752 3,678,423.4742 3,678,419.7763 3,678,416.0814 3,678,412.3897 3,678,408.7010 3,678,405.0153 3,678,401.3327 3,678,397.6532 3,678,393.9768 3,678,390.3034 3,678,386.6330 3,678,382.9657 3,678,379.3014 3,678,375.6402 3,678,371.9821 5,206,256.5995 5,206,260.1550 5,206,263.7111 5,206,267.2677 5,206,270.8250 5,206,274.3829 5,206,277.9414 5,206,281.5005 5,206,285.0602 5,206,288.6204 5,206,292.1813 5,206,295.7427 5,206,299.3048 5,206,302.8674 5,206,306.4306 5,206,309.9945 5,206,313.5589 5,206,317.1239 5,206,320.6894 5,206,324.2556 5,206,327.8224 5,206,331.3897 5,206,334.9577 5,206,338.5262 5,206,342.0953 5,206,345.6650 5,206,349.2353 5,206,352.8061 5,206,356.3776 5,206,359.9496 5,206,363.5222 5,206,367.0954 5,206,370.6692 5,206,374.2436 5,206,377.8185 5,206,381.3940 5,206,384.9701 8,884,762.2118 8,884,762.0010 8,884,761.7940 8,884,761.5906 8,884,761.3910 8,884,761.1952 8,884,761.0030 8,884,760.8146 8,884,760.6298 8,884,760.4488 8,884,760.2715 8,884,760.0979 8,884,759.9280 8,884,759.7618 8,884,759.5993 8,884,759.4405 8,884,759.2854 8,884,759.1339 8,884,758.9862 8,884,758.8421 8,884,758.7017 8,884,758.5650 8,884,758.4319 8,884,758.3025 8,884,758.1768 8,884,758.0547 8,884,757.9362 8,884,757.8215 8,884,757.7103 8,884,757.6028 8,884,757.4990 8,884,757.3988 8,884,757.3022 8,884,757.2093 8,884,757.1199 8,884,757.0342 8,884,756.9522 73 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3,265 3,266 3,267 3,268 3,269 3,270 3,271 3,272 3,273 3,274 3,275 3,276 3,277 3,278 3,279 3,280 3,281 3,282 3,283 3,2S4 3,285 3,286 3,287 2,971 2,972 2,973 2,974 2,975 2,976 2,977 2,978 2,979 2,980 2,981 2,982 2,983 2,984 2,985 2,986 2,987 3,678,368.3269 3,678,364.6748 3,678,361.0258 3,678,357.3797 3,678,353.7367 3,678,350.0968 3,678,346.4598 3,678,342.8259 3,678,339.1949 3,678,335.5670 3,678,331.9421 3,678,328.3202 3,678,324.7014 3,678,321.0855 3,678,317.4726 3,678,313.8627 3,678,310.2558 3,678,306.6519 3,678,303.0510 3,678,299.4531 3,678,295.8581 3,678,292.2662 3,678,288.6772 3,679,586.6701 3,679,581.9891 3,679,577.3122 3,679,572.6393 3,679,567.9704 3,679,563.3055 3,679,558.6447 3,679,553.9878 3,679,549.3350 3,679,544.6861 3,679,540.0412 3,679,535.4003 3,679,530.7634 3,679,526.1305 3,679,521.5016 3,679,516.8766 3,679,512.2556 5,206,388.5468 5,206,392.1240 5,206,395.7019 5,206,399.2803 5,206,402.8593 5,206,406.4388 5,206,410.0189 5,206,413.5997 5,206,417.1809 5,206,420.7628 5,206,424.3452 5,206,427.9282 5,206,431.5118 5,206,435.0960 5,206,438.6807 5,206,442.2660 5,206,445.8518 5,206,449.4383 5,206,453.0253 5,206,456.6129 5,206,460.2010 5,206,463.7897 5,206,467.3790 8,884,756.8737 8,884,756.7989 8,884,756.7276 8,884,756.6600 8,884,756.5960 8,884,756.5356 8,884,756.4787 8,884,756.4255 8,884,756.3759 8,884,756.3298 8,884,756.2874 8,884,756.2485 8,884,756.2132 8,884,756.1814 8,884,756.1533 8,884,756.1287 8,884,756.1077 8.884,756.0902 8,884,756.0763 8,884,756.0659 8,884,756.0591 8,884,756.0559 8,884,756.0562 5,204,131.7000 5,204,136.1411 5,204,140.5828 5,204,145.0252 5,204,149.4683 5,204,153.9120 5,204,158.3564 5,204,162.8014 5,204,167.2471 5,204,171.6935 5,204,176.1405 5,204,180.5881 5,204,185.0364 5,204,189.4853 5,204,193.9349 5,204,198.3851 5,204,202.8360 8,883,718.3701 8,883,718.1302 8,883,717.8951 8,883,717.6646 8,883,717.4387 8,883,717.2176 8,883,717.0011 8,883,716.7893 8,883,716.5821 8,883,716.3796 8,883,716.1817 8,883,715.9884 8,883,715.7998 8,883,715.6158 8,883,715.4365 8,883,715.2618 8,883,715.0916 74 Reproduced w ith permission of the copyright owner. Further reproduction prohibited without permission. 2,988 j 2,989 2,990 2,991 2,992 2,993 2,994 2,995 2,996 2,997 2,998 2,999 3,679,507.6386 3,679,503.0255 3,679,498.4164 3,679,493.8112 3,679,489.2100 3,679,484.6127 3,679,480.0194 3,679,475.4300 3,679,470.8445 3,679,466.2630 3,679,461.6854 3,679,457.1117 5,204,207.2876 5,204,211.7397 5,204,216.1926 5,204,220.6460 5,204,225.1002 5,204,229.5549 5,204,234.0103 5,204,238.4664 5,204,242.9230 5,204,247.3804 5,204,251.8383 5,204,256.2970 8,883,714.9261 8,883,714.7652 8,883,714.6089 8,883,714.4573 8,883,714.3102 8,883,714.1676 8,883,714.0297 8,883,713.8964 8,883,713.7676 8,883,713.6434 8,883,713.5238 8,883,713.4087 3,000 3,001 3,002 3,003 3,004 3,005 3,006 3,007 3,008 3,009 3,010 3,011 3,012 3,013 3,014 3,015 3,016 3,017 3,018 1 3,0191 3,020 3,679,452.5419 3,679,447.9761 3,679,443.4141 3,679,438.8561 3,679,434.3020 3,679,429.7517 3,679,425.2054 3,679,420.6629 3,679,416.1244 3,679,411.5897 5,204,260.7562 5,204,265.2161 5,204,269.6766 5,204,274.1378 5,204,278.5996 5,204,283.0620 5,204,287.5251 5,204,291.9888 5,204,296.4532 5,204,300.9181 5,204,305.3837 5,204,309.8500 5,204,314.3169 5,204,318.7844 5,204,323.2525 5,204,327.7213 5,204,332.1907 5,204,336.6607 5,204,341.1313 5,204,345.6026 5,204,350.0745 5,204,354.5471 5,204,359.0202 5,204,363.4940 5,204,367.9684 5,204,372.4434 5,204,376.9191 5,204,381.3954 5,204,385.8723 5,204,390.3498 5,204,394.8279 8,883,713.2982 8,883,713.1922 8,883,713.0908 8,883,712.9939 8,883,712.9016 8,883,712.8138 8,883,712.7305 8,883,712.6517 8,883,712.5775 8,883,712.5078 8,883,712.4426 8,883,712.3819 8,883,712.3257 8,883,712.2740 8,883,712.2268 8,883,712.1841 8,883,712.1459 8,883,712.1121 8,883,712.0829 8,883,712.0581 8,883,712.0377 8,883,712.0219 8,883,712.0105 8,883,712.0035 8,883,712.0010 8,883,712.0029 8,883,712.0093 8,883,712.0201 8,883,712.0354 8,883,712.0550 3,021 3,022 3,023 3,024 3,025 3,026 3,027 3,028 3,029 3,030 3,679,407.0589 3,679,402.5319 3,679,398.0089 3,679,393.4897 3,679.388.9743 3,679,384.4628 3,679,379.9552 3,679,375.4515 3,679,370.9515 3,679,366.4555 3,679,361.9632 3,679,357.4748 3,679,352.9902 3,679,348.5095 3,679,344.0326 3,679,339.5595 3,679,335.0902 3,679,330.6247 3,679,326.1631 3,679,321.7052 3,679,317.2512 8,883,712.0791 75 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2,766 2,767 2,768 2,769 2,770 2,771 2,772 2,773 2,774 2,775 2,776 2,777 2,778 2,779 2,780 2,781 2,782 2,783 2,784 2,785 2 786 2,787 2,788 2,789 2,790 2,791 2,792 2,793 2,794 2,795 2,796 2,797 2,798 2,799 2,800 2,801 2,802 2,803 2,804 2,805 2,806 2,807 2,808 3,680,637.6132 3,680,632.0127 3,680,626.4172 3,680,620.8266 3,680,615.2410 3,680,609.6604 5,202,769.2160 5,202,774.5456 5,202,779.8759 5,202,785.2069 5,202,790.5387 3,680,604.0848 5,202,801.2042 5,202,806.5381 5,202,811.8727 5,202,817.2079 5,202,822.5439 5,202,827.8806 5,202,833.2180 5,202,838.5561 5,202,843.8949 5,202,849.2344 5,202,854.5746 5,202,859.9155 5,202,865.2571 5,202,870.5994 5,202,875.9424 5,202,881.2861 5,202,886.6305 3,680,598.5141 3,680,592.9484 3,680,587.3876 3,680,581.8318 3,680,576.2809 3,680,570.7350 3,680,565.1939 3,680,559.6578 3,680,554.1267 3,680,548.6004 3,680,543.0790 3,680,537.5626 3,680,532.0510 3,680,526.5443 3,680,521.0425 3,680,515.5456 5,202,795.8711 3,680,510.0536 5,202,891.9756 3,680,504.5665 3,680,499.0842 3,680,493.6067 3,680,488.1341 3,680,482.6664 3,680,477.2035 3,680,471.7455 3,680,466.2923 3,680,460.8439 3,680,455.4003 3,680,449.9616 3,680,444.5277 3,680,439.0985 3,680,433.6742 3,680,428.2547 3,680,422.8400 3,680,417.4300 3,680,412.0249 3,680,406.6245 5,202,897.3214 5,202,902.6678 5,202,908.0150 5,202,913.3629 5,202,918.7114 5,202,924.0607 5,202,929.4106 5,202,934.7613 5,202,940.1126 5,202,945.4646 5,202,950.8173 5,202,956.1707 5,202,961.5248 5,202,966.8796 5,202,972.2350 5,202,977.5912 5,202,982.9480 5,202,988.3055 5,202,993.6637 8,883,406.8292 8,883,406.5583 8,883,406.2931 8,883,406.0335 8,883,405.7797 8,883,405.5315 8,883,405.2890 8,883,405.0522 8,883,404.8211 8,883,404.5956 8,883,404.3757 8,883,404.1615 8,883,403.9530 8,883,403.7500 8,883,403.5527 8,883,403.3610 8,883,403.1750 8,883,402.9945 8,883,402.8197 8,883,402.6504 8,883,402.4867 8,883,402.3286 8,883,402.1761 8,883,402.0292 8,883,401,8878 8,883,401.7520 8,883,401.6217 8,863,401.4970 8,883,401.3779 8,883,401.2642 8,883,401.1561 8,883,401.0536 8,883,400.9565 8,883,400.8650 8,883,400.7789 8,883,400.6984 8,883,400.6233 8,883,400.5538 8,883,400.4897 8,883,400.4311 8,883,400,3780 8,883,400.3303 8,883,400.2882 76 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2,809 2,810 2,811 2,812 2,813 2,814 2,815 2,816 2,817 2,818 2,819 2,820 2,821 2,822 2,823 2,824 2,825 3,680,401.2289 3,680,395.8380 3,680,390.4520 3,680,385.0706 3,680,379.6941 3,680,374.3223 3,680,368.9552 3,680,363.5928 3,680,358.2352 3,680,352.8823 3,680,347.5342 3,680,342.1907 3,680,336.8520 3,680,331.5180 3,680,326.1886 3,680,320.8640 3,680,315.5441 5,202,999.0225 5,203,004.3821 5,203,009.7423 5,203,015.1032 5,203,020.4648 5,203,025.8271 5,203,031.1900 5,203,036.5536 5,203,041.9179 5,203,047.2829 5,203,052.6485 5,203,058.0148 5,203,063.3818 5,203,068.7495 5,203,074.1178 5,203,079.4868 5,203,084.8565 8,883,400.2514 8,883,400.2201 8,883,400.1943 8,883,400.1739 8,883,400.1589 8,883,400.1493 8,883,400.1452 8,883,400.1464 8,883,400.1531 8,883,400.1652 8,883,400.1827 8,883,400.2056 8,883,400.2338 8,883,400.2674 8,883,400.3065 8,883,400.3508 8,883,400.4006 2,597 2,598 2,599 2,600 2,601 2,602 2,603 2,604 2,605 2,606 2,607 2,608 2,609 2,610 2,611 2,612 2,613 2,611 2,615 2,616 2,617 2,618 2,619 3,681,660.4642 3,681,653.9362 3,681,647.4143 3,681,640.8984 3,681,634.3885 3,681,627.8846 3,681,621.3867 3,681,614.8948 3,681,608.4089 3,681,601.9289 3,681,595.4549 3,681,588.9869 3,681,582.5249 3,681,576.0688 3,681,569.6186 3,681,563.1744 3,681,556.7361 3,681,550.3037 3,681,543.8772 3,681,537.4567 3,681,531.0420 3,681,524.6333 3,681,518.2304 5,201,897.8884 5,201,904.1080 5,201,910.3284 5,201,916.5496 5,201,922.7715 5,201,928.9942 5,201,935.2177 5,201,941.4419 5,201,947.6669 5,201,953.8926 5,201,960.1191 5,201,966.3464 5,201,972.5744 5,201,978.8032 5,201,985.0327 5,201,991.2630 5,201,997.4940 5,202,003.7258 5,202,009.9583 5,202,016.1916 5,202,022.4257 5,202,028.6604 5,202,034.8960 8,883,558.3526 8,883,558.0442 8,883,557.7427 8,883,557.4480 8,883,557.1600 8,883,556.8788 8,883,556.6044 8,883,556.3367 8,883,556.0758 8,883,555.8216 8,883,555.5741 8,883,555.3333 8,883,555.0993 8,883,554.8719 8,883,554.6513 8,883,554.4374 8,883,554.2301 8,883,554.0295 8,883,553.8356 8,883,553.6483 8,883,553.4677 8,883,553.2937 8,883,553.1264 77 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2,620 2,621 2,622 2,623 2,624 2,625 2,626 2,627 2,628 2,629 2,630 2,631 2,632 2,633 2,634 2,635 2,636 2,637 2,638 2,639 2,640 2,641 2,642 2,643 2,644 2,645 2,646 2,647 2,648 2,649 2,650 2,651 2,652 2,653 2,654 2,655 2,656 3,681,511.8334 3,681,505.4423 3,681,499.0571 3,681,492.6777 3,681,486.3041 3,681,479.9365 3,681,473.5746 3,681,467.2186 3,681,460.8685 3,681,454.5241 3,681,448.1856 3,681,441.8528 3,681,435.5259 3,681,429.2048 3,681,422.8894 3,681,416.5799 3,681,410.2761 3,681,403.9781 3,681,397.6858 3,681,391.3993 3,681,385.1186 3,681,378.8436 3,681,372.5743 3,681,366.3107 3,681,360.0529 3,681,353.8008 3,681,347.5544 3,681,341.3137 3,681,335.0787 3,681,328.8494 3,681,322.6258 3,681,316.4079 3,681,310.1956 3,681,303.9890 3,681,297.7881 3,681,291.5928 3,681,285.4031 5,202,041.1323 5,202,047.3693 5,202,053.6070 5,202,059.8456 5,202,066.0848 5,202,072.3248 5,202,078.5656 5,202,084.8070 5,202,091.0493 5,202,097.2922 5,202,103.5359 5,202,109.7804 5,202,116.0255 5,202,122.2714 5,202,128.5181 5,202,134.7655 5,202,141.0136 5,202,147.2624 5,202,153.5120 5,202,159.7623 5,202,166.0133 5,202,172.2651 5,202,178.5176 5,202,184.7708 5,202,191.0248 5,202,197.2795 5,202,203.5349 5,202,209.7910 5,202,216.0478 5,202.222.3054 5,202,228.5637 5,202,234.8227 5,202,241.0824 5,202,247.3429 5,202,253.6041 5,202,259.8660 5,202,266.1286 03i2,456 2,457 3,682,644.7400 3,682,637.2873 5,201,369.7222 5,201,376.8342 8,883,552.9657 8,883,552.8116 8,883,552.6641 8,883,552.5232 8,883,552.3890 8,883,552.2613 8,883,552.1402 8,883,552.0257 8,883,551.9177 8,883,551.8163 8,883,551.7215 8,883,551.6332 8,883,551.5514 8,883,551.4762 8,883,551.4075 8,883,551.3453 8,883,551.2897 8,883,551.2405 8,883,551.1978 8,883,551.1616 8,883,551.1319 8,883,551.1087 8,883,551.0919 8,883,551.0816 8,883,551.0777 8,883,551.0803 8,883,551.0893 8,883,551.1047 8,883,551.1266 8,883,551.1548 8,883,551.1895 8,883,551.2.306 8,883,551.2780 8,883,551.3319 8,883,551.3921 8,883,551.4587 8,883,551.5317 : CcfX,0) ” 8,884,014.4621 8,884,014.1215 78 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2,458 2459 2,460 2,461 2,462 2,463 2,464 2,465 2,466 2,467 2,468 2,469 2,470 2,471 2,472 2,473 2,474 2,475 2,476 2,477 2,478 2,479 2,480 2,481 2,482 2,483 2,484 2,485 2,486 2,487 2,488 2,489 2,490 2,491 2,492 2,493 2,494 2,495 2,496 2,497 2,498 2,499 2,500 3,682,629.8418 3,682,622.4034 3,682,614.9721 3,682,607.5479 3,682,600.1308 3,682,592.7208 3,682,585.3178 3,682,577.9219 3,682,571.5331 3,682,563.1513 3,682,555.7765 3,682,548.4088 3,682,541.0481 3,682,533.6944 3,682,526.3477 3,682,519.0080 3,682,511.6753 3,682,504.3496 3,682,497.0308 3,682,489.7190 3,682,482.4142 3,682,475.1163 3,682,467.8253 3,682,460.5412 3,682,453.2641 3,682,445.9939 3,682,438.7306 3,682,431.4742 3,682,424.2246 3,682,416.9820 3,682,409.7462 3,682,402.5173 3,682,395.2952 3,682,388.0800 3,682,380.8716 3,682,373.6700 3,682,366.4753 3,682,359.2873 3,682,352.1062 3,682,344.9319 3,682,337.7643 3,682,330.6035 3,682,323.4495 5,201,383.9470 5,201,391.0606 5,201,398.1751 5,201,405.2903 5,201,412.4064 5,201,419.5233 5,201,426.6410 5,201,433.7594 5,201,440.8787 5,201,447.9988 5,201,455.1197 5,201,462.2415 5,201,469.3640 5,201,476.4873 5,201,483.6114 5,201,490.7363 5,201,497.8620 5,201,504.9885 5,201,512.1158 5,201,519.2439 5,201,526.3728 5,201,533.5025 5,201,540.6330 5,201,547.7643 5,201,554.8963 5,201,562.0292 5,201,569.1628 5,201,576.2972 5,201,583.4325 5,201,590.5685 5,201,597.7052 5,201,604.8428 5,201,611.9812 5,201,619.1203 5,201,626.2602 5,201,633.4009 5,201,640.5424 5,201,647.6847 5,201,654.8277 5,201,661.9715 5,201,669.1161 5,201,676,2614 5,201,683.4076 8,884,013.7888 8,884,013.4640 8,884,013.1472 8,884,012.8383 8,884,012.5372 8,884,012.2441 8,884,011.9588 8,884,011.6814 8,884,011.4118 8,884,011.1501 3,884,010.8963 8,884,010.6503 8,884,010.4121 8,884,010.1817 8,884,009.9591 8,884,009.7443 8,884,009.5373 8,884,009.3381 8,884,009.1466 8,884,008.9629 8,884,008.7870 8,884,008.6188 8,884,008.4583 8,884,008.3055 8,884,008.1604 8,884,008.0231 8,884,007.8934 8,884,007.7714 8,884,007.6571 8,884,007.5504 8,884,007.4514 8,884,007.3601 8,884,007.2764 8,884,007.2003 8,884,007.1318 8,884,007.0709 8,884,007.0177 8,884,006.9720 8,884,006.9339 8,884,006.9034 8,884,006.8804 8,884,006.8650 8,884,006.8571 79 Reproduced with permission o f th e c o p y rig h t ow n er. Further reproduction prohibited without permission. 2,501 2,502 2,503 2,504 2,505 2,506 2,507 2,508 2,509 2,510 2,511 2,512 2.513 2,514 2,515 3,682,316.3023 3,682,309.1618 3,682,302.0281 3,682,294.9011 3,682,287.7808 3,682,280.6672 3,682,273.5604 3,682,266.4603 3,682,259.3668 3,682,252.2801 3,682,245.2000 3,682,238.1266 3,682,231.0599 3,682,223.9999 3,682,216.9464 5,201,690.5545 5,201,697.7022 5,201,704.8506 5,201,711.9998 5,201,719.1498 5,201,726.3006 5,201,733.4521 5,201,740.6044 5,201,747.7574 5,201,754.9113 5,201,762.0659 5,201,769.2212 5,201,776.3773 5,201,783.5342 5,201,790.6918 8,884,006.8568 8,884,006.8640 8,884,006.8787 8,884,006.9009 8,884,006.9306 8,884,006.9678 8,884,007.0125 8,884,007.0647 8,884,007.1243 8,884,007.1914 8,884,007.2659 8,884,007.3478 8,884,007.4372 8,884,007.5340 8,884,007.6383 60 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix 1.8 Case III No Cordlnation Space: The table below shows the total retailers cost (Cr(Q)) tabulated against the value of each order quantity (Q). 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 wrnmÜ R ffi 2709010 2708816 2708623 2708431 2708239 2708048 2707858 541 542 543 544 545 546 2707669 547 2707481 2707293 2708906 2708719 2708533 2708348 2708164 2707981 2707798 2707615 2707434 2707253 548 549 540 550 551 552 553 554 555 556 557 558 5fQ 2708873 2708693 2703514 2708336 2708159 2707982 2707805 2707630 2707455 2707281 2708907 2708734 2708562 2708390 2708219 2708048 2707878 2707709 2707540 2707372 81 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Case III No Coordination Space: Figure below illustrates the change in the retailers cost with the change in Q Retailer's C ost Function 2709500 2709000 ■4i ■iiÿÿ&âasiMà&t'k 2708500 5 2708000 ^ 2707500 ■Cr(Q)i 2707000 2706500 2706000 Q (Units) 82 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix 1.9 Case III No Coordination Space: Table below shows the variation the supplier cost with the change in the lot size multiplier (X) ^ s É s to . 1 2 3 4 5 6 1,559,115 1,512,867 1,551,774 1,611,971 1,680,682 1,752,152 The figure below illustrates the variation the suppliers cost (Cs(X,Qe)) with the lot size multiplier (X.) Supplier C ost Function 1,800,000 1.750.000 1.700.000 1.650.000 1.600.000 a 6 1.550.000 (5 1.500.000 1.450.000 1.400.000 1.350.000 TOC ■Cs(A,Q) 2 3 4 5 Lot S ize Multiplier (A) 83 Reprodficed with permission of the copyright owner. Further reproduction prohibited without permission.
https://openalex.org/W2131903215	https://europepmc.org/articles/pmc2222674?pdf=render	English	null	Seizures as the first manifestation of chromosome 22q11.2 deletion syndrome in a 40-year old man: a case report	Journal of medical case reports	2,007	cc-by	3,245	BioMed Centra Journal of Medical Case Reports Open Acces Case report Seizures as the first manifestation of chromosome 22q11.2 deletion syndrome in a 40-year old man: a case report Adriano R Tonelli1, Kalyan Kosuri1, Sainan Wei2 and Davoren Chick1 Address: 1Department of Internal Medicine, Michigan State University, East Lansing, Michigan, USA and 2DNA Diagnostic and Cytogenetics Laboratories, Department of Pediatrics and Human Development. Michigan State University, East Lansing, Michigan, USA Email: Adriano R Tonelli - Adriano.Tonelli@medicine.ufl.edu; Kalyan Kosuri - Kalyan.Kosuri@ht.msu.edu; Sainan Wei - weisaina@msu.edu; Davoren Chick - Davoren.Chick@hc.msu.edu * Corresponding author Abstract Background: The microdeletion of chromosome 22q11.2 is the most common human deletion syndrome It typically presents early in life and is rarely considered in adult patients As part of the Published: 3 December 2007 Journal of Medical Case Reports 2007, 1:167 doi:10.1186/1752-1947-1-167 Received: 14 April 2007 Accepted: 3 December 2007 This article is available from: http://www.jmedicalcasereports.com/content/1/1/167 © 2007 Tonelli et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. BioMed Central Published: 3 December 2007 Published: 3 December 2007 Journal of Medical Case Reports 2007, 1:167 doi:10.1186/1752-1947-1-167 This article is available from: http://www.jmedicalcasereports.com/content/1/1/167 © 2007 Tonelli et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Abstract Background: The microdeletion of chromosome 22q11.2 is the most common human deletion syndrome. It typically presents early in life and is rarely considered in adult patients. As part of the manifestations of this condition, patients can have parathyroid glandular involvement ranging from hypocalcemic hypoparathyroidism to normocalcemia with normal parathryroid hormone levels. The first manifestation of the syndrome might be seizures due to profound hypocalcemia. Case presentation: A 40-year-old man without significant past medical history presented with a new-onset generalized tonic-clonic seizure. He had no personal history of hypocalcemia or seizures. Physical examination was remarkable for short stature, hypertelorism, prominent forehead and nasal voice. His initial laboratory examination showed hypocalcemia (Calcium 5.2 mg/ dl and Calcium ionized 0.69 mmol/l) with hypoparathyroidism (Parathyroid hormone intact < 2.5 pg/ml. NV: 14–72 pg/ml). Urine Calcium was 3 mg/dl on a spot and 88 mg in a 24-hour urine collection (NV: 100–300 mg/24 hs). The electrocardiogram showed a prolonged corrected QT interval. Echocardiogram, abdominal ultrasound and electroencephalogram were normal. A computer tomography of the brain showed basal ganglia calcification. The subtle physical findings and the presence of idiopathic hypoparathyroidism motivated the performance of fluorescent in situ hybridization which demonstrated a microdeletion on one of the homologs 22q11.2. The patient was treated with calcium citrate and calcitriol with good response. Conclusion: Microdeletion of chromosome 22q11.2 is among the most clinically variable syndromes, with more than 180 features associated with the deletion. It has a variable phenotypical expression, requiring a high level of awareness for its early diagnosis. Seizures, related to marked hypocalcemia due to idiopathic hypoparathyroidism, might be the presenting feature in an adult patient with this syndrome. somal dominant inheritance observed in 15% of cases [3]. The deletion has been identified in most patients with DiGeorge syndrome, velocardiofacial syndrome and conotruncal anomaly face syndrome. The chromosome Open Acc Case report Seizures as the first manifestation of chromosome 22q11.2 deletion syndrome in a 40-year old man: a case report Adriano R Tonelli*1, Kalyan Kosuri1, Sainan Wei2 and Davoren Chick1 Open Access Received: 14 April 2007 Accepted: 3 December 2007 Received: 14 April 2007 Accepted: 3 December 2007 Background Background The microdeletion of chromosome 22q11.2 is the most common human deletion syndrome (1 in 4000 live births) [1,2]. Most deletions occur de novo, with auto- Page 1 of 5 (page number not for citation purposes) Page 1 of 5 (page number not for citation purposes) http://www.jmedicalcasereports.com/content/1/1/167 http://www.jmedicalcasereports.com/content/1/1/167 Journal of Medical Case Reports 2007, 1:167 22q11.2 deletion syndrome is characterized by a congen- ital failure in the development of the derivatives of various pharyngeal arches and pouches with absent or small par- athyroid glands [2]. It typically presents early in life and is rarely considered in adult patients. The classical features are conotruncal cardiac defects, dysmorphic facies, velo- pharyngeal insufficiency, immunodeficiency, learning disabilities and parathyroid dysfunction. of the nose over the bridge. The etiology of the seizures was attributed to the marked hypocalcemia noted on his initial laboratory evaluation (Calcium 5.2 mg/dl and Cal- cium ionized 0.69 mmol/l). Further work-up revealed hypoparathyroidism (Parathyroid hormone intact < 2.5 pg/ml. NV: 14–72 pg/ml) and a normal total 25 hydroxy- vitamin D (30 ng/ml. NV: 25–80 ng/ml). Thyroid stimu- lation hormone was normal. Urine Calcium was 3 mg/dl on a spot and 88 mg in a 24-hour urine collection (NV: 100–300 mg/24 hs). Parathyroid dysfunction in adults is generally due to acquired conditions such as autoimmune processes or injury of the parathyroid glands during surgery. Less com- monly a genetic disorder like chromosome 22q11.2 dele- tion syndrome can be the culprit condition. In this syndrome the parathyroid glandular dysfunction ranges from hypocalcemia with hypoparathyroidism to normo- calcemia with normal parathryroid hormone levels [1-3]. Various forms of hypoparathryoidism have been reported including late onset, transient, latent and recurrent [2]. The 12-lead electrocardiogram showed a prolonged cor- rected QT interval. Echocardiogram, abdominal ultra- sound and electroencephalogram were normal. A computer tomography of the brain showed basal ganglia calcification (figure 1). The presence of hypoparathy- roidism and subtle dysmorphic facial features made chro- mosome 22q11.2 deletion syndrome a possible diagnosis. For this reason, fluorescent in situ hybridiza- tion was performed using DNA fluorescent probes for the DiGeorge/velocardiofacial syndrome critical region (TUPLE1) at 22q11.2 and a control probe, arylsulfatase-A (ARSA), at 22q13.3, from a commercially available source (Vysis). This is a direct-labeled dual-color probe mixture with TUPLE1 (HIRA) probe labeled in orange and ARSA probe green (figure 2). At least 50 metaphase cells and 100 interphase cells were scored for both TUPLE1 and Case presentation A 40-year-old man presented with a new-onset general- ized tonic-clonic seizure. He did not have any history of hypocalcemia, seizures, psychiatric disorders or obvious cognitive impairments. Physical examination was remark- able for short stature, hypertelorism, hooded upper eye- lids, prominent forehead, hypernasal speech and fullness Brain computer tomography cuts of the patient, demonstrating basal ganglia and periventricular calcification Figure 1 Brain computer tomography cuts of the patient, demonstrating basal ganglia and periventricular calcification. mputer tomography cuts of the patient, demonstrating basal ganglia and periventricular calcification 1 mputer tomography cuts of the patient, demonstrating basal ganglia and periventricular calcification. Page 2 of 5 (page number not for citation purposes) p g p y p , g g g p g Brain computer tomography cuts of the patient, demonstrating basal ganglia and periventricular calcification. Page 2 of 5 (page number not for citation purposes) Journal of Medical Case Reports 2007, 1:167 http://www.jmedicalcasereports.com/content/1/1/167 Result of FISH analysis using LSI probe (TUPLE 1) from DiGeorge/velocardiofacial syndrome critical region Figure 2 Result of FISH analysis using LSI probe (TUPLE 1) from DiGeorge/velocardiofacial syndrome critical region. TUPLE 1 (HIRA) probe was labeled in Spectrum Orange and Arylsulfatase A (ARSA) in SpectrumGreen as control. Absence of the orange signal indicates deletion of the TUPLE 1 locus at 22q11.2. 22q11.2 Del 22q11.2 Del ●Tuple1 ●ARSA NORMAL NORMAL ●Tuple1 ●ARSA M E T A P H A S E I N T E R P H A S E NORMAL NORMAL ●Tuple1 ●ARSA 22q11.2 Del 22q11.2 Del ●Tuple1 ●ARSA M E T A P H A S E NORMAL NORMAL ●Tuple1 ●ARSA NORMAL NORMAL ●Tuple1 ●ARSA I N T E R P H A S E Result of Figure 2 Result of FISH analysis using LSI probe (TUPLE 1) from DiGeorge/velocardiofacial syndrome critical region Figure 2 Result of FISH analysis using LSI probe (TUPLE 1) from DiGeorge/velocardiofacial syndrome critical region. TUPLE 1 (HIRA) probe was labeled in Spectrum Orange and Arylsulfatase A (ARSA) in SpectrumGreen as control. Absence of the orange signal indicates deletion of the TUPLE 1 locus at 22q11.2. ARSA signals. All metaphase and interphase cells analyzed show a deletion of TUPLE1 at 22q11.2 locus. The patient was treated with calcium citrate and calcitriol. He remained asymptomatic with normalization of his serum calcium level. Case presentation that 21% of the patients with chromosome 22q11.2 dele- tion syndrome (62/290) had seizures, and at least 68% (42/62) of the seizures were hypocalcemic in origin [8]. There are only a few reports in the medical literature which describe new onset seizures caused by hypocal- cemia in adolescence or adulthood (one report [7]), in patients not previously diagnosed with chromosome 22q11.2 deletion syndrome [1,3,7,9,10]. Page 3 of 5 (page number not for citation purposes) Consent The patient provided written informed consent for publi- cation. SW: Carried out the molecular genetic studies. Revised the manuscript; Patients with chromosome 22q11.2 deletion syndrome can have a variety of brain abnormalities when assessed by neuroimaging. Basal ganglia calcification, similar to that seen in the patient presented in this case report, have been rarely described in the literature. Possible explana- tions include a low incidence of this finding or a lack of computer tomography studies being performed on such patients or being performed in patients who were too young to have developed this abnormality. Most of the patients with chromosome 22q11.2 syndrome were stud- ied with magnetic resonance imaging, possibly missing any calcification. The significance of this finding is unclear [1,10-13]. DC: Contributed to conception and design. Revised the manuscript. DC: Contributed to conception and design. Revised the manuscript. Discussion Microdeletion of chromosome 22q11.2 is among the most clinically variable syndromes, with more than 180 features associated with the deletion [1,4]. The phenotype develops over time; therefore the clinical presentation may change [4]. It has a variable phenotypic expression, with no pathognomonic or obligatory clinical features, and requires a high level of awareness for its early diagno- sis [1,5]. One of the clinical features of this syndrome is seizures, which can be related to hypocalcemia, stroke, cerebellar or cerebral cortical atrophy, or transient or chronic ischemia [1,4,6-8]. A large cohort study showed Some patients present with asymptomatic hypocalcemia and inappropriately low parathyroid hormone levels that leads to fluorescence in situ hybridization and the diagno- sis of chromosome 22q11.2 deletion syndrome [5,7,8]. The majority of patients are diagnosed soon after birth, when the transport of calcium from mother to fetus is abruptly interrupted. The low serum calcium level gener- ally improves over the first year of life as the parathyroid gland hypertrophies and the dietary calcium intake Page 3 of 5 (page number not for citation purposes) Page 3 of 5 (page number not for citation purposes) http://www.jmedicalcasereports.com/content/1/1/167 Journal of Medical Case Reports 2007, 1:167 http://www.jmedicalcasereports.com/content/1/1/167 increase. For this reason few older patients require ongo- ing calcium supplementation [7,8,11]. hypocalcemia, have their calcium-parathyroid hormone axis periodically checked and receive genetic counseled as there is a 50% risk of having an affected offspring [1,5]. Of note is that even when hypocalcemia is considered one of the cardinal features of the syndrome, the prevalence of this electrolyte abnormality depends on the selection cri- teria used, ranging from 13 to 72%. Patients with pheno- typic characteristics of the DiGeorge anomaly are more likely to have clinical evidence of hypocalcemia or to have calcium levels measured. A lower prevalence of hypocal- cemia is seen in clinical presentations more consistent with the velocardiofacial syndrome in which calcium lev- els may not be routinely checked [11]. In a large European cohort study of 558 patients known to have 22q11.2 dele- tions, hypocalcemia was noted in 60% of the subjects of whom 39% had presented with seizures [8]. Competing interests The author(s) declare that they have no competing inter- ests. Hypoparathyroidism can result in transient, persistent, late-onset or latent hypocalcemia. Patients who are diag- nosed later in life may have a latent form of hypoparath- yroidism [2,10]. There are insufficient data to describe the natural course of asymptomatic hypocalcemia diagnosed in older patients. In spite of this it is clear that in some patients, seizures may be the first manifestation of their hypocalcemia, either spontaneously or precipitated dur- ing periods of increased metabolic demand [3]. References References 1. Hiéronimus S, Bec-Roche M, Pedeutour F, Lambert JC, Wagner-Mal- her K, Mas JC, Sadoul JL, Fénichel P: The spectrum of parathyroid gland dysfunction associated with the microdeletion 22q11. European Journal of Endocrinology 2006, 155:47-52. 2. Al-Jenaidi F, Makitie O, Grunebaum E, Sochett E: Parathyroid gland dysfunction in 22q11.2 deletion syndrome. Horm Res 2007, 67:117-122. 3. Maalouf N, Sakhaee K, Odvina C: A case of chromosome 22q11 deletion syndrome diagnosed in a 32-year-old man with hypoparathyroidism. J Clin Endocrinol Metab 2004, 89:4817-4820. 4. Robin N, Shprintzen R: Defining the clinical spectrum of dele- tion 22q11.2. J Pediatr 2005, 147:90-96. 5. Taylor SC, Morris G, Wilson D, Davies SJ, Gregory JW: Hypopar- athyroidism and 22q11 deletion syndrome. Arch Dis Child 2003, 88:520-522. 6. Shprintzen FJ: Velo-cardio-facial syndrome. In Management of genetic syndromes 2nd edition. Edited by: Cassidy SB, Allanson J. New York: Wiley; 2004:615-632. 7. Kar PS, Ogoe B, Meeking D: Di-George syndrome presenting with hypocalcaemia in adulthood: two case reports and a review. J Clin Pathol 2005, 58:655-657. 8. Ryan AK, Goodship JA, Wilson DI, Philip N, Levy A, Seidel H, Schuffenhauer S, Oechsler H, Belohradsky B, Prieur M, Aurias A, Ray- References 1. Hiéronimus S, Bec-Roche M, Pedeutour F, Lambert JC, Wagner-Mal- her K, Mas JC, Sadoul JL, Fénichel P: The spectrum of parathyroid gland dysfunction associated with the microdeletion 22q11. European Journal of Endocrinology 2006, 155:47-52. 1. Hiéronimus S, Bec-Roche M, Pedeutour F, Lambert JC, Wagner-Mal- her K, Mas JC, Sadoul JL, Fénichel P: The spectrum of parathyroid gland dysfunction associated with the microdeletion 22q11. European Journal of Endocrinology 2006, 155:47-52. Treatment of severe symptomatic hypocalcemia requires prompt administration of parental calcium. In contrast, asymptomatic hypocalcemia may be treated with oral cal- cium and vitamin D supplements. Serum calcium levels should be maintained in the low-normal range to mini- mize hypercalciuria and the risk of development of renal calculi [11]. p J f gy , 2. Al-Jenaidi F, Makitie O, Grunebaum E, Sochett E: Parathyroid gland dysfunction in 22q11.2 deletion syndrome. Horm Res 2007, 67:117-122. p J f gy 2. Al-Jenaidi F, Makitie O, Grunebaum E, Sochett E: Parathyroid gland dysfunction in 22q11.2 deletion syndrome. Horm Res 2007, 67:117-122. 3. Maalouf N, Sakhaee K, Odvina C: A case of chromosome 22q11 deletion syndrome diagnosed in a 32-year-old man with hypoparathyroidism. J Clin Endocrinol Metab 2004, 89:4817-4820. 4. Conclusion Chromosome 22q11.2 deletion syndrome has a variable phenotypic expression. This diagnosis should be consid- ered in adult patients presenting with idiopathic hypopar- athyroidism. The absence of the classical features of this condition should not exclude this pathology. Seizures, related to marked hypocalcemia, might be the initial pres- entation of this deletion. Treatment is relatively simple and can bring about profound clinical improvement. Authors' contributions AT: Involved in drafting of the manuscript, contributions to conception and design, analysis of data; KK: Involved in acquisition of data, conception and design. Revised the manuscript; SW: Carried out the molecular genetic studies. Revised the manuscript; SW: Carried out the molecular genetic studies. Revised the manuscript; References Robin N, Shprintzen R: Defining the clinical spectrum of dele- tion 22q11.2. J Pediatr 2005, 147:90-96. 3. Maalouf N, Sakhaee K, Odvina C: A case of chromosome 22q11 deletion syndrome diagnosed in a 32-year-old man with hypoparathyroidism. J Clin Endocrinol Metab 2004, 89:4817-4820. 4. Robin N, Shprintzen R: Defining the clinical spectrum of dele- tion 22q11.2. J Pediatr 2005, 147:90-96. q J 5. Taylor SC, Morris G, Wilson D, Davies SJ, Gregory JW: Hypopar- athyroidism and 22q11 deletion syndrome. Arch Dis Child 2003, 88:520-522. q J 5. Taylor SC, Morris G, Wilson D, Davies SJ, Gregory JW: Hypopar- athyroidism and 22q11 deletion syndrome. Arch Dis Child 2003, 88:520-522. In summary, the diagnosis of chromosome 22q11.2 dele- tion syndrome during adulthood is uncommon. The con- dition is probably under diagnosed. It should be considered in adult patients presenting with idiopathic hypoparathyroidism. Patients with this syndrome should be informed of the symptoms that might occur with 6. Shprintzen FJ: Velo-cardio-facial syndrome. In Management of genetic syndromes 2nd edition. Edited by: Cassidy SB, Allanson J. New York: Wiley; 2004:615-632. 6. Shprintzen FJ: Velo-cardio-facial syndrome. In Management of genetic syndromes 2nd edition. Edited by: Cassidy SB, Allanson J. New York: Wiley; 2004:615-632. y 7. Kar PS, Ogoe B, Meeking D: Di-George syndrome presenting with hypocalcaemia in adulthood: two case reports and a review. J Clin Pathol 2005, 58:655-657. y 7. Kar PS, Ogoe B, Meeking D: Di-George syndrome presenting with hypocalcaemia in adulthood: two case reports and a review. J Clin Pathol 2005, 58:655-657. J 8. Ryan AK, Goodship JA, Wilson DI, Philip N, Levy A, Seidel H, Schuffenhauer S, Oechsler H, Belohradsky B, Prieur M, Aurias A, Ray- J 8. Ryan AK, Goodship JA, Wilson DI, Philip N, Levy A, Seidel H, Schuffenhauer S, Oechsler H, Belohradsky B, Prieur M, Aurias A, Ray- J 8. Ryan AK, Goodship JA, Wilson DI, Philip N, Levy A, Seidel H, Schuffenhauer S, Oechsler H, Belohradsky B, Prieur M, Aurias A, Ray- Page 4 of 5 (page number not for citation purposes) Page 4 of 5 (page number not for citation purposes) http://www.jmedicalcasereports.com/content/1/1/167 http://www.jmedicalcasereports.com/content/1/1/167 Journal of Medical Case Reports 2007, 1:167 http://www.jmedicalcasereports.com/content/1/1/167 mond FL, Clayton-Smith J, Hatchwell E, McKeown C, Beemer FA, Dallapiccola B, Novelli G, Hurst JA, Ignatius J, Green AJ, Winter RM, Brueton L, Brøndum-Nielsen K, Scambler PJ, et al.: Spectrum of clinical features associated with interstitial chromosome 22q11 deletions: a European collaborative study. References J Med Genet 1997, 34:798-804. mond FL, Clayton-Smith J, Hatchwell E, McKeown C, Beemer FA, Dallapiccola B, Novelli G, Hurst JA, Ignatius J, Green AJ, Winter RM, Brueton L, Brøndum-Nielsen K, Scambler PJ, et al.: Spectrum of clinical features associated with interstitial chromosome 22q11 deletions: a European collaborative study. J Med Genet 1997, 34:798-804. 9. Adachi M, Tachibana K, Masuno M, Makita Y, Maesaka H, Okada T, Hizukuri K, Imaizumi K, Kuroki Y, Kurahashi H, Suwa S: Clinical characteristics of children with hypoparathyroidism due to 22q11.2 microdeletion. Eur J Pediatr 1998, 157:34-38. 10. Scirè G, Dallapiccola B, Iannetti P, Bonaiuto F, Galasso C, Mingarelli R, Boscherini B: Hypoparathyroidism as the major manifesta- tion in two patients with 22q11 deletions. Am J Med Genet 1994, 52:478-482. 11. Weinzimer SA: Endocrine aspects of the 22q11.2 deletion syn- drome. Genetics in Medicine 2001, 3:19-22. 12. Sieberer M, Haltenhof H, Haubitz B, Pabst B, Miller K, Garlipp P: Basal ganglia calcification and psychosis in 22q11.2 deletion syndrome. Eur Psychiatry 2005, 20:567-569. y y y 13. Chow EW, Mikulis DJ, Zipursky RB, Scutt LE, Weksberg R, Bassett AS: Qualitative MRI findings in adults with 22q11 deletion syndrome and schizophrenia. Biol Psychiatry 1999, 46:1436-1442. Publish with BioMed Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp BioMedcentral Page 5 of 5 (page number not for citation purposes) Publish with BioMed Central and every scientist can read your work free of charge
https://openalex.org/W4285765929	https://bmcmedgenet.biomedcentral.com/counter/pdf/10.1186/s12881-019-0942-4	English	null	Post-lingual non-syndromic hearing loss phenotype: a polygenic case with 2 biallelic mutations in MYO15A and MITF	Research Square (Research Square)	2,019	cc-by	5,746	Khalil et al. BMC Medical Genetics (2020) 21:1 https://doi.org/10.1186/s12881-019-0942-4 Khalil et al. BMC Medical Genetics (2020) 21:1 https://doi.org/10.1186/s12881-019-0942-4 Open Access © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Abstract Background: Hearing loss (HL) represents the most common congenital sensory impairment with an incidence of 1–5 per 1000 live births. Non-syndromic hearing loss (NSHL) is an isolated finding that is not part of any other disorder accounting for 70% of all genetic hearing loss cases. Methods: In the current study, we reported a polygenic mode of inheritance in an NSHL consanguineous family using exome sequencing technology and we evaluated the possible effect of the detected single nucleotide variants (SNVs) using in silico methods. Results: Two bi-allelic SNVs were detected in the affected patients; a MYO15A (. p.V485A) variant, and a novel MITF (p.P338L) variant. Along with these homozygous mutations, we detected two heterozygous variants in well described hearing loss genes (MYO7A and MYH14). The novel MITF p. Pro338Leu missense mutation was predicted to change the protein structure and function. Conclusion: A novel MITF mutation along with a previously described MYO15A mutation segregate with an autosomal recessive non-syndromic HL case with a post-lingual onset. The findings highlight the importance of carrying whole exome sequencing for a comprehensive assessment of HL genetic heterogeneity. Keywords: Congenital hearing loss, Non-syndromic hearing loss, MITF, MYO15A, Whole exome sequ Post-lingual non-syndromic hearing loss phenotype: a polygenic case with 2 biallelic mutations in MYO15A and MITF Athar Khalil1†, Samer Bou Karroum1†, Rana Barake2, Gabriel Dunya2, Samer Abou-Rizk2, Amina Kamar1, Georges Nemer1,3* and Marc Bassim2* Background either in an autosomal recessive manner (75–80%), auto- somal dominant manner (20–25%), X- linked or in rare cases by mitochondrial inheritance (1–2%) [4]. Up to date, over 115 genes have been linked to non-syndromic HL with GJB2, SLC26A4, MYO15A, OTOF, and CDH23 being considered as the most commonly identified genes. Some of these genes were shown to be associated with both re- cessive and dominant form of the disease [5, 6]. With a prevalence of 1 to 5 per 1000 births, hearing loss (HL) represents the most common congenital sensory impairment. Congenital hearing loss could be either due to hereditary/non-hereditary genetic factors, or due to certain complications during pregnancy and childbirth [1]. Most of the cases (~ 60%) are attributed to genetic causes with more than 150 genes identified to be associated with either syndromic or non-syndromic form of this disease [2, 3]. Non-syndromic hearing loss (NSHL) accounts for 70% of genetic HL cases that are usually not associated with other signs and symptoms. NSHL can be inherited With the advent of next-generation sequencing (NGS), genetic mapping within large, clinically well-characterized families with NSHL provides a powerful approach for mapping critical chromosomal intervals which when mu- tated could be responsible for this phenotype. In the Mid- dle East, the high rate of consanguineous marriages favors the incidence of autosomal recessive diseases such as that of NSHL [7]. Unfortunately, despite this high prevalence, the needed genetic linkage studies using NGS technolo- gies are still not very well established [8]. * Correspondence: gn08@aub.edu.lb; mb81@aub.edu.lb †Athar Khalil and Samer Bou Karroum contributed equally to this work. 1Departments of Biochemistry and Molecular Genetics, Faculty of Medicine, American University of Beirut, Beirut, Lebanon 2Otolaryngology – Head and Neck Surgery, Faculty of Medicine, American University of Beirut, Beirut, Lebanon Full list of author information is available at the end of the article * Correspondence: gn08@aub.edu.lb; mb81@aub.edu.lb †Athar Khalil and Samer Bou Karroum contributed equally to this work. 1Departments of Biochemistry and Molecular Genetics, Faculty of Medicine, American University of Beirut, Beirut, Lebanon 2Otolaryngology – Head and Neck Surgery, Faculty of Medicine, American University of Beirut, Beirut, Lebanon Full list of author information is available at the end of the article Methods Subjects Two young siblings presented to the Department of Oto- laryngology - Head and Neck Surgery at American Univer- sity of Beirut (AUB) with a complaint of late-onset HL. These patients, along with their consanguineous family, were included in the ongoing study of the genetic basis of HL in Lebanon. Family members received a complete oto- laryngologic examination, in addition to pure tone audiom- etry testing. They were also referred to Ophthalmology, Cardiology and Nephrology for identification of possible other congenital abnormalities and ruling out syndromic HL. A follow-up examination was done for one available af- fected patient (II.5) and her parents after 4 years from the first visit. The study was approved by the Institutional Re- view Board (IRB) at the American University of Beirut (protocol number:OTO.MB1.02). A follow-up audiogram for the patient (II.5) indicated a stable hearing after 4 years from the initial diagnosis. In addition no features of any syndromic disease that are usually initiated after puberty were detected. Clinical manifestation The family consists of consanguineous parents with two sisters diagnosed with post-lingual hearing impairment and four unaffected brothers (Fig. 1). HL was noted in the two sisters (II.5/II.6) at the age of six and twelve, respect- ively. Physical examination did not demonstrate any dys- morphic features suggestive of a syndromic disease. Both patients were reported not to have any pigmentary changes in hair, eyes, or skin. No visual complaints includ- ing night blindness, visual field loss and decrease in cen- tral vision were detected. Audiogram analysis of this family revealed that the two siblings had a bilateral HL. Puretone audiometry for patients revealed approximately similar pattern of a “cookie-bite audiogram” with mild HL in the low frequencies, sloping to borderline severe in mid frequencies, and rising to moderate in high frequencies (Fig. 2). Word discrimination score was excellent for both patients at the time of referral. Page 2 of 8 Page 2 of 8 Page 2 of 8 Khalil et al. BMC Medical Genetics (2020) 21:1 Khalil et al. BMC Medical Genetics (2020) 21:1 In this study, we report a polygenic mode of inherit- ance in an NSHL consanguineous family using exome sequencing analysis. Accordingly, we propose for the first time the involvement of a novel MITF variant along with a previously described MYO15A mutation in non- syndromic HL disease with post-lingual onset. Exome sequencing Exome sequencing of the four family members achieved approximately (95%) mean exome coverage, at coverage of (8X). From a total number of around 58,000 variants, we only analyzed those that occur in the coding regions of the genes. We filtered variants via a list of 155 genes used for clinical diagnosis of HL while including only missense, frameshift, splice and stop gained alterations with a minor allele frequency (MAF) of < 0.01 (Additional file 1: Table Blood samples were collected from the family members and DNA extraction was performed using the QIAamp Blood Midi Kit (Qiagen Sciences, Inc., Germantown, MD), using the manufacturer’s instructions. DNA quantification was also performed through the NanoDrop (Thermo Fisher Scientific, Inc., Waltham, MA) at the molecular core facility at AUB. One microgram of coded DNA samples from both parents and the two patients were shipped to Macrogen (South Korea), where exome sequencing was performed using the V5 SureSelect Target Enrichment Capture system from Agilent on a HiSeq 4000 platform from Illumina. Fig. 1 Family’s phenotype and genotype. The pedigree of the enrolled family, with affected individuals marked in grey. Possible causative variants of the affected sisters and those of the parents are listed Data analysis Primary analysis was done at Macrogen. Generated FASTQ files were mapped to the reference genome using the SureCall software from Agilent technologies. The Illimuna Variant Studio was used for annotation and variant calls. The Integrative Genomics Viewer (IGV) was also used as a high-performance visualization tool for genomic annotations [9]. To assess the patho- genicity of possible candidates, we used SIFT (http://sift. jcvi.org/), PolyPhen2 (http://genetics.bwh.harvard.edu/ pph2/),MutationTaster (http://www.mutationtaster.org/), and GERP++ (http://mendel.stanford.edu/ SidowLab/ downloads/gerp/) scores to predict deleterious variants. To predict the effect of the detected mutations on the protein structure and stability, we used DUET software (http://biosig.unimelb.edu.au/duet/stability). Fig. 1 Family’s phenotype and genotype. The pedigree of the enrolled family, with affected individuals marked in grey. Possible causative variants of the affected sisters and those of the parents are listed Khalil et al. BMC Medical Genetics Khalil et al. BMC Medical Genetics (2020) 21 Page 3 of 8 (2020) 21:1 Fig. 2 Audiograms of the affected probands. The audiograms show mild to severe progressive hearing loss in both ears for both affected individuals (II.5) and (II.6). The audiograms were taken at the time of diagnosis w mild to severe progressive hearing loss in both ears for both affected of diagnosis Fig. 2 Audiograms of the affected probands. The audiograms show mild to severe progressive hearing loss in both ears for both affected individuals (II.5) and (II.6). The audiograms were taken at the time of diagnosis Fig. 2 Audiograms of the affected probands. The audiograms show mild to severe progressive hearing loss in both ears for both affected individuals (II.5) and (II.6). The audiograms were taken at the time of diagnosis S1). Possible causative variants for each patient were sum- marized in (Additional file 2: Tables S2 and Add- itional file 3: Tables S3). The strong candidate variants that might underlie the mild to moderate NSHL in the two patients were those detected in MITF, MYO15A, MYO7A, and MYH14 genes (Fig. 1, 10]. effect of the = p. Pro338Leu missense variant on the struc- ture and function of the MITF protein using different in silico predictive software. The proline residue at position 338 lies within the α-helix of the bHLH motif domain (Fig. 4). The amino acid substitution in the MITF protein is predicted to be damaging by Polyphen2 (score 1; range 0–1 with 0 = benign and 1 = probably damaging). Data analysis SIFT predicts that the substitution is tolerated (score 0.92; a score ≤0.05 predicts the change to be damaging and > 0.05 predicts it to be tolerated). However, mutation taster predicts the substi- tution to be disease causing with a probability of 1 (0–1) (Table 1). In order to better assess this perturbation on the protein structure and its DNA binding activity, we per- formed an in silico protein stimulation assay, using the modelled-crystal structure of the bHLH domain of MITF (Fig. 4a) bound to DNA (PDB#4ATI). Interestingly, both the murine and human MITF proteins shared high identity in their amino acids bHLH domain including the Proline residue at position 338 which is highly conserved among species (Fig. 4b). Molecular modeling predicts that substitu- tion of proline for leucine can destabilize the protein (NMA Based Predictions ΔΔG ENCoM: 0.207 kcal/mol) (Fig. 5). Therefore, it is expected that this missense muta- tion changes the structure of the protein, thus, affecting protein function either by disrupting its homotypic/hetero- typic dimerization, its DNA binding affinities, or its inter- action with partners. Two bi-allelic single nucleotide variants (SNVs) were de- tected in the two patients; a previously described MYO15A (NM_016239.3:c.1454 T > C) mutation and a novel MITF variant (NM_198159.2:c.1013C > T) resulting in the mis- sense mutations p.V485A and p.P338L respectively (Add- itional file 2: Table S2). Moreover, on the top of the variants that were detected amongst the known HL genes were: 1- a mono-allelic variant in MYO7A (NM_000260.3: c.5563C > T) resulting in the nonsense mutation p.Q1855* inherited from the mother, and2- a heterozygous variant in MYH14 (NM_001145809.1:c.1150G > T) inherited from the father. (Fig. 1 and Additional file 2: Table S2). Finally, a search for unbiased bi-allelic mutations in the family did not yield additional variants with a MAF < 1% except for TRPV2 (rs756373391). The latter is a close mem- ber of the TRPV4 gene that is implicated in some cases of HL (Additional file 4: Tables S4 and Additional file 5: Tables S5). In silico prediction and modulation for the novel MITF variant We focused our analysis on the NM_198159.2:c.1013C > T variant in MITF because it lies on the boundary of exon8 and as such could lead either to a missense mutation and/ or alternative splicing (Fig. 3). We evaluated the possible Discussion Although consanguinity can facilitate the discovery of novel genes associated with many diseases, yet it challenges the Khalil et al. BMC Medical Genetics (20 Page 4 of 8 Fig. 3 Chromosomal localization of the MITF missense mutation. The NM_198159.2:c.1013C > T variant on chromosome 3 is visualized Using the IGV software. Both parents (I.1 and I.2) carry the heterozygous form (blue and red), whereas both affected daughters carry the homozygous form (red). The amino acids are shown in the lower panel below their corresponding codons, whereas a straight blue line was shown under the nucleotides that correspond to the intronic region Fig. 3 Chromosomal localization of the MITF missense mutation. The NM_198159.2:c.1013C > T variant on chromosome 3 is visualized Using the IGV software. Both parents (I.1 and I.2) carry the heterozygous form (blue and red), whereas both affected daughters carry the homozygous form (red). The amino acids are shown in the lower panel below their corresponding codons, whereas a straight blue line was shown under the nucleotides that correspond to the intronic region Fig. 3 Chromosomal localization of the MITF missense mutation. The NM_198159.2:c.1013C > T variant on chromosome 3 is visualized Using the IGV software. Both parents (I.1 and I.2) carry the heterozygous form (blue and red), whereas both affected daughters carry the homozygous form (red). The amino acids are shown in the lower panel below their corresponding codons, whereas a straight blue line was shown under the nucleotides that correspond to the intronic region Fig. 4 Structural Characterization of the P338 residue. The mouse bHLH amino acid sequence (a) used for depicting the crystal structure of MITF bound to DNA showing the position of the corresponding P338 residue (red circle and arrow) is highly identical to the human sequence (b). The position of the proline residue at position 338 (referred to as Pro 237) is to the outside of the interface of the dimerization interface between two molecules of the mouse MITF bHLH domain (c). (adapted from https://www.rcsb.org/structure/4ATI) Fig. 4 Structural Characterization of the P338 residue. The mouse bHLH amino acid sequence (a) used for depicting the crystal structure of MITF bound to DNA showing the position of the corresponding P338 residue (red circle and arrow) is highly identical to the human sequence (b). Discussion The position of the proline residue at position 338 (referred to as Pro 237) is to the outside of the interface of the dimerization interface between two molecules of the mouse MITF bHLH domain (c). (adapted from https://www.rcsb.org/structure/4ATI) Fig. 4 Structural Characterization of the P338 residue. The mouse bHLH amino acid sequence (a) used for depicting the crystal structure of MITF bound to DNA showing the position of the corresponding P338 residue (red circle and arrow) is highly identical to the human sequence (b). The position of the proline residue at position 338 (referred to as Pro 237) is to the outside of the interface of the dimerization interface between two molecules of the mouse MITF bHLH domain (c). (adapted from https://www.rcsb.org/structure/4ATI) Khalil et al. BMC Medical Genetics (2020) 21:1 Page 5 of 8 Khalil et al. BMC Medical Genetics Table 1 Pathogenicity scores of detected variants assessed by SIFT, PolyPhen2, Mutation Taster, and GERP++ software Allelic Variant Zygosity Amino acid change SIFT Polyphen-2 MutationTaster GERP++ MITF NM_198159.2 c.1013C > T Homozygous P338L Tolerated Score:0.92 Probably damaging Score:1 Disease causing Score:1 Conserved Score:5.06 MYO15A NM_016239 c.1454 T > C Homozygous V485A Damaging Score:0 Probably damaging Score:0.9 Disease causing Score:0.9 Conserved Score:5.1 MYH14 NM_001145809 c.1150G > T Heterozygous G384C Damaging Score: 0 Probably damaging Score:1 Disease causing Score:1 Conserved Score:3.42 MYO7A NM_000260.3 c.5835 C > T Heterozygous Q1855* NA NA Disease causing Score:1 NA concept of single causative genetic variant [3]. Interestingly, in this study we revealed a polygenic inheritance of NSHL with the liaison of two independent homozygous alterations in well-known HL genes. To the best of our knowledge, this is the first study to report the implication of a novel MITF variant in an NSHL case with an autosomal recessive mode of inheritance and a post-lingual onset. terminal domain are thought to be associated with a milder form of HL since they can affect only one of the two major isoforms of the gene [11]. Although the p.V485A mutation is located within the N-terminal do- main, our indexed patients suffer from a mild to severe phenotype. In addition, two healthy individuals from the Gnomad Exome database harbor this variant which argue against a major role for this mutation in the af- fected individuals. Accordingly, we postulate that other players might be linked, in collaboration or independent of MYO15A, to the underlying phenotype. Polygenic inheritance Although most genetic deafness cases result from muta- tions in a single gene, an emerging number of examples are being documented where recessive mutations at two loci are being involved. For example, the digenic inter- action that underlies the cause of deafness in individuals carrying a single mutation at the GJB2 locus along with a deletion in the functionally related GJB6 gene [18]. Moreover, a study done by Legar.et al. on twelve patients with MITF mutations demonstrated a large range of variability in phenotype among these patients which argue for the possible interaction with modifier loci [19]. Herein, we propose a polygenic form of inheritance mainly through the implication of both MITF and MYO15A variants coupled with two detected heterozy- gous variants in MYO7A and MYH14 genes. Different compound heterozygous or homozygous mutations re- lated to MYO7A have been reported in a variety of auto- somal recessive Usher Syndrome families [20]. However, mutations in MYH14 gene are associated with autosomal dominant hearing impairment [21]. Thus, we speculate an involvement of the detected MYH14 and MYO7A mutations in the observed phenotype but not as the dir- ect independent cause of HL since the parents presented as healthy carriers. Further functional studies are needed to assess the independent and combined effect of these mutations on the development of HL. Patients who previously presented with HL as the only phenotypic feature were thought to have NSHL. In conse- quence, only mutations in genes associated with this type of HL were investigated. On the other hand, some SHL cases require special confirmatory tests since the pene- trance of secondary features is either incomplete or age dependent. One example is the Usher syndrome which is presented as an NSHL case early-on in life as the onset of the secondary symptom (retinitis pigmentosa) does not ap- pear until puberty. This might cause a false clinical classifi- cation of some patients with SHL who can benefit from the appropriate implementation of visual rehabilitation at early stages [6]. Therefore, it is very critical to categorize genes and variants that are either specific to each type or involved in both forms of HL. Another example is the heterozygous MITF (p.R110X) variant that was specifically associated with SHL cases but was recently detected in an NSHL case that presented in the absence of WS2 common features (no pigmentary changes in hair, eyes, or skin) [15]. Polygenic inheritance Originally in-vivo studies on the phenotypic variation seen with the different alleles of the mouse MITF gene referred to as mi gene suggests that mutations in the human MITF gene may also manifest themselves in different ways. This proposed a possibility for detecting phenotypes different from the characteristic WS2 phenotype among patients with MITF mutations [16]. Combining these facts with our results, we propose expanding the implications of MITF variants from syndromic to non-syndromic HL cases while associating it with an autosomal recessive mode of inheritance. Finally, we could not rule out other genetic/epigenetic modifiers that could be associated with the underlying phenotype, especially that a growing number of studies have showed that copy number variation (CNV) is widely encountered in syndromic and non-syndromic HL cases [22–24]. Such studies would require a case- control study with a substantial number of patients with SHL, NSHL, and controls. MYO15A and MITF homozygous alterations: the dilemma of predictive tools? MYO15A encodes for XVA myosin protein which plays a vital role in the elongation and development of stereo- cilia and actin filaments. More than forty MYO15 muta- tions have been reported in the motor domain of the protein with generally autosomal recessive HL impair- ment characterized by a profound phenotype at all frequencies [10]. The detected homozygous MYO15A mutation, p.V485A, was previously associated with a HL phenotype in an Iranian family [3]. Mutations in the N- We therefore considered the second shared bi-allelic novel MITF gene mutation p. P338L between the two sisters. MITF encodes the melanocyte-specific promoter of microphthalmia-associated bHLH transcription factor. A total of more than forty MITF mutations have been verified to be disease-causing in patients with either the Waarden- burg’s syndrome type 2)WS2) (OMIM#193510) or the Tietz syndrome (OMIM #103500, 12]. Both syndromes are Fig. 5 Protein structure prediction of the novel MITF variant (p.P338L). In silico modeling (a) the effect of the MITF mutation using the DUET software shows a general destabilization of the structure (b). Wild-type and mutant residues are colored in light-green and are also represented as sticks alongside with the surrounding residues which are involved on any type of interactions (a). The magnitude of the fluctuation is represented by thin to thick tube colored blue (low), white (moderate) and red (high) (b) Fig. 5 Protein structure prediction of the novel MITF variant (p.P338L). In silico modeling (a) the effect of the MITF mutation using the DUET software shows a general destabilization of the structure (b). Wild-type and mutant residues are colored in light-green and are also represented as sticks alongside with the surrounding residues which are involved on any type of interactions (a). The magnitude of the fluctuation is represented by thin to thick tube colored blue (low), white (moderate) and red (high) (b) Fig. 5 Protein structure prediction of the novel MITF variant (p.P338L). In silico modeling (a) the effect of the MITF mutation using the DUET software shows a general destabilization of the structure (b). Wild-type and mutant residues are colored in light-green and are also represented as sticks alongside with the surrounding residues which are involved on any type of interactions (a). The magnitude of the fluctuation is represented by thin to thick tube colored blue (low), white (moderate) and red (high) (b) Khalil et al. BMC Medical Genetics (2020) 21:1 Khalil et al. MYO15A and MITF homozygous alterations: the dilemma of predictive tools? BMC Medical Genetics (2020) 21:1 Page 6 of 8 autosomal dominant and are characterized by overlapping phenotypes that encompass HL and pigmentary abnormal- ities with variable penetrance. To the best of our know- ledge, only 2 homozygous MITF cases were detected in WS2 and WS4 [13, 14]. In the present study, the detected homozygous p.P338L missense mutation was neither re- ported in the dbSNP database, nor in the Gnomad Exome/ Genome database. It was also absent from more than 300 Lebanese exomes. The heterozygous frequency of this vari- ant is less than 0.00001 in these databases as it is only present in 3 individuals. Since the detected MITF misssense mutation is localized in the bHLH DNA-binding domain and since the in-silico analysis revealed a deleterious effect prediction, we accordingly hypothesize that this mutation is disease causing (Table 1). Thus, structural and functional assays are compulsory to assess the effect of this mutation on the ability of MITF to heterodimerize, bind DNA, and/ or translocate to the nucleus. as autosomal recessive loci causing stable post-lingual hearing impairment rather than a progressive pre- lingual one. Conclusion The present study describes a rare form of hereditary non-syndromic autosomal recessive post-lingual sensori- neural HL that is associated with polygenic inheritance mode of bi- and mono- allelic variants. In this study, we unraveled the association of a novel MITF variant in NSHL along with a previously described mutation in MYO15A associated with a mild form of HL. We highlighted the importance of clinical exome sequen- cing for a comprehensive addressing of genetic hetero- geneity of HL and in detecting novel variants associated with NSHL. Additionally, it is widely known that most mutations in autosomal dominant loci cause post-lingual hearing impairment (including MYO7A and MYH14) while mu- tations in autosomal recessive HL cases with delayed childhood onset are rare clinical findings [17]. Herein, we are the first to propose MITF and MYO15A variants Page 7 of 8 Page 7 of 8 Page 7 of 8 Khalil et al. BMC Medical Genetics Khalil et al. BMC Medical Genetics (2020) 21:1 (2020) 21:1 Supplementary information Supplementary information accompanies this paper at https://doi.org/10. 1186/s12881-019-0942-4. Additional file 1: Table S1. List of hearing loss genes used for filtering variants Additional file 2: Table S2. Filtering results from whole exome sequencing for patient II.5 using the 150 genes panel from Supplementary Table 1 Additional file 3: Table S3. Filtering results from whole exome sequencing for patient II.6 using the 150 genes panel from Supplementary Table 1 Additional file 4: Table S4. Filtering results from whole exome sequencing for patient II.5 showing only homozygous mutations with a MAF < 5% Additional file 5: Table S5. Filtering results from whole exome sequencing for patient II.6 showing only homozygous mutations with a MAF < 5%. Supplementary information Received: 6 July 2019 Accepted: 24 December 2019 Ethics approval and consent to participate 16. Steingrímsson E, Nii A, Fisher DE, Ferré-D’Amaré AR, McCormick RJ, Russell LB, et al. The semidominant mi(b) mutation identifies a role for the HLH domain in DNA binding in addition to its role in protein dimerization. EMBO J. 1996;15(22):6280–9. The entire procedure was approved by the Institutional Review Board (IRB) at the American University of Beirut (protocol number:OTO.MB1.02) and carried out with written informed consent of the patients, the parents, and/or legal guardians. The entire procedure was approved by the Institutional Review Board (IRB) at the American University of Beirut (protocol number:OTO.MB1.02) and carried out with written informed consent of the patients, the parents, and/or legal guardians. out with written informed consent of the patients, the parents, and/or legal guardians. 17. Shearer AE, Hildebrand MS, Smith RJ. Hereditary Hearing Loss and Deafness Overview. GeneReviews®. Seattle: University of Washington; 1993. Funding Th This project was funded by the Medical Practice Plan (MPP) at the American University of Beirut, Beirut, Lebanon. The funding body played no role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript. 12. Chen L, Guo W, Ren L, Yang M, Zhao Y, Guo Z, et al. A de novo silencer causes elimination of MITF-M expression and profound hearing loss in pigs. BMC Biol [Internet]. 2016;14(1):52 Available from: http://bmcbiol. biomedcentral.com/articles/10.1186/s12915-016-0273-2, [cited 2019 Oct 7]. 13. Potrony M, Puig-Butille JA, Aguilera P, Badenas C, Tell-Marti G, Carrera C, et al. Prevalence of MITF p.E318K in Patients With Melanoma Independent of the Presence of CDKN2A Causative Mutations. JAMA Dermatol. 2016;152(4):405. Availability of data and materials The datasets used and analyzed during the current study are available from the corresponding author upon a reasonable request. Exome sequencing files are available for sharing with any researcher or research team through a direct request process to the corresponding authors. The novel MITF mutation was submitted to ClinVar under accession number: SCV001035077. 14. Pang X, Zheng X, Kong X, Chai Y, Wang Y, Qian H, et al. A homozygous MITF mutation leads to familial Waardenburg syndrome type 4. Am J Med Genet Part A. 2018;179(2):ajmg.a.60693. 15. Bademci G, Cengiz FB, Foster J II, Duman D, Sennaroglu L, Diaz-Horta O, et al. Variations in multiple Syndromic deafness genes mimic non- syndromic hearing loss. Sci Rep. 2016;6(1):31622. Acknowledgments g We thank the patients and the family for their contribution to this research. We thank the patients and the family for their contribution to this research. 9. Robinson JT, Thorvaldsdóttir H, Winckler W, Guttman M, Lander ES, Getz G, et al. Integrative genomics viewer. Nat Biotechnol [Internet]. 2011;29(1):24–6 Available from: http://www.nature.com/articles/nbt.1754, [cited 2019 Oct 13]. References 1. Korver AMH, Smith RJH, Van Camp G, Schleiss MR, Bitner-Glindzicz MAK, Lustig LR, et al. Congenital hearing loss. Nat Rev Dis Primers. 2017;3:16094. 1. Korver AMH, Smith RJH, Van Camp G, Schleiss MR, Bitner-Glindzicz MAK, Lustig LR, et al. Congenital hearing loss. Nat Rev Dis Primers. 2017;3:16094. 2. Abou Tayoun AN, Al Turki SH, Oza AM, Bowser MJ, Hernandez AL, Funke BH, et al. Improving hearing loss gene testing: a systematic review of gene evidence toward more efficient next-generation sequencing–based diagnostic testing and interpretation. Genet Med. 2016;18(6):545–53. 2. Abou Tayoun AN, Al Turki SH, Oza AM, Bowser MJ, Hernandez AL, Funke BH, et al. Improving hearing loss gene testing: a systematic review of gene evidence toward more efficient next-generation sequencing–based diagnostic testing and interpretation. Genet Med. 2016;18(6):545–53. 3. Sloan-Heggen CM, Babanejad M, Beheshtian M, Simpson AC, Booth KT, Ardalani F, et al. Characterising the spectrum of autosomal recessive hereditary hearing loss in Iran. 2015. 3. Sloan-Heggen CM, Babanejad M, Beheshtian M, Simpson AC, Booth KT, Ardalani F, et al. Characterising the spectrum of autosomal recessive hereditary hearing loss in Iran. 2015. 4. Venkatesh MD, Moorchung N, Puri B. Genetics of non syndromic hearing loss. Med J Armed Forces India. 2015;71(4):363–8. 4. Venkatesh MD, Moorchung N, Puri B. Genetics of non syndromic hearing loss. Med J Armed Forces India. 2015;71(4):363–8. 5. Hu S, Sun F, Zhang J, Tang Y, Qiu J, Wang Z, et al. Genetic etiology study of ten Chinese families with Nonsyndromic hearing loss. Neural Plast. 2018 Jul; 2018:1–7. 5. Hu S, Sun F, Zhang J, Tang Y, Qiu J, Wang Z, et al. Genetic etiology study of ten Chinese families with Nonsyndromic hearing loss. Neural Plast. 2018 Jul; 2018:1–7. 6. Likar T, Hasanhodžić M, Teran N, Maver A, Peterlin B, Writzl K. Diagnostic outcomes of exome sequencing in patients with syndromic or non- syndromic hearing loss. Janecke AR, editor. PLoS One. 2018;13(1):e0188578. 6. Likar T, Hasanhodžić M, Teran N, Maver A, Peterlin B, Writzl K. Diagnostic outcomes of exome sequencing in patients with syndromic or non- syndromic hearing loss. Janecke AR, editor. PLoS One. 2018;13(1):e0188578. Abbreviations 7. Barake R, Abou-Rizk S, Nemer G, Bassim M. The OTOGL p.Arg925* variant is associated with moderate hearing loss in a Syrian nonconsanguineous family. Genet Test Mol Biomarkers. 2017;21(7):445–9. 7. Barake R, Abou-Rizk S, Nemer G, Bassim M. The OTOGL p.Arg925* variant is associated with moderate hearing loss in a Syrian nonconsanguineous family. Genet Test Mol Biomarkers. 2017;21(7):445–9. HL: Hearing loss; NGS: Next Generation Sequencing; NSHL: Non-syndromic hearing loss; SNVs: Single Nucleotide Variants; WS: Waardenburg’s Syndrome HL: Hearing loss; NGS: Next Generation Sequencing; NSHL: Non-syndromic hearing loss; SNVs: Single Nucleotide Variants; WS: Waardenburg’s Syndrome 8. Mustapha M. Autosomal recessive non-syndromic hearing loss in the Lebanese population: prevalence of the 30delG mutation and report of two novel mutations in the connexin 26 (GJB2) gene. J Med Genet. 2001;38(10):36e–36. Consent for publication 18. Nance WE. The genetics of deafness. Ment Retard Dev Disabil Res Rev. 2003 Jan;9(2):109–19. All patients, parents and legal guardians signed an informed consent for data publication. All patients, parents and legal guardians signed an informed consent for data publication. 19. Léger S, Balguerie X, Goldenberg A, Drouin-Garraud V, Cabot A, Amstutz- Montadert I, et al. Novel and recurrent non-truncating mutations of the MITF basic domain: genotypic and phenotypic variations in Waardenburg and Tietz syndromes. Eur J Hum Genet [Internet]. 2012;20(5):584–7 Available from: http://www.nature.com/articles/ejhg2011234, [cited 2019 Oct 13]. Author’s contribution AK and SBK: did all the experiments, and participate in the analysis and writing up of the manuscript. RB,GD,SAR, and AK did the clinical assessment and recruitment of the patients and their family members. MB and GN designed the study, secured the funding, analyzed the data, and wrote up the manuscript. All authors read and approved the final manuscript. 10. Chang MY, Kim AR, Kim NKD, Lee C, Lee KY, Jeon W-S, et al. Identification and clinical implications of novel MYO15A mutations in a non- consanguineous Korean family by targeted exome sequencing. Mol Cells. 2015;38(9):781–8. 11. Crespo M, van Dalum G, Ferraldeschi R, Zafeiriou Z, Sideris S, Lorente D, et al. Androgen receptor expression in circulating tumour cells from castration-resistant prostate cancer patients treated with novel endocrine agents. Br J Cancer. 2015;112(7):1166–74. 20. Ayub Medical College. R, Ayub M. Journal of Ayub Medical College Abbottabad : JAMC. Vol. 29, Journal of Ayub Medical College Abbottabad. Ayub Medical College; 1988. 671–676 p. Competing interests Competing interests The authors declare that they have no competing interests. Author details 1 20. Ayub Medical College. R, Ayub M. Journal of Ayub Medical College Abbottabad : JAMC. Vol. 29, Journal of Ayub Medical College Abbottabad. Ayub Medical College; 1988. 671–676 p. Author details 1Departments of Biochemistry and Molecular Genetics, Faculty of Medicine, American University of Beirut, Beirut, Lebanon. 2Otolaryngology – Head and Neck Surgery, Faculty of Medicine, American University of Beirut, Beirut, Lebanon. 3Genomics and Precision Medicine Program, College of Health and Life Siences, Hamad Bin Khalifa University, Doha, Qatar. 21. Kim K-Y, Kovács M, Kawamoto S, Sellers JR, Adelstein RS. Disease-associated mutations and alternative splicing alter the enzymatic and motile activity of nonmuscle myosins II-B and II-C. J Biol Chem. 2005;280(24):22769–75. Page 8 of 8 Page 8 of 8 Khalil et al. BMC Medical Genetics (2020) 21:1 Khalil et al. BMC Medical Genetics (2020) 21:1 22. Shearer A, Kolbe DL, Azaiez H, Sloan CM, Frees KL, Weaver AE, et al. Copy number variants are a common cause of non-syndromic hearing loss. Genome Med [Internet]. 2014;6(5):37 Available from: http://www.ncbi.nlm. nih.gov/pubmed/24963352, [cited 2019 Oct 13]. 23. Haraksingh RR, Jahanbani F, Rodriguez-Paris J, Gelernter J, Nadeau KC, Oghalai JS, et al. Exome sequencing and genome-wide copy number variant mapping reveal novel associations with sensorineural hereditary hearing loss. BMC Genomics [Internet]. 2014;(1):15, 1155 Available from: http://www.ncbi.nlm.nih.gov/pubmed/25528277, [cited 2019 Oct 13]. 24. Vona B, MAH H, Schröder J, Shehata-Dieler W, Nanda I, Haaf T. Hereditary hearing loss SNP-microarray pilot study. BMC Res Notes [Internet]. 2018; 11(1):391 Available from: https://bmcresnotes.biomedcentral.com/articles/1 0.1186/s13104-018-3466-7, [cited 2019 Oct 13]. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
https://openalex.org/W2612082204	https://europepmc.org/articles/pmc5439072?pdf=render	English	null	Surface Characteristics and Biofilm Development on Selected Dental Ceramic Materials	International journal of dentistry	2,017	cc-by	4,514	Kyoung H. Kim, Carolina Loch, J. Neil Waddell, Geoffrey Tompkins, and Donald Schwass Sir John Walsh Research Institute, Faculty of Dentistry, University of Otago, Dunedin, New Zealand Kyoung H. Kim, Carolina Loch, J. Neil Waddell, Geoffrey Tompkins, and Donald Schwass Sir John Walsh Research Institute, Faculty of Dentistry, University of Otago, Dunedin, New Zealand Correspondence should be addressed to Donald Schwass; donald.schwass@otago.ac.nz Received 15 December 2016; Accepted 18 April 2017; Published 8 May 2017 Academic Editor: Spiros Zinelis Received 15 December 2016; Accepted 18 April 2017; Published 8 May 2017 Academic Editor: Spiros Zinelis Copyright © 2017 Kyoung H. Kim et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Background. Intraoral adjustment and polishing of dental ceramics often affect their surface characteristics, promoting increased roughness and consequent biofilm growth. This study correlated surface roughness to biofilm development with four commercially available ceramic materials. Methods. Four ceramic materials (Vita Enamic, Lava Ultimate, Vitablocs Mark II, and Wieland Reflex) were prepared as per manufacturer instructions. Seventeen specimens of each material were adjusted and polished to simulate clinical intraoral procedures and another seventeen remained unaltered. Specimens were analysed by SEM imaging, confocal microscopy, and crystal violet assay. Results. SEM images showed more irregular surface topography in adjusted specimens than their respective controls. Surface roughness (𝑅𝑎) values were greater in all materials following adjustments. All adjusted materials with the exception of Vitablocs Mark II promoted significantly greater biofilm growth relative to controls. Conclusion. Simulated intraoral polishing methods resulted in greater surface roughness and increased biofilm accumulation. Hindawi International Journal of Dentistry Volume 2017, Article ID 7627945, 6 pages https://doi.org/10.1155/2017/7627945 Hindawi International Journal of Dentistry Volume 2017, Article ID 7627945, 6 pages https://doi.org/10.1155/2017/7627945 Research Article Surface Characteristics and Biofilm Development on Selected Dental Ceramic Materials Kyoung H. Kim, Carolina Loch, J. Neil Waddell, Geoffrey Tompkins, and Donald Schwass Sir John Walsh Research Institute, Faculty of Dentistry, University of Otago, Dunedin, New Zealand Correspondence should be addressed to Donald Schwass; donald.schwass@otago.ac.nz Received 15 December 2016; Accepted 18 April 2017; Published 8 May 2017 2. Material and Methods 2.1. Specimen Preparation. Four ceramic materials (Table 1) were studied. Each of the CAD materials was sectioned into rectangular blocks with a diamond disc (Komet 918B, Komet Dental, Germany) attached to a straight handpiece. The veneering porcelain material (Wieland Reflex veneering porcelain) was formed into disks using a stainless steel mould and fired (Dekema Austromat M, Dekema, Germany), followed by further glaze firing according to manufacturer guidelines. Thirty-four specimens of variable dimensions were prepared for each material, being ultrasonically cleaned for 10 min and air dried, and the surfaces were polished following manufacturer guidelines. The number of specimens analysed was determined taking into account material avail- ability and preparation constraints. The bacterial strain Streptococcus gordonii C219 (from the University of Otago culture collection) was grown for 24 h in 10 ml of brain heart infusion (BHI; BD Biosciences, Franklin Lakes, United States) and 30 𝜇L aliquots used to inoculate each well of the specimen-containing wells, which each contained BHI (7.5 mL). Culture trays were incubated for 72 h at 37∘C and the medium was replaced every 24 h. Specimens were removed from the impression material, rinsed thoroughly with phosphate buffered saline (PBS) on an orbital shaker (42 rpm for 15 minutes), and placed into new culture trays. Crystal violet (0.1% in 25% methanol) was added (5 mL) to each well and after 15 minutes, the trays with embedded specimens were washed (8 L of water per tray), avoiding direct application onto the specimen. The trays with specimens were dried overnight.h y p p Seventeen specimens of each ceramic material were adjusted and polished to simulate clinical intraoral proce- dures. The polishing sequences were carried out using a Pow- ertorque Lux 646B high speed handpiece (KaVo, Biberach, Germany) under water irrigation. Adjustments used a “rugby ball” diamond bur, followed by polishing with ISO standard grit size red (30 𝜇m), yellow (15 𝜇m), and white (8 𝜇m) fine finishing burs (B260, B261, and B262). Final polishing used an extrafine porcelain bur (Cerapol Plus, Edenta, Hauptstrasse, Switzerland). The other 17 specimens of each material were unaltered to use as controls. Physical dimensions of each specimen were measured in order to calculate the surface areas. The unprepared surfaces of each specimen were coated with resin and light-cured to prevent crystal violet solubiliza- tion because biofilm growth also occurred on the unexposed surfaces despite covering with PVS. 1. Introduction or polymer-infiltrated-ceramic-networks are 3M ESPE Lava Ultimate and Vita Enamic. 3M ESPE Lava Ultimate (3M ESPE, USA) is a resin nanoceramic material containing a combination of aggregated 20 nm silica and 4 to 11 nm zirco- nia clusters in a resin matrix. Vita Enamic (Vita, Germany) is a hybrid ceramic containing 86% feldspathic porcelain and an interpenetrating polymer network 14% by weight. Among glass-based systems or feldspathic porcelains are Vitablocs Mark II and Wieland Reflex. Vitablocs Mark II (Vita, Germany) is a fine-grained, homogeneous feldspathic porcelain with an average particle size of 4 𝜇m. Wieland Reflex is a feldspathic porcelain which contains homogenous nanoleucite crystals [7–9].t Dental ceramics are the restorative material of choice for indirect restorations, mainly due to their biocompatibility, low thermal conductivity, color stability, and aesthetics [1]. Dental ceramics are used in restorative dentistry because of their success rate as well as diverse range of chemical and structural compositions, resulting from recent improvements in biomaterial technology [2, 3]. These materials commonly consist of both glassy and crystalline phases, which are usually heat-treated to provide desirable properties [4]. While the glassy phase contributes to the aesthetics of the ceramics [5], the crystalline phase is responsible for the mechanical properties of the material [4]. Fixed prostheses such as ceramic crowns often require adjustments before and after cementation, which generally removes the smooth external glazed surface layer [10]. The exposed porcelain layer is commonly roughened in com- parison to the normal smooth glazed surface. In addition, abrasive machining processes in CAD/CAM systems often induce damage, generating the need for final finishing in oral conditions using a dental handpiece and diamond burs [6]. Surface roughness, a component of surface texture, influences Structurally, dental bioceramics cover a wide spectrum of glass–ceramics, reinforced porcelains, oxide ceramics (zirconia, alumina, and spinel), fiber-reinforced ceramic composites, and multilayered ceramic structures. In the last two decades, computer-assisted design and manufacturing (CAD/CAM) technologies have replaced the laborious and time-consuming conventional restoration fabrication meth- ods [6]. Two of the new dual network hybrid ceramics International Journal of Dentistry 2 how a material interacts with the environment [11]. A higher degree of surface roughness of ceramic restorations often gen- erates greater wear on the opposing dentition, compromises aesthetics of the restoration, and increases biofilm adhesion and growth [12]. palladium for scanning electron microscopy (SEM) observa- tion. 1. Introduction Specimens were embedded in polyvinyl siloxane (PVS) impression material, leaving only the prepared surfaces exposed to biofilm growth. The trays were then irradiated under UV light in a laminar flow cabinet for 15 min. 1. Introduction SEM images were obtained in a Field Emission SEM (JSM-6700F, JEOL Ltd., Tokyo, Japan) operating at 10 kV and 10 𝜇A. Magnifications ranged from 25x to 2000x. Material topography was evaluated through qualitative assessment of the surface characteristics of different materials observed via SEM imaging. Microscopic irregularities caused by cracks, grooves, and abrasion defects often lead to greater surface roughness, becoming common sites for bacterial adhesion [13]. These surface irregularities provide shelter for bacteria from shear forces generated in the oral cavity, allowing them to form stronger bonds to the substratum [14]. Greater biofilm for- mation on dental surfaces has been implicated in the devel- opment of both gingivitis and dental caries [15]. For these rea- sons, material predilection to biofilm formation is an impor- tant consideration in the selection of restorative materials. 2.3. Surface Roughness. One further control and one adjusted specimen of each material were examined under a confo- cal laser-scanning microscope (LSM 510 Upright Axioplan, Carl Zeiss Inc., Oberkochen, Germany). Stacks of confocal microscope images were edited using 3D Slicer [19] to level the images which were analysed using Fiji and additional plugins including “extended depth of field” and “surfaceChJ 1q” (ImageJ, NIH, Bethesda, Maryland, USA). Height maps were generated, and peak-to-valley surface roughness (𝑅𝑎) values were measured at three spots for each specimen at randomly selected locations and averaged. Most evidence supporting clinical use of newer ceramic materials is from information provided by manufacturers regarding the chemical composition and physical properties (e.g., [16–18]). Information on surface characteristics and biofilm formation on new materials is sparse, particularly with respect to the impact of intraoral polishing.h The aims of this study were to characterize the surface ultrastructure and roughness of four ceramic materials and to assess their promotion of biofilm development following adjustments simulating clinical intraoral polishing. 2.4. Biofilm Growth. The remaining control and adjusted specimens of each material were placed in the wells of a tissue-culture tray (CELLSTAR 6-well, Greiner Bio- one, Kremsm¨unster, Austria). Specimens were embedded in polyvinyl siloxane (PVS) impression material, leaving only the prepared surfaces exposed to biofilm growth. The trays were then irradiated under UV light in a laminar flow cabinet for 15 min. 2.4. Biofilm Growth. The remaining control and adjusted specimens of each material were placed in the wells of a tissue-culture tray (CELLSTAR 6-well, Greiner Bio- one, Kremsm¨unster, Austria). 3. Results 3.1. Surface Characterization by SEM. For all materials anal- ysed, SEM images revealed that the surfaces which were adjusted and polished to simulate clinical intraoral pro- cedures had more irregular surface topography than the respective controls (Figure 1). Parallel scratch marks and small pits were more often seen on control specimens, which was consistent with normal preparation and polishing arte- facts. Specimens which were adjusted and polished displayed coarse pits and irregularities, as evidenced by a rougher sur- face. The control and adjusted Vitablocs Mark II specimens exhibited the roughest surfaces, whereas Wieland Reflex glazed porcelain recorded the smoothest control surfaces, and 3M ESPE Lava Ultimate had the smoothest adjusted surfaces, with relatively small pits and scratches. SEM indicated that adjusted surfaces had more surface irregularities compared to control surfaces for all materials analysed and that these features were dependent on the spe- cific material, reflecting the diverse chemical compositions and varied fabrication methods. 3M ESPE Lava Ultimate, Vita Enamic, and Vitablocs Mark II were prefabricated CAD/CAM blocks, presenting more regular surfaces than Wieland Reflex. These machinable ceramics may involve less volume defects than the laboratory-fabricated Wieland Reflex porcelain [20]. It seems that if intensive intraoral polishing procedures are utilized, then restorations prepared from CAD/CAM blocks would be expected to have a smoother finish than those of adjusted surfaces of other ceramics, such as the Wieland Reflex porcelain. Surface porosities and irregularities cannot be easily eliminated by conventional intraoral polishing processes [12, 21].i 3.2. Surface Roughness. 𝑅𝑎values were higher for all adjusted surfaces in comparison to controls (Figure 2). For both hybrid ceramics (Vita Enamic and 3M ESPE Lava Ultimate) adjust- ment resulted in minor increases in 𝑅𝑎values in comparison to controls. Conversely, adjusted surfaces of Vitablocs Mark II and Wieland Reflex porcelain showed much higher 𝑅𝑎 values than controls. The highest 𝑅𝑎values were recorded for the adjusted Vitablocs Mark II specimen (1.597 𝜇m ± 1.22), while the lowest was the control Wieland Reflex porcelain (0.260 𝜇m ± 0.09). Measures of surface roughness (𝑅𝑎values) were not tested statistically due to the small sample sizes available for this analysis. In accordance with our findings, studies of intraoral polishing methods generally report increases in the rough- ness of ceramic surfaces [22–24], but there is not yet a consensus. Reports are inconsistent because of different measuring parameters, combinations of polishing methods, and the variety of tested ceramic materials [25]. 2. Material and Methods Sealing the unexposed surfaces eliminated possible effects of this growth on the assessment of the test surface biofilm. Acetic acid (3 mL at 30%) was added to each well to elute the crystal violet from the biofilm on the exposed surfaces. The adsorption (𝐴570) of the eluted stain was determined using an Ultrospec 6300 Pro (Amersham Biosciences, Buckinghamshire, UK), as a measure of relative biomass. Due to the different sizes of the specimens analysed, absorbance values (𝐴570) obtained were then standardised by dividing 𝐴570 values by the surface area of each specimen. 2.2. Surface Characterization. One control and one adjusted specimen of each ceramic material were coated with gold 3 International Journal of Dentistry Table 1: Ceramic materials tested in this study. Product Type Manufacturer Vita Enamic Hybrid ceramic Vita Zahnfabrik, Bad S¨ackingen, Germany 3M ESPE Lava Ultimate Hybrid ceramic 3M ESPE, Minnesota, USA Vitablocs Mark II Leucite-reinforced glass ceramic Vita Zahnfabrik, Bad S¨ackingen, Germany Wieland Reflex veneering porcelain Nanoleucite-reinforced glass ceramic Wieland Dental, Pforzheim, Germany Table 1: Ceramic materials tested in this study. Manufacturer Vita Zahnfabrik, Bad S¨ackingen, Germany 3M ESPE, Minnesota, USA Vita Zahnfabrik, Bad S¨ackingen, Germany Wieland Dental, Pforzheim, Germany 4. Discussion 2.5. Statistical Analyses. The nonparametric Mann–Whitney U Test for two independent samples was employed to test differences in the biofilm biomass between pairs of adjusted versus control surfaces for each material. Statistical signifi- cance was set at the 5% probability level. Tests were performed with BioStat 2009 (AnalystSoft, Alexandria, VA, USA). 2.5. Statistical Analyses. The nonparametric Mann–Whitney U Test for two independent samples was employed to test differences in the biofilm biomass between pairs of adjusted versus control surfaces for each material. Statistical signifi- cance was set at the 5% probability level. Tests were performed with BioStat 2009 (AnalystSoft, Alexandria, VA, USA). This study explored the impact of simulated intraoral adjust- ment and polishing procedures on surface ultrastructural characteristics, surface roughness, and biofilm growth for four commercially available ceramic materials, including two new hybrid materials. For most ceramic materials, finishing or contouring involves adjustments required to achieve the final shape of the restoration, generally using ultrafine dia- monds and/or coarse rubber abrasives. Polishing is required to achieve the final surface smoothness of the restoration with minimal changes to surface shape [9]. 3.3. Biofilm Development. With the exception of Vitablocs Mark II, adjustment of the surface promoted significantly greater biofilm growth: Lava U = 216.5, 𝑝< 0.001; Vita Enamic U = 168, 𝑝= 0.021; Wieland Reflex U = 183.5, 𝑝= 0.003 (Mann–Whitney U test, 95% confidence interval). Surface adjustment of Vitablocs Mark II did not influence biofilm accumulation (U = 92.5, 𝑝= 0.407). Adjusted 3M ESPE Lava Ultimate promoted the greatest biofilm accumu- lation (0.003 ± 0.0004 𝐴570/mm2), while adjusted Vitablocs Mark II was least supportive of biofilm development (0.0014 ± 0.0004 𝐴570/mm2) (Figure 3). Conflicts of Interest The authors declare that there are no conflicts of interest regarding the publication of this paper. Figure 3: Comparison of biofilm development on control and adjusted surfaces of ceramic materials (average ± SD). ∗Statistically significant (𝑝< 0.05, 95% confidence interval). 3. Results 5 International Journal of Dentistry Control Adjusted 0.00 0.50 1.00 1.50 2.00 2.50 3.00 Lava Ultimate Mark II Refex Vita Enamic Ra values (휇m) Figure 2: Surface roughness (𝑅𝑎) of control and adjusted surfaces of ceramic materials (average ± SD). and development of residual compressive stresses in the porcelain surfaces [6]. Rough surfaces facilitate the adhesion of bacteria by providing niches where bacteria can adhere and grow pro- tected from brushing, muscular action, and salivary flow [23]. Commonly employed finishing/polishing procedures have clinical relevance, particularly when procedures are conducted close to the cervical area of indirect restorations, where biofilm formation can result in either dental caries or periodontal disease [23]. For both surface characterization via SEM and surface roughness analyses, only one specimen per group was used. Although it is expected that the features described here were typical of each material, the results were indicative and the conclusions were made with caution. In addition, this in vitro study used a monoculture biofilm, whereas oral biofilms are complex communities of microorganisms [32]. This investigation also did not account for the influ- ences of saliva, temperature, and pH changes on biofilm growth. Despite these limitations, this study demonstrated that simulated intraoral polishing methods resulted in greater surface roughness and increased biofilm growth for the four materials investigated. Future studies in a clinical setting will further elucidate the significance of these findings. Figure 2: Surface roughness (𝑅𝑎) of control and adjusted surfaces of ceramic materials (average ± SD). Figure 2: Surface roughness (𝑅𝑎) of control and adjusted surfaces of ceramic materials (average ± SD). ∗ ∗ ∗ 0.0000 0.0005 0.0010 0.0015 0.0020 0.0025 0.0030 0.0035 0.0040 Absorbance per area (A570/mm2) Control Adjusted Lava Ultimate Mark II Refex Vita Enamic Figure 3: Comparison of biofilm development on control and adjusted surfaces of ceramic materials (average ± SD). ∗Statistically significant (𝑝< 0.05, 95% confidence interval). Acknowledgments The authors wish to acknowledge Vita Zahnfabrik H. Rauter BmfH & Co., KG, Germany, for their donation of the Vita CAD blocks used in this study. Likewise they also acknowl- edge Henry Schein Dental, New Zealand, for the donation of the 3M ESPE Lava Ultimate CAD blocks. The authors also acknowledge the facilities as well as technical assistance from staff at the Otago Centre for Electron Microscopy (OCEM) and the Otago Centre for Confocal Microscopy (OCCM) at the University of Otago. Kyoung H. Kim was supported by an Auckland Dental Association Honours scholarship. The effect of surface roughness on biofilm growth has been assessed by both in vitro and in vivo assays [14, 24, 28, 29]. In the current study, with the exception of Vitablocs Mark II, adjustment resulted in increased biofilm development on all surfaces, which accords with a previous report [22]. Though it is difficult to determine the relative contributions, surface roughness, surface free energy, and chemical composition all influence biofilm development [14, 22, 23, 28, 30]. Decreasing abrasive machining-induced surface and sub- surface damage in CAD/CAM and intraoral finishing pro- cesses is a major challenge to bioceramics engineering. Surface roughness can negatively influence ceramic strength as brittle cracks can lead to either earlier or catastrophic failure of enamel and prosthetic components [31]. In addition, roughened surfaces increase abrasion to antagonistic tooth structures [6, 25]. On the other hand, polishing could improve not only aesthetic surface texture but also the strength of ceramics, due to the elimination of surface flaws Additional Points Practical Implications. Intraoral adjustment and polishing of dental ceramics may result in greater surface roughness which can contribute to biofilm adhesion and growth. 3. Results Glazed microfilled feldspar ceramics present less surface roughness than do materials submitted to glaze followed by diamond bur polishing. However, when ceramic surfaces are treated with a bur and polished with rubber tips, roughness values decrease [23]. Our study revealed that all materials tested had greater 𝑅𝑎values following polishing adjustments. In partic- ular, adjustment of Vitablocs Mark II and Wieland Reflex porcelain increased 𝑅𝑎values by 1.197 𝜇m and 0.789 𝜇m, respectively. For the hybrid ceramics Vita Enamic, and 3M ESPE Lava Ultimate, the increases in 𝑅𝑎values were not as pronounced. 3.3. Biofilm Development. With the exception of Vitablocs Mark II, adjustment of the surface promoted significantly greater biofilm growth: Lava U = 216.5, 𝑝< 0.001; Vita Enamic U = 168, 𝑝= 0.021; Wieland Reflex U = 183.5, 𝑝= 0.003 (Mann–Whitney U test, 95% confidence interval). Surface adjustment of Vitablocs Mark II did not influence biofilm accumulation (U = 92.5, 𝑝= 0.407). Adjusted 3M ESPE Lava Ultimate promoted the greatest biofilm accumu- lation (0.003 ± 0.0004 𝐴570/mm2), while adjusted Vitablocs Mark II was least supportive of biofilm development (0.0014 ± 0.0004 𝐴570/mm2) (Figure 3). Fasbinder and Neiva [9] report that a resin nanoce- ramic material (3M ESPE Lava Ultimate) is smoother than hybrid ceramic (Vita Enamic), and both are smoother than glazed-fired leucite-reinforced ceramics. Manually polished 4 International Journal of Dentistry Control Adjusted Mark II Vita Enamic Lava Ultimate Refex Figure 1: SEM images (magnification 1000x) of control and adjusted surfaces of ceramic materials (scale bar = 1 Mark II Mark II Vita Enamic Lava Ultimate Refex Figure 1: SEM images (magnification 1000x) of control and adjusted surfaces of ceramic materials (scale bar = 10 𝜇m). storage and tooth brushing despite lower overall 𝑅𝑎values for 3M ESPE Lava Ultimate [26].i CAD/CAM materials are often smoother than conventional ceramics after either polishing or glazing [9]. Flury et al. [26] also evaluated 3M ESPE Lava Ultimate and Vita Enamic by roughening specimens in a standardised manner followed by polishing with three different systems. Surface roughness and microhardness were measured immediately after polishing and after six-month storage with monthly artificial tooth brushing. The surface of Vita Enamic was less affected by Studies aiming at developing optimal finishing tech- niques report surface roughness values as low as 0.171 𝜇m [27]. However, this involves intensive polishing procedures incorporating diamond pastes, whereas our study con- strained polishing to simulate clinical conditions, resulting in greater roughness. International Journal of Dentistry 6 [3] I. Denry and J. R. Kelly, “Emerging ceramic-based materials for dentistry,” Journal of Dental Research, vol. 93, no. 12, pp. 1235– 1242, 2014. [21] W. D. Sulik and E. J. Plekavich, “Surface finishing of dental porcelain,” The Journal of Prosthetic Dentistry, vol. 46, no. 2, pp. 217–221, 1981. [22] F. Aykent, I. Yondem, A. G. Ozyesil, S. K. Gunal, M. C. Avunduk, and S. Ozkan, “Effect of different finishing techniques for restorative materials on surface roughness and bacterial adhesion,” Journal of Prosthetic Dentistry, vol. 103, no. 4, pp. 221– 227, 2010. [4] A. Shenoy and N. Shenoy, “Dental ceramics: an update,” Journal of Conservative Dentistry, vol. 13, no. 4, pp. 195–203, 2010. [5] N. D. Ruse and M. J. Sadoun, “Resin-composite blocks for dental CAD/CAM applications,” Journal of Dental Research, vol. 93, no. 12, pp. 1232–1234, 2014. [6] L. Yin, X. F. Song, Y. L. Song, T. Huang, and J. Li, “An overview of in vitro abrasive finishing & CAD/CAM of bioceramics in restorative dentistry,” International Journal of Machine Tools and Manufacture, vol. 46, no. 9, pp. 1013–1026, 2006. [23] A. S. Brentel, K. Z. Kantorski, L. F. Valandro, S. B. F´ucio, R. M. Puppin-Rontani, and M. A. Bottino, “Confocal laser microscopic analysis of biofilm on newer feldspar ceramic,” Operative Dentistry, vol. 36, no. 1, pp. 43–51, 2011. [7] H. J. Conrad, W.-J. Seong, and I. J. Pesun, “Current ceramic materials and systems with clinical recommendations: a system- atic review,” Journal of Prosthetic Dentistry, vol. 98, no. 5, pp. 389–404, 2007. [24] S. B. Haralur, “Evaluation of efficiency of manual polishing over autoglazed and overglazed porcelain and its effect on plaque accumulation,” Journal of Advanced Prosthodontics, vol. 4, no. 4, pp. 179–186, 2012. [8] D. J. Fasbinder, “Materials for chairside CAD/CAM restora- tions,” Compendium of Continuing Education in Dentistry, vol. 31, no. 9, pp. 702–704, 2010. [25] S. Flury, A. Lussi, and B. Zimmerli, “Performance of different polishing techniques for direct CAD/CAM ceramic restora- tions,” Operative Dentistry, vol. 35, no. 4, pp. 470–481, 2010. [9] D. J. Fasbinder and G. F. Neiva, “Surface evaluation of polishing techniques for new resilient CAD/CAM restorative materials,” Journal of Esthetic and Restorative Dentistry, vol. 28, no. 1, pp. 56–66, 2016. [26] S. Flury, A. Peutzfeldt, and A. International Journal of Dentistry Lussi, “Surface roughness and microhardness of two recent CAD/CAM-materials after storage,” in Proceedings of International Association for Dental Research/Pan European Region (IADR/PER), Dubrovnik, Croa- tia, September 2014. [10] J. D. Brewer, D. A. Garlapo, E. A. Chipps, and L. A. Tedesco, “Clinical discrimination between autoglazed and polished porcelain surfaces,” The Journal of Prosthetic Dentistry, vol. 64, no. 6, pp. 631–635, 1990. [27] J. M. C. Boaventura, R. Nishida, A. A. Elossais et al., “Effect finishing and polishing procedures on the surface roughness of IPS Empress 2 ceramic,” Acta Odontologica Scandinavica, vol. 71, no. 3-4, pp. 438–443, 2013. [11] M. I. Al-Marzok and H. J. Al-Azzawi, “The effect of the surface roughness of porcelain on the adhesion of oral Streptococcus mutans,” Journal of Contemporary Dental Practice, vol. 10, no. 6, pp. 17–24, 2009. [28] S. Hahnel, M. Rosentritt, G. Handel, and R. B¨urgers, “Surface characterization of dental ceramics and initial streptococcal adhesion in vitro,” Dental Materials, vol. 25, no. 8, pp. 969–975, 2009. [12] S. Sethi, D. Kakade, S. Jambhekar, and V. Jain, “An in vitro investigation to compare the surface roughness of auto glazed, reglazed and chair side polished surfaces of Ivoclar and Vita feldspathic porcelain,” Journal of Indian Prosthodontist Society, vol. 13, no. 4, pp. 478–485, 2013. [29] R. Meier, I. Hauser-Gerspach, H. L¨uthy, and J. Meyer, “Adhesion of oral streptococci to all-ceramics dental restorative materials in vitro,” Journal of Materials Science: Materials in Medicine, vol. 19, no. 10, pp. 3249–3253, 2008. [13] T. Lie and F. Gusberti, “Replica study of plaque formation on human tooth surfaces,” Acta Odontologica Scandinavica, vol. 37, no. 2, pp. 65–72, 1979. [30] M. Quirynen and C. M. Bollen, “The influence of surface roughness and surface-free energy on supra- and subgingival plaque formation in man. A review of the literature.,” Journal of Clinical Periodontology, vol. 22, no. 1, pp. 1–14, 1995. [14] W. Teughels, N. Van Assche, I. Sliepen, and M. Quirynen, “Effect of material characteristics and/or surface topography on biofilm development,” Clinical Oral Implants Research, vol. 17, no. 2, pp. 68–81, 2006. [31] C. W. Chang, J. N. Waddell, K. M. Lyons, and M. V. Swain, “Cracking of porcelain surfaces arising from abrasive grinding with a dental air turbine,” Journal of Prosthodontics, vol. 20, no. 8, pp. 613–620, 2011. [15] B. Rosan and R. J. Lamont, “Dental plaque formation,” Microbes and Infection, vol. 2, no. 13, pp. 1599–1607, 2000. [32] S. J. References [1] A. Raut, P. L. Rao, and T. Ravindranath, “Zirconium for esthetic rehabilitation: an overview,” Indian Journal of Dental Research, vol. 22, no. 1, pp. 140–143, 2011. [2] I. Sailer, B. E. Pjetursson, M. Zwahlen, and C. H. F. H¨ammerle, “A systematic review of the survival and complication rates of all-ceramic and metal-ceramic reconstructions after an obser- vation period of at least 3 years. Part II: fixed dental prostheses,” Clinical Oral Implants Research, vol. 18, no. 3, pp. 86–96, 2007. International Journal of Dentistry International Journal of Dentistry Ahn, S. N. Park, Y. J. Lee et al., “In vitro antimi- crobial activities of 1-methoxyficifolinol, licorisoflavan a, and 6,8-diprenylgenistein against Streptococcus mutans,” Caries Research, vol. 49, no. 1, pp. 78–89, 2015. [16] G. Datzmann, CEREC Vitablocs Vitablocs Mark II Machin- able Ceramic. CAD/CAM in Aesthetic Dentistry, Quintessence, Chicago, Ill, USA, 1996. [17] E. Bojem¨uller and A. Coldea, VITA VITA ENAMIC technical- scientific documentation, VITA Zahnfabrik, Bad S¨ackingen, Germany, 2012. [18] A. Awada and D. Nathanson, “Mechanical properties of resin- ceramic CAD/CAM restorative materials,” Journal of Prosthetic Dentistry, vol. 114, no. 4, pp. 587–593, 2015. [19] A. Fedorov, R. Beichel, J. Kalpathy-Cramer et al., “3D slicer as an image computing platform for the quantitative imaging network,” Magnetic Resonance Imaging, vol. 30, no. 9, pp. 1323– 1341, 2012. [20] H. Y. Chen, R. Hickel, J. C. Setcos, and K. H. Kunzelmann, “Effects of surface finish and fatigue testing on the fracture strength of CAD-CAM and pressed-ceramic crowns,” The Journal of Prosthetic Dentistry, vol. 82, no. 4, pp. 468–475, 1999.
https://openalex.org/W2903867284	https://ntb.gpntb.ru/jour/article/download/163/164	Russian	null	The departmental publisher’s and sci-tech library within the single division	Naučnye i tehničeskie biblioteki	2,017	cc-by	2,733	Andrey Krulev Krylov State Research Center, St. Petersburg, Russia А. А. Крулёв А. А. Крулёв Крыловский государственный научный центр, Санкт-Петербург ру Крыловский государственный научный центр, Санкт-Петербург UDC 655.4.5+026 UDC 655.4.5+026 Ведомственное издательство и научно-техническая библиотека в составе единого подразделения Освещено объединение научно-технической библиотеки и редакционно-изда- тельского отдела Крыловского государственного научного центра в одно подразделение – Информационно-издательский центр. Отмечено, что качественное улучшение деятельно- сти обеих структур по ряду показателей убедительно доказывает целесообразность такой практики. Акцентировано внимание не на процессе объединения, а на достигнутых ре- зультатах. Рекомендации по объединению издательства и НТБ адресованы в том числе и руководству различных предприятий и высших учебных заведений, так как представ- ленные мероприятия могут быть реализованы независимо от направления деятельности или профиля подготовки специалистов. Ключевые слова: научно-техническая библиотека, издательство, цитирование, те- заурус, презентация, информационное обслуживание, Крыловский государственный научный центр. UDC 655.4.5+026 ОРГАНИЗАЦИЯ РАБОТЫ УДК 655.4.5+026 The departmental publisher’s and sci-tech library within the single division Integrating the Sci-tech library and the Editorial and Publishing Department of Krylov State Research Center into a single division – The Information and Publishing Center is dis- cussed. The author emphasizes that the efficiency of this solution is proved by several improved indicators. He focuses rather on the achievements than on the integration process itself. He also offers his recommendations on integrating publishing services and sci-tech library resources to top managers of organizations and universities and argues that the accomplished integration procedure is applicable to any institutions or professional training area. words: sci-tech library, publishing house, citation, thesaurus, information services. Науч. и техн. б-ки, 2017, № 2 103 The corporate library of the Krylov State Research Center for shipbuilding exists for more than 120 years. Researchers of the center have been regularly pub- lished in the "Shipbuilding" publishing house. Now these two units – the library and the publishing house are merged into a single Information and Publishing Center. We consider the benefits in the chain "publisher-library-reader”, that were brought by this merge. The main reason for combining these two successful oper- ating units is the development of electronic communication. Methods of indexing by keywords became available for employees of the publishing house. Consulta- tion with the authors of publications while preparation for release accelerate the process, and allows effectively use refillable thesaurus. Editors receive a deeper knowledge of the thematic areas of the publications issued since the thesaurus helps to better navigate in the existing collections documents and make compari- sons. In turn, the bibliography does not spend time on the processing of internal publications and can concentrate on processing the incoming external documents. This is important because such documents are much more difficult to be indexed using the internal thesaurus. Another joint area of collaboration is an analysis of the citations. The library provides consulting support to the bibliometric analyses. To summarize briefly the combined work of librarians and publishers this is the result of combining the two units: 1. Efficiency and control the transmission of copies of new publications in the library collections. 2. Collaboration in indexing of published materials, increasing the competence of the editors, achievement more active dialogue between the authors and editors at the stage of preparation of the manuscript for publication. 3. Науч. и техн. б-ки, 2017, № 2 104 The departmental publisher’s and sci-tech library within the single division Advising authors in the field of rating sources, as well as providing information on the relevance of subjects of pub- lished material on the basis of analysis of global databases. This information can be made available to representatives of the academic council for policy in the field of publishing. 4. Presentations in a specially equipped library premises. 5. Release of information products with the involvement of professionals in the field of pre-press editing. И в вузах, и в научно-исследовательских организациях, в составе кото- рых есть НТБ, – своя специфика организационной структуры. Поэтому речь в статье пойдёт в основном о результатах, достигнутых после объединения этих подразделений в единый Информационно-издательский центр (ИИЦ). НТБ ФГУП «Крыловский государственный научный центр» (далее – Крыловский центр) существует столько же, сколько и сам центр – вот уже более 120 лет. За долгие годы менялись и состав справочно-информа- ционного фонда, и численность персонала, и подчинённость подразделения, однако функции НТБ в целом оставались неизменными, а перечень услуг Науч. и техн. б-ки, 2017, № 2 104 менялся в зависимости от финансовых возможностей Крыловского центра и запросов специалистов. Издательская деятельность – один из традиционных участков работы Крыловского центра: его сотрудники регулярно публиковались, в первую очередь в издательстве «Судостроение». Существовало ведомственное из- дательство, выполняющее весь комплекс работ по выпуску научно-техни- ческой литературы. р ур Основная причина объединения этих двух успешно функционирую- щих подразделений – развитие средств электронной коммуникации. Н. В. Соколова на одной из конференций РБА говорила о неизбежных изменениях в работе библиотек и издательств: взаимодействие (например, создание классификационных индексов предстоящих публикаций) суще- ствовало всегда, а в условиях развития информационных технологий оно расширится за счёт создания интегрированных информационных систем. Работа в таких системах позволит не только автоматизировать и ускорить коммуникационные процессы, но и совершать их максимально открыто для всех участников [1]. Открытость взаимодействия в цепочке «издательство – библиотека – читатель» делает этот процесс максимально объективным. Слияние первых двух участников цепочки – не универсальный, но весьма эффективный способ оптимизировать процесс взаимодействия. Прежде чем говорить о расширении взаимодействия, обусловленного в том числе техническим прогрессом, который меняет и издательскую дея- тельность, и библиотечно-библиографические процессы, следует обозна- чить его традиционные участки. Всем публикуемым материалам, включая научные статьи, сотрудники НТБ присваивали классификационные индексы. По факту издания книги или сборника статей сигнальные экземпляры направлялись в фонд, а полу- чающие их библиографы создавали библиографические записи. Этот поря- док работы изменился в связи с созданием электронной библиотеки. The departmental publisher’s and sci-tech library within the single division ЭБ Крыловского центра – это информационная система, предназна- ченная для хранения электронных документов и обеспечивающая доступ к ним через локальную сеть предприятия. Работа с ЭБ ведётся в браузере. С помощью этого сервиса доступны следующие электронные ресурсы: элек- тронный каталог НТБ; электронная коллекция (полнотекстовые электрон- ные документы, обладающие однотипными формальными признаками и содержащие фактографическую информацию); актуализируемый перечень периодических изданий, поступающих в НТБ по подписке; раздел «Новости», в котором регулярно обновляются сведения об изданиях, размещаются интер- активные ссылки на электронные ресурсы [2. С. 29]. Науч. и техн. б-ки, 2017, № 2 105 В НТБ внедрена АБИС «Руслан», благодаря чему все библиотечно- библиографические процессы автоматизированы. На её базе функционирует ЭБ, предоставляющая пользователям локальной сети возможность доступа не только к ЭК, но и к электронным документам, включая полнотекстовые [2. С. 28]. АБИС пришла на смену внедрённой в 1990-е гг. программе MARC, ко- торая также позволяла создавать ЭК, однако доступ к нему был только у специалистов НТБ. Читатели не могли самостоятельно пользоваться этим каталогом для поиска документов, что приводило к издержкам как в обла- сти создания библиографических записей, так и в области их предметиза- ции. Сегодня, когда доступ к ЭК и полнотекстовым электронным докумен- там возможен с рабочего места, требования читателя к системе поиска неизбежно растут. При индексировании нельзя руководствоваться только личным опытом. Если раньше библиограф в лучшем случае разово консуль- тировался с представителями подразделения, то теперь этот процесс стал более открытым, а следовательно, и объективным. Создано методическое пособие, которое доступно всем в печатном и электронном виде. Оно пред- ставляет собой справочник терминов – информационно-поисковый тезаурус. р р ф р уру Обязанности по предметизации, безусловно, не возлагаются на авто- ров или читателей. Но в новых условиях возможен и актуален активный диалог, а мнение специалиста, который является в то же время и читателем, очень важно. Методика индексирования ключевыми словами стала доступна и для сотрудников издательства. Консультация с авторами на стадии подготовки издания к выпуску позволяет не только ускорить работу (вместе с докумен- том библиографы получают выбранные автором совместно с редактором рубрики, ключевые слова и т.д.), но и максимально объективно использо- вать пополняемый тезаурус. Без преувеличения можно сказать, что редакто- ры получают более глубокие знания в тематических областях издаваемых публикаций, поскольку тезаурус помогает лучше ориентироваться в фонде документов и проводить сопоставления. у р В свою очередь, библиографы не тратят время на обработку внутрен- них изданий и могут сконцентрироваться на работе с поступающими внеш- ними документами. Это важно, поскольку такие документы значительно сложнее индексировать, используя внутренний тезаурус. Науч. и техн. б-ки, 2017, № 2 106 The departmental publisher’s and sci-tech library within the single division Ещё один совместный участок работы – анализ цитируемости публи- каций. Это направление, коснувшееся работников вузов в большей степени, чем работников научно-исследовательских организаций, – сложное, но в то же время одно из самых перспективных в деятельности и издательств, и библиотек. Многие НТБ не только оказывают консультационную поддерж- ку в области наукометрии и библиометрии, но и включают эти услуги в свои обязанности. А издатели могут очень эффективно использовать сведе- ния о цитируемости, предоставляемые библиотекарями. же время одно из самых перспективных в деятельности и издательств, и библиотек. Многие НТБ не только оказывают консультационную поддерж- ку в области наукометрии и библиометрии, но и включают эти услуги в свои обязанности. А издатели могут очень эффективно использовать сведе- ния о цитируемости, предоставляемые библиотекарями. ру р р Новые возможности таких международных баз цитирования, как Sco- pus и Web of Science, позволяют пользователям оценить цитируемость пуб- ликации и получить сведения о числе публикаций по выбранной теме в ми- ровых высокорейтинговых научных журналах. Издатель не может в полной мере взять на себя ответственность за принятие решений относительно вы- пуска того или иного издания. Очевидно, что такие решения – в зоне ответ- ственности учёного совета или редакционной коллегии. Но, получая объек- тивные данные, подтверждающие актуальность публикуемых материалов, издатель в глазах руководства может выступать не просто исполнителем, но и полноправным участником обсуждения, в результате которого утвержда- ется издательская политика организации. р С этой точки зрения также очень важны сведения об использовании литературы, имеющейся в фонде и доставляемой по МБА. При должном анализе издательство может обоснованно настаивать на корректировке пла- на выпуска продукции. Необходимо сказать несколько слов о поступлении документов в фонд НТБ. Из личного опыта знаю, что вопросы, задаваемые библиотекарям (раз- личных профильных организаций) относительно новинок изданий, выпус- каемых учреждением, в состав которого входят НТБ, порой вызывают за- мешательство. И это далеко не всегда связано с техническими аспектами (отсутствие ЭК и т.п.). К сожалению, библиотекарь нередко не знает о том, какая литература вышла или скоро выйдет в ведомственном издательстве. Был случай, когда научная литература, выпущенная в одном из вузов, по- ступила в фонд НТБ Крыловского центра раньше, чем в фонд библиотеки того вуза. На вопрос о возможностях использования полнотекстовых материалов в ЭБ сотрудники библиотек зачастую тоже не могут оперативно ответить, поскольку оформлением нормативных документов, регламентирующих по- рядок работы по электронной доставке и т.д., занимаются издатели. Справедливости ради нужно сказать, что представители издательств также часто не обладают информацией о судьбе выпускаемой ими продук- ции. The departmental publisher’s and sci-tech library within the single division Вопросы о том, как получить то или иное издание, нередко остаются без ответов. Сводить функции издательства только к процессу подготовки рукопи- 107 Науч. и техн. б-ки, 2017, № 2 си к печати и руководству полиграфическим производством означает отка- зываться от анализа того, как используется издаваемая литература. Это, в свою очередь, значительно обесценивает роль издательского подразделе- ния в организации, особенно когда нет прямой коммерческой составляю- щей. В этом случае вполне объяснимо решение руководства обратиться к контрагенту. Ещё один положительный пример совместной деятельности НТБ и из- дательского центра – организация и проведение презентаций вышедших в свет изданий. Такие мероприятия, как правило, проходят в читальном зале НТБ, оснащённом современным демонстрационным оборудованием. Не исключаю, что презентации можно проводить и непосредственно в изда- тельстве, однако это, скорее, исключение, чем правило. Очевидное преимущество таких мероприятий – возможность пооб- щаться с автором, получить экземпляр издания с автографом. Именно бла- годаря подобным встречам может состояться настоящий диалог, поскольку в одном помещении собираются не просто коллеги, а учёные, проявившие профессиональный интерес к выпущенной книге. Нередко презентация про- водится в формате, близком к конференции. Не будет преувеличением ска- зать, что такие мероприятия наилучшим образом раскрыли возможности ведомственной НТБ как места для подобных встреч. N. Kranich, M. Lotts и G. Springs называют университетскую библиоте- ку местом, где члены университетского сообщества могут общаться без границ, в спокойной атмосфере [3]. Основной перспективной задачей биб- лиотеки авторы считают именно организацию пространства для коммуни- каций. Важно отметить: несмотря на развитие электронных ресурсов, биб- лиотека должна сохранить свой статус места встречи, диалога читателя и библиотекаря. В России многие университетские библиотеки сохранили этот статус, однако в силу разных причин это направление незаслуженно забыто, а работа НТБ сводится к хранению документов и организации до- ступа к ним. у В Крыловском центре НТБ сегодня – практически единственное место, где все сотрудники могут собраться для обсуждения любого вопроса. В от- личие от тематических конференций, совещаний с жёстким регламентом и ограниченным количеством участников, встречи в библиотеке, которые орга- низовать гораздо проще, становятся наиболее предпочтительными мероприя- тиями для сотрудников. Такие мероприятия полезны и для представителей издательства: они могут узнать реакцию читателей на новое издание, проана- лизировав число посетителей, активность дискуссии и т.п. Науч. и техн. б-ки, 2017, № 2 108 Одно из главных преимуществ слияния в единое подразделение – вы- пуск библиотекой собственных информационных продуктов, в том числе в печатном виде. Благодаря совместному труду библиотекарей и издателей сотрудники регулярно получают информационные бюллетени о новых по- ступлениях в НТБ и другие печатные продукты. The departmental publisher’s and sci-tech library within the single division Они прекрасно дизайнер- ски и полиграфически оформлены, что имеет весьма существенное значе- ние. Раньше библиотекарь, не имеющий опыта в области вёрстки, дизайна и полиграфии, готовил по запросу клиента библиографическую справку в текстовом редакторе. Кто-то может возразить: перечисленные направления работы (переда- ча документов от издателей в библиотеку, проведение мероприятий, выпуск информационной продукции) можно развивать и без объединения подраз- делений. Однако опыт показывает, что именно единоначалие лучшим обра- зом способствует организации работы, в том числе и потому, что издержки, связанные с координацией деятельности сотрудников (консультации и т.п.), сведены к минимуму. Подводя итоги, кратко перечислим новые возможности совместной деятельности, которые появились в результате объединения двух подразде- лений. Оперативность и контроль при передаче экземпляров новых изданий в фонд НТБ. Эта работа способствует установлению более тесных связей с коллегами из смежных организаций благодаря оповещениям. Совместная работа по предметизации публикуемых материалов, по- вышение компетенции редакторов, более активный диалог с авторами уже на стадии подготовки рукописи к изданию. Оперативное создание библио- графических записей. Консультирование авторов в области рейтинговой оценки используе- мых источников, а также предоставление сведений об актуальности темати- ки публикуемых материалов на основе анализа мировых баз научного цити- рования. Эти сведения могут быть предоставлены представителям учёного совета для определения политики в издательской деятельности. Проведение презентаций в специально оборудованном помещении НТБ. Единоначалие в этой деятельности (как и в остальных случаях) не яв- ляется определяющим условием, но помогает свети к минимуму организа- ционные издержки. Выпуск информационных продуктов с привлечением профессионалов в области допечатной подготовки и полиграфии. Выгоды от совместной, а точнее единоначальной деятельности не ис- черпываются перечисленными возможностями. В статье представлены Науч. и техн. б-ки, 2017, № 2 109 только некоторые из тех направлений работы, которые уже успешно реали- зуются в ИИЦ Крыловского центра. только некоторые из тех направлений работы, которые уже успешно реали- зуются в ИИЦ Крыловского центра. Непрерывный рост средств электронной коммуникации, увеличение количества электронных документов, другими словами – технический про- гресс, диктуют новые условия работы и издательствам, и библиотекам, ос- новным объектом деятельности которых на протяжении веков была печат- ная книга. Происходящие процессы не могут не сказаться и на управлении этими структурами. Возможно, объединение – это не только способ повы- сить эффективность деятельности, но и единственное средство выжить в новую эпоху. СПИСОК ИСТОЧНИКОВ 1. Соколова Н. В. Библиотеки и издательства – новые формы сотрудничества в едином информационном пространстве / Н. В. Соколова. – Режим доступа: http://www.rba.ru/content/ activities/section/12/mag/mag08/23.pdf. g g p Sokolova N. V. Biblioteki i izdatelstva – novye formy sotrudnichestva v edinom informatsion- nom prostranstve / N. V. Sokolova. 2. Крулёв А. А. Системный рубрикатор и тезаурус ведомственной технической библио- теки: особенности, возможности и перспективы / А. А. Крулёв // НТИ. Сер. 2. – 2016. – № 3. – С. 28–33. Krulev A. A. Sistemnyy rubrikator i tezaurus vedomstvennoy tehnicheskoy biblioteki: osoben- nosti, vozmozhnosti i perspektivy / A. A. Krulev // NTI. Ser. 2. – 2016. – № 3. – S. 28–33. 3. Kranich N. The promise of academic libraries / N. Kranich, M. Lotts, G. Springs // College and Research Libraries News. – 2014. – Vоl. 75. – № 4. – P. 182–186. 3. Kranich N. The promise of academic libraries / N. Kranich, M. Lotts, G. Springs // College and Research Libraries News. – 2014. – Vоl. 75. – № 4. – P. 182–186. Andrey Krulev, Head, Sci-tech Information Processing Group, Krylov State Research Center; Andrey Krulev, Head, Sci-tech Information Processing Group, Krylov State Research Center; Науч. и техн. б-ки, 2017, № 2 110
https://openalex.org/W4308971900	https://www.mdpi.com/2076-2615/12/22/3128/pdf?version=1668758495	English	null	Feeding Strategies for Adapting Lake Sturgeon (Acipenser fulvescens) Larvae to Formulated Diets at Early Life Stages	Animals	2,022	cc-by	13,021	Feeding Strategies for Adapting Lake Sturgeon (Acipenser fulvescens) Larvae to Formulated Diets at Early Life Stages Seunghyung Lee 1,2 , Shaowei Zhai 1 , Dong-Fang Deng 1,, Yuquan Li 1, Patrick Christopher Blaufuss 1, Bradley T. Eggold 3 and Fred Binkowski 1 2 , Shaowei Zhai 1 , Dong-Fang Deng 1,, Yuquan Li 1, Patrick Christopher Blaufuss 1, and Fred Binkowski 1 1 School of Freshwater Sciences, University of Wisconsin-Milwaukee, Milwaukee, WI 53204, USA 2 Major in Aquaculture and Applied Life Sciences, Division of Fisheries Life Sciences, Pukyong National University, 45 Yongso-ro, Nam-gu, Busan 48513, Korea 3 Great Lakes Research Facility, Department of Natural Resources, 600 E. Greenﬁeld Avenue, Milwaukee, WI 53204, USA * Correspondence: dengd@uwm.edu 1 School of Freshwater Sciences, University of Wisconsin-Milwaukee, Milwaukee, WI 53204, USA 2 Major in Aquaculture and Applied Life Sciences, Division of Fisheries Life Sciences, Pukyong National University, 45 Yongso-ro, Nam-gu, Busan 48513, Korea 3 Great Lakes Research Facility, Department of Natural Resources, 600 E. Greenﬁeld Avenue, Milwaukee, WI 53204, USA * Correspondence: dengd@uwm.edu * Correspondence: dengd@uwm.edu Simple Summary: Failure of larvae and juveniles to transition from live feed to prepared diets is a common cause of signiﬁcant mortality in many ﬁsh species, including lake sturgeon (Acipenser fulvescens). Our study investigated feeding strategies that adapt lake sturgeon transition to formulated diets. The results showed that co-feeding formulated diets with live feed for periods of 3 or 4 weeks can improve growth and survival during this transition. Our ﬁnding also suggested that introducing formulated diets early possesses a potential to improve tolerance to environmental hypoxia, which may be due to balanced nutrient proﬁles. Abstract: Cost-effective feeding management is required to support conservation hatcheries for lake sturgeon (Acipenser fulvescens), an ecologically important species in the Great Lakes region. This study investigated an approach to transition lake sturgeon larvae from live feed (Artemia) to formulated feed and its effect on growth performance, survival, and response to acute hypoxia stress. The ﬁrst experiment showed that sturgeon had similar (p > 0.05) growth and survival when fed Artemia or the combined feeding of Artemia with the commercial diet (crude protein, 551 g/kg diet). Feeding solely on the commercial or lab-made (crude protein, 491 g/kg diet) diet signiﬁcantly reduced growth and survival (p < 0.05). In the second experiment, the growth performance of sturgeon (14 days post-hatch, DPH) fed with either Artemia only or combined feeding different feeding durations of two, three, and four weeks followed by a complete transition to the commercial diet. Citation: Lee, S.; Zhai, S.; Deng, D.-F.; Li, Y.; Blaufuss, P.C.; Eggold, B.T.; Binkowski, F. Feeding Strategies for Adapting Lake Sturgeon (Acipenser fulvescens) Larvae to Formulated Diets at Early Life Stages. Animals 2022, 12, 3128. https://doi.org/ 10.3390/ani12223128 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. Copyright: © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Keywords: Artemia; combined feeding; growth; larval fish culture; larval feed; survival; stock enhancement Feeding Strategies for Adapting Lake Sturgeon (Acipenser fulvescens) Larvae to Formulated Diets at Early Life Stages At the end of six weeks, the 3- and 4-week combined feeding periods resulted in signiﬁcantly higher body weight and survival compared to the 2-week combined and the Artemia only feeding treatments. In the last experiment, sturgeons (27 DPH) were fed only with Artemia or combined feeding of Artemia with the commercial diet for four weeks followed by the complete transition to the commercial diet for two weeks. Eighteen ﬁsh from each treatment were investigated the response to acute hypoxic conditions (gradual decrease in dissolved oxygen level from 8 to 2.3 mg/L at the rate of 1 mg/L per hour). When the dissolved oxygen was between 3 and 4 mg/L, the mortality rate of the combination- fed sturgeon (11.7%) was signiﬁcantly lower than those fed only Artemia (83.3%). These results clearly demonstrate that a commercial diet can partially replace Artemia at early life stages to improve growth, survival, and hypoxia tolerance and thus its co-feeding with Artemia is recommended. animals animals 1. Introduction Lake sturgeon (Acipenser fulvescens) are ecologically important species of ﬁsh in the Laurentian Great Lakes and Mississippi River basins and are currently listed as a either https://www.mdpi.com/journal/animals Animals 2022, 12, 3128. https://doi.org/10.3390/ani12223128 Animals 2022, 12, 3128 2 of 15 species of special concern or threatened in different states of the Great Lake Region [1–3]. The total population of lake sturgeon was estimated to be more than ﬁfteen million at its peak, but it is now reduced to less than 1% of the historic population. Different stressors including overﬁshing, habitat deterioration, reduced water quality, climate change, and invasive species have contributed to lake sturgeon extirpation in different regions [4,5]. Maturing at a late age (12–15 years for males and 18–27 years for females), it has a prolonged reproductive cycle [6]. With this special life history, a long-term management plan is needed to establish sufﬁcient juvenile and adult ﬁsh to reach an adequate population restoration [7]. The stocking hatchery-reared juvenile sturgeon is one strategy that can enhance the recovery of lake sturgeon populations [8]. Many have explored feeding strategies to improve culture success in lake sturgeon hatcheries [9–13]. Overall, live Artemia remains an essential live feed for maintaining good growth and survival during the early life stages. Bauman et al. [10] reported that partial replacement of Artemia by formulated feed could maintain high survival, but reduced growth appeared within two weeks of feeding. In contrast to larval ﬁsh, juvenile lake sturgeon (56 days post-hatch: DPH) that had been weaned to formulated diet obtain better growth and have higher nutrient retention than those fed with bloodworm [12]. For the grow-out lake sturgeon, the formulated feed could also support good growth and survival observed under lab conditions [13]. This suggests that lake sturgeon can be raised on formulated feed if a proper weaning process is implemented during the larval stage. Thus, the objective of this study is to investigate feeding strategies to manage lake sturgeon weaning to formulated feed. The ultimate goal is to reduce labor and feed costs associated with juvenile production and produce healthy ﬁsh to support its population restoration. 2.1. Feeding Trial I 2.1. Feeding Trial I Two thousand and one hundred 14-DPH larvae (33 ± 4 mg; mean ± SD) were dis- tributed into 21 polypropylene circular tanks (45 cm diameter, 50 cm height, 60 L water volume), with 100 larvae per tank, supplied with ﬂow-through degassed water (ﬂow rate: ca. 0.9 L/min) for acclimation of the larvae to the experimental set-up. During the one-week acclimation period, commercial feed (Otohime B1, Marubeni Nisshin Feed Co., LTD, Tokyo, Japan) was provided at a feeding rate of 13% body weight per day (BW/d) using 24 h automatic feeders (Lifegard Automatic Fish Feeder, Lifegard Aquatics, Cerritos, California, USA). Artemia nauplii were also fed to the larvae at the same rate (based on the estimation of the dry weight of the nauplii) six times per day (8:30, 10:00, 11:30, 13:00, 14:30, 16:00 h). The overall feeding rate, 26% BW/d, was adopted from a previous study conducted on white sturgeon during a similar life stage [15]. g g g At the end of the acclimation period, all larvae were pooled into a large tank, and larvae (21-DPH, 62 ± 4 mg) were re-distributed into 15 experimental tanks at 75 larvae per tank. The tanks were randomly assigned to 5 diets with 3 replicated tanks per treatment (Table 1): A, Artemia nauplii; B, commercial larval feed (Otohime B1); L, a lab-made diet; A + B; and A + L. The lab-made diet was formulated to contain (g/1000 g): casein (280), gelatin (70), squid meal (60), wheat gluten (10), egg white (80), dextrin (113.3), wheat starch (60), a 1:1:1 mixture of soy oil: ﬁsh oil: lard (80), lecithin (60), cholesterol (1.2), sodium alginate (20), carboxyl methylcellulose (10), canthaxanthin (1), ascorbyl palmitate (0.5), choline chloride (0.5), glucosamine (1.5), potassium phosphate (10), sodium phosphate (5), Haematococcus (algae) powder (20), vitamin premix (15), mineral premix (1.5), and betaine hydrochloride (1.5). All ingredients were obtained from Sigma-Aldrich, Inc., St. Louis, MO, USA) except vitamin and mineral premixes, which were provided by the Bozeman Fish Technology Center (Bozeman, MT, USA) and shared the same formulation as that in Sealey et al. [16]. The algae powder was from Cyanotech Corporation, Kona Hawaii, USA. The lab-made diet was processed following a cold extruding method described by Jiang et al. [17]. Brieﬂy, all dry ingredients were fully mixed to obtain a homogenous mixture before the oil and lecithin were added for further mixing. 2. Materials and Methods Three feeding experiments were conducted using the same culture system at the School of Freshwater Sciences, University of Wisconsin-Milwaukee. In the ﬁrst experiment (Feeding trial I), a 4-week feeding trial was conducted to compare the effect of feeding Artemia, a commercial diet, a lab-made diet, or a combination of Artemia with either formu- lated diet on the growth performance of lake sturgeon larvae. In the second experiment (Feeding trial II), a 6-week feeding trial was conducted to investigate the impact of Artemia or combined feeding (Artemia and a commercial diet) on the growth and survival of lake sturgeon weaned to the commercial feed only at different points (2, 3, and 4 weeks). In the third experiment (Feeding trial III), a 6-week feeding trial was conducted to compare the tolerance to hypoxia of lake sturgeon fed with Artemia or combined feed (Artemia and a commercial feed) for four weeks before being weaned to the commercial diet only. The experiments followed an animal care protocol approved by the Institute Animal Care and Use Committee (15-16 #60) at the University of Wisconsin-Milwaukee. ( ) y Seven-DPH larvae were stocked in circular ﬁberglass tanks (86 cm diameter, 61 cm height, 260 L water volume) supplied with ﬂow-through degassed water at a rate of 5 L/min. Water temperature was gradually increased from 18 ◦C at stocking to 22 ◦C (0.5 ◦C increment per day). Dissolved oxygen was measured daily (YSI Pro1020, YSI Life Sciences, Yellow Springs, OH, USA) and was maintained at >8 mg/L throughout the rearing period. Once excretion of the melanin plugs from spiral valves was ob- served (11-12 DPH), a commercial feed (Otohime B1, Marubeni Nisshin Feed Co., LTD, Tokyo, Japan) and newly hatched Artemia nauplii (Great Salt Lake strain, Artemia In- ternational LLC, Fairview, TX, USA) were provided daily for acclimating the larvae to exogenous feeding. The Artemia nauplii were prepared by incubating decapsulated Artemia cysts (Great Salt Lake, Artemia International LLC, Fairview, Texas, USA). Decapsulation was performed according to the methodology described in Sorgellos et al. [14]. Cysts were hatched in 25 g/L salinity water at 28 ◦C. Artemia nauplii (incubated for 24–32) were collected to feed the larvae 6 times daily at 8:30, 10:00, 11:30, 13:00, 14:30, and 16:00 h. 3 of 15 3 of 15 Animals 2022, 12, 3128 2.1. Feeding Trial I Total ammonia nitrogen and pH were measured weekly, using a test kit (Ammonia Test Strips, Hach, Loveland, CO, USA) and a pH meter (YSI Pro1020, YSI Animals 2022, 12, 3128 4 of 15 Life Sciences, Yellow Springs), respectively, and their levels were maintained at <0.08 mg/L and 7.5–7.6, respectively. The photoperiod was maintained at 12L:12D (12 h light and 12 h dark) throughout the trial. Table 1. Experimental treatments of Feeding trial I. Treatment Feeding Rate (% BW) during the 4 Weeks 1 2 3 4 Artemia (A) 10 7.5 6.5 4.8 Commercial diet (B) 20 No more feeding Lab diet (L) 20 No more feeding A + B A: 5, B: 10 A: 3.75, B: 7.5 A: 3.25, B: 6.5 A: 2.4, B: 4.8 A + L A: 5, L: 10 A: 3.75, L: 7.5 A: 3.25, L: 6.5 A: 2.4, L: 4.8 The initial average weight of lake sturgeon larvae was 62 mg ± 1.1 (n = 15, 21-day post-hatch). Table 1. Experimental treatments of Feeding trial I. Table 1. Experimental treatments of Feeding trial I. Treatment Feeding Rate (% BW) during the 4 Weeks 1 2 3 4 Artemia (A) 10 7.5 6.5 4.8 Commercial diet (B) 20 No more feeding Lab diet (L) 20 No more feeding A + B A: 5, B: 10 A: 3.75, B: 7.5 A: 3.25, B: 6.5 A: 2.4, B: 4.8 A + L A: 5, L: 10 A: 3.75, L: 7.5 A: 3.25, L: 6.5 A: 2.4, L: 4.8 The initial average weight of lake sturgeon larvae was 62 mg ± 1.1 (n = 15, 21-day post-hatch). Table 2. Nutritional composition of test diets for Feeding trial I. Test Diet Analysis Artemia nauplii Commercial Feed (Otohime B) Lab-Made Diet Proximate composition (g/kg, dry matter basis) Moisture 946 65 110 Protein 513 551 491 Lipid 164 180 139 Ash 98 129 76 Major fatty acid (g/kg of total fat) 16:0 113.4 176.6 187 16:1 31.9 41.8 49.7 18:0 40.2 22.8 56.2 18:1 n-9 187.0 100.4 138.3 18:2 n-6 56.3 53.1 218.1 18:3 n-3 308.9 44.2 47 20:5 n-3 14.9 123.3 54.5 22:6 n-3 0.3 107.8 38.1 Table 2. Nutritional composition of test diets for Feeding trial I. 2.1. Feeding Trial I Then, the mixture was mixed with warm deionized water (80 ◦C) before extruding into 2 mm strands, which were post-cooked at 80 ◦C for 15 min to improve binding. The feed was air dried overnight at room temperature and then ground and sieved to a size suitable for the larvae depending on the ﬁsh sizes. The nutritional composition of test diets is presented in Table 2. The feeding for treatment B and L only last for one week and the rest treatments lasted for 4 weeks. During the ﬁrst week of the trial, high mortality (18–20%) was observed in the larvae fed the formulated feed only (B and L diets). In addition, the remaining larvae showed a very poor response to feeding. Thus, the two treatments were terminated at the end of the ﬁrst week. The remaining treatments were continued until the end of the 4-week feeding trial. The newly hatched Artemia nauplii were provided four times daily at 8:30, 11:00, 13:30, and 16:00 h. The prepared diets were loaded into the same 24 h automatic feeder described previously and dispensed feed 8 times daily. The central standpipe was wrapped with a 1 mm2-opening nylon mesh screen and was cleaned daily. Tanks were siphoned out daily before the ﬁrst feeding of the day. During the 4-week feeding trial, all larvae in each tank were batch-weighed weekly, and then the feeding rate was adjusted accordingly. The larvae were not fed for 20 h prior to weighing to reduce any stress caused by handling. The feeding rate was 20%, 7.5%, 6.5%, and 4.8% BW/d for the ﬁrst, second, third, and last weeks of the feeding trial, respectively. The feeding rates were adopted from a previous study conducted on white sturgeon larvae [15] because relevant information on an optimum feeding rate for lake sturgeon at this life stage was not available. Water temperature was measured twice daily in the morning and afternoon and was maintained at 21.8–23.5 ◦C during the trial. Dissolved oxygen was measured daily using a meter (YSI Pro1020, YSI Life Sciences, Yellow Springs) and was maintained at >8 mg/L throughout the trial. 2.1. Feeding Trial I Test Diet Analysis Artemia nauplii Commercial Feed (Otohime B) Lab-Made Diet Proximate composition (g/kg, dry matter basis) Moisture 946 65 110 Protein 513 551 491 Lipid 164 180 139 Ash 98 129 76 Major fatty acid (g/kg of total fat) 16:0 113.4 176.6 187 16:1 31.9 41.8 49.7 18:0 40.2 22.8 56.2 18:1 n-9 187.0 100.4 138.3 18:2 n-6 56.3 53.1 218.1 18:3 n-3 308.9 44.2 47 20:5 n-3 14.9 123.3 54.5 22:6 n-3 0.3 107.8 38.1 Table 2. Nutritional composition of test diets for Feeding trial I. At the end of the 4-week feeding trial, all ﬁsh in each tank were batch-weighed to obtain growth performance indices: weight gain, feed conversion ratio, condition factor, and protein efﬁciency ratio. These measurements were calculated using the following equations: WG, %) = 100 × (ﬁnal body weight (mg) −initial body weight (mg))/initial body weight (mg) Feed conversion ratio (FCR) = dry feed weight provided (mg)/wet weight gain (mg) Condition factor (CF) = 100 × Body weight (mg)/total body length (mm)3 Protein efﬁciency ratio (PER) = Fish weight gain (mg)/protein provided (mg) Prior to weighing and sampling, the feed was withheld from ﬁsh for 20 h. After the weighing, all ﬁsh from each tank were euthanized with an overdose (500 mg/L) of MS- 222 (Western Chemical, Ferndale, WA, USA) and rinsed with deionized water for sample collection. Ten ﬁsh per tank were collected for the assays of RNA/DNA ratio in the whole ﬁsh, and 19–25 ﬁsh per tank were used for measuring individual body weight and total length. The latter group of ﬁsh was also used for the analysis of whole-body proximate composition. All the above samples were snap-frozen in liquid nitrogen and stored at −80 ◦C until analysis RNA/DNA ratio is widely used as an indicator of the growth and nutritional status of ﬁshes [18,19]. RNA/DNA ratio in the whole larva was measured following the protocol established for larval ﬁsh, using a ﬂuorometric method [20]. Brieﬂy, 10 pooled larvae from each tank were pulverized into powder using a liquid nitrogen-cooled mortar and Animals 2022, 12, 3128 5 of 15 5 of 15 pestle. 2.1. Feeding Trial I Approximately 20 mg of each of the ground samples was mixed with 150 µL of 1% sarcosil Tris-EDTA buffer solution (STEB: 1% N-lauroylsarcosine (w/v), 5 mM Tris-HCl, 0.5 mM EDTA, pH 7.5) for 1 h at room temperature at 1200 rpm using a mixer (Ther- momixer R, Eppendorf, Hamburg, Germany). Then, 1.35 mL of Tris-EDTA buffer solution was added to the above mixture, which was mixed 40 times by manual inversion. The samples were then centrifuged at 14,000 rpm for 15 min at room temperature, and the super- natant collected. DNA (originated from calf thymus; Product Code: D4764, Sigma-Aldrich, St. Louis, MO, USA) and RNA (16S- and 23S-ribosomal originated from E. coli MRE 600; Product Code: 10206938001, Roche Diagnostics, Indianapolis, IN, USA) standards were used to develop a calibration standard curve for estimation of DNA and RNA concentra- tions. The ﬁrst reading (total concentration of RNA and DNA) was recorded at 525 nm excitation wavelength and 600 nm emission wavelength using a microplate ﬂuorescence reader (Synergy H4 Hybrid Reader, BioTek, Winooski, VT, USA). Then, RNase (from bovine pancreas; Product Code: R6513, Sigma-Aldrich) solution (20 U/mL) was added to degrade the RNA to obtain a second reading of DNA concentration. Concentrations (µg) were calculated based on the ﬂuorescent readings and the standard calibration curves, with the concentration of RNA calculated by subtracting the concentration of DNA from the total concentration. The proximate composition of feed and whole larvae was determined following the methods of the Association of Ofﬁcial Analytical Chemists [21]. Moisture was determined by drying a sample in a freeze dryer overnight and then in a 105 ◦C convection oven for 24 h. Total nitrogen and carbon contents were measured following the Dumas method using an elemental combustion system (Costech Analytical Technologies Inc., Valencia, CA, USA), with crude protein estimated by N × 6.25. Ash was measured in a mufﬂe furnace at 600 ◦C for 6 h [21]. 2.2. Feeding Trial II There were 7 feeding treatments, including: A-R, feeding Artemia at 50% of estimated saturation for 4 weeks followed by two weeks of feeding with the commercial diet B (treat- ment 1); A2 two weeks of Artemia feeding (treatment 2) or AB2, a combination of Artemia and commercial diet B (treatment 3) followed by two weeks of commercial diet B; A3, three weeks of Artemia feeding (treatment 4) or a combination of Artemia and commercial diet B (treatment 5) followed by 3 weeks of commercial diet B; and A4, four weeks of Artemia feeding (treatment 6) or a combination of Artemia and commercial diet B (treatment 7) followed by 2 weeks of commercial diet B. The actual feeding was presented in Table 2. The preparation of Artemia and fish maintenance followed the same protocols used in feeding trial I. p g Lake sturgeon larvae (12 DPH) were randomly assigned to one of the 7 treatments (Table 3) with 3 replicated tanks per treatment and 120 fish (average body weight, 34 ± 1.8 mg, n = 21) per tank. Artemia was fed at 9:00, 11:00, 13:00, and 15:00 h, and the commercial feed (Otohime B1) was fed eight meals daily. Water temperature was maintained between 21.5–23.2 ◦C, dissolved oxygen > 7.6–8.1 mg/L, pH 7.5–7.9, and TAN < 0.08 mg/L. The treatments fed A2 and AB2 were terminated at the end of 4 weeks because the overall mortality was 50% or greater. During the 6-week feeding trial, all larvae in each tank were batch-weighed weekly, and feeding rates were adjusted according to fish growth (Table 2). The final body weight and survival of fish from each tank were determined following the same method described in Feeding trial I. Table 3. Feeding rate (based on a percentage of body weight) for lake sturgeon during 6 weeks of feeding (Feeding trial II). Dietary Treatment Feeding Duration (Week) 1 2 3 4 5 6 Feeding Rate 8 A-R 1 10 A 5 A 3 A 3 A 9 B 9 B Table 3. Feeding rate (based on a percentage of body weight) for lake sturgeon during 6 weeks of feeding (Feeding trial II). Animals 2022, 12, 3128 6 of 15 Table 3. Cont. 2.3. Feeding Trial III Sturgeon larvae (27 DPH) were raised for 4 weeks with either Artemia or a combination of Artemia and the commercial diet used in the last two trials (2:1 ratio) following a similar protocol described in feeding trial II. Then, all ﬁsh were transitioned to the commercial diet for two more weeks. Water temperature was maintained at 20.2–22.3 ◦C, dissolved oxygen 7.8–8.1 mg/L, pH was 7.5–8.0, and TAN was <0.02 mg/L. g p g At the end of the 6-week trial, the fish were exposed to acute hypoxic stress. Fish were withheld from feeding and then transferred to 6 cages (width × length × height = 10 cm × 10 cm × 6 cm). Each cage held six fish, and three cages were assigned to each dietary treatment. The water temperature was similar to that in the previous culture system. The fish were acclimated to the system for 6 h, and then dissolved oxygen was decreased by injecting nitrogen gas. The level of dissolved oxygen was decreased from 8 mg/L to 2.3 mg/L at a rate of 1 mg/L hourly and mortality was monitored. 2.2. Feeding Trial II 3 A3: ﬁsh were fed to the estimated satiation for three weeks with artemia (A) before they were switched to the commercial diet (B). 4 A4: ﬁsh were fed to the estimated satiation for four weeks with artemia (A) before they were switched to the commercial diet (B). 5 AB2, 6 AB3, and 7 AB4: ﬁsh were co-fed with artemia and the commercial diet (B) for 2, 3, or 4 weeks, respectively, before they were switched to the commercial diet (B). The treatment AB2 was ﬁnished at the end of the 4th week. 8 The numbers before each capital letter represent the feeding rate (based on % body weight) and the capital letters represent two different diets (A for artemia and B for the commercial diet). The initial body weight of the ﬁsh was 34 ± 1.8 mg, n = 21, 14-day post-hatch. 2.4. Data Analysis Results were subjected to a one-way analysis of variance with signiﬁcance at p < 0.05, with pairwise comparisons between different treatment means performed with Tukey’s studentized range (HSD) test. Tests for assumptions of normality and homogeneity of vari- ance were performed using Shapiro–Wilk and Levene’s tests, respectively. Data violating either assumption (p < 0.05) were log-transformed. A second-order polynomial regression analysis was applied to estimate the relationship between the feeding duration and body weight increase (Trial I). Results for Trial III were subjected to t-test to compare responses of two diet treatments (A and A + B) (p < 0.05). 2.2. Feeding Trial II Dietary Treatment Feeding Duration (Week) 1 2 3 4 5 6 A2 2 20 A 10 A 9 B 9 B A3 3 20 A 10 A 6 A 9 B 6 B 6 B A4 4 20 A 10 A 6 A 6 A 6 B 6 B AB2 5 10 A + 10 B 5 A + 10 B 9 B 9 B AB3 6 10 A + 10 B 5 A + 10 B 3 A + 6 B 9 B 6 B 6 B AB4 7 10 A + 10 B 5 A + 10 B 3 A + 6 B 3 A + 6 B 6 B 6 B 1 A-R: ﬁsh were fed to 50% of the estimated satiation with artemia for our weeks before they were transited to feeding the commercial diet (B). 2 A2: ﬁsh were fed to the estimated satiation for two weeks with artemia (A) before they were switched to the commercial diet (B). This treatment was ﬁnished at the end of the 4th week. 3 A3: ﬁsh were fed to the estimated satiation for three weeks with artemia (A) before they were switched to the commercial diet (B). 4 A4: ﬁsh were fed to the estimated satiation for four weeks with artemia (A) before they were switched to the commercial diet (B). 5 AB2, 6 AB3, and 7 AB4: ﬁsh were co-fed with artemia and the commercial diet (B) for 2, 3, or 4 weeks, respectively, before they were switched to the commercial diet (B). The treatment AB2 was ﬁnished at the end of the 4th week. 8 The numbers before each capital letter represent the feeding rate (based on % body weight) and the capital letters represent two different diets (A for artemia and B for the commercial diet). The initial body weight of the ﬁsh was 34 ± 1.8 mg, n = 21, 14-day post-hatch. 1 A-R: ﬁsh were fed to 50% of the estimated satiation with artemia for our weeks before they were transited to feeding the commercial diet (B). 2 A2: ﬁsh were fed to the estimated satiation for two weeks with artemia (A) before they were switched to the commercial diet (B). This treatment was ﬁnished at the end of the 4th week. 3.1. The Potential of Substituting Artemia with Formulated Feed Growth performances of lake sturgeon larvae fed the different diets for one week (Trial I) 1. 1 Means ± SEM (n = 3) with different superscripts within each row are signiﬁcantly (p < 0.05) different by Tukey’s studentized range test. 2 Treatment: A, Artemia; B, commercial diet; L, lab diet; A + B, a combination of A and B; A + L, a combination of A and L; Treatment B and L were terminated at the end of the ﬁrst week feeding due to the high mortality observed in these treatments. 3 Weight gain (%) = 100 × (ﬁnal body weight (mg) −initial body weight (mg))/initial body weight (mg). 4 Feed conversion ratio = Dry feed weight provided (mg)/weight gain (mg). 5 Protein efﬁciency ratio = Fish weight gain (mg)/protein provided (mg). At the end of the 4th week feeding trial (Trial I), the survival of larvae was similar among the three remaining treatments (Table 5). The ﬁsh fed with treatment A or A + B had a similar weight gain and body weight, but those fed treatment A + L had signiﬁcantly lower growth than the ﬁsh fed diet A (Table 5). The feed conversion ratio was the lowest and the protein efﬁciency ratio was the highest for ﬁsh fed diet A, followed by those fed A + B and then A + L. A signiﬁcantly positive correlation was observed between ﬁsh weight and feeding duration for all three groups of ﬁsh (Figure 1). Although the ﬁnal body weight was similar between ﬁsh-fed A and A + B, based on the regression model calculation it is likely that the ﬁsh fed the A + B will be larger than the ﬁsh fed A if feeding had continued (Figure 1). The moisture content was signiﬁcantly lower, but the ratio of C/N was higher in ﬁsh fed A + B when compared to those fed A or A + L. The lowest level of ash was detected in the ﬁsh fed with A + L and the highest level was measured in the A-fed ﬁsh. No signiﬁcant difference was observed in the condition factor, protein level, and RNA/DNA ratio of whole ﬁsh in response to different feedings. Table 5. Growth performances and whole-body proximate composition of lake sturgeon fed with different diets for 4 weeks (Trial I) 1. 3.1. The Potential of Substituting Artemia with Formulated Feed The growth performance (Trial I) of lake sturgeon larvae fed different diets during the ﬁrst week is shown in Table 4. The larvae fed solely the formulated diets (treatment B and L) showed signiﬁcantly (p < 0.05) lower survival in comparison to the larvae fed Artemia (A) or the combinations treatment (A + B and A + L), which showed 99 to 100% survival. Feeding of the sole formulated diet generally showed poor performance in weight gain, feed conversion ratio, and protein efﬁciency ratio in comparison to feeding of Artemia or the combinations. In addition, the ﬁsh-fed diet B or L had low locomotor activity (observation) and thus these two treatments were terminated at the end of the ﬁrst week. Animals 2022, 12, 3128 7 of 15 Table 4. Growth performances of lake sturgeon larvae fed the different diets for one week (Trial I) 1. Treatment 2 Measurement A B L A + B A + L Survival (%) 100.0 ± 0.0 A 78.2 ± 2.7 B 80.9 ± 7.2 B 99.1 ± 0.9 A 99.1 ± 0.9 A Final body weight (mg) 137 ± 3 A 93 ± 6 C 81 ± 1 C 121 ± 3 AB 103 ± 8 BC Weight gain 3 (%) 120.2 ± 4.3 A 49.3 ± 10.2 C 29.9 ± 1.9 C 95.1 ± 5.6 AB 66.4 ± 12.9 BC Feed conversion ratio 4 0.4 ± 0.0 C 2.6 ± 0.7 AB 3.6 ± 0.2 A 0.8 ± 0.1 C 1.2 ± 0.3 BC Protein efﬁciency ratio 5 4.74 ± 0.17 A 0.74 ± 0.15 C 0.53 ± 0.03 C 2.07 ± 0.12 B 1.65 ± 0.32 B 1 Means ± SEM (n = 3) with different superscripts within each row are signiﬁcantly (p < 0.05) different by Tukey’s studentized range test. 2 Treatment: A, Artemia; B, commercial diet; L, lab diet; A + B, a combination of A and B; A + L, a combination of A and L; Treatment B and L were terminated at the end of the ﬁrst week feeding due to the high mortality observed in these treatments. 3 Weight gain (%) = 100 × (ﬁnal body weight (mg) −initial body weight (mg))/initial body weight (mg). 4 Feed conversion ratio = Dry feed weight provided (mg)/weight gain (mg). 5 Protein efﬁciency ratio = Fish weight gain (mg)/protein provided (mg). Table 4. 1 Means ± SEM (n = 3) with different superscripts within each row are signiﬁcantly (p < 0.05) different by Tukey’s HSD test. 2 Deﬁnition of treatment sees Table 4 for a footnote. 3.1. The Potential of Substituting Artemia with Formulated Feed y = 18.111x2 + 51.444x + 63.508 R² = 0.9999 y = 30.82x2 + 6.2912x + 70.203 R² = 0.9943 y = 16.871x2 + 10.863x + 67.394 R² = 0.9911 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 0 1 2 3 4 Body weight (mg) Feeding duration (week) A A+B A+L Figure 1. Average body weight (n = 3) of sturgeon fed different test diets during a 4-week feeding (Trial-I). A, Artemia; A + B, combined feeding of Artemia and commercial diet. A + L, combined feeding of Artemia and lab diet. Figure 1. Average body weight (n = 3) of sturgeon fed different test diets during a 4-week feeding (Trial-I). A, Artemia; A + B, combined feeding of Artemia and commercial diet. A + L, combined feeding of Artemia and lab diet. 3.2. Effect of Different Feedings on the Adaptation of Lake Sturgeon to Formulated Feed 3.2. Effect of Different Feedings on the Adaptation of Lake Sturgeon to Formulated Feed 3.2. Effect of Different Feedings on the Adaptation of Lake Sturgeon to Formulated Feed 3.2. Effect of Different Feedings on the Adaptation of Lake Sturgeon to Formulated Feed The results (Trial II) presented in Figure 2 compared the mortality and body weight of sturgeon-fed Artemia for 4 weeks (A-R and A4), sturgeon-fed Artemia (A2), or combined feeding (Artemia and the commercial diet, AB2) for 2 weeks before changing to feeding solely on the commercial diet only during weeks 3 and 4. Sturgeon raised in treatment A- R and AB2 were fed only 50% of Artemia that was provided to A2 or A4. Despite different feedings, mortality (Figure 2a) and body weight (Figure 2b) were similar among all of the treatments at the end of the 2nd week. During weeks 3 and 4, the mortality remained low (1.3 to 3.3%) for the fish continually fed with Artemia (treatments A and A-R). This agrees with the results observed in Trial I, which showed that Artemia feeding is sufficient to maintain survival during this early life stage. A reduced feeding of Artemia (A-R) led to significantly reduced growth fish during the third and fourth weeks when compared to the A4 fish. 3.1. The Potential of Substituting Artemia with Formulated Feed A significant increase in mortality occurred in fish previously fed with either Artemia (A2) or combined feeding (AB2) when their feed switched to the commercial diet at the end of the second week (Figure 2a). The mortality and body weight of fish were similar between these two treatments at the end of third week (the first week after changing feed), but at the end of the fourth week, the mortality was significantly higher and body weight was lower in treatment A2 than those observed in treatment AB2. The fish from the A-R treatment were smaller than any of the other feeding treatments. The results (Trial II) presented in Figure 2 compared the mortality and body weight of sturgeon-fed Artemia for 4 weeks (A-R and A4), sturgeon-fed Artemia (A2), or combined feeding (Artemia and the commercial diet, AB2) for 2 weeks before changing to feeding solely on the commercial diet only during weeks 3 and 4. Sturgeon raised in treatment A-R and AB2 were fed only 50% of Artemia that was provided to A2 or A4. Despite different feedings, mortality (Figure 2a) and body weight (Figure 2b) were similar among all of the treatments at the end of the 2nd week. During weeks 3 and 4, the mortality remained low (1.3 to 3.3%) for the ﬁsh continually fed with Artemia (treatments A and A-R). This agrees with the results observed in Trial I, which showed that Artemia feeding is sufﬁcient to maintain survival during this early life stage. A reduced feeding of Artemia (A-R) led to signiﬁcantly reduced growth ﬁsh during the third and fourth weeks when compared to the A4 ﬁsh. A signiﬁcant increase in mortality occurred in ﬁsh previously fed with either Artemia (A2) or combined feeding (AB2) when their feed switched to the commercial diet at the end of the second week (Figure 2a). The mortality and body weight of ﬁsh were similar between these two treatments at the end of third week (the ﬁrst week after changing feed), but at the end of the fourth week, the mortality was signiﬁcantly higher and body weight was lower in treatment A2 than those observed in treatment AB2. The ﬁsh from the A-R treatment were smaller than any of the other feeding treatments. fish from the A R treatment were smaller than any of the other feeding treatments. 3.1. The Potential of Substituting Artemia with Formulated Feed Treatment 2 A A + B A + L Growth performance Survival (%) 96.7 ± 0.7 94.7 ± 2.9 92.7 ± 1.8 Final body weight (mg) 560 ± 20 A 599 ± 27 A 389 ± 22 B Weight gain (%) 803.4 ± 31.8 A 866.8 ± 43.6 A 527.3 ± 35.4 B Condition factor 0.34 ± 0.01 0.34 ± 0.00 0.33 ± 0.00 Feed conversion ratio 0.5 ± 0.0 C 0.7 ± 0.0 B 0.8 ± 0.0 A Protein efﬁciency ratio 3.83 ± 0.11 A 2.55 ± 0.10 B 2.22 ± 0.08 B Proximate composition Moisture (g/kg) 895 ± 3 A 880 ± 3 B 900 ± 3 A Crude protein (g/kg) 78 ± 1 83 ± 1 76 ± 3 Crude ash (g/kg) 14 ± 1 A 13 ± 0 AB 11 ± 1 B C/N ratio 3.74 ± 0.05 B 4.25 ± 0.13 A 3.70 ± 0.02 B RNA/DNA ratio 1.12 ± 0.09 1.19 ± 0.08 1.26 ± 0.14 1 Means ± SEM (n = 3) with different superscripts within each row are signiﬁcantly (p < 0.05) different by Tukey’s HSD test. 2 Deﬁnition of treatment sees Table 4 for a footnote. Table 5. Growth performances and whole-body proximate composition of lake sturgeon fed with different diets for 4 weeks (Trial I) 1. 1 Means ± SEM (n = 3) with different superscripts within each row are signiﬁcantly (p < 0.05) different by Tukey’s HSD test. 2 Deﬁnition of treatment sees Table 4 for a footnote. Animals 2022, 12, 3128 8 of 15 14 different Figure 1. Average body weight (n = 3) of sturgeon fed different test diets during a 4-week feeding (Trial-I). A, Artemia; A + B, combined feeding of Artemia and commercial diet. A + L, combined feeding of Artemia and lab diet. y = 18.111x2 + 51.444x + 63.508 R² = 0.9999 y = 30.82x2 + 6.2912x + 70.203 R² = 0.9943 y = 16.871x2 + 10.863x + 67.394 R² = 0.9911 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 0 1 2 3 4 Body weight (mg) Feeding duration (week) A A+B A+L Figure 1. Average body weight (n = 3) of sturgeon fed different test diets during a 4-week feeding (Trial-I). A, Artemia; A + B, combined feeding of Artemia and commercial diet. A + L, combined feeding of Artemia and lab diet. 3.1. The Potential of Substituting Artemia with Formulated Feed These results indicate that co-feeding Artemia with the commercial feed should beneﬁt sturgeon larvae weaning to the commercial diet. However, the timing of this transition is important for achieving the best results. Based on this study, weaning off Artemia at the end of 2nd week led to about 50% mortality for combined-fed ﬁsh and over 70% mortality for Artemia-fed ﬁsh. This suggests that complete weaning from Artemia two weeks after exogenous feeding begins is too early. For ﬁsh weaned at 3 or 4 weeks, mortality was low for all treatments before the ﬁsh were completely weaned to the commercial diet (Figure 3a). At the end of the sixth week, morality was the highest for A-R and A4. The mortality of ﬁsh from A3 and AB3 treatments was similar and reached a plateau two weeks after the ﬁsh were weaned to the commercial diet. The lowest mortality was observed with the AB4 feeding, but this was not statistically different from the mortality of AB3. 9 of 15 Animals 2022, 12, 3128 2, x (a) (b) -10.0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 1 2 3 4 Accumulated mortality (%) Feeding duration (week) A-R A2 AB2 A4 a,x a,x a,x a,x a,x a,x a,x a,x a,x a,x b,y b,y a,x a,x c,z b,z 0 100 200 300 400 500 600 700 800 900 1000 1 2 3 4 Body weight (mg) Feeding duration (week) A-R A2 AB2 A4 c,z a,y b,y b,y c,Y a,xy bc,x ab,y a,x a,x a,x a,x a,x a,x a,x a,x (a) (b) ccumulated mortality (a) and average body weight (b) of sturgeon fed with different tegies for four weeks (Trial II). Data are presented as the mean of three replications. hin the same feeding time are compared using letters a, b, and c to indicate significant among feeding treatments determined by Tukey’s HSD test (p < 0.05). Letters x, y, and z e difference of different sampling times within the same feeding treatment. See the Table 3 for the information on feeding treatments. 3.1. The Potential of Substituting Artemia with Formulated Feed -10.0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 1 2 3 4 Accumulated mortality (%) Feeding duration (week) A-R A2 AB2 A4 a,x a,x a,x a,x a,x a,x a,x a,x a,x a,x b,y b,y a,x a,x c,z b,z 0 100 200 300 400 500 600 700 800 900 1000 1 2 3 4 Body weight (mg) Feeding duration (week) A-R A2 AB2 A4 c,z a,y b,y b,y c,Y a,xy bc,x ab,y a,x a,x a,x a,x a,x a,x a,x a,x Figure 2. Accumulated mortality (a) and average body weight (b) of sturgeon fed with different dietary strategies for four weeks (Trial II). Data are presented as the mean of three replications. Means within the same feeding time are compared using letters a, b, and c to indicate signiﬁcant differences among feeding treatments determined by Tukey’s HSD test (p < 0.05). Letters x, y, and z compare the difference of different sampling times within the same feeding treatment. See the footnote of Table 3 for the information on feeding treatments. (a) (b) cumulated mortality (a) and average body weight (b) of sturgeon fed with different gies for four weeks (Trial II). Data are presented as the mean of three replications. n the same feeding time are compared using letters a, b, and c to indicate significant mong feeding treatments determined by Tukey’s HSD test (p < 0.05). Letters x, y, and z difference of different sampling times within the same feeding treatment. See the able 3 for the information on feeding treatments. Figure 2. Accumulated mortality (a) and average body weight (b) of sturgeon fed with different dietary strategies for four weeks (Trial II). Data are presented as the mean of three replications. Means within the same feeding time are compared using letters a, b, and c to indicate signiﬁcant differences among feeding treatments determined by Tukey’s HSD test (p < 0.05). Letters x, y, and z compare the difference of different sampling times within the same feeding treatment. See the footnote of Table 3 for the information on feeding treatments. results indicate that co-feeding Artemia with the commercial feed should geon larvae weaning to the commercial diet. However, the timing of this s important for achieving the best results. Based on this study, weaning off he end of 2nd week led to about 50% mortality for combined-fed fish and over ity for Artemia-fed fish. 3.1. The Potential of Substituting Artemia with Formulated Feed This suggests that complete weaning from Artemia two exogenous feeding begins is too early. h weaned at 3 or 4 weeks, mortality was low for all treatments before the fish etely weaned to the commercial diet (Figure 3a) At the end of the sixth week The differences in body weight among treatments increased with the duration of feeding (Figure 3b). No difference was observed during the ﬁrst two weeks. At the end of 3rd week, the body weight was the highest for AB3 or AB4 ﬁsh but was not statistically different for the A3 or A4 ﬁsh. From weeks 4 to 6, the weight of AB3 and AB4 increased rapidly and reached about triple the weight of A3 and A4. The lowest body weight was observed for the ﬁsh from A-R feeding. In addition, body weight was not different between AB3 and AB4 or A3 and A4 treatments. 10 of 15 Animals 2022, 12, 3128 2, 12, x (a) (b) -10.0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 1 2 3 4 5 6 Accumulated mortality (%) Feeding duration (week) A-R A3 AB3 A4 AB4 a,x ab,x a,x a,x c,z b,y a,x b,y a,xy a,x a,x ab,x c,y a,x a,y a,y c,z c,z a,x c,z b,z ab,z a,z a,x a,x a,x a,x a,x a,x a,x 0 250 500 750 1000 1250 1500 1750 2000 2250 2500 1 2 3 4 5 6 Body weight (mg) Feeding duration (week) A-R A3 AB3 A4 AB4 a,w a,x a,w a,v a,x a,wx a,vw a,vw a,w a,wx b,w ab,wx abw b,x bc,y ab,x a,xy c,x d,x a,y b,y b,y c,y d,y a,z b,z b,z c,z c,z a,v (a) (b) Accumulated mortality (a) and average body weight (b) of sturgeon fed with different trategies for six weeks (Trial II). Data are presented as the mean of three replicates. Means he same feeding time were compared using letters a, b, and c to indicate significant es among feeding treatments determined by Tukey’s HSD test (p < 0.05). Letters x, y, and z the difference of different sampling times within the same feeding treatment. See the of Table 3 for the information on feeding treatments. 3.1. The Potential of Substituting Artemia with Formulated Feed -10.0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 1 2 3 4 5 6 Accumulated mortality (%) Feeding duration (week) A-R A3 AB3 A4 AB4 a,x ab,x a,x a,x c,z b,y a,x b,y a,xy a,x a,x ab,x c,y a,x a,y a,y c,z c,z a,x c,z b,z ab,z a,z a,x a,x a,x a,x a,x a,x a,x 0 250 500 750 1000 1250 1500 1750 2000 2250 2500 1 2 3 4 5 6 Body weight (mg) Feeding duration (week) A-R A3 AB3 A4 AB4 a,w a,x a,w a,v a,x a,wx a,vw a,vw a,w a,wx b,w ab,wx abw b,x bc,y ab,x a,xy c,x d,x a,y b,y b,y c,y d,y a,z b,z b,z c,z c,z a,v Figure 3. Accumulated mortality (a) and average body weight (b) of sturgeon fed with different dietary strategies for six weeks (Trial II). Data are presented as the mean of three replicates. Means within the same feeding time were compared using letters a, b, and c to indicate signiﬁcant differences among feeding treatments determined by Tukey’s HSD test (p < 0.05). Letters x, y, and z compare the difference of different sampling times within the same feeding treatment. See the footnote of Table 3 for the information on feeding treatments. (b) Accumulated mortality (a) and average body weight (b) of sturgeon fed with different ategies for six weeks (Trial II). Data are presented as the mean of three replicates. Means e same feeding time were compared using letters a, b, and c to indicate significant s among feeding treatments determined by Tukey’s HSD test (p < 0.05). Letters x, y, and z he difference of different sampling times within the same feeding treatment. See the f Table 3 for the information on feeding treatments. Figure 3. Accumulated mortality (a) and average body weight (b) of sturgeon fed with different dietary strategies for six weeks (Trial II). Data are presented as the mean of three replicates. Means within the same feeding time were compared using letters a, b, and c to indicate signiﬁcant differences among feeding treatments determined by Tukey’s HSD test (p < 0.05). Letters x, y, and z compare the difference of different sampling times within the same feeding treatment. See the footnote of Table 3 for the information on feeding treatments. differences in body weight among treatments incr Figure 3b) No difference was observed during the fi 3.3. Effect of Different Feedings on Hypoxia Tolerance Figure 3b). No difference was observed during the first two weeks. At the end of , the body weight was the highest for AB3 or AB4 fish but was not statistically for the A3 or A4 fish. From weeks 4 to 6, the weight of AB3 and AB4 increased nd reached about triple the weight of A3 and A4. The lowest body weight was for the fish from A-R feeding. In addition, body weight was not different AB3 and AB4 or A3 and A4 treatments. of Different Feedings on Hypoxia Tolerance The mortality of sturgeon exposed to hypoxia (Trial III) is presented in Figure 4. Our results show that larvae fed exclusively on Artemia before being weaned to the commercial diet, had >80% mortality when the dissolved oxygen was between 3 and 4 mg/L. In contrast, mortality was only 11% in ﬁsh fed with 1/3 Artemia plus 2/3 commercial diet before weaning. The remaining ﬁsh in the two treatments all died when dissolved oxygen was reduced below 3 mg/L. This indicates that larvae co-fed with Artemia had better tolerance to hypoxia. 11 of 15 11 of 15 11 of 15 11 of 15 Animals 2022, 12, 3128 Animals 2022, 12, x Figure 4. Mortality of sturgeon challenged by acute hypoxia after 6-week feeding (Trial III). Data were presented by means of mortality based on survival fish from three replications. Different letters indicated the significant difference between the two dietary treatments, with X and Y comparing the mortality that occurred at 3–4 mg/L dissolved oxygen: x and y comparing the mortality at 2–3 dissolved oxygen. A: sturgeon fed Artemia for 4 weeks before weaning to a commercial diet. AB: sturgeon co-fed Artemia and the commercial diet for 4 weeks before weaning to a commercial diet. 0 20 40 60 80 100 120 >5 4-5 3-4 2-3 Mortality (%) Dissolved oxygen (mg/L) A AB X y x Y Figure 4. Mortality of sturgeon challenged by acute hypoxia after 6-week feeding (Trial III). Data were presented by means of mortality based on survival ﬁsh from three replications. Different letters indicated the signiﬁcant difference between the two dietary treatments, with X and Y comparing the mortality that occurred at 3–4 mg/L dissolved oxygen: x and y comparing the mortality at 2–3 dissolved oxygen. A: sturgeon fed Artemia for 4 weeks before weaning to a commercial diet. 4. Discussion 4. Discussion The results of this study demonstrated that neither the commercial diet nor the lab diet alone could support good survival of larvae at early life stages up to 49 DPH. A combination of the commercial diet (B) and a reduced level of Artemia (50% ratio of the Artemia group) did not cause a negative impact on the survival and the growth of the sturgeon. This finding suggests that the commercial diet can partially replace Artemia to support the nutrients required by the larval fish. In a previous study by Lee et al. [12], lake sturgeon juveniles obtained good growth when fed this lab-made diet, verifying that it meets the nutritional and physical quality requirements for juvenile fish. Low survival and poor growth of the larvae reared solely on the formulated diets assessed in the current study are comparable to previous reports [9,10,22,23]. DiLauro et al. [9] commented that poor acceptance of an artificial diet by the larvae may have been attributed to a lack of imprinting. This may partially explain why the inferior performance of sturgeon larvae fed the lab-made diet compared to the fish fed the commercial diet, which had been used to condition the larvae for one week before the feeding trial. On the other hand, based on the proximate compositions of the three diets, the lab-made diet had lower protein, lipid, and ash contents than the other two diets (Table 2). A much higher n-6 fatty acid was measured in the lab diet than in the other two diets. Thus, the difference in the nutritional composition of the lab-made diet and the commercial diet may explain the poor performance of larvae fed the lab diet. The commercial diet and Artemia were similar in proximate composition but possess different fatty acid profiles. The level of linolenic acid was higher in Artemia, but eicosapentaenoic acid (EPA, 20:5n-3) and docosahexaenoic acid (DHA, 22:6n-3) were lower when compared to the commercial diet. The lab-made diet also had low levels of EPA and DHA compared to the commercial diet. 4. Discussion 4. Discussion Although the nutritional requirement of lake sturgeon has not been investigated, both n-3 and n-6 long- chain unsaturated fatty acids as well as highly unsaturated fatty acids are required by th t i [24] F th i ti ti i t th di t f tt id i t The results of this study demonstrated that neither the commercial diet nor the lab diet alone could support good survival of larvae at early life stages up to 49 DPH. A combination of the commercial diet (B) and a reduced level of Artemia (50% ratio of the Artemia group) did not cause a negative impact on the survival and the growth of the sturgeon. This ﬁnding suggests that the commercial diet can partially replace Artemia to support the nutrients required by the larval ﬁsh. In a previous study by Lee et al. [12], lake sturgeon juveniles obtained good growth when fed this lab-made diet, verifying that it meets the nutritional and physical quality requirements for juvenile ﬁsh. Low survival and poor growth of the larvae reared solely on the formulated diets assessed in the current study are comparable to previous reports [9,10,22,23]. DiLauro et al. [9] commented that poor acceptance of an artiﬁcial diet by the larvae may have been attributed to a lack of imprinting. This may partially explain why the inferior performance of sturgeon larvae fed the lab-made diet compared to the ﬁsh fed the commercial diet, which had been used to condition the larvae for one week before the feeding trial. On the other hand, based on the proximate compositions of the three diets, the lab-made diet had lower protein, lipid, and ash contents than the other two diets (Table 2). A much higher n-6 fatty acid was measured in the lab diet than in the other two diets. Thus, the difference in the nutritional composition of the lab-made diet and the commercial diet may explain the poor performance of larvae fed the lab diet. The commercial diet and Artemia were similar in proximate composition but possess different fatty acid proﬁles. The level of linolenic acid was higher in Artemia, but eicosapentaenoic acid (EPA, 20:5n-3) and docosahexaenoic acid (DHA, 22:6n-3) were lower when compared to the commercial diet. The lab-made diet also had low levels of EPA and DHA compared to the commercial diet. differences in body weight among treatments incr Figure 3b) No difference was observed during the fi 3.3. Effect of Different Feedings on Hypoxia Tolerance AB: sturgeon co-fed Artemia and the commercial diet for 4 weeks before weaning to a commercial diet. 4 Discussion 0 20 40 60 80 100 120 >5 4-5 3-4 2-3 Mortality (%) Dissolved oxygen (mg/L) A AB X y x Y Figure 4. Mortality of sturgeon challenged by acute hypoxia after 6-week feeding (Trial III). Data were presented by means of mortality based on survival fish from three replications. Different letters indicated the significant difference between the two dietary treatments, with X and Y comparing the mortality that occurred at 3–4 mg/L dissolved oxygen: x and y comparing the mortality at 2–3 dissolved oxygen. A: sturgeon fed Artemia for 4 weeks before weaning to a commercial diet. AB: sturgeon co-fed Artemia and the commercial diet for 4 weeks before weaning to a commercial diet. Figure 4. Mortality of sturgeon challenged by acute hypoxia after 6-week feeding (Trial III). Data were presented by means of mortality based on survival ﬁsh from three replications. Different letters indicated the signiﬁcant difference between the two dietary treatments, with X and Y comparing the mortality that occurred at 3–4 mg/L dissolved oxygen: x and y comparing the mortality at 2–3 dissolved oxygen. A: sturgeon fed Artemia for 4 weeks before weaning to a commercial diet. AB: sturgeon co-fed Artemia and the commercial diet for 4 weeks before weaning to a commercial diet. 4. Discussion 4. Discussion The difference in ﬁsh ash content reﬂected the dietary ash levels, with the highest in diet A and the lowest in L. Besides the nutritional quality, physical qualities such as water stability are critical because small particle sizes increase the risk of nutrient leaching [32]. Based on our observations, the lab-made diet had poorer water stability than the commercial diet. Consequently, the nutrient availability of the lab-made diet may have been reduced as a result. This may explain the decreased protein efﬁciency ratio, higher feed conversion ratio, and poorer growth of sturgeon fed the A + L combined diet. The challenges are nutrient leaching especially difﬁcult to overcome in slow-feeding ﬁsh such as sturgeon. p y g g We observed that feeding solely Artemia to the sturgeon can support survival and growth during the ﬁrst 4 weeks of exogenous feeding. Similar ﬁndings were reported by Bauman et al. [10] and Valentine et al. [11] through 14 days and 35 days post-exogenous feeding, respectively. The results of these two studies and our current ﬁndings support the rationale that the nutritional proﬁle of Artemia nauplii can be used as a baseline for developing larval feed for lake sturgeon. However, it is particularly critical to note that the nutritional requirements required to support growth and physiological function may not parallel the requirements for optimizing stress tolerance [33]. As we discussed above, the commercial diet contained a high level of EPA and DHA, which was low or not present in Artemia. Thus, compared to Artemia-feeding larvae, sturgeon larvae under combined feeding might accumulate a higher level of EPA or DHA, which have been found to improve digestive activity, survival, and growth of lake sturgeon, and the tolerance of larval ﬁsh to hypoxia in different ﬁsh species, such as Adriatic sturgeon (Acipenser naccarii), Dove Sole (Solea solea) and European eel (Anguilla anguilla), owning to lowered metabolic rate and oxygen consumption in ﬁsh fed these fatty acids [34–37]. This may partially explain why combined feeding enhanced the tolerance of lake sturgeon to the hypoxia challenge in the current study. In contrast, the Artemia-fed sturgeon might not accumulate substantial EPA and DHA to obtain this beneﬁt because of only being on the commercial diet for two weeks. Further investigation is needed to test this hypothesis and understand mechanisms for lake sturgeon tolerance to hypoxia. 4. Discussion 4. Discussion Although the nutritional requirement of lake sturgeon has not been investigated, both n-3 and n-6 long-chain unsaturated fatty acids as well as highly unsaturated fatty acids are required by other sturgeon species [24]. Further investigation into the dietary fatty acid requirements of lake sturgeon larvae is warranted. A i il t i l l i l l ﬁh b d i th ﬁh l th diff t other sturgeon species [24]. Further investigation into the dietary fatty acid requirements of lake sturgeon larvae is warranted. A similar protein level in larval ﬁsh was observed in the ﬁsh larvae across the different treatments, indicating that protein supply should be similar from the three feeding treat- Animals 2022, 12, 3128 12 of 15 12 of 15 ments. The RNA/DNA ratio, an indicator of protein biosynthesis, has been used as an indicator to determine growth rates in larval ﬁsh [20,25]. It is also a biomarker of nutritional status of ﬁsh in response to feeding [13,26–28]. The similar ratio of RNA/DNA of sturgeon fed the diet A, A + B, or A + L agrees with the results for growth and protein content, which were similar among all treatments. This indicates that the three feeding regimes provided sufﬁcient nutrition to support their growth at this stage. Lipid is one major energy reserve present in a ﬁsh body and provides an important indication of the ﬁsh’s nutritional status. The analysis of lipid content involves a time-consuming process and requires an adequate sample size to achieve reliable results, which is challenging when a limited sample is available. In general, lipid molecules contain a very low number or no nitrogen with carbon as their major element. Thus, C/N ratios in bulk tissue positively correlate with lipid levels. This provides the rationale for applying C/N ratios to predict the lipid content of ﬁsh in previous studies [29–31]. In our previous study, we also observed a positive relationship between lipid content and C/N ratios in lake sturgeon [13]. Thus, in the current study, based on the results of C/N ratios of whole ﬁsh tissue we predict that the lipid content was lower in ﬁsh fed Artemia or Artemia + Lab diet than in Artemia + Commercial diet, suggesting that the feeding A + B resulted in higher energy reserves. 5. Conclusions The current study demonstrates that feeding solely on Artemia can support good survival and growth of lake sturgeon during 28 days of post-exogenous feeding, but co- feeding with the commercial diet beneﬁts weaning to a formulated diet based on growth, survival, and tolerance to hypoxia. Co-feeding of lake sturgeon larvae for 3 or 4 weeks after initial exogenous feeding is recommended before lake sturgeon larvae are weaned to the commercial diet to receive these beneﬁts. Partial substitution of Artemia by the commercial feed did not cause adverse impacts on the growth and survival of the lake sturgeon. Our results indicate that nutrient enrichment should be considered in live feed commonly used for lake sturgeon rearing, including Artemia (lacking eicosapentaenoic acid and docosahexaenoic acid) to enhance ﬁsh health and tolerance to stress. A successful transition to alternative feed is dependent on both feed quality and the capacity of digestive enzymes, which warrants future research. The use of formulated feeds should offer a cost- effective approach to lake sturgeon larviculture, but more research is needed to address the inﬂuences on ﬁsh health, behavior, and stress tolerance, considering that lake sturgeon is cultured to support population conservation efforts. Author Contributions: Conceptualization, D.-F.D., S.L. and F.B.; methodology, D.-F.D., S.L. and S.Z.; formal analysis, S.L., Y.L. and S.Z.; investigation, S.L., Y.L., S.Z. and B.T.E.; data curation, S.L., D.-F.D. and S.Z.; writing—original draft preparation, S.L. and D.-F.D.; writing—D.-F.D. and P.C.B. project administration, D.-F.D.; funding acquisition, D.-F.D. All authors have read and agreed to the published version of the manuscript. Funding: The current study was funded by the University of Wisconsin-Milwaukee (Project 150-25- 3150-PRJ93WQ). Funding: The current study was funded by the University of Wisconsin-Milwaukee (Project 150-25- 3150-PRJ93WQ). Institutional Review Board Statement: The animal study protocol was approved by the Institutional Animal Care and Use Committee of the University of Wisconsin-Milwaukee (protocol 15-16 #60). Institutional Review Board Statement: The animal study protocol was approved by the Institutional Animal Care and Use Committee of the University of Wisconsin-Milwaukee (protocol 15-16 #60). Informed Consent Statement: Not applicable, as this research did not involve humans. Informed Consent Statement: Not applicable, as this research did not involve humans. Data Availability Statement: The data presented in this study are available on request from the corresponding author. Acknowledgments: We acknowledge the great support from the Wisconsin Department of Natural Resources by providing lake sturgeon to this research. 4. Discussion 4. Discussion In the current study, we did not investigate the digestive enzyme activities of the gastrointestinal tract, which play pivotal roles in ﬁsh feeding. The composition and overall activity of digestive enzymes have been observed to change with oncogenic stages and are also inﬂuenced by exogenous feeding [34,38–41]. For example, during Stelate sturgeon (Acipenser stellatus) larval development, the activity of trypsin, chymotrypsin, and amylase was decreased while the activity of pepsin, alkaline phosphatase, and lipase increased, suggesting that these larval ﬁsh become dependent on lipid to meet their metabolic energy requirements but spare protein for maintenance and tissue synthesis, thus appropriate lipid supply from feeding at this stage is necessary to optimize their growth [40]. Furthermore, Animals 2022, 12, 3128 13 of 15 the appearance and activity of digestive enzymes are important factors supporting the diet weaning of larval ﬁsh. It was reported that about 2 weeks post exogenous feeding (the ﬁrst-month post-hatching) in Adriatic sturgeon (Acipenser naccarii), digestive amylase and protease activities reached stabilization with a fully developed digestive structure [39]. For Persian sturgeon (Acipenser persicus), the maturation of their digestion functions was found to be achieved by 19–24 days post-hatching when diet weaning is recommended [38]. Recent ﬁndings by Yoon et al. [34] also discovered that enriched feeding with DHA resulted in increased lipase activity and survival of lake sturgeon during the weaning period. Thus, it is highly likely that dietary EPA and DHA have made a signiﬁcant contribution to enhancing the digestive enzyme activity and thus growth and survival of the lake sturgeon co-fed with the commercial diet in the current study [41]. 5. Conclusions Shaowei Zhai was partially supported by China Agriculture Research System (CARS-46). Yuquan Li was supported by Qingdao Agricultural University, Shandong, China. Patrick C. Blaufuss was supported by the Wisconsin Sea Grant Keillor Fellowship. We also thank Wendy M. Sealey for providing the vitamin and mineral premixes for this study and Peter Shep for assistance in sampling. Conﬂicts of Interest: The authors declare no conﬂict of interest. 14 of 15 14 of 15 Animals 2022, 12, 3128 References 2019, 2 g p g y g Rates for Juvenile Lake Sturgeon (Acipenser fulvescens) Fed a Formulated Dry Diet. Aquac. Nutr. 2019, 25, 1 14. Sorgeloos, P.; Bossuyt, E.; Laviña, E.; Baeza-Mesa, M.; Persoone, G. Decapsulation of Artemia Cysts: A Simple Technique of the Improvement of the Use of Brine Shrimp in Aquaculture. Aquaculture 2019, 12, 311–315. [CrossRef] p p q q 15. Deng, D.F.; Koshio, S.; Yokoyama, S.; Bai, S.C.; Shao, Q.; Cui, Y.; Hung, S.S.O. Effects of Feeding Rate on Growth Performance of White Sturgeon (Acipenser transmontanus) Larvae. Aquaculture 2003, 217, 589–598. [CrossRef] 16. Sealey, W.M.; O’Neill, T.J.; Peach, J.T.; Gaylord, T.G.; Barrows, F.T.; Block, S.S. Reﬁning Inclusion Levels of Grain Distiller’s Dried Yeast in Commercial Type and Plant Based Diets for Juvenile Rainbow Trout, Oncorhynchus mykiss. J. World Aquac. Soc. 2015, 46, 434–444. [CrossRef] 17. Jiang, M.; Zhao, H.; Zai, S.; Shepherd, B.; Wen, H.; Deng, D.F. A Defatted Microalgae Meal (Haematococcus pluvialis) as a Partial Protein Source to Replace Fishmeal for Feeding Juvenile Yellow Perch Perca ﬂavescens. J. Appl. Phycol. 2019, 31, 1197–1205. [CrossRef] ne, E.M.; Ong, T.L. RNA–DNA Ratio and Other Nucleic Acid-based Indicators for Growth and Condition of obiologia 1999, 401, 265–277. [CrossRef] 18. Buckley, L.; Caldarone, E.M.; Ong, T.L. RNA–DNA Ratio and Other Nucleic Acid-based Indicators Marine Fishes. Hydrobiologia 1999, 401, 265–277. [CrossRef] 19. Deng, D.F.; Wang, C.; Lee, S.; Bai, S.; Hung, S.S.O. Feeding Rates Affect Heat Shock Protein Levels in Liver of Larval White Sturgeon (Acipenser transmontanus). Aquaculture 2009, 287, 223–226. [CrossRef] g p q 20. Caldarone, E.M.; Wagner, M.; St. Onge-Burns, J.; Buckley, L.J. Protocol and Guide for Estimating Nucleic Acids in Larval Fish Using a Fluorescence Microplate Reader; Ref. Doc. 01-11; Northeast Fisheries Science Center: Woods Hole, MA, USA, 2001; 22p. 21. Association of Official Analytical Chemists (AOAC). Official Methods of Analysis; Heldrich, K., Ed.; AOAC: 21. Association of Official Analytical Chemists (AOAC). Official Methods of Analysis; Heldrich, K., Ed.; AOAC: Arlington, VA, USA, 2000. 22. Anderson, E.R. Artiﬁcial Propagation of Lake Sturgeon Acipenser fulvescens (Raﬁnesque), Under Hatchery Conditions in Michigan; Fisheries Report 1898; Michigan Department of Natural Resources, Fisheries Division: Ann Arbor, MI, USA, 1984. 22. Anderson, E.R. Artiﬁcial Propagation of Lake Sturgeon Acipenser fulvescens (Raﬁnesque), Under Hatchery Conditions in Michigan; Fisheries Report 1898; Michigan Department of Natural Resources, Fisheries Division: Ann Arbor, MI, USA, 1984. g 23. ´Ceskleba, D.G.; AveLallement, S.; Thuemler, T.F. References [CrossRef] g y Hatchery Rearing Environment Affect Levels of Genetic Diversity in Early Life Stages of Lake Stu Aquaculture 2011, 310, 312–324. [CrossRef] 9. DiLauro, M.N.; Krise, W.F.; Fynn-Aikins, K. Growth and Survival of Lake Sturgeon Larvae Fed Formulated Diets. Prog. Fish Cult. 1998, 60, 293–296. [CrossRef] 10. Bauman, J.M.; Woodward, B.M.; Baker, E.A.; Marsh, T.L.; Scribner, K.T. Effects of Family, Feeding Frequency, and Alternate Food Type on Body Size and Survival of Hatchery-Produced and Wild-Caught Lake sturgeon Larvae. N. Am. J. Aquac. 2016, 78, 136–144. [CrossRef] 11. Valentine, S.A.; Bauman, J.M.; Scriber, K.T. Effects of Alternative Food Types on Body Size and Survival of Hatchery-Reared Lake Sturgeon. N. Am. J. Aquac. 2017, 79, 275–282. [CrossRef] g q 12. Lee, S.; Zhao, H.; Li, Y.; Binkowski, F.P.; Deng, D.F.; Shepherd, B.S.; Hung, S.S.O.; Bai, S.C. Evaluation of Formulated Feed for Juvenile Lake Sturgeon Based on Growth Performance and Nutrient Retention. N. Am. J. Aquac. 2018, 80, 223–236. [CrossRef] h h h d k k l f l d 12. Lee, S.; Zhao, H.; Li, Y.; Binkowski, F.P.; Deng, D.F.; Shepherd, B.S.; Hung, S.S.O.; Bai, S.C. Evaluation of Formulated Feed for Juvenile Lake Sturgeon Based on Growth Performance and Nutrient Retention. N. Am. J. Aquac. 2018, 80, 223–236. [CrossRef] 13 Yang S ; Zhai S W ; Shepherd B S ; Binkowski F P; Hung S S O ; Sealey W M ; Deng D F Determination of Optimal Feeding Juvenile Lake Sturgeon Based on Growth Performance and Nutrient Retention. N. Am. J. Aquac. 2018, 80, 223–236. [CrossRef] 13. Yang, S.; Zhai, S.W.; Shepherd, B.S.; Binkowski, F.P.; Hung, S.S.O.; Sealey, W.M.; Deng, D.F. Determination of Optimal Feeding R f J il L k S (A i f l ) F d F l d D Di A N t 2019 25 1171 1182 [C R f] Juvenile Lake Sturgeon Based on Growth Performance and Nutrient Retention. N. Am. J. Aquac. 2018, 13. Yang, S.; Zhai, S.W.; Shepherd, B.S.; Binkowski, F.P.; Hung, S.S.O.; Sealey, W.M.; Deng, D.F. Determina Rates for Juvenile Lake Sturgeon (Acipenser fulvescens) Fed a Formulated Dry Diet. Aquac. Nutr. 2019, 2 13. Yang, S.; Zhai, S.W.; Shepherd, B.S.; Binkowski, F.P.; Hung, S.S.O.; Sealey, W.M.; Deng, D.F. Determina Rates for Juvenile Lake Sturgeon (Acipenser fulvescens) Fed a Formulated Dry Diet. Aquac. Nutr. References 1. Schneeberger, P.J.; Elliott, R.F.; Jonas, J.L.; Hart, S. Benthivores. In The State of Lake Michigan in 2000; Holey, M.E., Trudeau, T.N., Eds.; Great Lakes Fisheries Commission: Ann Arbor, MI, USA, 2005; ISSN 1090-1051. 2. Michigan Natural Features Inventory. Acipenser fulvescens. Availa Acipenser_fulvescens.pdf (accessed on 1 November 2020). 2. Michigan Natural Features Inventory. Acipenser fulvescens. Available online: https://mnﬁ.anr.msu.edu/abstracts/zoology/ Acipenser_fulvescens.pdf (accessed on 1 November 2020). p ( ) nlan, H.R. Population Status and Demographics of Lake Sturgeon in the Bad and White rivers, Wisconsin. J 019, 10, 442–457. [CrossRef] p p 3. Schloesser, J.T.; Quinlan, H.R. Population Status and Demographics of Lake Sturgeon in the Bad and W Fish Wildl. Manag. 2019, 10, 442–457. [CrossRef] g 4. Schram, S.T.; Lindgren, J.; Evrard, L. Reintroduction of Lake Sturgeon in the St. Louis River, Western Lake Superior. N. Am. J. Fish Manag. 1999, 19, 815–823. [CrossRef] g 5. Haxton, T.; Whelan, G.; Bruch, R. Historical Biomass and Sustainable Harvest of Great Lakes Lake Sturgeon (Acipenser fulvescens Raﬁnesque, 1817). J. Appl. Ichthyol. 2014, 30, 1371–1378. [CrossRef] q ) pp y eterson, D.L.; Vecsei, P.; Jennings, C.A. Ecology and Biology of the Lake Sturgeon: A Synthesis of Curren hreatened North American Acipenseridae. Rev. Fish Biol. Fish. 2007, 17, 59–76. [CrossRef] q pp y 6. Peterson, D.L.; Vecsei, P.; Jennings, C.A. Ecology and Biology of the Lake Sturgeon: A Synthesis of Th t d N th A i A i id R Fi h Bi l Fi h 2007 17 59 76 [C R f] 6. Peterson, D.L.; Vecsei, P.; Jennings, C.A. Ecology and Biology of the Lake Sturgeon: A Synthesis of Current Knowledge of a Threatened North American Acipenseridae. Rev. Fish Biol. Fish. 2007, 17, 59–76. [CrossRef] g gy gy g y Threatened North American Acipenseridae. Rev. Fish Biol. Fish. 2007, 17, 59–76. [CrossRef] 7. Noakes, D.L.G.; Beamish, F.W.H.; Rossiter, A. Conservation Implications of Behaviour and Growth of the Lake Sturgeon, Acipenser fulvescens, in Northern Ontario. Environ. Biol. Fish. 1999, 55, 135–144. [CrossRef] thern Ontario. Environ. Biol. Fish. 1999, 55, 135–144. [C f 8. Crossman, J.A.; Scribner, K.T.; Duong, Y.T.; Davis, C.A.; Forsythe, P.S.; Baker, E.A. Gamete and La 8. Crossman, J.A.; Scribner, K.T.; Duong, Y.T.; Davis, C.A.; Forsythe, P.S.; Baker, E.A. Gamete and Larval Collection Methods and Hatchery Rearing Environment Affect Levels of Genetic Diversity in Early Life Stages of Lake Sturgeon (Acipenser fulvescens). Aquaculture 2011, 310, 312–324. 27. Kono, N.; Tsukamoto, Y.; Zenitani, H. RNA/DNA Ratio for Diagnosis of the Nutritional Condition of Jap japonicus Larvae During the Firstfeeding Stage. Fish. Sci. 2003, 69, 1096–1102. [CrossRef] ukamoto, Y.; Zenitani, H. RNA/DNA Ratio for Diagnosis of the Nutritional Condition of Japanese Anchovy E vae During the Firstfeeding Stage. Fish. Sci. 2003, 69, 1096–1102. [CrossRef] References Rapid Commun. Mass Spectrom. 2015, 29, 2069–2077. [CrossRef] 32. D’Abramo, L.R. Challenges in Developing Successful Formulated Feed for Culture of Larval Fish and Crustaceans. In Avances en Nutrición Acuícola VI. Memorias del VI Simposium Internacional de Nutrición Acuícola. Cancún, Quintana Roo, México, 3–6 September 2002 Cru Suáre L E Ricque Marie D Tapia Sala ar M Gaxiola Cortés M G Simoes N Eds Uni ersidad Autónoma de 32. D’Abramo, L.R. Challenges in Developing Successful Formulated Feed for Culture of Larval Fish and Crustaceans. In Avances en Nutrición Acuícola VI. Memorias del VI Simposium Internacional de Nutrición Acuícola. Cancún, Quintana Roo, México, 3–6 September 2002; Cruz-Suárez, L.E., Ricque-Marie, D., Tapia-Salazar, M., Gaxiola-Cortés, M.G., Simoes, N., Eds.; Universidad Autónoma de Nuevo León: Monterrey, NL, México, 2002. L.E., Ricque-Marie, D., Tapia-Salazar, M., Gaxiola-Cortés, M.G., Simoes, N., Eds.; Universidad Autónoma de rrey, NL, México, 2002. y 33. Tocher, D.R. Fatty Acid Requirements in Ontogeny of Marine and Freshwater Fish. Aquac. Res. 2010, 41, 717–732. [CrossRef] 34 Y G R L l k A B I A A d W G Eff f Di Shif O i D l f M b li R 33. Tocher, D.R. Fatty Acid Requirements in Ontogeny of Marine and Freshwater Fish. Aquac. Res. 2010, 41, 717–732. [CrossRef] 34. Yoon, G.R.; Laluk, A.; Bouyoucos, I.A.; Anderson, W.G. Effects of Dietary Shifts on Ontogenetic Development of Metabolic Rates in Age 0 Lake Sturgeon (Acipenser fulvescens). Physiol. Biochem. Zool. 2022, 95, 135–151. [CrossRef] y q g y q 34. Yoon, G.R.; Laluk, A.; Bouyoucos, I.A.; Anderson, W.G. Effects of Dietary Shifts on Ontogenetic Development of Metabolic Rates in Age 0 Lake Sturgeon (Acipenser fulvescens). Physiol. Biochem. Zool. 2022, 95, 135–151. [CrossRef] 35. McKenzie, D.; Piraccini, G.; Papini, N. Oxygen Consumption and Ventilatory Reﬂex Responses are in Sturgeon. Fish. Physiol. Biochem. 1997, 16, 365–379. [CrossRef] cini, G.; Papini, N. Oxygen Consumption and Ventilatory Reﬂex Responses are Inﬂuenced by Dietary Lipids hysiol. Biochem. 1997, 16, 365–379. [CrossRef] 36. McKenzie, D.J.; Lund, L.; Pedersen, P.B. Essential Fatty Acids Inﬂuence Metabolic Rate and Tolerance of Hypoxia in Dover Sole (Solea solea) Larvae and Juveniles. Mar. Biol. 2008, 154, 1041–1051. [CrossRef] 37. McKenzie, D.J.; Piraccini, G.; Piccolella, M.; Steffensen, J.F.; Bolis, C.L.; Taylor, E.W. Effects of Dietary Fatty Acid Composition on Metabolic Rate and Responses to Hypoxia in the European Eel, Anguilla anguilla. Fish. Physiol. Biochem. 2000, 22, 281–296. [CrossRef] 38. References Artiﬁcial Spawning and Rearing of Lake Sturgeon, Acipenser fulvescens, in Wild Rose State Fish Hatchery, Wisconsin, 1982-1983. Environ. Biol. Fish. 1985, 14, 79–85. [CrossRef] 23. ´Ceskleba, D.G.; AveLallement, S.; Thuemler, T.F. Artiﬁcial Spawning and Rearing of Lake Sturgeon, A Rose State Fish Hatchery, Wisconsin, 1982-1983. Environ. Biol. Fish. 1985, 14, 79–85. [CrossRef] y 24. Hung, S.S.O. Recent Advances in Sturgeon Nutrition. Anim. Nutr. 2017, 3, 191–204. [CrossRef] 25. Clemmesen, C. Importance and Limits of RNA/DNA Ratios as a Measure of Nutritional Condition in Fish Larvae. In Survival Strategies in Early Life Stages of Marine Resources: Proceedings of the International Workshop, Tsukuba, Japan, 11–15 November 1996; 25. Clemmesen, C. Importance and Limits of RNA/DNA Ratios as a Measure of Nutritional Condition in Fish Larvae. In Survival Strategies in Early Life Stages of Marine Resources: Proceedings of the International Workshop, Tsukuba, Japan, 11–15 November 1996; Watanabe, Y., Yamashita, Y.A., Oozeki, Y., Eds.; Balkema: Rotterdam, The Netherlands, 1996; pp. 67–82. ISBN 90-5410-637-9. 26. Chung, K.S.; Deb, M.S.; Donaldson, E.M. RNA:DNA Ratio as Physiological Condition of Rainbow Trout Fry Fasted and Fed. Ital. Strategies in Early Life Stages of Marine Resources: Proceedings of the International Workshop, Tsukuba, Japan, 11–15 November 1996; Watanabe, Y., Yamashita, Y.A., Oozeki, Y., Eds.; Balkema: Rotterdam, The Netherlands, 1996; pp. 67–82. ISBN 90-5410-637-9. 26. Chung, K.S.; Deb, M.S.; Donaldson, E.M. RNA:DNA Ratio as Physiological Condition of Rainbow Trout Fry Fasted and Fed. Ital. J. Zool. 1998, 65, 517–519. [CrossRef] 27. Kono, N.; Tsukamoto, Y.; Zenitani, H. RNA/DNA Ratio for Diagnosis of the Nutritional Condition of japonicus Larvae During the Firstfeeding Stage. Fish. Sci. 2003, 69, 1096–1102. [CrossRef] 15 of 15 15 of 15 Animals 2022, 12, 3128 28. Zehra, S.; Khan, M.A. Dietary Leucine Requirement of Fingerling Catla catla (Hamilton) Based on Growth, Feed Conversion Ratio, RNA/DNA Ratio, Leucine Gain, Blood Indices and Carcass Composition. Aquac. Int. 2014, 23, 577–595. [CrossRef] p q 29. Barnes, C.; Sweeting, C.J.; Jennings, S.; Barry, J.T.; Polunin, N.V.C. Effect of Temperature and Ration Size on Carbon and Nitrogen Stable Isotope Trophic Fractionation. Funct. Ecol. 2007, 21, 356–362. [CrossRef] p p 30. Abrantes, K.G.; Semmens, J.M.; Lyle, J.M.; Peter, D. Normalisation Models for Accounting for Fat Content in Stable Isotope Measurements in Salmonid Muscle Tissue. Mar. Biol. 2012, 159, 57–64. [CrossRef] 31. Hoffman, J.; Sierszen, M.; Cotter, A. Fish Tissue Lipid-C:N Relationships for Correcting ä13C Values and Estimating Lipid Content in Aquatic Food Web Studies. References Babaei, S.S.; Abedian-Kenari, A.; Nazari, R.; Gisbert, E. Developmental Changes of Digestive Enzymes in Persian sturgeon (Acipenser persicus) During Larval Ontogeny. Aquaculture 2011, 318, 138–144. [CrossRef] g g y 39. Sanz, A.; Llorente, J.I.; Furné, M.; Carmona, M.R.; Domezain, A.; Hidalgo, M.C. Digestive Enzymes During Ontogeny of the Sturgeon Acipenser naccarii: Intestine and Pancreas Development. J. Appl. Ichthyol. 2011, 27, 1139–1146. [CrossRef] g p p pp y 40. Ghasemi, N.; Imani, A.; Noori, F.; Shahrooz, R. Ontogeny of Digestive Tract of Stellate Sturgeon (Acipenser stellatus) from Hatching to Juvenile Stage: Digestive Enzymes Activity, Stomach and Proximal Intestine. Aquaculture 2020, 519, 734–751. [CrossRef] 41. Gwangseok, R.Y.; Amjad, H.; Weinrauch, A.M.; Laluk, A.; Suh, M.; Anderson, W.G. Long-Term Effects of EPA and DHA Enriched Diets on Digestive Enzyme Activity, Aerobic Scope, Growth and Survival in Age-0 Lake Sturgeon (Acipenser fulvescens). Aquaculture 2022, 552, 737972.
https://openalex.org/W2765290536	https://russjcardiol.elpub.ru/jour/article/download/946/2163	Russian	null	SINO-ATRIAL NODE TELOCYTES IN THE HUMAN HEART. MORPHOLOGICAL EVIDENCE FOR PACEMAKING AND CONDUCTION PROPERTIES	Rossijskij kardiologičeskij žurnal	2,017	cc-by	5,339	Российский кардиологический журнал № 9 (149) \| 2017 Российский кардиологический журнал № 9 (149) \| 2017 Ключевые слова: телоциты, синусный узел человека, иммуногистохимиче- ское исследование, электронная и конфокальная микроскопия. Ключевые слова: телоциты, синусный узел человека, иммуногистохимиче- ское исследование, электронная и конфокальная микроскопия. Синусный узел (СУ) состоит преимущественно из пейсмекерных Р-клеток, переходных Т-клеток, а также из Пуркинье-подобных клеток, располагающихся по периферии узла. Р-клетки, особенно в центре СУ, окружены плотной грубово- локнистой фиброзной тканью и не контактируют с Т-клетками. Мы выдвигаем гипотезу, что в СУ есть телоциты, которые могут играть роль в проведении электрического импульса от пейсмекерных клеток к рабочему миокарду. 1ФГБУ Северо-Западный федеральный медицинский исследовательский центр им. В. А. Алмазова Минздрава России, Санкт-Петербург; 2ФГБУ НИИ гриппа, Санкт-Петербург; 3НИИ акушерства, гинекологии и репродуктологии имени Д. О. Отта, Санкт-Петербург, Россия. Цель. Морфологический анализ телоцитов в составе СУ. Материал и методы. Гистологическое и иммуногистохимическое исследова- ние СУ было проведено в 10 аутопсийных случаях. Метод двойной метки использовался с комбинациями первичных антител к HCN4/connexin43 и CD34/connexin43. Конфокальная лазерная микроскопия с 4 коктейлями антител к CD34/S100, CD34/connexin43, SMA/connexin43, S100/vimentin про- водилась в 3 из 10 случаев. Дополнительные тканевые образцы СУ от 3 других пациентов подверглись электронной микроскопии и электронной иммуноци- тохимии с HCN4. Митрофанова Л. Б.* — д. м.н., зав. НИЛ патоморфологии, Горшков А. Н. — к. б.н., зав. лаборатории внутриклеточного сигналинга и транспорта, с. н.с. НИЛ патоморфологии, Коновалов П. В. — к. м.н., врач-патологоанатом патоло- гоанатомического отделения клиники, Крылова Ю. С. — к. м.н., с. н.с лаборато- рии клеточной биологии отдела патоморфологии, Полякова В. О. — д. б.н., профессор, зав. лабораторией клеточной морфологии отдела патоморфоло- гии, Кветной И. М. — д. м.н., профессор, зав. отдела патоморфологии. Результаты. Во всех исследованных СУ были обнаружены клетки с иммуно- фенотипом телоцитов. В центре узла их было в 2 раза больше, чем на перифе- рии (20,3±4,8 против 10,8±4,4 клеток при х400). Телоциты имели тесные кон- такты с Р-клетками, сократительным миокардом, сосудами и экспрессиро- вали HCN4. Их ультраструктурные характеристики полностью соответствовали телоцитам, обнаруженным в других органах и других отделах сердца. Автор, ответственный за переписку (Corresponding author): Автор, ответственный за переписку (Corresponding author): *Автор, ответственный за переписку (Corresponding author): lubamitr@yandex.ru ИГХ — иммуногистохимия, КЛСМ — конфокальная лазерная сканирующая микроскопия, ПСС — проводящая система сердца, СУ — синусный узел, HCN — hyperpolarization-activated cyclic nucleotide-gated channels. Заключение. Был обнаружен еще один тип клеток, способных проводить и, возможно, генерировать электрический импульс в СУ. На наш взгляд, элект рическая гетерогенность СУ может быть объяснена присутствием телоцитов. Ключевые слова: телоциты, синусный узел человека, иммуногистохимиче- ское исследование, электронная и конфокальная микроскопия. Рукопись получена 11.11.2016 Рецензия получена 30.11.2016 Принята к публикации 02.12.2016 Рукопись получена 11.11.2016 Рецензия получена 30.11.2016 Принята к публикации 02.12.2016 Российский кардиологический журнал 2017, 9 (149): 42–49 http://dx.doi.org/10.15829/1560-4071-2017-9-42-49 Российский кардиологический журнал 2017, 9 (149): 42–49 http://dx.doi.org/10.15829/1560-4071-2017-9-42-49 Российский кардиологический журнал 2017, 9 (149): 42–49 http://dx.doi.org/10.15829/1560-4071-2017-9-42-49 Mitrofanova L. B. 1, Gorshkov А. N. 1,2, Konovalov P. V. 1, Krylova Yu. S. 3, Polyakova V. О. 3, Kvetnoy I. М. 3 The sinus node (SN) is built predominantly from pacemaking P-cells, transient T-cells, and Purkinje-like cells located on the periphery of the node. P-cells, especially in the center of SN, are surrounded by dense hardfiber fibrous tissue and do not contact with T-cells. We come up with a hypothesis that in the SN there are telocytes that may play role in electrical impulse conduction from pacemaking cells to contracting myocardium. 10,8±4,4 cells in х400). Telocytes had dense contacts with P-cells, contracting myocardium, vessels, and expressed HCN4. Their ultrastructural characteristics completely resembled telocytes that are found in other organs and other heart tissues. Conclusion. Another type of cells was found, able to conduct and, probably, generate electrical impulse in the SN. In our opinion, electrical heterogeneity of the SN might be explained by the presence of telocytes. Conclusion. Another type of cells was found, able to conduct and, probably, generate electrical impulse in the SN. In our opinion, electrical heterogeneity of the SN might be explained by the presence of telocytes. Aim. Morphological analysis of telocytes in SN. Material and methods. Histological and immune-histochemical study of SN done on 10 autopsies. The double trace method was used, with combinations of primary antibodies to HCN4/connexin43 and CD34/connexin43. Confocal laser microscopy with 4 cocktails of antibodies to CD34/S100, CD34/connexin43, SMA/connexin43, S100/vimentin was done in 3 among 10 cases. Additional tissue specimens from SN of 3 other patients underwent electronic microscopy and immune cytochemistry analysis with HCN4. http://dx.doi.org/10.15829/1560-4071-2017-9-42-49 http://dx.doi.org/10.15829/1560-4071-2017-9-42-49 Key words: telocytes, sinus node of the human, immune histochemistry, electronic and confocal microscopy. 1Federal Almazov North-West Medical Research Centre of the Ministry of Health, Saint-Petersburg; 2Research Institute of Influenza, Saint-Petersburg; 3Ott’s SRI of Obstetrics, Gynecology and Reproductology, Saint-Petersburg, Russia. Results. In all studied SN, there were cells with immune phenotype of telocytes. In the center of the node, their number was 2 times more than in periphery (20,3±4,8 versus абельности. Мало известно о человеческом СУ, например, о последовательности его активации, экс- прессии ионных каналов и региональных различиях [1]. Последнее имеет большое значение для хирургов, Несмотря на то, что сино-атриальный, или синус- ный, узел (СУ) был описан Keith A. и Flack М. в начале прошлого века, существует много разногла- сий в его топографии и строении из-за большой вари- 42 42 ОРИГИНАЛЬНЫЕ СТАТЬИ карде желудочков и предсердий и других органах, обладают пейсмекерной активностью [8]. Mitrofanova L. B. 1, Gorshkov А. N. 1,2, Konovalov P. V. 1, Krylova Yu. S. 3, Polyakova V. О. 3, Kvetnoy I. М. 3 В телоци- тах желудка кошки найдены Hyperpolarization- activated cyclic nucleotide-gated channels (HCN) [9], присутствующие в Р-клетках СУ. Это интегральные белки катионных каналов мембран клеток сердца, центральной и периферической нервной системы и фоторецепторов сетчатки глаза. HCN-каналы обо- значают как “каналы-водители ритма”, поскольку они участвуют в генерации ритмической активности клетками сердца и головного мозга. так как точное расположение узла и его артерии невозможно предсказать безошибочно. В любом слу- чае вся область соединения верхней полой вены с правым предсердием должна рассматриваться как зона риска при оперативных вмешательствах [2]. В большинстве случаев СУ находится субэпикар- диально, представлен плотной фиброзной тканью, состоящей из коллагеновых и эластических волокон, среди которых “замурованы” мелкие округлые свет- лые Р-клетки — водители ритма размером 5-10 мкм с небольшим количеством миофибрилл, миофила- ментов, случайно расположенными митохондриями, слабо развитым саркоплазматическим ретикулумом и лизосомами. Р-клетки образуют небольшие группы, которые окружены базальной мембраной, а каждая клетка — плазматической мембраной [3]. Связь между клетками осуществляется через соприкасаю- щиеся клеточные мембраны и лишь отчасти — через десмосомы. В СУ описывают также переходные Т-клетки с контактами конец-в-конец, конец-в-бок, бок-в-бок, с расположенными под углом большим количеством миофибрилл, чем в Р-клетках и разме- ром от 12 до 16 мкм, длиной от 40 до 250 мкм, а также крупные бледные Пуркинье-подобные клетки с мень- шим количеством миофибрилл и большим содержа- нием митохондрий диаметром от 20 до 50 мкм [4]. Большую часть клеточного состава узла составляют клетки-пейсмекеры (Р-клетки), меньшую — крупные клетки Пуркинье-подобного типа [5]. Последние находятся по периферии СУ. P-клетки, T-клетки и Пуркинье-подобные клетки являются специализи- рованными формами кардиомиоцитов в составе узла. Кроме того, в СУ описывают фибробласты, макро- фаги, тучные клетки, перициты, шванновские клетки, кровеносные сосуды и нервные волокна. Крупные пучки нервных волокон и ганглиев располагаются вокруг узла, в основном, в эпикарде. Считается, что генерируемый Р-клетками электрический импульс передается на рабочий миокард через Т-клетки [6]. На наш взгляд, интересным является факт, что клетки-пейсмекеры, особенно в центральной зоне СУ, разделены друг от друга грубой фиброзной тканью и не контактируют с Т-клетками. Остается непонят- ным, как же осуществляется проведение электриче- ского импульса. Таким образом, остаются вопросы: 1) как прово- дится электрический импульс от Р-клеток к сократи- тельному миокарду правого предсердия, 2) все ли мы знаем о клеточном составе СУ? Мы выдвигаем гипотезу о том, что в СУ находятся телоциты, которые, возможно, принимают участие в проведении электрического импульса к сократи- тельному миокарду. Цель исследования: морфологический анализ телоцитов в составе СУ. Таблица 1 Табли Клиническая характеристика пациентов, морфологические данные и методы исследования Таблица 1 Клиническая характеристика пациентов, морфологические данные и методы исследования № Пол Возраст Заболевание Причина смерти Масса сердца (г) Средняя длина и ширина СУ на гистологическом срезе (см) Методы исследования 1 М 65 ДКМП ХСН 563 1,0 x 0,2 Г, ИГХ, КЛСМ 2 М 68 ИБС ОИМ 463 1,2 x 0,2 Г, ИГХ, КЛСМ 3 Ж 28 Некомпактный миокард ЛЖ ТЭЛА 587 1,2 x 0,2 Г, ИГХ, КЛСМ 4 М 41 Аортальный стеноз ХСН 764 0,8 x 0,5 Г, ИГХ 5 М 55 ДКМП ХСН 678 0,8 x 0,1 Г, ИГХ 6 Ж 19 Лейкоз Пневмония 320 1,0 x 0,4 Г, ИГХ 7 М 63 Аортальный стеноз ХСН 704 0,7 x 0,3 Г, ИГХ 8 Ж 48 Рак матки Интоксикация 356 1,5 x 0,5 Г, ИГХ 9 М 69 ИБС ОИМ 438 1,2 x 0,3 Г, ИГХ 10 Ж 72 ИБС ОИМ 479 1,0 x 0,7 Г, ИГХ 11 М 65 ИБС ОИМ 478 1,0 x 0,2 Г, ЭМ 12 М 46 Лейкоз Пневмония 398 1,2 x 0,5 Г, ЭМ 13 М 49 Лейкоз Пневмония 378 1,0 x 0,3 Г, ЭМ Сокращения: Г — гистологическое исследование, ИГХ — иммуногистохимическое исследование, КЛСМ — конфокальная лазерная сканирующая микроскопия, ЭМ — электронная микроскопия, ДКМП — дилатационная кардиомиопатия, ЛЖ — левый желудочек, ОИМ — острый инфаркт миокарда, ТЭЛА — тромбоэмболия легочной артерии. Клиническая характеристика пациентов, морфологические данные и методы исследования Сокращения: Г — гистологическое исследование, ИГХ — иммуногистохимическое исследование, КЛСМ — конфокальная лазерная сканирующая микроскопия, ЭМ — электронная микроскопия, ДКМП — дилатационная кардиомиопатия, ЛЖ — левый желудочек, ОИМ — острый инфаркт миокарда, ТЭЛА — тромбоэмболия легочной артерии. объекты были дегидратированы в серии растворов этанола возрастающей концентрации, пропитаны ацетоном и заключены в эпоксидную смолу Эпон. Полутонкие срезы (1-2 мкм) были окрашены толуи- диновым синим для выбора зон СУ. Затем из блоков, содержащих клетки проводящей системы, на ультра- микротоме Leica UC7 были получены ультратонкие срезы толщиной 50-70 нм. Срезы были помещены на медные сетки и отконтрастированы в спиртовом растворе уранил-ацетата и водном растворе цитрата свинца. Электронномикроскопическое исследование срезов было выполнено на микроскопе Libra JEM 1011 (JEOL, Japan). морфометрическое исследование размеров узла и его клеток с помощью Leica LAS Image Analysis System (LeicaQWin Plus v3, Leica Microsystems IS, Cambridge, UK). Конфокальная лазерная сканирующая микроскопия. Конфокальная лазерная сканирующая микроскопия (КЛСМ) проводилась с 4 коктейлями первичных антител: CD34/S100, CD34/connexin43, SMA/ connexin43, vimentin/S100. Материал и методы Все аутопсии выполнялись в СЗФМИЦ имени В. А. Алмазова в соответствии с принципами Хель- синкской декларации и были одобрены Этическим комитетом Центра. Макроскопическое, гистологиче- ское и иммуногистохимическое исследование СУ выполнялось у 10 умерших от сердечно-сосудистых, онкологических и гематологических заболеваний пациентов (табл. 1), конфокальная микроскопия — у 3 из них. Дополнительные образцы СУ для элек- тронной микроскопии были получены от 3 других больных. Образцы забирались в течение 15-30 мин после смерти. Таким образом, исследовалось 13 СУ. Макроскопическое исследование СУ. Для исследо- вания СУ мы иссекали фрагмент передней стенки правого предсердия с правым ушком, 1 см верхней полой вены и crista terminalis, в центре которого нахо- дилась пограничная борозда (рис. 1). Затем перпен- дикулярно пограничной борозде, от правого ушка к верхней полой вене с шагом в 3 мм скальпелем выполнялись серийные срезы (рис. 2). Гистологическое и иммуногистохимическое исследо- вание. Парафиновые срезы СУ окрашивались гема- токсилином и эозином, по ван Гизону и трихромом Массона. Выполнялись традиционное иммуногисто- химическое исследование с одним первичным анти- телом к HCN4 и метод двойной метки с 2 первич- ными антителами: коктейль CD34/HCN4 и CD34/ connexin43. При этом использовались кроличье поли- клональное антитело к HCN4 (Alamone Labs, Jerusalem, Israel), мышиное моноклональное анти- тело к CD34 (clone QBEnd-10, DAKO, Denmark), кро- личье поликлональное антитело к connexin43 (Diagnostic Biosystems, USA). Затем выполнялось Телоциты — уникальный тип интерстициальных клеток со специфическими отростками — телоподи- ями, и дилатированными сегментами — подомами. Они имеют одновременно иммунофенотип интер- стициальных, эндотелиальных, гладкомышечных, нервных, тучных и гемопоэтических стволовых кле- ток, экспрессируют CD117, vimentin, CD34, SMA, S100, NSE, а также connexin43 — белок щелевых меж- клеточных контактов (gap junctions) [7]. Телоциты найдены в матке, маточных трубах, кишечнике, молочной железе, поджелудочной железе, коже, мио- 43 43 Российский кардиологический журнал № 9 (149) \| 2017 Таблица 1 Использовались вышеука- занные для иммуногистохимического исследования антитела, а также кроличье поликлональное антитело к S100 (DAKO, Denmark), мышиное моноклональное антитело к SMA (clone IA4, DAKO, Denmark), мыши- ное моноклональное антитело к виментину (clone V9, DAKO, Denmark). Для электронной иммуноцитохимии фрагменты СУ фиксировались в 4% растворе параформальдегида в фосфатном буфере с добавлением 0,2% глютараль- дегида в течение 1 часа. Далее объекты дегидратиро- вали в серии растворов этанола возрастающей кон- центрации и заключали в акриловую смолу LRWhite (Sigma inc.). Полимеризация смолы проходила в закрытых желатиновых капсулах при температуре +52º С. На ультрамикротоме Leica UC7 были полу- чены ультратонкие срезы изучаемого материала тол- щиной 50-70 нм. Срезы были собраны на никелевые сетки для электронной микроскопии. С целью бло- кирования неспецифического связывания антител срезы на сетках обрабатывали 1% раствором бычьего сывороточного альбумина (Sigma Inc.) в фосфатном буфере в течение 15 мин при комнатной температуре. HCN4 на ультратонких срезах выявляли непрямым методом. В качестве первичного было использовано кроличье поликлональное антитело HCN4 (Alamone Labs, Israel). Сеточки со срезами инкубировали в рас- В качестве вторичных антител применялись Alexa Fluor 647® goat anti-mouse (Abcam, UK), Alexa Fluor 488® goat anti-rabbit (Abcam, UK). После промывания срезы контрастировались с DAPI (appliChem). В результате мышиные антитела имели красную флюоресценцию, кроличьи — зеленую, двойное (одновременное) окрашивание антителами — оран- жево-желтую флюоресценцию, а ядра клеток окра- шивались в синий цвет. КЛСМ проводилась на Olympus FV1000D (Япония). Электронная микроскопия. Для электронной микроскопии образцы СУ фиксировались в 2,5%-ом растворе глютарового альдегида на 0,1 М фосфатном буфере в течение 1 ч при комнатной температуре, после чего промывались в трёх сменах фосфатного буфера. Далее выполнялась фиксация кусочков в 1% растворе тетроксида осмия на том же буфере, при той же температуре в течение 1 ч. После фиксации 44 44 ОРИГИНАЛЬНЫЕ СТАТЬИ Рис. 2. Серийные срезы передней стенки правого предсердия с синусным узлом. Зона синусного узла с артерией обведена кружком. Рис. 1. Фрагмент передней стенки правого предсердия с правым ушком и верхней полой веной. Пограничная борозда указана отрезком. Рис. 2. Серийные срезы передней стенки правого предсердия с синусным узлом. Зона синусного узла с артерией обведена кружком. Рис. 1. Фрагмент передней стенки правого предсердия с правым ушком и верхней полой веной. Пограничная борозда указана отрезком. Рис. 4. Иммуногистохимическое исследование. HCN4 в клетках проводящей системы сердца (коричневое окрашивание). Х200. Рис. 3. Синусный узел. Специализированные клетки-пейсмекеры прово- дящей системы сердца “замурованы” в грубую фиброзную ткань (указаны стрелками; центральная артерия указана звездочкой; трихром Массона, х100. Рис. 3. Синусный узел. Таблица 1 Наблюдается коэкспрессия виментина и S100 на отростках телоцитов (оранжевое окрашивание в центре); матово- зеленая автофлюоресценция тканей; х40 Рис. 8. Электронная иммуноцитохимия СУ. Скопления HCN4 преимущест- венно на мембране телоподии телоцита (выделен зеленым цветом) в виде кластеров коллоидного золота (красные стрелки). Б А В Рис. 7 (А, Б, В). Ультраструктура телоцитов в составе СУ. А — компактная ядросодержащая область телоцита и отходящая от нее телоподия; локальное утолщение (подома) в составе телоподии; локализация телоподии вблизи от кровеносного капилляра и P-клетки. Б — проникновение телоцита под базальную мембрану P-клетки (БМ, белые стрелки). Виден межклеточный кон- такт, в составе которого выявляются электронно-плотные структуры, соединя- ющие мембраны телоподии и P-клетки. Телоциты выделены зеленым цветом, межклеточные контакты указаны красными стрелками. В — межклеточный контакт телоподии с Р-клеткой в виде электронно-плотных палочковидных структур. Сокращения: Я — ядро телоцита, Tп — телоподии, П — подома, PК — P-клетка, БМ — базальная мембрана Р-клетки, Mф — миофиламенты. А Рис. 8. Электронная иммуноцитохимия СУ. Скопления HCN4 преимущест- венно на мембране телоподии телоцита (выделен зеленым цветом) в виде кластеров коллоидного золота (красные стрелки). Б В Рис. 7 (А, Б, В). Ультраструктура телоцитов в составе СУ. А — компактная ядросодержащая область телоцита и отходящая от нее телоподия; локальное утолщение (подома) в составе телоподии; локализация телоподии вблизи от кровеносного капилляра и P-клетки. Б — проникновение телоцита под базальную мембрану P-клетки (БМ, белые стрелки). Виден межклеточный кон- такт, в составе которого выявляются электронно-плотные структуры, соединя- ющие мембраны телоподии и P-клетки. Телоциты выделены зеленым цветом, межклеточные контакты указаны красными стрелками. В — межклеточный контакт телоподии с Р-клеткой в виде электронно-плотных палочковидных структур. Сокращения: Я — ядро телоцита, Tп — телоподии, П — подома, PК — P-клетка, БМ — базальная мембрана Р-клетки, Mф — миофиламенты. А Б А А В Б Рис. 8. Электронная иммуноцитохимия СУ. Скопления HCN4 преимущест- венно на мембране телоподии телоцита (выделен зеленым цветом) в виде кластеров коллоидного золота (красные стрелки). Рис. 7 (А, Б, В). Ультраструктура телоцитов в составе СУ. А — компактная ядросодержащая область телоцита и отходящая от нее телоподия; локальное утолщение (подома) в составе телоподии; локализация телоподии вблизи от кровеносного капилляра и P-клетки. Б — проникновение телоцита под базальную мембрану P-клетки (БМ, белые стрелки). Виден межклеточный кон- такт, в составе которого выявляются электронно-плотные структуры, соединя- ющие мембраны телоподии и P-клетки. Телоциты выделены зеленым цветом, межклеточные контакты указаны красными стрелками. В — межклеточный контакт телоподии с Р-клеткой в виде электронно-плотных палочковидных структур. Рис. 8. Электронная иммуноцитохимия СУ. Таблица 1 Специализированные клетки-пейсмекеры прово- дящей системы сердца “замурованы” в грубую фиброзную ткань (указаны стрелками; центральная артерия указана звездочкой; трихром Массона, х100. Рис. 4. Иммуногистохимическое исследование. HCN4 в клетках проводящей системы сердца (коричневое окрашивание). Х200. Рис. 5. Иммуногистохимическое исследование. Двойная метка connexin43 и CD34. Клетка с иммунофенотипом телоцита и телоподиями (бордовое окра- шивание) рядом с клеткой-пейсмекером (указана стрелкой). Х1000. творе указанных антител на фосфатном буфере (1:100) в течение 1 часа и отмывали в 0,05% растворе Твина- 20 в фосфатном буфере. В качестве вторичных анти- тел использовали антитела козы против иммуногло- булинов кролика, конъюгированные с коллоидным золотом диаметром 10 нм (Sigma; разведение 1:100, инкубация в течение 1 часа). После проведения иммунной реакции сетки со срезами были откон трастированы в водном растворе уранил-ацетата и водном растворе цитрата свинца. Электронно- микроскопическое исследование срезов выполня- лось на том же микроскопе JEOL JEM 1011. Цифро- вые электронные микрофотографии были получены с использованием камеры Morada (Digital Imaging Solutions Inc.). творе указанных антител на фосфатном буфере (1:100) в течение 1 часа и отмывали в 0,05% растворе Твина- 20 в фосфатном буфере. В качестве вторичных анти- тел использовали антитела козы против иммуногло- булинов кролика, конъюгированные с коллоидным золотом диаметром 10 нм (Sigma; разведение 1:100, инкубация в течение 1 часа). После проведения иммунной реакции сетки со срезами были откон трастированы в водном растворе уранил-ацетата и водном растворе цитрата свинца. Электронно- микроскопическое исследование срезов выполня- лось на том же микроскопе JEOL JEM 1011. Цифро- вые электронные микрофотографии были получены с использованием камеры Morada (Digital Imaging Solutions Inc.). Статистический анализ. Статистическая обработка проводилась с применением программы “Statistica 10” (StatSoft, USA). Для оценки количественных Рис. 5. Иммуногистохимическое исследование. Двойная метка connexin43 и CD34. Клетка с иммунофенотипом телоцита и телоподиями (бордовое окра- шивание) рядом с клеткой-пейсмекером (указана стрелкой). Х1000. 45 45 Российский кардиологический журнал № 9 (149) \| 2017 А Б Рис. 6 (А, Б). А — двойная иммунофлуоресценция CD34 (красная флуоресценция эндотелия сосудов) и S100 (ярко-зелёная флуоресценция) в СУ. Наблюдается коэкспрессия CD34 и S100 на отростках телоцитов (оранжевое окрашивание); х60. Б — двойная иммунофлуоресценция виментина (красная флуоресценция) и S100 (ярко-зелёная флуоресценция) в СУ. Наблюдается коэкспрессия виментина и S100 на отростках телоцитов (оранжевое окрашивание в центре); матово- зеленая автофлюоресценция тканей; х40. А Рис. 6 (А, Б). А — двойная иммунофлуоресценция CD34 (красная флуоресценция эндотелия сосудов) и S100 (ярко-зелёная флуоресценция) в СУ. Наблюдается коэкспрессия CD34 и S100 на отростках телоцитов (оранжевое окрашивание); х60. Б — двойная иммунофлуоресценция виментина (красная флуоресценция) и S100 (ярко-зелёная флуоресценция) в СУ. Результаты Световая микроскопия. Во всех 13 случаях был обнаружен СУ, средние размеры которого на гисто- логических срезах составили 1,0±0,2 х 0,3±0,2 см. Он был представлен округлыми Р-клетками с бледной цитоплазмой, диаметром 3,6-8,2 мкм, в среднем 4,6±3,1 мкм, которые располагались по 1-2 или груп- пами, были “замурованы” в плотную фиброзную ткань и на большем протяжении не имели контактов с другими специализированными клетками проводя- щей системы сердца (рис. 3). Т-клетки с контактами бок-в-бок и конец-в-бок имели диаметр 6-16 мкм, средний — 11,6±4,3 мкм. Клетки типа клеток Пур кинье располагались по периферии СУ, имели блед- ную цитоплазму, диаметр — от 19 до 40 мкм, в сред- нем — 29,2±10,7 мкм. Узел имел центральную арте- рию, был окружен жировой тканью, нервными волокнами и ганглиями. В 2 из 10 случаев артерия СУ располагалась не в центре, а на периферии. Телоподии часто располагаются непосредственно снаружи базальной мембраны, выстилающей специа- лизированные кардиомиоциты ПСС, на расстоянии 0,2-0,4 нм от их плазматической мембраны. В неко- торых случаях нами было выявлено проникновение телоцитов под базальную мембрану P-клеток, сбли- жение мембран телоподий и P-клетки на 20-30 нм и формирование межклеточных контактов в этой области (рис. 7Б). В данных контактах обнаружива- ются электронно-плотные палочковидные струк- туры, обеспечивающие структурную ассоциацию мембран в составе контакта (рис. 7В). Телоциты активно секретируют во внеклеточную среду вези- кулы (экзосомы) диаметром 0,1-0,2 нм, по-видимому, являющиеся средством коммуникации с соседними клетками. Иммуногистохимическое исследование. При имму- ногистохимическом исследовании с HCN4 (рис. 4) все специализированные клетки ПСС экспрессиро- вали данный маркер. Кроме этих клеток антиген экс- прессировали и клетки треугольной, овальной или булавовидной формы с длинными отростками (харак- терная структурная организация телоцитов) (рис. 4). Иммуногистохимическое исследование с двойной меткой выявило одновременную экспрессию CD34 и connexin43 в этих клетках (рис. 5). При этом, они имели тесный контакт с Р-клетками и рабочими кар- диомиоцитами. В этих клетках также одновременно экспрессировались HCN4 и CD34. Среднее количество этих клеток с иммунофеноти- пом телоцитов в одном поле зрения при х400 соста- вило 16,2±4,5. При этом, характер распределения клеток был неравномерный. Так, наибольшее среднее количество клеток (20,3±4,8) отмечалось в централь- ной части CУ. Напротив, по периферии узла, преиму- щественно среди жировой ткани, среднее количество клеток составляло 10,8±4,4. Средняя длина клеток была 29,2±12,4 мкм, а средний диаметр самой широ- кой (треугольной, овальной или булавовидной) части клеток был 2,6±0,6 мкм. Электронно-микроскопическая иммунодетекция HCN4 на ультратонких срезах СУ позволила обнару- жить данный белок в составе телоподий телоцитов, что соответствует нашим данным иммуногистохи- мии. Таблица 1 Скопления HCN4 преимущест- венно на мембране телоподии телоцита (выделен зеленым цветом) в виде кластеров коллоидного золота (красные стрелки). Сокращения: Я — ядро телоцита, Tп — телоподии, П — подома, PК — P-клетка, БМ — базальная мембрана Р-клетки, Mф — миофиламенты. 46 46 ОРИГИНАЛЬНЫЕ СТАТЬИ параметров с нормальным распределением вычисля- лись следующие показатели: среднее арифметическое (M), ошибка среднего арифметического (m), среднее квадратичное отклонение (δ). Для сравнения несколь- ких независимых групп по качественным признакам использовался непараметрический критерий Крас кила-Уоллиса. Достоверными считались различия, когда вероятность справедливости нулевой гипотезы (p) не превышала 0,05. параметров с нормальным распределением вычисля- лись следующие показатели: среднее арифметическое (M), ошибка среднего арифметического (m), среднее квадратичное отклонение (δ). Для сравнения несколь- ких независимых групп по качественным признакам использовался непараметрический критерий Крас кила-Уоллиса. Достоверными считались различия, когда вероятность справедливости нулевой гипотезы (p) не превышала 0,05. и vimentin в вышеописанных клетках с несколькими длинными, тонкими отростками, соответствующих иммунофенотипу телоцитов. Метод позволил выя- вить густую сеть из длинных отростков-телоподий с дихотомным делением и тесные контакты телопо- дий телоцитов с Р-клетками (рис. 6). Электронная микроскопия. Электронно-микроско- пическое исследование обнаружило, что в интерсти- ции СУ регулярно присутствуют клетки, имеющие компактную ядросодержащую область размером 6-12 мкм и образующие длинные тонкие, нередко извитые цитоплазматические отростки толщиной 0,1-0,3 мкм, длиной 40 мкм и более (рис. 7А). Досто- верно оценить истинную длину отростков на ультра- тонких срезах затруднительно, т. к. отростки имеют неправильную форму и частично могут уходить из плоскости среза. Ультраструктурная организация данных клеток полностью совпадает с неоднократно описанными ранее в других органах и других облас- тях сердца телоцитами [10]. Характерной особен ностью упомянутых цитоплазматических отростков (телоподий) являются периодически встречающиеся по их ходу утолщения (подомы) толщиной 0,4- 0,8 мкм. В составе СУ телоциты и их телоподии лока- лизованы преимущественно в близости от специали- зированных клеток проводящей системы сердца и от мелких кровеносных сосудов. Результаты Непрямое иммуномечение выявляет HCN4 в виде небольших скоплений коллоидного золота преимущественно на мембране телоподий (рис. 8). Обсуждение Таким образом, в результате исследования с использованием иммуногистохимического анализа, в том числе с двойной меткой, КЛСМ и электронной микроскопии, было доказано наличие телоцитов Конфокальная микроскопия. Конфокальная микро- скопия СУ позволила увидеть коэкспрессию CD34 и S100, CD34 и connexin43, SMA и connexin43, S100 47 47 Российский кардиологический журнал № 9 (149) \| 2017 Телоциты в миокарде были подробно описаны ранее другими авторами с помощью электронной микро- скопии [14]. В миокарде выявлена трехмерная сеть из телоподий телоцитов, густо обволакивающих кар- диомиоциты и контактирующих со всеми клетками и структурами [15]. Доказано модулирующее влияние телоцитов на миоциты, иммунные, сосудистые и нервные клетки [16]. Кроме gap junctions они имеют точечные, наноконтакты и плоские контакты с иммунными, нервными клетками, эндотелиоци- тами, перицитами, шванновскими клетками и кардиомиоцитами. Описывают, что часто телоциты близки к базальной пластинке кардиомиоцитов, но расстояние между двумя клеточными мембранами составляет около 150 нм, в то же время, есть точечные слияния (dot fusion), точечные узлы-контакты (dot junctions), соединяющие клеточные мембраны. В этих местах в цитоплазме кардиомиоцитов отмечают плот- ный материал (подобный Z-линии). Обычно кон- такты телоцитов с кардиомиоцитами располагаются в зоне вставочных дисков, реже вне этих зон. В СУ мы видели межклеточные контакты между клетками- пейсмекерами и телоцитами в виде параллельных друг к другу и перпендикулярных к мембранам элек- тронно-плотных наноконтактов, и плоские контакты телоподий с мембраной кардиомиоцита. При этом, телоцит находился под базальной мембраной, общей для двух Р-клеток. в СУ, которые одновременно экспрессировали connexin43 и CD34, connexin43 и SMA, vimentin и S100, HCN4 и CD34. Этот иммунофенотип может соответствовать только телоцитам. Факт наличия экспрессии коннексина43 в СУ противоречит резуль- татам исследований других авторов [11]. Это, воз- можно, связано с тем, что работы выполнялись на мелких объектах — мышах. в СУ, которые одновременно экспрессировали connexin43 и CD34, connexin43 и SMA, vimentin и S100, HCN4 и CD34. Этот иммунофенотип может соответствовать только телоцитам. Факт наличия экспрессии коннексина43 в СУ противоречит резуль- татам исследований других авторов [11]. Это, воз- можно, связано с тем, что работы выполнялись на мелких объектах — мышах. Региональную неоднородность узла описывают не только с точки зрения морфологии клеток, кото- рую мы в своем исследовании подтвердили, но и пейсмекерной активности, конфигурации потенциала действия и проводимости, плотности ионных токов, экспрессии белков gap junction (Cx40, Cx43 и Cx45), вегетативной регуляции и старения. С электрофизиологической точки зрения СУ явля- ется гетерогенной структурой, которая экспресси- рует уникальный набор ионных каналов, необходи- мых для генерации и распространения потенциала действия. Verker AO, et al. Литература 1. Schuessler RB, Boineau JP, Bromberg BI. Origin of the sinus impulse. J Cariovasc Electrophysiol 1996; 7(3): 263-74. 1. Schuessler RB, Boineau JP, Bromberg BI. Origin of the sinus impulse. J Cariovasc 1. Schuessler RB, Boineau JP, Bromberg BI. Origin o 11. Liu J, Dobrzynki H, Yanni J, Boyett MR, Lei M. Organization of the mouse sinoatrial node: structure and expression of HCN channels. Cardivasc Res 2007; 73: 729-38. 1. Schuessler RB, Boineau JP, Bromberg BI. Origin of the sinus 1. Schuessler RB, Boineau JP, Bromberg BI. Origin of the sinus impulse. J Cariovasc Electrophysiol 1996; 7(3): 263-74. Electrophysiol 1996; 7(3): 263-74. Electrophysiol 1996; 7(3): 263-74. Electrophysiol 1996; 7(3): 263-74. 12. Verkerk AO, Wilders R, van Borren MMGJ, Peters RJG, Broekhuis E, Lam K, Coronel R, de 2. Anderson KR, Ho SY, Anderson RH. Location and vascular supply of sinus node in human heart. Br Heart J 1979; 41:28-32. 2. Anderson KR, Ho SY, Anderson RH. Location and vascular supply of sinus node in human heart. Br Heart J 1979; 41:28-32. 12. Verkerk AO, Wilders R, van Borren MMGJ, Peters RJG, Broekhuis E, Lam K, Coronel R, de Bakker JMT, Tan HL. Pacemaker current (If) in the human sinoatrial node. European Hearth Journal 2007; 28: 2472-78. 12. Verkerk AO, Wilders R, van Borren MMGJ, Peters RJG, Broekhuis E, Lam K Bakker JMT, Tan HL. Pacemaker current (If) in the human sinoatrial node. European Hearth Journal 2007; 28: 2472-78. 3. Saunchez-Quintana D, Yen Ho S. Anatomy of cardiac and atrioventricular specialized conduction system. Rev Esp Cardiol 2003; 56(11): 1085-92. 13. Brioschi C, Micheloni S, Tellez JO, Pisoni G, Longhi R, Moroni P, Billeter R, Barbuti A, Dobrzynski H, Boyett MR, DiFrancesco D, Baruscotti M. Distribution of the pacemaker HCN4 channel mRNA and protein in the rabbit sinoatrial node. J Mol Cell Cardiol 2009; 47(2): 221-227. conduction system. Rev Esp Cardiol 2003; 56(11): 1085-92. 4. Waller BF, Gering LE, Branyas NA, Slack JD. Anatomy, histology, and pathology of the 4. Waller BF, Gering LE, Branyas NA, Slack JD. Anatomy, histology, and path cardiac conduction system: Part I. Clin Cardiol 1993; 16: 249-52. 4. Waller BF, Gering LE, Branyas NA, Slack JD. Anatomy, histolog cardiac conduction system: Part I. Clin Cardiol 1993; 16: 249-52. 5. James TN. Anatomy of the conduction system of the heart. In The Heirrt (Ed. Hurst JW). McGraw-Hill. New York, 1982: 46-56. 14. Kostin S. Обсуждение впервые на человеческом СУ продемонстрировали роль If тока в пейсмейкер- ной активности и определении частоты сердечных сокращений [12]. Ген ионных каналов, отвечающих за этот ток, принадлежит семейству генов HCN, состоящих из четырех изоформ. В СУ человека были описаны три его изоформы — HCN1, -2, -4. Счита- ется, что в сердце человека преобладают изоформы HCN1 и HCN4. Известно также, что HCN1-ток обладает наиболее быстрой кинетикой активации, а HCN4 — самой медленной. Феномен пейсмекерного сдвига (pacemaker shift) также демонстрирует, что СУ неоднороден. Исследо- вания Boyett MR, et al. выявили электрическую гете- рогенность СУ с замедленным проведением импульса в центре и более быстрым по периферии [17]. Ряд авторов считают, что СУ — сложный комплекс неод- нородной ткани, имеющий много зон для генерации импульса с доминирующими и дочерними фокусами, которые и обусловливают электрофизиологическую гетерогенность клеток СУ [18]. На наш взгляд, элек- трическую гетерогенность можно объяснить одно временным наличием 2 типов клеток, способных к проведению электрического импульса в СУ — Т-клеток и телоцитов, а также 2 типов клеток, спо- собных генерировать этот импульс — Р-клеток и телоцитов. Замедленное проведение импульса в центре может быть связано и с тем, что в централь- ной части СУ телоцитов в 2 раза больше, чем на пери- ферии. Густая 3D-сеть телоподий этих клеток может являться субстратом для mаcro reentry. Экспрессия телоцитами HCN4, доказанная нами с помощью иммуногистохимического исследования с двойной меткой и электронной иммуноцитохимии, на наш взгляд, свидетельствует о том, что эти клетки могут проводить электрический импульс от клеток- пейсмекеров к рабочему миокарду, и, возможно, даже генерировать электрический импульс. На наш взгляд, Brioschi С, et al. [13], описывающие в своей работе HCN4-позитивные мелкие клетки, организованные в группы, которые связаны между собой 3D сетью тонких отростков, на самом деле описывают Р-клетки, оплетенные сетью телоцитов. Эти же авторы утвер- ждают, что миоциты кроличьего СУ с интенсивной экспрессией HCN4 являются пейсмекерными клет- ками. Наше исследование демонстрирует, что высо- кая интенсивность (плотность) экспрессии HCN4 в центральной части СУ вызвана не только нахожде- нием там Р-клеток, но и большого количества тело- цитов. Таким образом, в результате исследования был выявлен еще один тип клеток, на наш взгляд, способ- ных генерировать и проводить электрический импульс в СУ. В нашей работе было показано, что телоциты находятся рядом не только с клетками-пейсмеке- рами, но и с рабочим миокардом на границе СУ. 48 48 ОРИГИНАЛЬНЫЕ СТАТЬИ Литература Cardiac telocytes in normal and diseased hearts. Semin Cell Dev Biol. 2016; 55: 22-30. McGraw-Hill. New York, 1982: 46-56. 6. James TN. Cardiac innervation: anatomic and pharmacologic relations. Bill N Y Acad Med 1967; 43(12): 1041-86. 15. Cretoiu D., Hummel E., Zimmermann H., Gherghiceanu M., Popescu L. M. Human cardiac telocytes: 3D imaging by FIB-SEM tomography. J. Cell. Mol. Med. 2014; 11(18): 2157-64. 7. Popescu LM, Nicolescu MI. Resident stem cells and r and Steam cells. New York: Elsevier, 2013: 205-31. 7. Popescu LM, Nicolescu MI. Resident stem cells and regenerative therapy. In: Telocytes and Steam cells. New York: Elsevier, 2013: 205-31. and Steam cells. New York: Elsevier, 2013: 205-31. 16. Popescu LM, Gherghiceanu M, Cretoiu D, Radu E. The connective connection: interstitial cells of Cajal (ICC) and ICC-like cells establish synapses with immunoreactive cells. Electron microscope study in situ. J Cell Mol Med 2005; 9:714–30. 8. Zhu YF, Wang XY, Lowie BJ, Parsons S, White L, Kunze W, Pawelka A, Huizinga JD. Enteric sensory neurons communicate with interstitial cells of Cajal to affect pacemaker activity in the small intestine. Pflugers Arch 2014; 466: 1467–75. 9. Si X, Huang L, Gong Y, Lu J, Lin L. Role of calcium in activation of hyperpolarization- activated cyclic nucleotide-gated channels caused by cholecystokinin octapeptide in interstitial cells of cajal. Diqestion 2012; 85(4): 266-75. 17. Boyett MR, Honjo H, Kodama I. The sinoatrial node, a heterogeneous pacemaker structure. Cardiovasc Res 2000; 47(4): 658-87. 18. Gomes JA, Winters SL. The origins of the sinus node pacemaker complex in man: demonstration of dominant and subsidiary foci. JACC 1987; 9(1): 45-52. 10. Kostin S. Myocardial telocytes: a specific new cellular entity. J Cell Mol Med 2010; 14(7): 1917-21. 10. Kostin S. Myocardial telocytes: a specific new cellular entity. J Cell Mol Med 2010; 14(7): 1917-21. 49 49 49 49
W4229066618.txt	https://www.cahiers-clsl.ch/article/download/1909/1664	fr		La mentalité hongroise et la langue hongroise dans la théorie de Sándor Karácsony (1891-1952)	Cahiers du CLSL	2,022	cc-by	5,594	Cahiers de l 'ILSL, N°8, 1 996, pp. 1 49 - 1 62 La mentalité hongroise et la langue hongroise dans la théorie de Sandor Karacsony (1891- 1952) Zsuzsa HETÉNYI Université de Budapest, Hongrie J'aimerais rappeler un colloque qui fut organisé en 1 925, à Oslo, sur le su jet «Mankind, Nation and Individual from a Linguistic Point of View», et dont les communications ont été éditées par OUo Jespersen sous le même titre 1 . Le colloque a examiné dans quelle mesure la langue est une institu tion sociale, et OUo Jespersen a exprimé ses doutes quant aux éléments psychosociologiques de la théorie de Saussure. En 1926 Charles Bally dans son article «Langue et parole» lui a répondu et a défendu les idées de Saussure. Charles Bally a commencé ses cours sur la théorie saussurienne à l ' Université de Genève en 1 9 1 3 , l 'année où le jeune Sandor Karacsony est venu à Genève. Ainsi l 'histoire se répète un peu : nous envisageons de nouveau le problème de la langue et de la nation, comme il y a 70 ans, et j ' aimerais vous présenter quelques idées de S andor Karacsony, qui s 'est inspiré, pour son système, de Charles Bally. Sandor Karacsony est né en 1 89 1 dans un petit village de l 'Est de la Hongrie. Il a fait ses études dans le célèbre lycée protestant de Debrecen. Une année de service militaire dans l 'armée d'Autriche-Hongrie a éveillé en lui l ' intérêt pour l 'étude du multiculturalisme et muItilinguisme. Après la première année à la Faculté des Lettres à Budapest, il est allé à Genève en tant qu'auditeur de Charles Bally, père de la stylistique fonctionnelle. Il a également participé aux sessions du Bureau International de l'Éducation. A Munich, Vienne et Graz i l a pris connaissance de la théorie de Wundt par l ' intermédiaire de son disciple Wilhelm Streitberg, et de la théorie de 1 Jespersen, 1 925. 1 50 Cahiers de l ' ILSL, N°8, 1 996 Hermann Paul, représentant de la linguistique psychoethnique. A la suite d ' une blessure pendant la Guerre (il est devenu invalide pour toute sa vie) il a travaillé dans des écoles militaires. Après la guerre, il a enseigné pen dant huit ans dans un lycée, afin de mettre en pratique ses innovations psychopédagogiques. Il écrivit des œuvres pédagogiques, ainsi que des pe tits livres utiles pour les adolescents (p. ex. comment faire les devoirs en évitant l'automatisme et les stupidités; il a même rédigé un journal intime pour les adolescents). Sandor Karacsony fut un membre actif de l 'organisa tion de la jeunesse religieuse œcumenique (YMCA); mais, en 1 93 1 , voyant l ' alignement des cercles religieux sur le pouvoir gouvernemental pro-allemand, il a abandonné son poste avant de quitter définitivement le YMCA en 1 939 . Comme Sandor Karacsony écrivait toujours sur la com préhension mutuelle entre les petits peuples de l'Europe Centrale, il est devenu une personne indésirable (la Hongrie s'est alliée à l 'Allemagne de crainte d 'être entourée des pays nommés «la petite entente» et a reçu tout de suite les territoires de ces petits peuples autour de la Hongrie avec l 'aide allemande). Après la guerre Karacsony a joué un rôle actif dans différentes organisations de la jeunesse et de la culture. Il sera à nouveau une personne indésirable, cette fois sous le régime communiste (le parti communiste a pris le pouvoir en 1 948 lors d'élections falsifiées) . Sandor Karacsony a été souvent attaqué, son département a été fermé, il a été privé de son poste u niversitaire et même du droit à la retraite. Il est mort en 1 952. La pre mière réédition d ' un de ses livres n'a vu le jour qu'il y a 1 0 ans. Karacsony se considérait avant tout comme un pédagogue, mais son système, fondé sur la psychologie sociale, embrasse tous les domaines de la vie et constitue un système philosophique. La théorie de S andor Karacsony est présentée dans neuf grands livres publiés entre 1 938 et 1 947 . (Il faut dire qu 'après 1 938 il a été coupé de toutes sortes de contact, soit scientifique, soit personnel, avec l'Europe). La théorie de S andor Karacsony réinterprète plusieurs catégories de la «Vôlkerpsychologie», la psychologie sociale de Wundt, mais ne les ac cepte pas sans critique. La plus grande différence peut se résumer dans le fait que Wundt considère la Science et l'Art comme des productions indi viduelles, c 'est-à-dire que pour lui un poème de Gœthe n'est pas un phé nomène déterminé par la psychologie sociale, tandis que Sandor Karacsony inclut Gœthe aussi bien que les chansons populaires dans le cadre des rela- Hetényi : La mentalité hongroise et la langue hongroise 151 tions sociales. Dans son premier volume intitulé Grammaire hongroise fondée sur la psychologie sociale, il part comme toujours de la langue comme moyen éducatif, mais, comme dans toutes ses œuvres, il accorde également une place importante aux particularités de la langue comme base de la communication : La langue de notre sophocratie (c' est-à-dire les intellectuels - Zs. H.) n'est pas une langue originale, mais traduite. Elle a été traduite après 1 825 et sur tout en 1 86 7 (l'année du compromis entre l'Autriche et la Hongrie). Elle étai t utilisée comme langue de l ' armée commune d' Autriche-Hongrie, langue de l 'enseignement (traduite à partir du latin et de l ' allemand), langue scientifique, langue des métiers, du commerce et de l' industrie, de la correspondance commerciale; langue du droit, de l'économie, de r adminis tration, du chemin de fer, de la poste, des journaux, etc. Ainsi, tout ce qui était le moyen et le résultat d 'un développement économique était traduit très rapidement et superficiellement. Le hongrois original n ' est resté intact que dans la langue du peuple et dans celle de la littérature. ( 1 985 : 235-236) Le problème est lié au fait que la langue de la sophocratie n'a pas été capable de digérer les éléments étrangers. Elle a donc perdu le contact avec la langue du peuple, en l'absence d 'une langue de la classe moyenne qui aurait dû jouer un rôle de médiateur. D'autre part, la langue du peuple s 'est figée dans son état du 1 7ème siècle, incapable de construire les dé terminations scientifiques et les thèses ou argumentations théoriques . Puisque la culture doit naître d'un mariage du peuple et de la sophocratie, et se manifester dans la classe moyenne, Sandor Karacsony pose la ques tion suivante : «la culture hongroise peut-elle exister si elle n ' a pas de langue ?» . En cherchant la réponse, Karacsony attache une importance pri mordiale à la langue du peuple comme seule base avec ses trois forces (sur lesquelles je reviendrai plus loin). «Cette langue a obtenu sa forme selon le contenu de la culture hongroise. Nous avons devant nous la forme qui nous permet de tirer des conclusions concernant son contenu essentiel»2 . Il est bien évident que Sandor Karacsony - en ce qui concerne ses idées linguistiques - est relativiste, puisqu 'il considère que l a diversité des langues reflète la diversité de la mentalité des peuples qui les parlent. De plus, c 'est un relativiste cognitif puisque selon sa thèse de départ l a 2 1 990 : 328. 1 52 Cahiers de l 'ILSL, N°8, 1 996 langue est formée par les circonstances et par la culture de la communauté qui la parle. Les trois forces de la langue hongroise, selon Karacsony, sont les suivantes : sa base d' articulation, c 'est-à-dire l ' intonation et la prononcia tion, son principe de parataxe (coordination, juxtaposition) et la symbo lique de ses figures primitives. Ces trois forces se mettent en action quand un élément vient d'une autre langue. Elles ont défendu notre langue pen dant des siècles, surtout du danger qui la menaçait de la part des langues indo-européennes qui l'encerclaient. Il faut rappeler que notre langue finno ougrienne a perdu le contact avec ses parents de langue il y a trois mille ans. Après les périodes ouralienne, finno-ougrienne et ougrienne, les tribus nomades «protohongroises» commencèrent leur migration de deux mille ans et leur langue subit l ' influence de plusieurs langues turques et ouralo altaïques . Arrivées dans le bassin des Carpates, les tribus se mélangèrent aux autres peuples voisins et leur langue fut influencée par les langues slaves et l ' allemand. La langue de la culture resta le latin jusqu' au 19ème siècle (Janos Bolyai publia en latin ses études sur la géométrie non-eucli dienne en 1 832). Mais depuis la contre-réforme du 17ème siècle, la langue de la culture écrite fut en fait l 'allemand (il suffit de citer Georg Lukacs, qui écrivit ses grandes œuvres philosophiques en allemand). Comment la langue hongroise fut-elle capable de survivre à toutes ces perturbations his toriques ? Quelles sont alors les particularités de cette langue, les particula rités qui caractérisent une communauté qu 'on appelle «nation», mais que Sandor Karacsony préfère nommer tout simplement «les Hongrois» ? Le hongrois est une langue jeune, pittoresque et imagée. Cela ne si gnifie pas qu 'elle est meilleure ou pire que les autre langues. Le hongrois donne la description des choses d'une manière telle qu'on peut les voir, côte à côte, comme elles coexistent dans la réalité. C ' est pourquoi on peut dire qu' il est à la foi primitif et objectif. Le hongrois préfère la parataxe, les constructions coordonnées, il n ' admet pas de subordinations, les membres de la phrase ont un lien faible. Le sujet parlant fait le moins d 'ef forts possible pendant qu' il parle : il ne fait pas d'abstraction, ne met pas en valeur ce qui est plus important, ne met pas les parties de la phrase en ordre logique, il ne donne pas son opinion, ne veut pas influencer son in terlocuteur. Dans les phrases coordonnées on n'a pas recours à des parti cules conjonctives. Un exemple curieux : la grossièreté dans la langue du Hetényi : La mentalité hongroise et la langue hongroise 1 53 peuple. Sandor Karacsony estime que les images obscènes et vulgaires sont largement utilisées en hongrois parce qu'elles font partie de la structure des images vivantes, plastiques, pittoresques, tandis que dans les langues indo-européennes des mots beaucoup moins rudes ne sont pas tolérés par la langue littéraire, puisque les structures utilisées pour exprimer des obscéni tés sont purement phraséologiques . J'essayerai de donner quelques exemples lexicaux (pas de cette sorte, bien entendu) pour montrer ce qu'il faut comprendre quand Sandor Karacsony parle de la force pittoresque de la langue hongroise. Tandis que dans la langue de la sophocratie le mot «Restaurant -W agen» est traduit mot à mot de l' allemand, le peuple l ' a ap pelé «bistrot à roues» . L'avion était au début «aéroplan », puis «machine volante» (cette dénomination est restée jusqu 'à maintenant), mais le peuple disait «dragon de Vienne». Pour montrer la différence entre les visions du monde hongroise et allemande, Sandor Karacsony prend un exemple très spécial : une information de journal de l 'entrevue entre Hitler et le gouver neur de la Hongrie, Mikl6s Horthy. Il prend une seule phrase et la décor tique pour montrer que sa structure et son point de vue sont pris de l'alle mand. Après deux pages d 'analyse ironisant sur Hitler et Horthy, Sandor Karacsony propose la variante qui convient pour la mentalité hongroise. Voici un autre exemple, plus simple : Karacsony cite l ' avis écrit sur les billets de bus : « Nous prions MM. les usagers de bien vouloir déchirer leur billet à la fin de leur trajet, afin qu'il soit évitable que les autres pas sagers fassent des abus des billets déjà utilisés.» Le texte «hongrois-hon grois» : «Si vous ne voyagez plus, déchirez votre billet, pour que les autres ne le ramassent pas». On peut objecter à cette théorie que toutes les langues du peuple sont aussi figuratives que la nôtre. Sandor Karacsony compare quelques proverbes hongrois et allemands pour démontrer que les figures en hongrois viennent vraiment de la vie concrète, tandis que les proverbes allemands sont abstraits. Proverbe hongrois : «Le chien est at teint de rage quand il n ' est pas à plaindre». C 'est vrai que le chien tombe malade de gastrobronchite quand il est trop bien nourri. L'équivalent en al lemand est le proverbe suivant : «Wenn ' s zurn Esel zu gut geht, geht er aufs Eis tanzen» . - où l ' image de l' âne dansant sur la glace est tout à fait abstraite3 . Sandor Karacsony ne se contente pas de montrer que les langues 3 ibid. : 288-289. 1 54 Cahiers de [ 'ILSL, N° S , 1 996 du peuple sont différentes, mais il propose une analyse comparative de dix pages de deux romans, le Waldheimat de l'Autrichien Peter Rosenegger et les récits de l 'écrivain hongrois, Mikszath Kalman, que je ne peux vous présenter ici. Karacsony est convaincu que la philosophie est aussi possible en hongrois. Dans notre langue, il est possible d'écrire les relations de la sphère la plus abstraite et intellectuelle, mais cette philosophie sera objec tive, figurative. Il essaie de montrer comment les images les plus concrètes servent de base pour les signifiants les plus abstraits dans le même champ sémantique : il consacre un chapitre entier à décrire le champ sémantique du verbe «aIl» (stehen, stay), et il donne des exemples de comparaison entre l 'allemand et le français : Le français ne peut pas même exprimer ce verbe, il ne le considère pas comme une action. Pour lui c'est une position exprimée en périphrase, avec un verbe d ' état. Se tenir, être debout, mettre debout, rester debout. «Debout 1 » crie-t-il pour le «Tagwache auf 1», quand le hongrois dit «le soleil brille déjà sur ton ventre».4 Dans ses exemples linguistiques, S andor Karacsony part du principe de Charles Bally, qui a lancé un projet de dictionnaire idéologique pour l 'enseignement secondaire5 . Malgré le fait que tous les deux accordent une place importante à l 'enseignement, il faut constater des différences. Tandis que BaUy veut moderniser l 'enseignement et introduire les nouveaux résul tats de la linguistique à l'école, et faire comprendre la nature de la langue française en général, Sandor Karacsony a deux tâches devant lui : premiè rement, avec la description de la langue hongroise, il essaie d'éliminer une grammaire construite sur la base du latin, c'est-à-dire sur une base indo-eu ropéenne, étrangère à la mentalité hongroise. Il voudrait établir une gram maire respectant l ' originalité du hongrois, et savoir établir les traits caracté ristiques du peuple hongrois à travers cette grammaire. Deuxièmement, avec les changements proposés, il veut rendre possible l 'accès à l 'ensei gnement secondaire pour les élèves issus des classes inférieures, pour qui même la langue de l 'enseignement est problématique (c'est un élément de la réforme sociale). Bally fait entrer en comparaison l' allemand et le fran4 ibid. : 309. 5 Bally, 1 9 1 1 . Hetényi : La mentalité hongroise et la langue hongroise 1 55 çais pour préciser les catégories linguistiques distinctives . S andor Kanicsony, en prenant les exemples de l' allemand, rejette et refuse catégo riquement l'influence forcée de la culture allemande après plusieurs siècles de colonisation autrichienne, et cela, à un moment (les années trente) où l 'influence allemande devient de plus en plus dangereuse en Europe et en Hongrie. En entrant dans tous les détails de la langue, Sandor Karacsony veut - comme deuxième tâche - aller au fond des problèmes de l ' his toire et du destin hongrois, au-delà de la psychologie sociale. Il espère trouver dans la langue des modèles de mentalité qui sont caractéristiques aussi des formes sociales. Karacsony a appliqué - mutatis mutandis - les vues générales de Bally à un autre idiome, mais il va beaucoup plus loin dans ses conclu sions . Dans ses articles «L'enseignement de la langue maternelle et la for mation de l'esprit» ( 1 92 1 ) et «L'expression des idées de sphère personnelle et de solidarité dans les langues indo-européennes» ( 1926), Bally signale que les différences entre les langues expriment les différences de la pensée. Il sous-entend comme allant de soi que chaque langue reflète une autre «Weltanschauung» . II écrit par exemple que «L'araignée évoque des repré sentations différentes selon que son nom est féminin, comme en français, ou masculin, comme en russe» 6 . Karacsony met en lumière le problème de la différence entre le sys tème de la flexion et celui de l 'agglutination : Cette différence n'est pas une cause, mais une conséquence. La différence entre les deux langues n'est pas due à leur système de flexions ou à celui de l'agglutination, mais c'est le fait qu'ils sont différents et à cause de cette différence l'une utilise les flexions, l' autre l' agglutination. [ ] La menta lité hongroise s'est creusé ce canal, la parataxe, pour rendre possible la parole7 . . . . Quand un Français dit «mon père», il se met en valeur par rapport à son père : il prend d' abord lui-même, et son père seulement après. En hon grois on dit «apam» : on exprime une relation dans le même corps de mot. Les mots «mon frère» et «ma sœur» montrent très bien cette relation : les mots «frère» et «sœur» n ' existent pas sans affixe possessif. Alors la para6 Bally, 1 927. 7 Op . cit. : 2 1 0, 267. 1 56 Cahiers de l 'ILSL, N°8, 1 996 taxe des deux personnes - du locuteur et de son père, frère, sœur - est évidente. Ainsi l 'agglutination suit aussi le principe de parataxe. La parataxe détermine la construction de la phrase. Les phrases comme navigare necesse est ou la vie est belle se construisent sans verbe, le sujet et le prédicat n'ont pas de lien grammatical. Un détail du hongrois - négligé par Karacsony - est que nous n ' avons pas de genre grammatical. Un Hongrois n ' a donc pas l 'association du féminin en disant terre ou lune, il ne divise pas le monde selon ce principe. (La traduction littéraire a alors bien des difficultés avec certains poèmes. Mais ce n ' est bien sûr pas une question purement littéraire, il s ' a git vraiment d 'une autre vision du monde.) L'ordre des mots en hongrois suit une logique diamétralement op posée à celle des langues indo-européennes. Après la partie connue (sujet) on met tout de suite le message essentiel, c'est-à-dire qu ' on ne met le pré dicat qu' après l 'élément nouveau, le tout précédant les éléments secon daires. Par exemple, les premiers mots de la Marseillaise Allons, enfants de la patrie, le jour de gloire est arrivé en hongrois sont rangés comme suit : Allons, patrie de / enfants /, arrivé est / gloire de / jour. Prenons un panneau d'interdiction, où le français change l 'ordre des mots pour attirer l 'attention : Défense de pénétrer dans le parc à bicyclette sous peine d 'a mende . Ce panneau a fait rire mon fils (il a compris peut-être que quel qu ' un qui met les mots d 'une telle manière doit concevoir le monde d'une façon totalement différente de la sienne) pour qui en hongrois ce tableau di rait: Amende de / peine sous / bicyclette à / le parc dans / pénétrer / dé fense, ici aussi dans un ordre inverse des mots pour attirer l' attention. Je peux encore ajouter que le hongrois est la seule langue en Europe où l ' on mette le prénom après le nom de famille (comme Karacsony Sandor, lequel Alexandre? - le Karacsony). Cet ordre des mots est une conséquence du principe essentiel de l 'intonation descendante. L' accent tombe sur la pre mière syllabe du mot. Le début de la phrase (où se trouve le message es sentiel) est le plus fort et est prononcé au registre le plus élevé. Dans les phrases coordonnées les conjonctions ne sont pas obligatoires, elles sont sous-entendues par la réflexion et sont exprimées par l' intonation . A l ' é poque de Sandor Karacsony, l'intonation montante était considérée comme influencée par le yiddish : son caractère étranger paraît dans les blagues et sur la scène de théâtre. Il faut penser plutôt à une influence de l 'allemand Hetényi : La mentalité hongroise et la langue hongroise 1 57 et, de nos jours, à ceIle de l' américain, de la musique populaire se déver sant sur l'Europe depuis les années 50, qui a laissé des traces définitives dans le langage de la jeunesse . Je signalerai en passant que l ' intonation descendante du slovaque, seule langue slave où elle existe, témoigne d 'une influence de l 'intonation hongroise. L' accent dynamique en hongrois donne l 'impression d'écouter de la poésie parce que les syllabes accentuées font les césures. (Pour être moins poétique, je peux dire qu'en écoutant un enregistrement hongrois à l'envers nous avons l 'impression d'entendre un aboiement de chien, à cause de cet accent fort et court). Il faut dire qu 'au 20ème siècle la force du vers iambique a cassé la structure de la poésie tra ditionnelle; les génies seuls trouvent les formes dans lesquelles le texte se lit à la foi comme vers métrique et vers accentué. Cette circonstance pose des problèmes essentiels pour la traduction littéraire et surtout des libretti d ' opéra, où l a musique ne permet pas de changer la rythmique. La musique populaire hongroise traditionnelle peut être décrite avec les mêmes catégories que ceIles que Sandor Karacsony a relevées comme les trois forces de la langue hongroise. C'est le résultat d'une recherche de deux décennies, faite parallèlement, mais indépendamment des études de Sandor Karacsony. Deux compositeurs, Béla Bart6k et Zoltan Kodaly trou vèrent que notre langue musicale originale a trois particularités : l 'isomé trie (les lignes isométriques dans les chansons des paysans avec la même quantité de syllabes); le rythme «parlando-rubato» (la musique suit le texte flexiblement, l ' accent parlé) et la gamme pentatonique (base musicale for mée de cinq tons). S andor Karacsony trouve tout de suite une correspon dance parfaite entre ce système et le sien. L 'isométrie est l'articulation et l ' intonation musicale qui ressemble au bruit répétitif de la nature, dans le principe «parlando-rubato» se manifeste la parataxe du texte et de la mélo die, et la pentatonie avec ses cinq tons de base exprime la vision du monde figurative, objective et primitive. En chantant, on utilise les tons qui se trouvent dans la nature sans rien ajouter (sans demi-tons ou notes alté rées) 8 . Bartok écrit : La musique du peuple dans son état actuel est un résultat d ' interdépen dance des petits peuples du bassin des Carpates qui est un territoire sans 8 ibid. : 327. 1 58 Cahiers de [ 'ILSL, N°S , 1 996 limites géographiques où les modèles des mélodies primitives s'échangent depuis des siècles. ( 1 9 37) Selon S andor Karacsony, la musique du peuple est tout aussi éloi gnée de la musique de la sophocratie que ce que l 'on a déjà vu dans le cas de la langue. Le génie de Bart6k a permis de jeter un pont à travers ce fossé séparant les deux couches de la culture, et de composer la musique en pre nant la base spirituelle de la source des chansons des paysans et exprimer un message universel en langue (musicale) de sa nation. Le même principe de parataxe qui organise les lignes des chansons et les phrases est présent dans les relations sociales : Notre pays est celui des petites communautés autonomes. Pour u n Hongrois, les autres sont toujours égaux o u même placés plus haut.[ . . . ] En Hongrie le principe minoritaire est devenu dominant, tandis qu 'en Europe c'est le principe majoritaire. ( 1 990 : 335-336) Cette conception nous semble aujourd'hui un peu idéaliste. Il est vrai que S andor Karacsony n'est pas le seul à évoquer la structure tradi tionnelle du village hongrois comme un microcosme paisible où les Juifs, Serbes, Croates et Slovaques vivaient côte à côte sans conflits; mais il faut bien voir que cette mentalité ne pouvait malheureusement entrer dans la vie politique, ni avant la guerre, ni après, et personne ne pouvait prévoir des changements si rapides dans la mentalité. Néanmoins, Karacsony a raison de dire que la Hongrie fut capable de survivre pendant les siècles difficiles grâce à ces petites autonomies. Les princes indépendants en Transsylvanie, les synodes des églises autonomes et les comitats étaient encore plus ou moins contrôlables; mais les petits villages seigneuriaux ou les conseils presbytériaux, coupés de la hiérarchie supérieure, fleurirent sous la domina tion turque et construisirent un système communautaire qui n ' a disparu qu' avec la première guerre mondiale. Mais au moment où cette commu nauté doit entrer dans les grandes structures d ' un comitat ou d'un État, le principe de la parataxe ne peut plus se manifester parce que la structure des comitats et celle de l'État sont le résultat d ' un système hiérarchique. C ' est pour cela peut-être que les Hongrois préféraient offrir le trône à l 'époque à des rois étrangers, se demande avec une certaine ironie Sandor Karacsony. Ensuite il passe en revue les traits les plus importants de la mentalité hon- Hetényi : La mentalité hongroise et la langue hongroise 1 59 groise : notamment le caractère chevaleresque avec son variant négatif la jobarderie, la naïveté, la résistance passive rejetant sans merci les éléments non-hongrois, et le fait que nous sommes un peuple épris de liberté; nous sommes des hommes des armes. Tous ces traits soi-disant «positifs» sont ramenés au principe d' objectivité et de primitivité . Ensuite il évoque les traits «négatifs», les «péchés» hongrois : nous sommes toujours en désac cord, nous tirons tout en longueur (c'est l ' inertie), nous sommes des vel léitaires qui commençons les choses avec un grand enthousiasme qui ne dure pas, enfin nous attendons que les alouettes tombent toutes rôties, que la fortune nous vienne en dormant. Selon sa logique, cette mentalité est le garant de survie dans les circonstances de la nation hongroise. Le désaccord exprime notre résistance à être esclaves, à vivre comme un troupeau; l'iner tie est la longue souffrance de la passivité entre les courtes périodes de l' ac tion de bravoure, du grand enthousiasme; et enfin nous attendons les mi racles - en fait «la mentalité hongroise n 'accepte ni le temps limité, ni l 'espace limité, ni le lien entre cause et conséquence, alors la spontanéité du miracle est pour nous la seule possibilité. Nous existons toujours en Europe en dépit de la logique de la cause et de la conséquence, c'est un mi racle. Notre mentalité asiatique regarde son destin ' sub specie aeternita tis ' » 9 . Le peuple hongrois n ' a pas disparu parce qu 'on a besoin de ce peuple ici ( = géographiquement) et jusqu 'à nos jours ( = historiquement). Ici - parmi les ethnies germaniques, romanes et slaves, subjectives et compliquées - nous sommes un noyau hétérogène, un peuple avec une mentalité objective et primitive. Historiquement, on a besoin d ' un peuple petit et dynamique comme point de rencontre entre les différents «Weltanschauungen» : le christianisme et l 'islamisme, l 'orthodoxie et la papauté, le catholicisme et le protestantisme, les Églises historiques et les petites sectes, l'Europe de Jésus Christ et l'Asie païenne lO. Rien de plus naturel que les partisans de la caractérologie nationale aient voulu exploiter les idées de Sandor Karacsony pour montrer que les Hongrois sont une race spéciale. De plus, les intellectuels qui voulaient défendre la Hongrie du fascisme, de crainte qu 'il ne menace l'existence du 9 ibid. : 339. 1 0 ibid. : 390. 1 60 Cahiers de l'ILSL, N°8, 1 996 pays, trouvaient dans les idées de Sandor Karacsony un moyen de renforcer la conscience nationale, pour construire une idéologie populiste de l 'hé roïsme romantique. Sandor Karacsony refusa ces appropriations en souli gnant qu'elles négligeaient les côtés pédagogiques; la logique de son livre n ' est d'ailleurs pas utilisable à des fins politiques. Après la guerre, S andor Kanicsony ajouta encore que les résultats de sa recherche avaient prouvé que les phénomènes qui lui semblaient être «spécifiquement hongrois» en 1 939, devaient être considérés en 1 947 comme «le niveau primitif de l ' humanité universelle». Sandor Karacsony donna ainsi u n nouveau sens au mot «hongrois» . Dans son livre La vision hongroise du monde, S andor Karacsony va encore plus loin, en contestant que le concept de la «Weltanschauung» de la pensée allemande ne puisse pas exister en Hongrie. «Les Hongrois ne regardent pas le monde de l'extérieur, mais doivent vivre dans le monde. Au lieu de se rapporter subjectivement, ils participent au monde objectif, réel» 1 . Par conséquent, la vision du monde n 'est pas une matière qu'on peut enseigner en répétant les idéaux abstraits de l'histoire nationale héroïque. Quelle conclusion peut-on tirer de l'œuvre de Sandor Karacsony, et notamment de son livre La mentalité hongroise ? L'enseignement de la langue maternelle est très important, car l 'appropriation de la langue ma ternelle est la première condition pour apprendre des langues étrangères. Et comme la langue n 'est pas un code génétique mais est apprise par la socia lisation, c'est par la socialisation qu'on devient membre d ' une nation . Bien que dans la définition de la nation la langue ne soit qu'un élément parmi d 'autres, la langue est la base qui fait naître le folklore, la culture, qui forme la communauté et par conséquent la nation. La langue est beaucoup plus concrète que la nation - l'idée de Sandor Karacsony de connaître la mentalité par l'intermédiaire de la langue est donc tout à fait justifiée et productive. Puisque l ' on devient membre d'une nation par éducation, il n 'est plus du tout pertinent de demander qui est le descendant des «protohongrois», venus dans le bassin des Carpates il y a 1 000 ans. S andor Pet6fi, le génie de la poésie hongroise du 1 9ème siècle, qui est resté sur le champ de bataille final de la lutte pour l'indépendance hon groise en 1 849, était le fils d'un tavernier serbe et d'une servante slovaque. «La question est de savoir si le membre de la nation est prêt à prendre la I l Kontra, 1 992 : 50. Hetényi : La mentalité hongroise et la langue hongroise 161 mission donnée par le destin hongrois» conclut Sandor Karacsony. Notre destin est de rester européens dans notre sophocratie - sinon, la Hongrie sera retardée; et de rester asiatiques dans notre peuple - sinon, la nation hongroise s ' absorberait dans le mer des grandes nations de l'Europe I 2 . Notre tâche est de trouver le lien entre eux par les classes moyennes, de réaliser les valeurs européennes dans les formes asiatiques pour être vrai ment universels. © Zsuzsa Hetényi ŒUVRES PRINCIPALES DE SANDOR KARACSONY - ( 1 939) : La mentalité hongroise et la réforme de notre éducation na tionale. Les bases sociopsychologiques de la pédagogie. - ( 1 985) : 2e édition corrigée et élargie. 1. L'éducation de la langue/par la langue et le mécanisme intellectuel de l 'esprit social - ( 1 938) : Grammaire hongroise fondée sur la psychologie sociale. - ( 1 94 1 ) : Le système des signes et des symboles. II. La limite supérieure de l'esprit social et l 'éducation de la transcendance ( 1 94 1 ) : L 'éducation idéologique. La vision hongroise du monde. - ( 1 943) : Le Dieu des Hongrois . L 'éducation religieuse. III. L'éducation sociale et le mécanisme de volonté de l'esprit social - ( 1 944) : Le trésor des Hongrois . Le système de valeurs et l 'axiologie . - - ( 1 946) : La jeunesse hongroise. Le système d 'actes et l 'éthique. - ( 1 942) : Le peuple hongrois qui se réveille. Le systèlne de coutumes et la pédagogie. IV. La limite inférieure de l 'esprit social et l'éducation de droit - ( 1 945) : La démocratie hongroise. L 'éducation d 'autonomie. - ( 1947) : La paix hongroise. L 'éducation de paix : après la guerre et vers des réformes . 1 2 Op . dt. : 386. 1 62 Cahiers de l 'ILSL, N°8, 1 996 RÉFÉRENCES BIBLIOGRAPHIQUES BALLY, Ch. ( 1 9 1 1 ) : La stylistique et l 'enseignement secondaire, Saint Blaise. - ( 1 927) : «La contrainte sociale dans le langage», Revue internationale de sociologie, 35ème année, 5-6, Mai-Juin. BARTOK, B. ( 1 937) : Népdalkutatas és nacionalizmus [Recherche de chanson folklorique et le nationalisme] . JESPERSEN O . ( 1 925) : Mankind, Nation, and lndividual from a Linguistic Point of View, Oslo. KARACS ONY, S . ( 1 985) : A magyar ész}aras, ( 1 939) Budapest : Magveto, [La mentalité hongroise] . - ( 1 990) : «A magyar észjaras c. mü ismertetése» [Extrait de la mentalité hongroise], (autoreferatum, 1 939), in Magyar Filoz6fiai Szemle, 1 990, 3-4 . KONTRA, GY. ( 1 992) : Karacsony Sandor, Budapest.
https://openalex.org/W3089121523	https://biblio.ugent.be/publication/8676376/file/8676377	Latin	null	Myocardial Function, Heart Failure and Arrhythmia in Marfan Syndrome: A Systematic Literature Review	Diagnostics	2,020	cc-by	10,361	Received: 04 September 2020; Accepted: 24 September 2020; Published: 25 September 2020 Abstract: Marfan syndrome (MFS) is a heritable systemic connective tissue disease with important cardiovascular involvement, including aortic root dilatation and mitral valve prolapse. Life expectancy in patients with MFS is mainly determined by cardiovascular complications, among which aortic dissection or rupture are most dreaded. In recent years, heart failure and ventricular arrhythmia have drawn attention as extra‐aortic cardiovascular manifestations and as additional reported causes of death. Imaging studies have provided data supporting a primary myocardial impairment in the absence of valvular disease or cardiovascular surgery, while studies using ambulatory ECG have demonstrated an increased susceptibility to ventricular arrhythmia. In this paper, current literature was reviewed in order to provide insights in characteristics, pathophysiology and evolution of myocardial function, heart failure and ventricular arrhythmia in MFS. Keywords: Marfan syndrome; heart failure; myocardial; ventricular arrhythmia e ie Myocardial Function, Heart Failure and Arrhythmia in Marfan Syndrome: A Systematic Literature Review Anthony Demolder 1,, Yskert von Kodolitsch 2, Laura Muiño‐Mosquera 1,3 and Julie De Backe 1 Centre for Medical Genetics, Ghent University Hospital, 9000 Ghent, Belgium; laura.muinomosquera@uzgent.be (L.M.‐M.); julie.debacker@ugent.be (J.D.B.) 1 Centre for Medical Genetics, Ghent University Hospital, 9000 Ghent, Belgium; laura.muinomosquera@uzgent.be (L.M.‐M.); julie.debacker@ugent.be (J.D.B.) 2 Department of Cardiology, University Heart Center, 20251 Hamburg, Germany; kodolitsch@uke.de 3 Department of Paediatrics, Division of Paediatric Cardiology, Ghent University Hospital, 9000 Ghent, Centre for Medical Genetics, Ghent University Hospital, 9000 Ghent, Belgium; aura.muinomosquera@uzgent.be (L.M.‐M.); julie.debacker@ugent.be (J.D.B.) Department of Cardiology, University Heart Center, 20251 Hamburg, Germany; kodolitsch@uke.de Department of Paediatrics, Division of Paediatric Cardiology, Ghent University Hospital, 9000 Ghent, Belgium laura.muinomosquera@uzgent.be (L.M.‐M.); julie.debacker@ugent.be (J.D.B.) 2 Department of Cardiology, University Heart Center, 20251 Hamburg, Germany; kodolitsch@uke.de 3 Department of Paediatrics, Division of Paediatric Cardiology, Ghent University Hospital, 9000 Ghen Belgium laura.muinomosquera@uzgent.be (L.M.‐M.); julie.debacker@ugent.be (J.D.B.) 2 Department of Cardiology, University Heart Center, 20251 Hamburg, Germany; kodolitsch@uke.de 3 D t t f P di t i Di i i f P di t i C di l Gh t U i it H it l 9000 Gh t 4 Department of Cardiology, Ghent University Hospital, 9000 Ghent, Belgium 4 Department of Cardiology, Ghent University Hospital, 9000 Ghent, Belgium Correspondence: anthony.demolder@ugent.be 4 Department of Cardiology, Ghent University Hospital, 9000 Ghent, Belgium * Correspondence: anthony.demolder@ugent.be R i d 04 S t b 2020 A t d 24 S t b 2020 P bli h d 25 S t b 2020 4 Department of Cardiology, Ghent University Hospital, 9000 Ghent, Belgium * Correspondence: anthony.demolder@ugent.be Received: 04 September 2020; Accepted: 24 September 2020; Published: 25 September 2020 1. Introduction Marfan syndrome (MFS) is a systemic connective tissue disease with autosomal dominant inheritance and a reported prevalence ranging from 1.5 to 17.2 per 100,000 individuals [1]. Cardiovascular, ocular and skeletal organ systems are most frequently involved in the Marfan phenotype. The most common clinical manifestations include aortic dilatation, mitral valve prolapse, lens luxation and skeletal abnormalities (disproportionally long limbs, scoliosis and pectus deformities). Other manifestations can be found in the integumental, pulmonary and central nervous organ systems. A wide phenotypic variability reflects the different extent to which various organ systems can be affected [1,2]. Diagnosis is based on the revised Ghent nosology, including aortic root dilatation and lens luxation as the two cardinal manifestations (Table 1) [2]. Table 1. Revised Ghent criteria for diagnosis of MFS [2]. Table 1. Revised Ghent criteria for diagnosis of MFS [2]. In the Absence of Family History of MFS: (1) Ao * (Z‐score ≥ 2) AND EL = MFS (2) Ao * (Z‐score ≥ 2) AND causal FBN1 mutation = MFS (3) Ao * (Z‐score ≥ 2) AND systemic score ≥ 7 points = MFS (4) EL AND causal FBN1 mutation with known Ao = MFS In the Presence of Family History of MFS: (5) EL AND family history of MFS = MFS www.mdpi.com/journal/diagnostics Diagnostics 2020, 10, 751; doi:10.3390/diagnostics10100751 Diagnostics 2020, 10, 751 2 of 21 (6) Systemic score ≥ 7 points AND family history of MFS = MFS (7) Ao * (Z‐score ≥ 2 above 20 years old, ≥ 3 below 20 years) + family history of MFS = MFS * Ao = Aortic diameter at the sinuses of Valsalva above indicated Z‐score or aortic root dissection. EL = ectopia lentis; MFS = Marfan syndrome. (6) Systemic score ≥ 7 points AND family history of MFS = MFS (6) Systemic score ≥ 7 points AND family history of MFS = MFS (7) Ao * (Z‐score ≥ 2 above 20 years old, ≥ 3 below 20 years) + family history of MFS = MFS In the majority of patients, a (likely) pathogenic variant is found in the FBN1 gene, encoding the extracellular matrix glycoprotein fibrillin‐1, an important element in the assembly of microfibrils. Microfibrils may perform a structural role individually (in the extracellular matrix of elastic and non‐ elastic tissues), or unified as a supporting scaffold for elastin, thereby forming elastic fibers [3]. 1. Introduction Elastic fibers play a central role in the structural integrity of connective tissues (e.g., in the aorta) by providing elasticity and tensile strength. In addition to the structural role, fibrillin‐1 also plays a communicative role in biosignaling (regulating local bioavailability of TGF‐β) and mechanosignaling (by interacting with mechanosensors and providing feedback to regulate the response to hemodynamic changes). Therefore, defects in fibrillin‐1 may alter the structural integrity of connective tissue and may result in abnormal cellular signaling [3–5]. Life expectancy in patients with MFS is mainly determined by cardiovascular complications. Progressive dilatation of the proximal aorta is an important manifestation, rendering these patients at risk of aortic dissection or fatal rupture [6]. Although the aortic sinus is most commonly affected, aneurysms and dissections in more distal aortic regions and in extra‐aortic arteries can also occur [7,8]. The reported prevalence of aortic root dilatation is slightly lower in children compared to adults (approx. 80% vs. 90%) [9,10]. Furthermore, data from the Genetically Triggered Thoracic Aortic Aneurysms and Cardiovascular Conditions (GenTAC) registry indicate that adult males are more likely than females to have aortic root dilatation (92% vs. 84%), aortic regurgitation (55% vs. 36%), and to have undergone prophylactic aortic root replacement (47% vs. 24%) [10]. Increased awareness, early detection, careful follow‐up, life‐style adjustments, pharmacological treatment and prophylactic surgery are currently established as the cornerstones of treatment in MFS. Implementation of these aspects in the treatment strategy has shown to substantially reduce the risk of type A dissection [6,11]. In patients with known (or suspected) MFS, echocardiography plays a central role in the identification, severity assessment and follow‐up of cardiovascular abnormalities [6]. In recent years, heart failure and ventricular arrhythmia have drawn attention as additional cardiovascular manifestations of MFS [12]. Several imaging studies have provided data supporting a (sub)clinical, primary myocardial impairment in the absence of valvular disease or cardiovascular surgery in patients with MFS. In addition, studies using ambulatory ECG have demonstrated an increased susceptibility to ventricular arrhythmia [13,14]. These manifestations are also reflected in studies reporting on survival in patients with MFS, with heart failure and arrhythmia or sudden cardiac death (SCD) included as additional causes of death [14–16]. In this paper, we review current literature in order to provide insights in characteristics, pathophysiology and evolution of myocardial function, heart failure and ventricular arrhythmia in MFS. 2. Methods We conducted a systematic review in accordance with the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) statement to evaluate the current literature on myocardial function and arrhythmia in MFS [17]. Cohort studies, cross‐sectional studies, case‐control studies, case series and case reports were eligible for inclusion. First, a search of Medline and Embase was performed using the interchangeable search terms “Marfan syndrome”, “myocardial”, “ventricular”, “function”, “arrhythmia”, “heart failure” and “cardiomyopathy” in June 2020 (Figure 1). Next, a search in PubMed was performed to identify literature published ahead of print using the same search terms. Additional references were sought by examining citations in papers obtained through the specific searches. After deduplication, 154 papers were screened based on the title or abstract. A total of 56 full text papers were eligible for inclusion. Two studies included other genetic connective tissue disorders and were excluded since the population with MFS could not be discerned. The final selection consisted of 35 publications on myocardial function, eight publications on arrhythmia, eight publications reporting heart failure and SCD among other causes of death and one 3 of 21 3 of 21 Diagnostics 2020, 10, 751 publication on both myocardial function, ventricular arrhythmia and SCD. Of the 35 publications on myocardial function, 22 studied myocardial function clinically, nine were reports on heart transplantation in MFS (seven case reports, one case series and one survey), one reported on the incidence of dilated cardiomyopathy after cardiovascular surgery and four studies assessed myocardial function in murine models of MFS. Results from the literature search are presented in tables and figures. Extracted information included author, study design, year, studied population, assessment methods and findings in MFS. Results of the articles were grouped, narratively synthesized and integrated with other relevant publications. Figure 1. Flow chart of literature search. MFS: Marfan syndrome, SCD: Sudden Cardiac Death, HF: Heart failure. Figure 1. Flow chart of literature search. MFS: Marfan syndrome, SCD: Sudden Cardiac Death, HF: Heart failure. 3. Myocardial Involvement 3.1. Left Ventricular Dimensions and Function: Evidence Obtained from Echocardiographic Studies The first mention of possible myocardial involvement in MFS can be found in a case‐report by Fujiseki et al. in 1984 [18]. Since then, various independent research groups hypothesized that myocardial impairment could be part of the MFS phenotype. Results from the studies conducted are summarized in Table 2. Several studies were designed to investigate left ventricular (LV) dimensions and systolic function in patients with MFS using echocardiography, with almost all of these studies excluding patients with significant valvular disease or previous aortic surgery [14,19–23]. Although in some of the early studies assessing LV dimensions and LV function [19–22], myocardial involvement was not clearly evidenced, subsequent studies reported the presence of increased LV dimensions in 7 to 68% (depending on the definition and the cohort), with mildly impaired LV systolic function (fractional shortening (FS) < 30%) present in approx. 10% of the patients [14,23]. In addition, mild impairment of diastolic function was demonstrated in multiple echocardiographic and cardiac magnetic resonance imaging (CMR) studies in adults and children with MFS [20,24–26]. Based on the coexistence of decreased ventricular compliance and reduced active myocardial relaxation, it was hypothesized that the impaired diastolic properties are attributable to reduced elastic recoil due to underlying connective tissue alterations [13]. 1 limits %)) FS ≥ 2 ble 2. Summary of studies assessing myocardial function in MFS. 3. Myocardial Involvement Controls Assessment Findings in MFS 59 age‐ and sex‐matched controls 59 age‐ and sex‐matched subjects with MVP M‐mode 2D echo Similar LV diameter and systolic function Increased LV mass 22 age‐matched healthy children M‐mode Doppler CMR Similar LV diameter and systolic function LV diastolic dysfunction 28 healthy, age and gender‐matched controls M‐mode 2D echo Doppler Similar LV diameter and systolic function LV diastolic dysfunction / 2D echo 68% had LV dilatation (LVEDD Z‐score > 2) 11% had LV systolic dysfunction (FS < 30%) / M‐mode 19% had LV dilatation (LVEDD > 95% above normal limi No change in LV dimensions during follow‐up No LV systolic dysfunction / M‐mode 2D echo 9% had mild LV dysfunction (FS 25–30%) 7% had LV dilatation (LVEDD > 117% (2SD + 5%)) 3% developed LV dilatation during follow‐up 1% had LVEDD > 112% and FS < 30% 26 age and sex‐matched controls 2D echo Doppler TDI CMR LV systolic and diastolic dysfunction LV dilatation 40 age and sex‐matched controls M‐mode Doppler Similar systolic function LV diastolic dysfunction LV dilatation 86 healthy controls 2D echo Doppler TDI LV diastolic dysfunction LV systolic dysfunction 61 healthy controls M‐mode 2D echo Doppler TDI 17% had LV dilatation (predicted LVEDD ≥ 112% and FS ≥ LV diastolic dysfunction LV systolic dysfunction 61 age, sex, height, weight, and BSA‐ matched normal volunteers M‐mode 2D echo RV systolic dysfunction RV dilatation 5 of Doppler TDI Increased right atrium area 73 age, sex, and BSA‐matched controls 2D echo Doppler TDI LV diastolic dysfunction RV diastolic dysfunction Atrial systolic and diastolic dysfunction / CMR 25% had reduced LV EF (below 95% CI for sex and age decile) 10% had reduced RV EF (below 95% CI for sex and age decile) LV dilatation RV dilatation 49 controls without significant differences in age, sex, height, weight, and BSA M‐mode 2D echo Doppler Strain rate imaging 20% had LV dilatation (predicted LVEDD ≥ 112% and FS ≥ 25%) LV diastolic dysfunction LV systolic dysfunction 19 healthy controls CMR 9% had reduced LV EF (<45%) LV systolic dysfunction RV systolic dysfunction 50 controls matched for age, sex, and BSA M‐mode 2D echo Doppler Strain rate imaging Similar LV and RV EF LV systolic dysfunction RV systolic dysfunction No changes in systolic or diastolic function during follow‐up 26 controls without significant differences in sex, race, age, weight, height, and BSA M‐mode 2D echo Doppler Strain rate imaging LV systolic dysfunction No significant differences in strain 40 age‐matched healthy controls M‐mode 2D and 3D echo Doppler 3D speckle tracking LV diastolic dysfunction LV systolic dysfunction Left atrial diastolic dysfunction No differences in left atrial systolic function No differences in M‐mode LVEDD, LVESD and FS 19 age and sex‐matched controls 2D echo Doppler TDI No changes in LV dimensions during follow‐up No changes in LV systolic or diastolic function during follow‐up 339 patients referred for suspected MFS (diagnosis ruled out according to the Gh. Diagnostics 2020, 10, 751 Diagnostics 2020, 10, 751 7 of 21 Discrepancies between some of the previously mentioned studies led to further debate on myocardial involvement in MFS. A small number of studies reported no differences in LV diameters or systolic function compared to controls. Furthermore, some follow‐up studies failed to detect changes in LV function or dimensions over time while others identified a small, yet significant, subgroup of patients presenting increased LV dimensions and reduced LV function [13,14,19–23,25]. These discrepancies can be attributed to (i) the small subgroup of patients affected, (ii) the mild degree of impairment in almost all of these affected patients and (iii) the lack of a uniform definition regarding myocardial involvement. Moreover, subtle changes in LV function are more difficult to identify and differentiate when using conventional (2D and M‐mode) echocardiography as compared to more sensitive and advanced imaging techniques. 3.2. Evidence Obtained from Advanced Imaging Techniques Additional insights in the LV function and volumes were provided in subsequent studies using tissue Doppler imaging (TDI), CMR and strain imaging. In 2006, De Backer et al. assessed diastolic filling in a small case‐control study using TDI and systolic function using CMR in patients with MFS [24]. Compared to age‐ and sex‐matched controls, patients with MFS showed signs of mild LV contractile dysfunction as expressed by a reduced ejection fraction (EF) (53.5 ± 9.0% vs. 59.6 ± 6.7%, p = 0.009), an increased indexed end‐systolic volume (36.0 ± 9.5 vs. 29.5 ± 6.7 mL/m2, p = 0.007), and reduced peak systolic velocities. Furthermore, impaired diastolic function was observed in MFS [24]. Soon after, these findings were confirmed by two larger case‐control studies [26,27]. g y g Three subsequent studies using CMR provided additional evidence for myocardial involvement, as demonstrated by the observation of a reduced LV EF in a subgroup of patients [28–30]. In 2010, Alpendurada et al. evaluated 68 patients with MFS without significant valvular disease or prior cardiovascular surgery [28]. In this study, 25% of the patients had reduced LV EF on CMR. The reduced LV EF found in patients with MFS was mild, being less than 10% below the 95% confidence interval (CI) for sex and age reference values in most of the cases. Only 2 patients (2.9%) were diagnosed with heart failure in this study [28]. The relatively high rate of reduced EF in MFS patients in this study, could be attributed to the proposed cut‐off value for reduced LV EF (below the 95% CI for sex and age decile). No association was found between reduced LV EF and age, gender, indexed aortic dimensions, presence of mitral valve prolapse or valve regurgitation, providing additional evidence that the impairment in ventricular function is inherent to the underlying connective tissue abnormality in MFS. Similar findings were observed in the CMR studies by de Witte et al. and by Winther et al., confirming that the reduced LV EF, is mostly mild but might affect a subgroup of patients in a more severe way [29,30]. By extending the detection prowess of conventional echocardiography, studies utilizing strain and strain rate imaging to assess and quantify changes in global and regional contractile function have confirmed the findings obtained from CMR [30–33]. 3. Myocardial Involvement nosology) M‐mode 2D echo Doppler Increased NT‐proBNP levels LV diastolic dysfunction LV dilatation nction 1 6 of signs of LV systolic dysfunction ortic stiffness index in MFS and ns‐TAAD ed LV end‐systolic elastance in MFS scular coupling index was abnormal in MFS ifference in LV stroke work in MFS 2% had reduced LV EF (≤55%) LV systolic dysfunction ; FS = Fractional shortening; LV = Left valve, myopia, Aorta, Skin and Skeletal neurysm and dissection; TDI = Tissue signs of LV syst rtic stiffness ind d LV end‐systol cular coupling i ference in LV st % had reduced LV systolic dy FS = Fraction valve, myopia, neurysm and d Diagnostics 2020, 10, 751 4.1. Intrinsic vs. Stressor‐Induced Problem 4.1. Intrinsic vs. Stressor‐Induced Problem Although almost all of the aforementioned studies reported myocardial impairment in the absence of previous cardiac surgery or significant valvular disease, very few patients were diagnosed with clinical heart failure. In contrast, there are several reports on end‐stage heart failure necessitating heart transplantation in patients with MFS (Table 3). In addition, heart failure is mentioned as one of the leading causes of death in MFS (see further) [15,16,40–49]. Whether myocardial impairment and development of heart failure in MFS is a primary intrinsic problem or a secondary, stressor‐induced problem remains an unanswered question. ble 3. Summary of studies reporting end‐stage heart failure necessitating heart transplantation in Table 3. Summary of studies reporting end‐stage heart failure necessitating heart transplantation in MFS. Table 3. Summary of studies reporting end‐stage heart failure necessitating heart transplantation MFS. Author Type of Study Number of Patients with MFS Prior Aortic Surgery Kesler et al. 1994 [42] Survey 11 Not stated Mullen et al. 1996 [43] Case report 1 Yes Varghese et al. 2006 [44] Case report 1 Yes Botta et al. 2006 [45] Case report 1 Yes Knosalla et al. 2007 [46] Case series 10 Yes (100%) Rajagopal et al. 2009 [47] Case report 1 Yes Audenaert et al. 2015 [49] Case report 1 Yes Rao et al. 2018 [40] Case report 1 Yes Ogawa et al. 2019 [41] Case report 1 No MFS = Marfan syndrome. 3.3. Involvement of the Right Ventricle and Atria Most studies have focused on the LV, but right ventricular (RV) involvement in MFS has also been suggested [34,35]. In the study by Kiotsekoglou et al., significant differences were found in tricuspid annular plane systolic excursion (TAPSE), rate of pressure rise (dp/dt) and pulsed TDI early filling measurements obtained over the lateral tricuspid valve corner, indicating impairment of RV function. In addition, atrial involvement was evidenced by reduced contractile, reservoir, and conduit function parameters for both atria [34,35]. The involvement of RV function was confirmed in the aforementioned CMR study by Alpendurada et al., showing that 10.3% of the patients also had a reduced RV EF [28]. Similarly, in the study by de Witte et al., RV EF was reduced compared to healthy controls (51% ± 7% vs. 56% ± 4%, p < 0.005) [29]. In both these studies, LV EF and RV EF were found to be strongly correlated, but the RV appears to be less frequently affected, possibly due to the higher workload imposed on the LV [28,29]. Diagnostics 2020, 10, 751 8 of 21 4.2. Valvular Disease, Surgery and Genotype‐Phenotype Relation Both studies reported that the observed impairment of LV function was independent of aortic stiffness (based on measurable derivatives of aortic elasticity) [29,39]. However, confirmation and longitudinal data on these findings are required. A decrease in aortic distensibility can also be observed after aortic surgery during which a Dacron tube is implanted [60]. The difference in compliance of the Dacron implant compared to the original aorta may be higher than the difference in aortic compliance between patients with MFS and healthy individuals. A Dacron implant could lead to a slight but significant increase in LV afterload and thereby result in long‐term cardiac stress. This hypothesis is supported by data reported by Nollen et al., demonstrating significantly lower distensibility in the tube graft compared to ascending aortic distensibility in patients without aortic root replacement [60]. Combined with the presence of a primary impairment of myocardial function in some patients, this could contribute to the prevalence of heart failure observed during long‐term follow‐up in patients after aortic surgery. Therefore, based on current data, a modulating role of aortic root replacement seems plausible [61]. In addition to these hemodynamic factors, it is conceivable that intrinsic factors play a role, including gene‐related factors. Several studies have already shown that the type of underlying FBN1 gene variant has an influence on aortic outcome [62–66]. Patients harbouring variants predicted to result in haploinsufficiency (HI) of fibrillin‐1 show a worse outcome than carriers of variants predicted to result in a dominant negative (DN) effect [67]. In the same vein, a genotype‐phenotype relationship in the myocardium can be suspected. Two studies have indicated a genotype‐phenotype relation between myocardial impairment and underlying FBN1 gene variants [33,68]. Aalberts et al. have shown an association between the type of underlying pathogenic FBN1 variant and the development of LV dilatation in MFS [68]. Patients carrying non‐missense variants (predicted to result in HI) more often demonstrated LV dilatation than those carrying missense variants (predicted to result in DN effect) [68]. Similarly, in a smaller study by Rahman et al. using three‐dimensional speckle tracking echocardiography, LV EF, global LV circumferential strain and global LV area strain were all significantly lower in patients with variants predicted to result in HI than in those variants predicted to result in DN effect (p < 0.05) [33]. 4.2. Valvular Disease, Surgery and Genotype‐Phenotype Relation One of the cardiac stressors which may contribute in the development of heart failure is regurgitant valvular disease, which is frequently encountered in MFS and may induce volume overload [50,51]. By the age of 30, more than half of the patients with MFS will have mitral valve regurgitation, with severe mitral valve regurgitation reported in up to 12% of the patients [51]. Aortic valve regurgitation attributed to the dilatation of the aortic valve annulus is observed in up to 1 in 3 adult patients [50]. Since the prevalence of valvular disease tends to increase with age, it is likely that some patients with MFS will face some form of chronic volume overload caused by aortic and/or mitral regurgitation, inducing enlargement of the LV end‐diastolic and end‐systolic volume which may not be adequately compensated in some patients [51]. An association between prior aortic or valvular surgery and the development of heart failure as long‐term complication in patients with MFS has also been suggested [48,52–54]. In several case reports and one case series describing patients with MFS undergoing orthotopic heart transplantation, almost all patients had a history of prior aortic or valvular surgery (Table 3) [40–47,49]. Similarly, a study by Hetzer et al. on a cohort of 421 patients with MFS who had undergone cardiac surgery reported cardiomyopathy in 11.2%. Only a minority of them already showed evidence of cardiomyopathy before the procedure. Even though occurrence of cardiomyopathy appeared to be independent of the type of myocardial protection and duration of ischemia, this study suggests that the performance of cardiovascular surgery on its own plays a role in the development of cardiomyopathy in patients with MFS [48]. Although these data point towards a relationship between heart failure and prior aortic surgery, it is also possible that this association reflects a subgroup of patients demonstrating a more severe phenotype, including a more vulnerable myocardium. A third potential cardiac stressor is decreased aortic distensibility, which may contribute to impairment of LV function by altering the hemodynamic load imposed on the LV [55]. Aortic elasticity has been shown to be decreased in patients with MFS [56–59]. The relation between aortic elasticity and LV function has been assessed in the study by de Witte et al. using CMR and in the study by 9 of 21 Diagnostics 2020, 10, 751 Loeper et al. using echocardiography. 4.2. Valvular Disease, Surgery and Genotype‐Phenotype Relation Different types of FBN1 gene variants may also have a different effect on aortic elasticity [67], thereby potentially further contributing to impairment of myocardial function, but this has not been studied yet. In the aforementioned studies by de Witte et al. and Loeper et al., the relation between aortic elasticity and predicted HI or DN effect of FBN1 gene variants was not evaluated [29,39]. Future studies should assess this relationship as the field of genotype‐phenotype correlations may hold valuable information with implications for personalized therapeutic approaches [67]. 4.4. Proposed Hypothesis Fibrillin‐1 and microfibrils can be found throughout the myocardium as components of the extracellular matrix. They are assumed to play a role in sustaining proper cardiac function by contributing to the diastolic and systolic properties of the myocardium [72–76]. Underlying abnormalities in the FBN1 gene are thought to result in abnormal mechanosignaling of the microfibrils which may cause inadequate compensation of cardiac stressors such as volume‐ or pressure overload in a subgroup of patients with MFS. Whether impaired elastic fiber function and/or impaired biosignaling is reflected in increased TGF‐β signaling remains to be elucidated [37]. It is possible that a combination of intrinsic abnormalities (possibly variant‐specific) renders the heart more vulnerable to cardiac stressors, resulting in an increased likelihood to develop myocardial dysfunction and ultimately heart failure (Figure 2). Figure 2. Proposed factors implicated in “Marfan cardiomyopathy”. FBN1: fibrillin‐1 gene, TGF‐β: transforming growth factor‐beta, LV: Left ventricular. Figure 2. Proposed factors implicated in “Marfan cardiomyopathy”. FBN1: fibrillin‐1 gene, TGF‐β: transforming growth factor‐beta, LV: Left ventricular. 4.3. Evidence Obtained from Mouse Models In the quest to unravel the pathophysiology underlying Marfan cardiomyopathy, mouse models have provided clues to possible underlying mechanisms and pathways [37,69–71]. The presence and extent of fibrillin networks in the LV has been evidenced in both human and mouse studies [72–74]. Findings reported by Steijns et al. indicate that in wild‐type mice, fibrillin‐1 is present in different regions of the myocardium, including the apex, mid‐ventricles and the atria [75]. These findings suggest that the mechanism of fibrillin‐1 deficiency most likely also underlies the reported atrial and biventricular involvement. In the Fbn1C1039G/+ mouse model studied by Campens et al., impairment of cardiac function and structure remained mild and subclinical, resembling the myocardial phenotype observed in patients with MFS. Histologic examination of the myocardium revealed upregulation of TGF‐β‐related pathways and consistent abnormalities of the microfibrillar network, implicating a role for microfibrils in the mechanical properties of the myocardium [37]. In the fibrillin‐1 deficient Fbn1mgR/mgR mouse model Cook et al. demonstrated that abnormal mechanosignaling by cardiomyocytes resulting from a deficient extracellular matrix caused dilated cardiomyopathy. The authors suggested that fibrillin‐1 is implicated in the physiological adaptation of the myocardium to an increased workload and that dilated cardiomyopathy is a primary manifestation of MFS mice [69]. Diagnostics 2020, 10, 751 Diagnostics 2020, 10, 751 10 of 21 10 of 21 Two studies using the Fbn1C1039G/+ mouse model tested the hypothesis that either pressure or volume overload on an already susceptible heart could result in a more severe dilated cardiomyopathy [70,71]. Valvular regurgitation and transverse aortic constriction ligation were shown to provoke dilated cardiomyopathy, while wild type controls remained fully compensated [70,71]. Taken together, these studies demonstrate the role of fibrillin‐1 contributing to the cardiac reserve of the LV in the setting of cardiac stress [70,71]. Arrhythmia in Marfan Syndrome In addition to aortic complications and cardiomyopathy, arrhythmia should be recognized as a relevant manifestation of the cardiac phenotype observed in MFS [14,52,54]. Several studies have associated MFS with an increased risk of arrhythmia, as summarized in Table 4. Studies based on data from ambulatory ECG in adults have demonstrated the presence of significant ventricular ectopy (defined as >10 premature ventricular contractions per hour) in 20–35% [14,77,78]. In children with MFS, the reported frequency of ventricular arrhythmia is much lower (7% demonstrating ventricular ectopy) [79]. Similarly, non‐sustained ventricular tachycardia (NSVT) is reported in 10–20% of the adult patients with MFS and appears to be very rare in children [14,77,78,80]. However, ventricular tachycardia (VT) and SCD have been reported in both adults and children with MFS [14,81]. Four studies reported life‐threatening arrhythmias in 7–9% of the patients and SCD, most likely due to arrhythmia, occurred in up to 4% [14,77,82,83]. Furthermore, fatal arrhythmias are reported in 12–19% of patients with MFS after aortic surgery, making it the 2nd most frequent cause of death in this setting [52–54]. of 2 n repola VF or A ns in e VF an Table 4. Summary of studies assessing arrhythmia in MFS. Controls Assessment Findings in MFS / M‐mode Resting ECG 33% presents at least 1 PVC on resting ECG 13% had VT QTc / QTUc prolongation was associated with ventricular arrhythm Combination of abnormal repolarization and MVP associated with ven arrhythmias ealthy age and atched controls M‐mode and 2D echo Ambulatory ECG Higher median number of PACs than controls (12/24 h vs. 6/24 h; p < Higher median number of PVCs than controls (17/24 h vs. Diagnostics 2020, 10, 751 Diagnostics 2020, 10, 751 13 of 21 13 of 21 The mechanisms underlying severe ventricular arrhythmia are multifactorial. Ventricular tachycardia and ventricular fibrillation (VF) usually arise from an initiating trigger in the presence of a proarrhythmogenic substrate (as observed in genetic channelopathies and cardiomyopathies), which allows the perpetuation of severe ventricular arrhythmias [84]. In patients with MFS, a proarrhythmogenic substrate may be present since subtle ECG changes have been identified independent of aortic root diameter, mitral and/or tricuspid valve prolapse or chamber dimension and function. Prolonged atrio‐ventricular conduction time and altered depolarization is suggested by longer PQ‐ and QTc‐intervals compared to healthy controls [80]. A (mildly) prolonged QTc‐interval (>440 ms) has been described in 16–20% and 9–20% of adults and children respectively, while almost no patients present with QTc‐intervals >500 ms [14,80,81,85]. The relevance of these subtle ECG changes remains understudied, but longer QTc‐intervals have been associated with ventricular arrhythmia in MFS [14]. Patients with mitral valve prolapse, mitral valve regurgitation and previous aortic surgery have been reported to have higher risk of ventricular arrhythmia in several studies [14,52–54,77,78]. Furthermore, decreased LV EF, increased LV dimensions, prolonged QTc/QTu interval and high levels of N‐terminal pro b‐type natriuretic peptide (NTproBNP) have also been associated with ventricular ectopy independently of valvular disease and aortic surgery [14,77,82]. However, with the exception of NTproBNP, all these factors failed to predict severe arrhythmic events [77,82]. In addition, one study found an association between abnormal heart rate turbulence parameters and VT [83]. Genotype‐ phenotype correlations for arrhythmia have been performed in several studies, but only an association between mutations in exons 24–32 and VT/SCD has been found thus far [77,78,83]. These findings show that predictors of ventricular ectopy can be found, but the factors associated with severe arrhythmic events are more elusive, most likely because of the small number of patients experiencing severe arrhythmic events. Arrhythmia in Marfan Syndrome 1/24 h; p < More frequently repolarization abnormalities than controls Longer PQ‐ and QTc‐intervals compared to controls 11% had NSVT° / 2D echo Resting ECG Ambulatory ECG 21% had ventricular ectopy (defined as >10 PVC/h) 6% had NSVT° 4% died from arrhythmias 16% had QTc prolongation and 60% had QTU prolongation Ventricular ectopy associated with LV size, MVP, and abnormalities of rep / 2D echo, Doppler and TDI Resting ECG and SAECG Ambulatory ECG 9% reached the composite endpoint (SCD, VT, VF or AS) 7% had VT 3% had SCD / M‐mode and 2D echo Doppler Resting ECG Ambulatory ECG 91% had PVCs with 35% having >10 PVC/h 11% had NSVT° 8% had ventricular tachycardia events (SCD, VT, VF 4% had SCD Ventricular tachycardia events associated with NTproBNP and mutations 32 / 2D echo Ambulatory ECG Heart rate turbulence 12% reached the primary endpoint (SCD, survived cardiac arrest, VT/VF 9% had VT 3% had SCD ge, BSA, sex‐ ched controls M‐mode and 2D echo Resting ECG Longer QTc intervals than controls / M‐mode and 2D echo Ambulatory ECG 7% had ventricular ectopy (defined as >10 PVC/h) 5% had supraventricular ectopy (defined as >10 PAC/h) 1% had both supraventricular and ventricular ectopy None had VT or supraventricular tachycardia of 2 VT° 1 ols (11/24 h vs. 2/24 h; p < 0.001) rols (8/24 h vs. 0/24 h; p < 0.001) T° ndependently associated with NS emature atrial complex; PVC maging; TVP = Tricuspid valv the U‐wave (if >50% of T‐wav Median ; ( ) = Range. ls (11/24 h vs. 2/24 ols (8/24 h vs. 0/24 T° dependently asso mature atrial co aging; TVP = Tr he U‐wave (if >5 Median ; ( ) = Ran Diagnostics 2020, 10, 751 6. Heart Failure and Arrhythmia as Additional Causes of Death Most of the improvement in life expectancy achieved throughout the years in patients with MFS was obtained by focusing on prophylactic treatment and prevention of aortic events [16]. In 1995, Silverman et al. showed that the mean age at death of patients with MFS had significantly increased compared to the mean age at death in 1972 (41 ± 18 years versus 32 ± 16 years, p = 0.0023). Furthermore, patients undergoing aortic surgery after 1980 demonstrated even longer life expectancy [16]. Heart failure and arrhythmia as causes of mortality have been reported in later studies [14–16,86–90]. In Figure 3, cardiovascular causes of death in MFS are displayed as percentages of the total amount of deaths (non‐cardiovascular causes of death were omitted). 14 of 21 Diagnostics 2020, 10, 751 Figure 3. Overview of studies reporting causes of death in patients with MFS. SCD: Sudden cardiac death. Figure 3. Overview of studies reporting causes of death in patients with MFS. SCD: Sudden cardiac death. As shown in Figure 3, the percentage of deaths attributed to heart failure and SCD (presumed arrhythmogenic) is approx. 30% and 20% respectively, with the exception of one study by Yetman et al. reporting SCD as the only cause of death in their patient cohort [15]. Although a wide variation in the reported numbers can be noted, it seems that, as the treatment and prevention strategies for aortic complications continue to improve, heart failure and arrhythmia constitute important cardiac manifestations requiring attention and awareness. When looking more specifically at post‐operative survival in patients with MFS with prior aortic surgery, the three largest studies report heart failure and arrhythmia among the major causes of death [52–54]. Since these three studies were performed by the same research group, a significant overlap should be taken into account when considering these numbers. In their most recent study in 2009, Cameron et al. reported results after aortic root replacement in 373 patients with MFS in a time period of more than 30 years [54]. In these three studies, dissection or rupture of the residual aorta remains the main cause of death, occurring in up to 19% of the patients. When looking at the extra‐aortic causes of death, arrhythmia stands out, occurring in 12–19% of the patients and thereby rivaling aortic events as the leading cause of death. 6. Heart Failure and Arrhythmia as Additional Causes of Death Heart failure on the other hand is reported as the cause of death in 2–10% of the patients, which appears to be less frequent compared to data from aforementioned studies reporting on survival in MFS. 7.2. The Intertwined Mechanism of Marfan Cardiomyopathy and Ventricular Arrhythmia 7.2. The Intertwined Mechanism of Marfan Cardiomyopathy and Ventricular Arrhythmia The relation between a reduced amount or quality of extracellular fibrillin in the myocardium, a primary impairment of myocardial function, increased likelihood of ventricular ectopy and possible alterations in the electrophysiological substrate remains unclear. It is possible that, due to the reduced amount or quality of fibrillin, mechanical forces imposed on the cardiomyocytes in patients with MFS may be less adequately compensated than in healthy individuals. Therefore, chronic or acute myocardial dilatation and associated stretch could perhaps induce (complex) ventricular ectopy more easily in these patients where subclinical myocardial impairment is noted. In addition, the impairment of myocardial function observed in some patients may also signify inherent abnormalities in the underlying electrophysiological substrate. The combination of (complex) ventricular ectopy together with the alterations in electrical and/or mechanical properties of the heart may be severe enough to induce SCD in some patients with MFS, as suggested in studies by Hoffmann et al. [82] and by Yetman et al. [14]. Furthermore, increased NT‐proBNP has been demonstrated as independent predictor of both diastolic dysfunction and severe arrhythmic events [38,82]. This may signify that long‐term mild myocardial stretch potentially predisposes these patients to (severe) ventricular arrhythmia [38,82]. 8. Current Limitations and Evidence Gaps To date, large multicentre studies reporting the overall incidence or prevalence of heart failure and severe arrhythmia in MFS are lacking. Therefore, identification of predisposing factors is limited. Additional studies are necessary to evaluate the clinical relevance of Marfan cardiomyopathy and ventricular ectopy, to elucidate the underlying mechanisms in MFS and to allow better risk stratification of patients with MFS. Information on these aspects could hold important implications for developing strategies to treat heart failure and ventricular arrhythmia in MFS. We should also take into account that—certainly in the case of older studies—some of the patients enrolled may have had some other form of Heritable Thoracic Aortic Disease, caused by pathogenic variants in genes other than FBN1. Advancing insight in recent years shows us that caution is advised in grouping all these conditions. 7.1. Current View on Marfan Cardiomyopathy Taken together, the aforementioned studies confirm that ventricular dimensions as well as systolic and diastolic function are well within normal limits in the vast majority of patients with MFS. However, even in the absence of cardiac surgery or significant valvular disease, a mild biventricular dilatation with diastolic and systolic dysfunction in a subgroup of patients with MFS has been repeatedly reported [20,23–26,28,35]. Since myocardial involvement was reported in the absence of any cardiac surgery or significant valvular abnormalities, this phenotypic expression was designated 15 of 21 15 of 21 Diagnostics 2020, 10, 751 an “intrinsic” or “inherent” dysfunction of the myocardium and was termed “Marfan cardiomyopathy”. Advanced imaging techniques (such as CMR, TDI, strain and strain rate imaging) appear to be more suited to detect these alterations. Despite these findings, almost no patients were diagnosed with clinical heart failure in the aforementioned studies. Follow‐up studies to better identify those patients at risk of clinically relevant myocardial dysfunction are still required. 9. Conclusions Myocardial involvement in the absence of valvular disease can be observed in patients with MFS, usually presenting as mild, asymptomatic impairment of LV systolic and diastolic function. In addition, some patients with MFS present (complex) ventricular arrhythmia as well as alterations in repolarization. A subgroup of patients with MFS tends to develop heart failure, severe arrhythmia and SCD, in which the effects of cardiac stressors may play an important role. Reduced myocardial function, heart failure and ventricular arrhythmia should be considered an essential concern of medical care for patients with MFS. Careful assessment of these features should be added to the standard aortic evaluation. Funding: This research received no external funding. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest. Abbreviations 16 of 21 Diagnostics 2020, 10, 751 FBN1 Fibrillin‐1 FS Fractional shortening HF Heart failure HI Haploinsufficiency LV Left ventricular LVEDD Left ventricular end‐diastolic diameter LVESD Left ventricular end‐systolic diameter MFS Marfan syndrome MVP Mitral valve prolapse NSVT Non‐sustained ventricular tachycardia NTproBNP N‐terminal pro b‐type natriuretic peptide PAC Premature atrial complex PVC Premature ventricular complex RV Right ventricular SCD Sudden cardiac death TDI Tissue Doppler imaging TGF‐β Transforming growth factor beta VF Ventricular fibrillation VT Ventricular tachycardia References 1. Von Kodolitsch, Y.; De Backer, J.; Schüler, H.; Bannas, P.; Behzadi, C.; Bernhardt, A.M.; Hillebrand, M.; Fuisting, B.; Sheikhzadeh, S.; Rybczynski, M.; et al. Perspectives on the revised Ghent criteria for the diagnosis of Marfan syndrome. Appl. Clin. Genet. 2015, 137–155, doi:10.2147/tacg.s60472. 1. Von Kodolitsch, Y.; De Backer, J.; Schüler, H.; Bannas, P.; Behzadi, C.; Bernhardt, A.M.; Hillebrand, M.; Fuisting, B.; Sheikhzadeh, S.; Rybczynski, M.; et al. Perspectives on the revised Ghent criteria for the diagnosis of Marfan syndrome. Appl. Clin. Genet. 2015, 137–155, doi:10.2147/tacg.s60472. 2. Loeys, B.; Dietz, H.C.; Braverman, A.C.; Callewaert, B.; De Backer, J.; Devereux, R.B.; Hilhorst‐Hofstee, Y.; Jondeau, G.; Faivre, L.; Milewicz, D.M.; et al. The revised Ghent nosology for the Marfan syndrome. J. Med. Genet. 2010, 47, 476–485, doi:10.1136/jmg.2009.072785. 2. Loeys, B.; Dietz, H.C.; Braverman, A.C.; Callewaert, B.; De Backer, J.; Devereux, R.B.; Hilhorst‐Hofstee, Y.; Jondeau, G.; Faivre, L.; Milewicz, D.M.; et al. The revised Ghent nosology for the Marfan syndrome. J. Med. Genet. 2010, 47, 476–485, doi:10.1136/jmg.2009.072785. 3. Ramirez, F.; Caescu, C.; Wondimu, E.; Galatioto, J. Marfan syndrome; A connective tissue disease at the crossroads of mechanotransduction, TGFβ signaling and cell stemness. Matrix Boil. 2018, 82–89, doi:10.1016/j.matbio.2017.07.004. 3. Ramirez, F.; Caescu, C.; Wondimu, E.; Galatioto, J. Marfan syndrome; A connective tissue disease at the crossroads of mechanotransduction, TGFβ signaling and cell stemness. Matrix Boil. 2018, 82–89, doi:10.1016/j.matbio.2017.07.004. 4. Humphrey, J.D.; Schwartz, M.A.; Tellides, G.; Milewicz, D.M. Role of mechanotransduction in vascular biology: Focus on thoracic aortic aneurysms and dissections. Circ. Res. 2015, 116, 1448–1461, doi:10.1161/CIRCRESAHA.114.304936. 4. Humphrey, J.D.; Schwartz, M.A.; Tellides, G.; Milewicz, D.M. Role of mechanotransduction in vascular biology: Focus on thoracic aortic aneurysms and dissections. Circ. Res. 2015, 116, 1448–1461, doi:10.1161/CIRCRESAHA.114.304936. 5. Chaudhry, S.S.; Cain, S.A.; Morgan, A.; Dallas, S.L.; Shuttleworth, C.A.; Kielty, C.M. Fibrillin‐1 regulates the bioavailability of TGFβ1. J. Cell Boil. 2007, 176, 355–367, doi:10.1083/jcb.200608167. 6. Child, A.H. Diagnosis and management of Marfan syndrome; 2016; ISBN 9781447154426. 7. Abbreviations Mariucci, E.; Lovato, L.; Rosati, M.; Palena, L.M.; Bonvicini, M.; Fattori, R. Dilation of peripheral vessels in Marfan syndrome: Importance of thoracoabdominal MR angiography. Int. J. Cardiol. 2013, 167, 2928–2931, doi:10.1016/j.ijcard.2012.08.001. 8. Schönhoff, F.; Yildiz, M.; Langhammer, B.; Jungi, S.; Wyss, T.R.; Makaloski, V.; Schmidli, J.; Carrel, T. The fate of nonaortic arterial segments in Marfan patients. J. Thorac. Cardiovasc. Surg. 2019, 157, 2150–2156, doi:10.1016/j.jtcvs.2018.10.089. 9. Stheneur, C.; Faivre, L.; Collod‐Beroud, G.; Gautier, E.; Binquet, C.; Bonithon–Kopp, C.; Claustres, M.; Child, A.H.; Arbustini, E.; Adès, L.C.; et al. Prognosis Factors in Probands with an FBN1 Mutation Diagnosed Before the Age of 1 Year. Pediatr. Res. 2011, 69, 265–270, doi:10.1203/pdr.0b013e3182097219. 10. Roman, M.J.; Devereux, R.B.; Preiss, L.R.; Asch, F.M.; Eagle, K.A.; Holmes, K.W.; Lemaire, S.A.; Maslen, C.L.; Milewicz, D.M.; Morris, S.A.; et al. Associations of Age and Sex with Marfan Phenotype. Circ. Cardiovasc. Genet. 2017, 10, doi:10.1161/CIRCGENETICS.116.001647. 11. Milleron, O.; Arnoult, F.; Delorme, G.; Detaint, D.; Pellenc, Q.; Raffoul, R.; Tchitchinadze, M.; Langeois, M.; Guien, C.; Beroud, C.; et al. Pathogenic FBN1 Genetic Variation and Aortic Dissection in Patients with Marfan Syndrome. J. Am. Coll. Cardiol. 2020, 75, 843–853, doi:10.1016/j.jacc.2019.12.043. 12. Von Kodolitsch, Y.; Demolder, A.; Girdauskas, E.; Kaemmerer, H.; Kornhuber, K.; Mosquera, L.M.; Morris, S.; Neptune, E.; Pyeritz, R.; Rand‐Hendriksen, S.; et al. Features of Marfan syndrome not listed in the Ghent nosology—The dark side of the disease. Expert Rev. Cardiovasc. Ther. 2019, 17, 883–915, doi:10.1080/14779072.2019.1704625. 17 of 21 17 of 21 Diagnostics 2020, 10, 751 13. Kiotsekoglou, A.; Sutherland, G.R.; Moggridge, J.C.; Nassiri, D.K.; Camm, A.J.; Child, A.H. The unravelling of primary myocardial impairment in Marfan syndrome by modern echocardiography. Heart 2009, 95, 1561– 1566, doi:10.1136/hrt.2008.152934. 14. Yetman, A.T.; Bornemeier, R.A.; McCrindle, B.W. Long‐term outcome in patients with Marfan syndrome: Is aortic dissection the only cause of sudden death? J. Am. Coll. Cardiol. 2003, 41, 329–332, doi:10.1016/s0735‐ 1097(02)02699‐2. 15. Diller, G.‐P.; Kempny, A.; González, R.A.; Swan, L.; Uebing, A.; Li, W.; Babu‐Narayan, S.; Wort, S.J.; Dimopoulos, K.; Gatzoulis, M.A. Survival Prospects and Circumstances of Death in Contemporary Adult Congenital Heart Disease Patients Under Follow‐Up at a Large Tertiary Centre. Circulation 2015, 132, 2118– 2125, doi:10.1161/circulationaha.115.017202. 16. Silverman, D.I.; Burton, K.J.; Gray, J.; Bosner, M.S.; Kouchoukos, N.T.; Roman, M.J.; Boxer, M.; Devereux, R.B.; Tsipouras, P. Life expectancy in the Marfan syndrome. Am. J. Cardiol. 1995, 75, 157–160, doi:10.1016/s0002‐9149(00)80066‐1. 17. Moher, D.; Liberati, A.; Tetzlaff, J.; Altman, D.G. Abbreviations Preferred reporting items for systematic reviews and meta‐ analyses: The PRISMA statement. BMJ 2009. 18. Fujiseki, Y.; Okuno, K.; Tanaka, M.; Shimada, M.; Takahashi, M.; Kawanishi, K. Myocardial involvement in the Marfan syndrome. Jpn. Heart J. 1985, 26, 1043–1050, doi:10.1536/ihj.26.1043. 19. Roman, M.J.; Devereux, R.B.; Kramer‐Fox, R.; Spitzer, M.C. Comparison of cardiovascular and skeletal features of primary mitral valve prolapse and Marfan syndrome. Am. J. Cardiol. 1989, 63, 317–321, doi:10.1016/0002‐9149(89)90338‐x. 20. Savolainen, A.; Nisula, L.; Keto, P.; Hekali, P.; Viitasalo, M.; Kaitila, L.; Kupari, M. Left ventricular function in children with the Marfan syndrome. Eur. Heart J. 1994, 15, 625–630, doi:10.1093/oxfordjournals.eurheartj.a060558. 21. Porciani, M.C.; Giurlani, L.; Chelucci, A.; Pepe, G.; Giusti, B.; Brunelli, T.; Attanasio, M.; Martinucci, P.; Fattori, R.; Abbatea, R.; et al. Diastolic subclinical primary alterations in marfan syndrome and marfan‐ related disorders. Clin. Cardiol. 2006, 25, 416–420, doi:10.1002/clc.4960250905. 22. Chatrath, R.; Beauchesne, L.M.; Connolly, H.M.; Michels, V.V.; Driscoll, D. Left ventricular function in the Marfan syndrome without significant valvular regurgitation. Am. J. Cardiol. 2003, 91, 914–916, doi:10.1016/s0002‐9149(03)00039‐0. 23. Meijboom, L.J.; Timmermans, J.; Van Tintelen, J.P.; Nollen, G.J.; De Backer, J.; Berg, M.P.V.D.; Boers, G.H.; Mulder, B.J. Evaluation of left ventricular dimensions and function in Marfan’s syndrome without significant valvular regurgitation. Am. J. Cardiol. 2005, 95, 795–797, doi:10.1016/j.amjcard.2004.11.042. 24. De Backer, J.; Devos, D.; Segers, P.; Matthys, D.; François, K.; Gillebert, T.; De Paepe, A.M.; De Sutter, J. Primary impairment of left ventricular function in Marfan syndrome. Int. J. Cardiol. 2006, 112, 353–358, doi:10.1016/j.ijcard.2005.10.010. 25. Das, B.; Taylor, A.; Yetman, A. Left Ventricular Diastolic Dysfunction in Children and Young Adults with Marfan Syndrome. Pediatr. Cardiol. 2006, 27, 256–258, doi:10.1007/s00246‐005‐1139‐5. 26. Rybczynski, M.; Koschyk, D.H.; Aydin, M.A.; Robinson, P.N.; Brinken, T.; Franzen, O.; Berger, J.; Hofmann, T.; Meinertz, T.; Von Kodolitsch, Y. Tissue Doppler imaging identifies myocardial dysfunction in adults with marfan syndrome. Clin. Cardiol. 2007, 30, 19–24, doi:10.1002/clc.3. 27. Kiotsekoglou, A.; Bajpai, A.; Bijnens, B.; Kapetanakis, V.; Athanassopoulos, G.; Moggridge, J.C.; Mullen, M.J.; Nassiri, D.K.; Camm, J.; Sutherland, G.R.; et al. Early impairment of left ventricular long‐axis systolic function demonstrated by reduced atrioventricular plane displacement in patients with Marfan syndrome. Eur. J. Echocardiogr. 2008, 9, 605–613, doi:10.1093/ejechocard/jen003. 28. Alpendurada, F.; Wong, J.; Kiotsekoglou, A.; Banya, W.; Child, A.; Prasad, S.K.; Pennell, D.J.; Mohiaddin, R.H. Evidence for Marfan cardiomyopathy. Eur. J. Heart. Fail. 2010, 12, 1085–1091, doi:10.1093/eurjhf/hfq127. 29. Abbreviations De Witte, P.; Aalberts, J.J.J.; Radonic, T.; Timmermans, J.; Scholte, A.J.; Zwinderman, A.H.; Mulder, B.J.M.; Groenink, M.; Berg, M.P.V.D. Intrinsic biventricular dysfunction in Marfan syndrome. Heart 2011, 97, 2063– 28. Alpendurada, F.; Wong, J.; Kiotsekoglou, A.; Banya, W.; Child, A.; Prasad, S.K.; Pennell, D.J.; Mohiaddin, R.H. Evidence for Marfan cardiomyopathy. Eur. J. Heart. Fail. 2010, 12, 1085–1091, doi:10.1093/eurjhf/hfq127. 29. De Witte, P.; Aalberts, J.J.J.; Radonic, T.; Timmermans, J.; Scholte, A.J.; Zwinderman, A.H.; Mulder, B.J.M.; Groenink, M.; Berg, M.P.V.D. Intrinsic biventricular dysfunction in Marfan syndrome. Heart 2011, 97, 2063– 2068, doi:10.1136/heartjnl‐2011‐300169. 30. Winther, S.; Williams, L.K.; Keir, M.; Connelly, K.A.; Bradley, T.J.; Rakowski, H.; Crean, A.M. Cardiovascular Magnetic Resonance Provides Evidence of Abnormal Myocardial Strain and Primary Cardiomyopathy in Marfan syndrome. J. Comput. Assist. Tomogr. 2019, 43, 410–415, doi:10.1097/rct.0000000000000863. 31. Kiotsekoglou, A.; Saha, S.; Moggridge, J.C.; Kapetanakis, V.; Govindan, M.; Alpendurada, F.; Mullen, M.J.; Nassiri, D.K.; Camm, J.; Sutherland, G.R.; et al. Impaired Biventricular Deformation in Marfan Syndrome: A 18 of 21 18 of 21 Diagnostics 2020, 10, 751 Strain and Strain Rate Study in Adult Unoperated Patients. Echocardiography 2011, 28, 416–430, doi:10.1111/j.1540‐8175.2010.01359.x. Strain and Strain Rate Study in Adult Unoperated Patients. Echocardiography 2011, 28, 416–430, doi:10.1111/j.1540‐8175.2010.01359.x. 32. Angtuaco, M.J.; Vyas, H.V.; Malik, S.; Holleman, B.N.; Gossett, J.M.; Sachdeva, R. Early detection of cardiac dysfunction by strain and strain rate imaging in children and young adults with marfan syndrome. J. Ultrasound Med. 2012, 31, 1609–1616, doi:10.7863/jum.2012.31.10.1609. 33. El Rahman, M.A.; Haase, D.; Rentzsch, A.; Olchvary, J.; Schäfers, H.‐J.; Henn, W.; Wagenpfeil, S.; Abdul‐ Khaliq, H. Left Ventricular Systolic Dysfunction in Asymptomatic Marfan Syndrome Patients Is Related to the Severity of Gene Mutation: Insights from the Novel Three Dimensional Speckle Tracking Echocardiography. PLoS ONE 2015, 10, e0124112, doi:10.1371/journal.pone.0124112. 34. Kiotsekoglou, A.; Sutherland, G.R.; Moggridge, J.C.; Kapetanakis, V.; Bajpai, A.; Bunce, N.; Mullen, M.J.; Louridas, G.; Nassiri, D.K.; Camm, J.; et al. Impaired right ventricular systolic function demonstrated by reduced atrioventricular plane displacement in adults with Marfan syndrome. Eur. J. Echocardiogr. 2008, 10, 295–302, doi:10.1093/ejechocard/jen239. 35. Kiotsekoglou, A.; Moggridge, J.C.; Bijnens, B.; Kapetanakis, V.; Alpendurada, F.; Mullen, M.J.; Saha, S.; Nassiri, D.K.; Camm, J.; Sutherland, G.R.; et al. Biventricular and atrial diastolic function assessment using conventional echocardiography and tissue‐Doppler imaging in adults with Marfan syndrome. Eur. J. Echocardiogr. 2009, 10, 947–955, doi:10.1093/ejechocard/jep110. 36. Scherptong, R.W.; Vliegen, H.W.; Van Der Wall, E.E.; Hilhorst‐Hofstee, Y.; Bax, J.J.; Scholte, A.J.; Delgado, V. Abbreviations Biventricular Performance in Patients with Marfan Syndrome without Significant Valvular Disease: Comparison to Normal Subjects and Longitudinal Follow‐Up. J. Am. Soc. Echocardiogr. 2011, 24, 1392–1399.e1, doi:10.1016/j.echo.2011.09.004. 37. Campens, L.; Renard, M.; Trachet, B.; Segers, P.; Mosquera, L.M.; De Sutter, J.; Sakai, L.; De Paepe, A.; De Backer, J. Intrinsic cardiomyopathy in Marfan syndrome: Results from in‐vivo and ex‐vivo studies of the Fbn1C1039G/+ model and longitudinal findings in humans. Pediatr. Res. 2015, 78, 256–263, doi:10.1038/pr.2015.110. 38. Gehle, P.; Robinson, P.N.; Heinzel, F.; Edelmann, F.; Yigitbasi, M.; Berger, F.; Falk, V.; Pieske, B.; Wellnhofer, E. NT‐proBNP and diastolic left ventricular function in patients with Marfan syndrome. IJC Heart Vasc. 2016, 12, 15–20, doi:10.1016/j.ijcha.2016.05.003. 39. Loeper, F.; Oosterhof, J.; Dorpel, M.V.D.; Van Der Linde, D.; Lu, Y.; Robertson, E.N.; Hambly, B.D.; Jeremy, R.W. Ventricular‐Vascular Coupling in Marfan and Non‐Marfan Aortopathies. J. Am. Heart Assoc. 2016, 5, e003705, doi:10.1161/jaha.116.003705. 40. Rao, P.; Keenan, J.B.; Rajab, T.K.; Kim, S.; Smith, R.; Amabile, O.; Khalpey, Z. Total artificial heart implantation in a young Marfan syndrome patient. Int. J. Artif. Organs 2018, 41, 175–177, doi:10.1177/0391398817752297. 41. Ogawa, Y.; Choi, C.W.; Shudo, Y.; Woo, Y.‐P.J. Successful orthotopic heart transplantation in a patient with Marfan syndrome. J. Card. Surg. 2019, 34, 875–876, doi:10.1111/jocs.14129. 42. Kesler, K.A.; Hanosh, J.J.; O’Donnell, J.; Faust, S.; Turrentine, M.W.; Mahomed, Y.; Brown, J.W. Heart transplantation in patients with Marfan’s syndrome: A survey of attitudes and results. J. Heart Lung Transplant. 1994, 13. 43. Mullen, J.C.; Lemermeyer, G.; Bentley, M.J. Recurrent Aortic Dissection After Orthotopic Heart Transplantation. Ann. Thorac. Surg. 1996, 62, 1830–1831, doi:10.1016/s0003‐4975(96)00615‐7. 44. Varghese, J.; Gilani, S.; Arentzen, C.; Jennison, S. Rupture of chronic thoracic aortic dissection in a Marfan syndrome patient after heart transplantation. J. Heart Lung Transplant. 2006, 25, 610–611, doi:10.1016/j.healun.2005.12.008. 45. Botta, L.; Russo, V.; Grigioni, F.; Arpesella, G.; Rocchi, G.; Di Bartolomeo, R.; Fattori, R. Unusual Rapid Evolution of Type B Aortic Dissection in a Marfan Patient Following Heart Transplantation: Successful Endovascular Treatment. Eur. J. Vasc. Endovasc. Surg. 2006, 32, 358–360, doi:10.1016/j.ejvs.2006.04.014. 46. Knosalla, C.; Weng, Y.‐G.; Hammerschmidt, R.; Pasic, M.; Schmitt‐Knosalla, I.; Grauhan, O.; Dandel, M.; Lehmkuhl, H.B.; Hetzer, R. Orthotopic Heart Transplantation in Patients with Marfan Syndrome. Ann. Thorac. Surg. 2007, 83, 1691–1695, doi:10.1016/j.athoracsur.2007.01.018. g j 47. Rajagopal, K.; Rogers, J.G.; Lodge, A.J.; Gaca, J.G.; McCann, R.L.; Milano, C.A.; Hughes, G.C. Abbreviations Two‐Stage Total Cardioaortic Replacement for End‐Stage Heart and Aortic Disease in Marfan Syndrome: Case Report and Review of the Literature. J. Heart Lung Transplant. 2009, 28, 958–963, doi:10.1016/j.healun.2009.05.012. 19 of 21 19 of 21 Diagnostics 2020, 10, 751 48. Hetzer, R.; Siegel, G.; Walter, E.M.D. Cardiomyopathy in Marfan syndrome. Eur. J. Cardio‐Thoracic Surg. 2015, 49, 561–568, doi:10.1093/ejcts/ezv073. 49. Audenaert, T.; De Pauw, M.; François, K.; De Backer, J. Type B aortic dissection triggered by heart transplantation in a patient with Marfan syndrome. BMJ Case Rep. 2015, 2015, doi:10.1136/bcr‐2015‐211138. 50. Porciani, M.C.; Attanasio, M.; Lepri, V.; Lapini, I.; Demarchi, G.; Padeletti, L.; Pepe, G.; Abbate, R.; Gensini, G.F. [Prevalence of cardiovascular manifestations in Marfan syndrome]. Ital. Heart journal. Suppl. Off. J. Ital. Fed. Cardiol. 2004, 5. 51. Rybczynski, M.; Mir, T.S.; Sheikhzadeh, S.; Bernhardt, A.M.; Schad, C.; Treede, H.; Veldhoen, S.; Groene, E.F.; Kühne, K.; Koschyk, D.; et al. Frequency and Age‐Related Course of Mitral Valve Dysfunction in the Marfan Syndrome. Am. J. Cardiol. 2010, 106, 1048–1053, doi:10.1016/j.amjcard.2010.05.038. 52. Gott, V.L.; Greene, P.S.; Alejo, D.E.; Cameron, D.E.; Naftel, D.C.; Miller, D.C.; Gillinov, A.M.; Laschinger, J.C.; Borst, H.G.; Cabrol, C.E.; et al. Replacement of the Aortic Root in Patients with Marfan’s Syndrome. New Engl. J. Med. 1999, 340, 1307–1313, doi:10.1056/nejm199904293401702. 53. Gott, V.L.; Cameron, D.E.; Alejo, D.E.; Greene, P.S.; Shake, J.G.; Caparrelli, D.J.; Dietz, H.C. Aortic root replacement in 271 Marfan patients: A 24‐year experience11This manuscript was adapted in part from the William, W.L. Glenn Lecture presented by Dr Gott at the American Heart Association Meeting, New Orleans, LA, Nov 13, 2000. Ann. Thorac. Surg. 2002, 73, 438–443, doi:10.1016/s0003‐4975(01)03336‐7. 54. Cameron, D.E.; Alejo, D.E.; Patel, N.D.; Nwakanma, L.U.; Weiss, E.S.; Vricella, L.A.; Dietz, H.C.; Spevak, P.J.; Williams, J.A.; Bethea, B.T.; et al. Aortic Root Replacement in 372 Marfan Patients: Evolution of Operative Repair Over 30 Years. Ann. Thorac. Surg. 2009, 87, 1344–1350, doi:10.1016/j.athoracsur.2009.01.073. 55. Ikonomidis, I.; Aboyans, V.; Blacher, J.; Brodmann, M.; Brutsaert, D.L.; Chirinos, J.A.; De Carlo, M.; Delgado, V.; Lancellotti, P.; Lekakis, J.; et al. The role of ventricular–arterial coupling in cardiac disease and heart failure: Assessment, clinical implications and therapeutic interventions. A consensus document of the European Society of Cardiology Working Group on Aorta & Peripheral Vascular Diseases, European Association of Cardiovascular Imaging, and Heart Failure Association. Eur. J. Heart Fail. 2019, 21, 402–424, doi:10.1002/ejhf.1436. 56. Abbreviations De Backer, J.; Campens, L.; Mosquera, L.M. Looking for the Missing Links. Circ. Genom. Precis. Med. 201 doi:10.1161/circgen.118.002185. 67. Rurali, E.; Perrucci, G.L.; Pilato, C.A.; Pini, A.; Gaetano, R.; Nigro, P.; Pompilio, G. Precise Therapy for Thoracic Aortic Aneurysm in Marfan Syndrome: A Puzzle Nearing Its Solution. Prog. Cardiovasc. Dis. 2018, 61, 328–335, doi:10.1016/j.pcad.2018.07.020. 68. Aalberts, J.J.J.; Van Tintelen, J.P.; Meijboom, L.J.; Polko, A.; Jongbloed, J.D.; Van Der Wal, H.; Pals, G.; Osinga, J.; Timmermans, J.; De Backer, J.; et al. Relation between genotype and left‐ventricular dilatation in patients with Marfan syndrome. Gene 2014, 534, 40–43, doi:10.1016/j.gene.2013.10.033. 69. Cook, J.R.; Carta, L.; Benard, L.; Chemaly, E.R.; Chiu, E.; Rao, S.K.; Hampton, T.G.; Yurchenco, P.D.; Consortium, G.R.; Costa, K.D.; et al. Abnormal muscle mechanosignaling triggers cardiomyopathy in mice with Marfan syndrome. J. Clin. Investig. 2014, 124, 1329–1339, doi:10.1172/JCI71059. 70. Tae, H.‐J.; Petrashevskaya, N.N.; Marshall, S.; Krawczyk, M.; Talan, M. Cardiac remodeling in the mouse model of Marfan syndrome develops into two distinctive phenotypes. Am. J. Physiol. Circ. Physiol. 2015, 310, H290–9, doi:10.1152/ajpheart.00354.2015. 71. Rouf, R.; Macfarlane, E.G.; Takimoto, E.; Chaudhary, R.; Nagpal, V.; Rainer, P.P.; Bindman, J.G.; Gerber, E.E.; Bedja, D.; Schiefer, C.; et al. Nonmyocyte ERK1/2 signaling contributes to load‐induced cardiomyopathy in Marfan mice. JCI Insight 2017, 2, doi:10.1172/jci.insight.91588. 72. Vracko, R.; Thorning, D.; Frederickson, R.G. Spatial arrangements of microfibrils in myocardial scars: Application of antibody to fibrillin. J. Mol. Cell. Cardiol. 1990, 22, 749–757, doi:10.1016/0022‐2828(90)90087‐i. 73. Bouzeghrane, F.; Reinhardt, D.P.; Reudelhuber, T.L.; Thibault, G. Enhanced expression of fibrillin‐1, a constituent of the myocardial extracellular matrix in fibrosis. Am. J. Physiol. Circ. Physiol. 2005, 289, H982– H991, doi:10.1152/ajpheart.00151.2005. 74. Lockhart, M.; Wirrig, E.; Phelps, A.; Wessels, A. Extracellular matrix and heart development. Birth Defects Res. Part. A Clin. Mol. Teratol. 2011, 91, 535–550, doi:10.1002/bdra.20810. 75. Steijns, F.; Van Hengel, J.; Sips, P.Y.; De Backer, J.; Renard, M. A heart for fibrillin: Spatial arrangement in adult wild‐type murine myocardial tissue. Histochem. Cell Boil. 2018, 150, 271–280, doi:10.1007/s00418‐018‐ 1686‐5. 76. Parker, K.K.; Ingber, D.E. Extracellular matrix, mechanotransduction and structural hierarchies in heart tissue engineering. Philos. Trans. R. Soc. B Boil. Sci. 2007, 362, 1267–1279, doi:10.1098/rstb.2007.2114. 77. Aydin, A.; Adsay, B.A.; Sheikhzadeh, S.; Keyser, B.; Rybczynski, M.; Sondermann, C.; Detter, C.; Steven, D.; Robinson, P.N.; Berger, J.; et al. Observational Cohort Study of Ventricular Arrhythmia in Adults with Marfan Syndrome Caused by FBN1 Mutations. PLoS ONE 2013, 8, e81281, doi:10.1371/journal.pone.0081281. 78. Abbreviations Vitarelli, A.; Conde, Y.; Cimino, E.; D’Angeli, I.; D’Orazio, S.; Stellato, S.; Padella, V.; Caranci, F. Aortic Wall Mechanics in the Marfan Syndrome Assessed by Transesophageal Tissue Doppler Echocardiography. Am. J. Cardiol. 2006, 97, 571–577, doi:10.1016/j.amjcard.2005.09.089. 57. Adams, J.N.; Brooks, M.; Redpath, T.W.; Smith, F.W.; Dean, J.C.; Gray, J.; Walton, S.; Trent, R.J. Aortic distensibility and stiffness index measured by magnetic resonance imaging in patients with Marfan’s syndrome. Heart 1995, 73, 265–269, doi:10.1136/hrt.73.3.265. y 58. Yin, F.C.; Brin, K.P.; Ting, C.T.; Pyeritz, R.E. Arterial hemodynamic indexes in Marfan’s syndrome. Circulation 1989, 79, 854–862, doi:10.1161/01.cir.79.4.854. 59. Kiotsekoglou, A.; Saha, S.K.; Moggridge, J.C.; Kapetanakis, V.; Bijnens, B.; Mullen, M.J.; Camm, J.; Sutherland, G.R.; Wilkinson, I.B.; Child, A.H. Effect of aortic stiffness on left ventricular long‐axis systolic function in adults with Marfan syndrome. Hell. J. Cardiol. HJC = Hell. Kardiologike Epitheorese 2010, 51. 60. Nollen, G.J.; Meijboom, L.J.; Groenink, M.; Timmermans, J.; Barentsz, J.O.; Merchant, N.; Webb, G.D.; Lamb, H.J.; Tijssen, J.G.; Van Der Wall, E.E.; et al. Comparison of aortic elasticity in patients with the marfan syndrome with and without aortic root replacement. Am. J. Cardiol. 2003, 91, 637–640, doi:10.1016/s0002‐ 9149(02)03330‐1. 61. Mortensen, K.; Aydin, M.; Bernhardt, A.M.J.; Appenzeller, V.; Robinson, P.N.; Berger, J.; Reichenspurner, H.; Von Kodolitsch, Y.; Kai, M.; Muhammet, A.; et al. Arterial mechanical properties after replacement or reconstruction of the aortic root. World J. Cardiovasc. Dis. 2012, 2, 8–13, doi:10.4236/wjcd.2012.21002. 62. Franken, R.; Groenink, M.; De Waard, V.; Feenstra, H.M.; Scholte, A.J.; Berg, M.P.V.D.; Pals, G.; Zwinderman, A.H.; Timmermans, J.; Mulder, B.J.M. Genotype impacts survival in Marfan syndrome. Eur. Heart J. 2016, 37, 3285–3290, doi:10.1093/eurheartj/ehv739. 63. Becerra‐Muñoz, V.M.; Gomez‐Doblas, J.; Porras, C.; Such, M.; Crespo‐Leiro, M.G.; Villa, R.B.; De Teresa‐ Galván, E.; Jimenez‐Navarro, M.; Cabrera‐Bueno, F. The importance of genotype‐phenotype correlation in the clinical management of Marfan syndrome. Orphanet J. Rare Dis. 2018, 13, 16, doi:10.1186/s13023‐017‐0754‐ 6. 64. Takeda, N.; Inuzuka, R.; Maemura, S.; Morita, H.; Nawata, K.; Fujita, D.; Taniguchi, Y.; Yamauchi, H.; Yagi, H.; Kato, M.; et al. Impact of Pathogenic FBN1 Variant Types on the Progression of Aortic Disease in Patients with Marfan Syndrome. Circ. Genom. Precis. Med. 2018, 11, doi:10.1161/circgen.117.002058. 20 of 21 20 of 21 Diagnostics 2020, 10, 751 65. Baudhuin, L.M.; Kotzer, K.E.; Lagerstedt, S.A. Increased frequency of FBN1 truncating and splicing variants in Marfan syndrome patients with aortic events. Genet. Med. 2014, 17, 177–187, doi:10.1038/gim.2014.91. 66. Abbreviations Mosquera, L.M.; De Wilde, H.; Devos, D.; Babin, D.; Jordaens, L.; Demolder, A.; De Groote, K.; De Wolf, D.; De Backer, J. Myocardial Disease and Ventricullar Arrhythmia in Marfan Syndrome—A Prospective Study. Orphanet J. Rare Dis 2020, doi:10.21203/rs.3.rs‐37970/v1. 79. Mah, D.Y.; Sleeper, L.A.; Crosson, J.E.; Czosek, R.J.; Love, B.A.; McCrindle, B.W.; Muiño‐Mosquera, L.; Olson, A.K.; Pilcher, T.A.; Tierney, E.S.S.; et al. Frequency of Ventricular Arrhythmias and Other Rhythm Abnormalities in Children and Young Adults With the Marfan Syndrome. Am. J. Cardiol. 2018, 122, 1429– 1436, doi:10.1016/j.amjcard.2018.07.006. 80. Savolainen, A.; Kupari, M.; Toivonen, L.; Kaitila, I.; Viitasalo, M. Abnormal ambulatory electrocardiographic findings in patients with the Marfan syndrome. J. Intern. Med. 1997, 241, 225–230, doi:10.1046/j.1365‐ 2796.1997.115125000.x. 81. Chen, S.‐C.; Fagan, L.F.; Nouri, S.; Donahoe, J.L. Ventricular Dysrhythmias in Children With Marfan’s Syndrome. Arch. Pediatr. Adolesc. Med. 1985, 139, 273–276, doi:10.1001/archpedi.1985.02140050067024. 82. Hoffmann, B.A.; Rybczynski, M.; Rostock, T.; Servatius, H.; Drewitz, I.; Steven, D.; Aydin, A.; Sheikhzadeh, S.; Darko, V.; Von Kodolitsch, Y.; et al. Prospective risk stratification of sudden cardiac death in Marfan’s syndrome. Int. J. Cardiol. 2013, 167, 2539–2545, doi:10.1016/j.ijcard.2012.06.036. 83. Schaeffer, B.; Rybczynski, M.; Sheikhzadeh, S.; Akbulak, R.Ö.; Moser, J.; Jularic, M.; Schreiber, D.; Daubmann, A.; Willems, S.; Von Kodolitsch, Y.; et al. Heart rate turbulence and deceleration capacity for risk prediction of serious arrhythmic events in Marfan syndrome. Clin. Res. Cardiol. 2015, 104, 1054–1063, doi:10.1007/s00392‐ 015‐0873‐9. 21 of 21 21 of 21 Diagnostics 2020, 10, 751 84. Priori, S.G.; Blomstrom‐Lundqvist, C.; Mazzanti, A.; Bloma, N.; Borggrefe, M.; Camm, J.; Elliott, P.M.; Fitzsimons, D.; Hatala, R.; Hindricks, G.; et al. 2015 ESC Guidelines for the management of patients with ventricular arrhythmias and the prevention of sudden cardiac death the Task Force for the Management of Patients with Ventricular Arrhythmias and the Prevention of Sudden Cardiac Death of the Europea. Eur. Heart J. 2015. 85. Arunamata, A.; Nguyen, C.T.; Ceresnak, S.R.; Dubin, A.M.; Olson, I.L.; Murphy, D.J.; Tierney, E.S.S. Utility of serial 12‐lead electrocardiograms in children with Marfan syndrome. Cardiol. Young 2018, 28, 1009–1013, doi:10.1017/s1047951118000707. 86. Murdoch, J.L.; Walker, B.A.; Halpern, B.L.; Kuzma, J.W.; McKusick, V.A. Life Expectancy and Causes of Death in the Marfan Syndrome. N. Engl. J. Med. 1972, 286, 804–808, doi:10.1056/nejm197204132861502. 87. Marsalese, D.L.; Moodie, D.S.; Vacante, M.; Lytle, B.W.; Gill, C.C.; Sterba, R.; Cosgrove, D.M.; Passalacqua, M.; Goormastic, M.; Kovacs, A. Marfan’s syndrome: Natural history and long‐term follow‐up of cardiovascular involvement. J. Am. Coll. Cardiol. © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). Abbreviations 1989, 14, 422–428, doi:10.1016/0735‐1097(89)90197‐6. 88. Chiu, H.‐H.; Wu, M.‐H.; Chen, H.‐C.; Kao, F.‐Y.; Huang, S.‐K. Epidemiological Profile of Marfan Syndrome in a General Population: A National Database Study. Mayo Clin. Proc. 2014, 89, 34–42, doi:10.1016/j.mayocp.2013.08.022. 89. Vanem, T.T.; Geiran, O.R.; Krohg‐Sørensen, K.; Røe, C.; Paus, B.; Rand‐Hendriksen, S. Survival, causes of death, and cardiovascular events in patients with Marfan syndrome. Mol. Genet. Genom. Med. 2018, 6, 1114– 1123, doi:10.1002/mgg3.489. 90. Groth, K.A.; Stochholm, K.; Hove, H.; Andersen, N.H.; Gravholt, C. Causes of Mortality in the Marfan Syndrome(from a Nationwide Register Study). Am. J. Cardiol. 2018, 122, 1231–1235, doi:10.1016/j.amjcard.2018.06.034. © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
https://openalex.org/W3035109245	https://europepmc.org/articles/pmc7297829?pdf=render	English	null	Scientific perspectivism in the phenomenological tradition	European journal for philosophy of science	2,020	cc-by	17,867	Abstract Abstract In current debates, many philosophers of science have sympathies for the project of introducing a new approach to the scientific realism debate that forges a middle way between traditional forms of scientific realism and anti-realism. One promis- ing approach is perspectivism. Although different proponents of perspectivism differ in their respective characterizations of perspectivism, the common idea is that scientific knowledge is necessarily partial and incomplete. Perspectivism is a new position in current debates but it does have its forerunners. Figures that are typically mentioned in this context include Dewey, Feyerabend, Leibniz, Kant, Kuhn, and Putnam. Interestingly, to my knowledge, there exists no work that discusses similarities to the phenomenological tradition. This is surprising because here one can find systematically similar ideas and even a very similar terminology. It is startling because early modern physics was noticeably influenced by phe- nomenological ideas. And it is unfortunate because the analysis of perspectival approaches in the phenomenological tradition can help us to achieve a more nuanced understanding of different forms of perspectivism. The main objective of this paper is to show that in the phenomenological tradition one finds a well- elaborated philosophy of science that shares important similarities with current versions of perspectivism. Engaging with the phenomenological tradition is also of systematic value since it helps us to gain a better understanding of the distinctive claims of perspectivism and to distinguish various grades of perspectivism. Keywords Scientificrealism . Perspectivism. Phenomenology.Husserl . Merleau-Ponty. QBism This article belongs to the Topical Collection: Perspectivism in science: metaphysical and epistemological reflections Guest Editor: Michela Massimi In current debates, many philosophers of science have sympathies for the project of introducing a new approach to the scientific realism debate that forges a middle way between traditional forms of scientific realism and anti-realism. One promis- ing approach is perspectivism. Although different proponents of perspectivism differ in their respective characterizations of perspectivism, the common idea is that scientific knowledge is necessarily partial and incomplete. Perspectivism is a new position in current debates but it does have its forerunners. Figures that are typically mentioned in this context include Dewey, Feyerabend, Leibniz, Kant, Kuhn, and Putnam. Interestingly, to my knowledge, there exists no work that discusses similarities to the phenomenological tradition. This is surprising because here one can find systematically similar ideas and even a very similar terminology. It is startling because early modern physics was noticeably influenced by phe- nomenological ideas. PAPER IN GENERAL PHILOSOPHY OF SCIENCE PAPER IN GENERAL PHILOSOPHY OF SCIENCE Open Access Scientific perspectivism in the phenomenological tradition Philipp Berghofer1 Received: 18 October 2019 /Accepted: 13 May 2020 # The Author(s) 2020 /Published online: 16 June 2020 * Philipp Berghofer philipp.berghofer@uni–graz.at https://doi.org/10.1007/s13194-020-00294-w European Journal for Philosophy of Science (2020) 10: 30 https://doi.org/10.1007/s13194-020-00294-w European Journal for Philosophy of Science (2020) 10: 30 https://doi.org/10.1007/s13194-020-00294-w European Journal for Philosophy of Science (2020) 10: 30 1 Perspectivism in current debates Advocating scientific realism, broadly speaking, means adopting “a positive epistemic attitude toward the content of our best theories and models, recommending belief in both observable and unobservable aspects of the world described by the sciences” (Chakravartty 2017). It is safe to say that scientific realism is the dominant view among the public. This is particularly due to the undeniable success of the sciences in making predictions and in contributing to the technological advancements we witness on a daily basis. Also among philosophers of science, the main motivation for adopting a realist position is the notorious miracle argument which quotes scientific realism as the best explanation for the obvious success of our scientific theories. This success would seem miraculous if our successful theories were misleading. Despite their initial plausibility, both scientific realism and the miracle argument have been attacked on many fronts. Objections based on the underdetermination of scientific theories by empirical data and the pessimistic meta induction gained such prominence in the twentieth century that Arthur Fine was prompted to declare: “Realism is dead” (Fine 1984, 83). Currently, realism is again the dominant stance, but most realists concede that the attacks from the anti-realist camp require clarification and some sort of constraint of realist commitments. The most common response to the problems surrounding realism is to adopt a form of selective realism. Here entity realism and structural realism are the most prominent examples, but each version of realism suffers from certain shortcom- ings and by now scientific realism has splintered into a variety of different positions each being vigorously attacked not only by anti-realists but also by other realists. Accordingly, prominent voices have pointed out that the debate between realists and anti-realists has come to a “stalemate” (cf., e.g., Chakravartty 2018, 233; Forbes 2017, 3327; Frost-Arnold 2010, 56). So where does this leave us? In recent debates, a new version of realism has emerged that is distinct from traditional versions of realism as well as from new versions of selective realism. This is perspectival realism or scientific perspectivism, in short, perspectivism. Its focus is not on certain parts of scientific theories (as it is the case for selective realism) but it aims at rethinking the nature and scope of scientific theories and models. The main works promoting perspectivism are Giere 2006, Massimi 2012, 2018a, b, and Teller 2001, 2011. Abstract And it is unfortunate because the analysis of perspectival approaches in the phenomenological tradition can help us to achieve a more nuanced understanding of different forms of perspectivism. The main objective of this paper is to show that in the phenomenological tradition one finds a well- elaborated philosophy of science that shares important similarities with current versions of perspectivism. Engaging with the phenomenological tradition is also of systematic value since it helps us to gain a better understanding of the distinctive claims of perspectivism and to distinguish various grades of perspectivism. Keywords Scientificrealism . Perspectivism. Phenomenology.Husserl . Merleau-Ponty QBism This article belongs to the Topical Collection: Perspectivism in science: metaphysical and epistemological reflections Guest Editor: Michela Massimi This article belongs to the Topical Collection: Perspectivism in science: metaphysical and epistemologic reflections Guest Editor: Michela Massimi * Philipp Berghofer philipp.berghofer@uni–graz.at * Philipp Berghofer philipp.berghofer@uni–graz.at 1 Department for Philosophy, University of Graz, Heinrichstraße 26/5, 8010, Graz, Austria 30 Page 2 of 27 European Journal for Philosophy of Science (2020) 10: 30 1 De Caro lists several contemporary proponents of the view that “only one true and complete description of the world exists, which is typically regarded as being offered by the natural sciences, especially physics” in (De Caro 2020, 58). 1 Perspectivism in current debates Proponents of perspectivism typically view their position as a via media between objectivist realism and all forms of anti-realism. However, there is no unified picture of perspectivism; different proponents of perspectivism differ in their respective accounts. Here I will focus on the depiction of perspectivism offered in Giere 2006 since it seems to be closest to the ideas developed in the phenomenological tradition. Giere contrasts his perspectival realism with what he calls objectivist realism: I will be arguing that there is a kind of realism that applies to scientific claims that is more limited than this full-blown objective realism. Thus, in the end, I wish to reject objective realism but still maintain a kind of realism, a perspectival realism, which I think better characterizes realism in science. For a perspectival realist, the strongest claims a scientist can legitimately make are of a qualified, conditional European Journal for Philosophy of Science (2020) 10: 30 Page 3 of 27 30 form: ‘According to this highly confirmed theory (or reliable instrument), the world seems to be roughly such and such.’ There is no way legitimately to take the further objectivist step and declare unconditionally: ‘This theory (or instru- ment) provides us with a complete and literally correct picture of the world itself.’ (Giere 2006, 5f.) Here we find three claims about scientific theories that Giere’s perspectivism subscribes to and that he takes to be in opposition to objective realism. & P1: Scientific theories are fallible. & P2: Scientific theories cannot be interpreted literally in their entirety. & P3: Scientific theories cannot provide a complete picture of the world. However, P1 and P2 are well-accepted claims and do not constitute a distinctive version of scientific realism. P1 is a claim any plausible form of scientific realism must accept. P2, as we have seen above, is advocated by all proponents of selective realism. Accordingly, what is distinctive about perspectivism should be encapsulated by P3. It is to be noted that P3 is an interesting and controversial claim that has been explicitly denied by prominent voices.1 Wilfrid Sellars expresses a widespread view within analytic philosophy of science when he says: But, speaking as a philosopher, I am quite prepared to say that the common sense world of physical objects in Space and Time is unreal-that is, that there are no such things. 1 Perspectivism in current debates His argument for the genuinely perspectival character of scientific observation resembles his argument concerning color vision. Humans and various other electromagnetic detectors respond differently to dif- ferent electromagnetic spectra. Moreover, humans and various other electromag- netic detectors may face the same spectrum of electromagnetic radiation and yet have different responses to it. In all cases, the response of any particular detector, including a human, is a function of both the character of the particular electro- magnetic spectrum encountered and the character of the detector. Each detector views the electromagnetic world from its own perspective. Every observation is perspectival in this sense. (Giere 2006, 48) For Giere, scientific observation is observation via instruments. But scientific instru- ments are structurally similar to human color vision in that human eyes as well as scientific instruments only engage with a limited scope of the electromagnetic spec- trum. How we perceive the world, what we observe, and what data we gain crucially depends on the make-up of the respective methods we use to observe. Giere illustrates this by discussing how different types of telescopes produce very different images of the Milky Way (Giere 2006, 45–49). The problem with Giere’s line of reasoning, as discussed so far, is that when we look at the examples he discusses, it remains unclear why there should be a clash with standard or objectivist scientific realism. What is the distinctive claim of perspectivism, embodied by the examples Giere provides, that standard forms of scientific realism must reject? Objectivist realists can clearly accept the fact that human vision and scientific observation only reveal limited aspects of nature. What they would insist on, however, is that the physical objects in question objectively and truly have the features ascribed to them. So when we say that from perspective X the object O has the feature F1 and from perspective Y the same object O has the feature F2, but F1 and F2 are not inconsistent, then there is no problem for objectivist realism. In this context, Chakravartty indicates that in order to be interesting and at odds with realism, perspectivism must amount to “one or another form of relativism” (Chakravartty 2010, 406). However, this objection is problematic since Giere explicitly rejects “the unanalyzed assumption that a robust scientific realism must be objectivist realism because otherwise it slides into constructivism or relativism” (Giere 2006, 92). 1 Perspectivism in current debates Or, to put it less paradoxically, that in the dimension of describing and explaining the world, science is the measure of all things, of what is that it is, and of what is not that it is not. (Sellars 1963, 173) As we will see below, this attitude that the common sense world is mere illusion and that science is the measure of reality, providing an exhaustive picture of all there is, is the main target of Husserl’s phenomenological philosophy of science. Sellars’ claim is also clearly opposed to P3. Let us take a closer look at how Giere motivates P3. Giere focuses on three sources of human knowledge, arguing that all three of them only deliver perspectival knowledge. These three sources are color vision, scientific observation, and scientific theories. Concerning color vision, Giere summarizes his line of reasoning as follows: For my purposes, maybe the most important feature of perspectives is that they are always partial. When looking out at a scene, a typical human trichromat is visually affected by only a narrow range of all the electromagnetic radiation available. In particular, the nearby wavelengths in the ultraviolet and infrared are 1 De Caro lists several contemporary proponents of the view that “only one true and complete description of the world exists, which is typically regarded as being offered by the natural sciences, especially physics” in (De Caro 2020, 58). European Journal for Philosophy of Science (2020) 10: 30 30 Page 4 of 27 simply not visually detected. And of course there is no possibility of visually detecting gamma rays or neutrinos. (Giere 2006, 35) simply not visually detected. And of course there is no possibility of visually detecting gamma rays or neutrinos. (Giere 2006, 35) simply not visually detected. And of course there is no possibility of visually detecting gamma rays or neutrinos. (Giere 2006, 35) Giere is obviously right in stating that human color vision is perspectival in the sense of being partial and incomplete. What we call light, i.e., the visible spectrum, is only a very limited portion of the electromagnetic spectrum. The question is whether this is a philosophically interesting fact that suggests preferring perspectival realism over ob- jectivist realism. Before we turn to this question in more detail, let us have a look at Giere’s further arguments based on his accounts of scientific observation and scientific theories. 1 Perspectivism in current debates The problem is that Giere is not very precise in specifying the distinctive feature of European Journal for Philosophy of Science (2020) 10: 30 Page 5 of 27 30 30 perspectival realism that makes his form of realism non-objectivist but also non- relativist. perspectival realism that makes his form of realism non-objectivist but also non- relativist. Chakravartty aims at clarification by making the following distinction: As a philosophically controversial thesis, then, perspectivism would seem to take the form of either one or possibly both of the following claims: (In the sciences, in connection with representations such as theories and models, ...) P1. We have knowledge of perspectival facts only, because non-perspectival facts are beyond our epistemic grasp. P2. We have knowledge of perspectival facts only, because there are no non- perspectival facts to be known. (Chakravartty 2010, 407) P2. We have knowledge of perspectival facts only, because there are no non- perspectival facts to be known. (Chakravartty 2010, 407) Chakravartty argues that none of the arguments and examples Giere provides support either of these strong theses. What is more, Chakravartty claims that there are simple counter-examples to P1 and P2. Such counter-examples are any facts that are intrinsi- cally non-perspectival and can be known via scientific observation. “It is a non- perspectival fact about charged bodies, for example, that they exert electrostatic forces on other charged bodies” (Chakravartty 2010, 407). Chakravartty argues that none of the arguments and examples Giere provides support either of these strong theses. What is more, Chakravartty claims that there are simple counter-examples to P1 and P2. Such counter-examples are any facts that are intrinsi- cally non-perspectival and can be known via scientific observation. “It is a non- perspectival fact about charged bodies, for example, that they exert electrostatic forces on other charged bodies” (Chakravartty 2010, 407). It is hard to tell how Giere would respond to such examples. This is because Giere does not talk in terms of facts. Terms such as “scientific fact” or “perspectival fact” do not occur in Giere 2006. Instead, he stresses that scientific knowledge is always incomplete. With respect to naked-eye observation and scientific observation, he argues that they are always partial in the sense that they only capture certain aspects of the observed object as they appear from a certain perspective. This is a plausible claim. 2 “But surely, it will be objected, scientists draw conclusions going beyond their instrumentation. Indeed they do. But they do so only by moving to a broader theoretical perspective” (Giere 2006, 49). 3 “My reply is that theoretical claims are also perspectival. The basic idea is that conception is a lot like perception, or, that theorizing is a lot like observing. More specifically, in creating theories, I will argue, scientists create perspectives within which to conceive of aspects of the world” (Giere 2006, 50). 4 “[T]he picture of science that emerges is an agent-based picture” (Giere 2006, 63). Cf. also Giere 2010. 5 We will revisit this distinction between methodological and ontological claims towards the end of Sect. 4. 6 “Like Putnam’s internal realism, perspectivism too is reacting against metaphysical realism and the so-called God’s eye view […] There cannot be an objective, unique, true description of the way the world is as soon as we acknowledge that our scientific knowledge is always from a specific vantage point” (Massimi 2018b, 165). 1 Perspectivism in current debates Importantly, this claim is not at odds with the claim that there are objective facts and that we can have knowledge of these facts. From my perspective, it appears that there is a laptop on my desk, this visual experience justifies me in believing that there is laptop on my desk, and it is objectively true that there is a laptop on my desk. We can think of similar examples regarding scientific observations. Accordingly, Giere’s claim that observation is perspectival and never provides a complete picture of the observed is not at odds with realism. However, we can see how Giere’s perspectivism might be at odds with objectivist realism. By objectivist realism, apparently, Giere understands the view that the sciences deliver a complete picture of the world that is free from all subjective factors. It is not only the case that the world really has the features ascribed to it by science, there is nothing to know about the world over and above what is described by our best (possible) scientific theories. This means that when Giere discusses the perspectival character of scientific theories (Giere 2006, chapter 4), he argues that scientific theories are systematically similar to color vision and scientific observation. Unfortunately, Giere’s discussion of scientific theories is a bit messy. He captures the basic claim “that theoretical claims are also perspectival” in many different ways, suggesting many nonequivalent formulations and theses. Although he provides a general picture, the individual theses are not fleshed out in much detail. It would go beyond the scope of the present paper to discuss these controversial theses in detail. In what follows in this section, I will list some of these theses that I consider particularly interesting. In what follows in this paper, I will show what role these claims play in the phenomenological tradition. European Journal for Philosophy of Science (2020) 10: 30 30 Page 6 of 27 (T1) Science itself is only a certain perspective we can have on the world. To describe or explain certain phenomena by virtue of a scientific theory, means to adopt a scientific perspective.2 p p (T2) Science and scientific theories can never deliver an exhaustive account of the world.3 (T2) Science and scientific theories can never deliver an exhaustive account of the world.3 (T3) If two scientific theories are inconsistent with each other, it does not follow that at least one of them is false. 1 Perspectivism in current debates They might both shed light on different aspects of nature (cf. Giere 2006, 62, 94). (T4) There is not and there cannot be one all-encompassing scientific theory to which all the other scientific theories can be reduced. (T5) Science cannot be totally detached from the scientist who is doing science.4 In the following section, I will shed light on Husserl’s conception of horizontal intentionality, highlighting systematic similarities to Giere’s account of the perspectival character of observation. In Sect. 3, I illustrate that the claims T1, T2, and T5 play important roles in the phenomenological tradition. In Sect. 4, I discuss Merleau-Ponty’s partial realism that can be considered a radicalization of T5. 2 “But surely, it will be objected, scientists draw conclusions going beyond their instrumentation. Indeed the do. But they do so only by moving to a broader theoretical perspective” (Giere 2006, 49). 3 7 Two new volumes on perspectivim were published in 2020 (Cretu & Massimi 2020; Massimi & McCoy 2020). None of the contributions addresses the phenomenological tradition. 8 For instance, Husserl’s phenomenology had a great influence on Hermann Weyl and the development of his gauge principle (cf. Ryckman 2005; and Ryckman forthcoming). 9 Here, many of Husserl’s insights are in agreement with the findings of early experimental psychologists such as Gestalt psychologists and the members of the Graz school. These ideas have been picked up in the recent movement of experimental phenomenology (cf. Albertazzi 2013). Although neglected for a long time, in the analytic tradition, there have recently been attempts to capture this distinctive character of perceptual experiences (e.g., Church 2013, 50). Particularly notable works in this context that blur the artificial distinction between analytic philosophy and phenomenology are Madary 2017; Smith 2010. 2 Horizontal intentionality As elaborated in the previous section, many philosophers of science have sympathies for the project of introducing a new approach to the scientific realism debate that forges a middle way between traditional forms of scientific realism and anti-realism. One such approach is perspectivism. Although different proponents of perspectivism differ in their respective characterizations of perspectivism, the common idea is that scientific knowledge is necessarily partial and incomplete. This epistemological claim can be supplemented by the stronger ontological thesis that the world itself is perspectival in the sense that at least some scientific facts are perspectival. Also, one can supplement it with the strong methodological claim that science must become aware of the fact that it is an agent-based endeavor in that it incorporates the first-person perspective into science. Science must incorporate the scientist into science.5 Perspectivism is a new position in recent debates but surely it has its forerunners. Massimi, for instance, points out that perspectivism “has a distinguished philosophical pedigree back to Leibniz, Kant, Nietzsche, and even Wittgenstein” (Massimi 2012, 25; cf. also Teller 2020). Matthew Brown argues that similar ideas can be found in Feyerabend and Dewey (Brown 2009; cf. also Giere 2016). Obviously, there are important similarities to ideas we find in Kuhn (Giere 2006, 2013) and in Putnam (De Caro 2020; Massimi 2018b).6 Interestingly, to my knowledge, there exists no work European Journal for Philosophy of Science (2020) 10: 30 Page 7 of 27 30 30 that discusses similarities to the phenomenological tradition.7 This is surprising because here one can find systematically similar ideas and even a very similar terminology. It is startling because early modern physics was noticeably influenced by phenomenological ideas.8 And it is unfortunate because the analysis of perspectival approaches in the phenomenological tradition can help us to get a more nuanced understanding of different forms of perspectivism. The main objective of this paper is historical: to show that in the phenomenological tradition one finds a well-elaborated philosophy of science that shares important similarities with current versions of perspectivism. How- ever, there is also a systematic value in shedding light on the philosophy of science in the phenomenological tradition because it helps us to gain a better understanding of the distinctive claims of perspectivism and to distinguish various grades of perspectivism. 2 Horizontal intentionality Husserl characterizes the perspectival character of perception as follows: Of necessity a physical thing can be given only ‘one-sidedly;’ […] A physical thing is necessarily given in mere ‘modes of appearance’ in which necessarily a core of ‘what is actually presented’ is apprehended as being surrounded by a horizon of ‘co-givenness,’ which is not givenness proper. (Husserl 1982, 94). This means that perception always “implies a plus ultra” (Husserl 2001, 48). According to Husserl, the perspectival character and horizontal structure of perception is not simply a result of the imperfection of human beings but an essential property of perception. Not even a god could change that perceptual experiences present their physical objects always in perspectives (Husserl 1982, 95). When Husserl illuminates the perspectival character of perception, he not only stresses that perception is incomplete but also that physical objects in perception always appear from a certain point of view. All orientation is thereby related to a null-point of orientation, or a null-thing, a function which my own body has, the body of the perceiver. And again, the perspectival mode of givenness of every perceptual thing and of each of its perceptual determinations – on the other hand, also of the entire unitary field of perception, that of the total spatial perception – is something new. The differences of perspective clearly are inseparably connected with the subjective differences of orientation and of the modes of givenness in sides.10 (Husserl 1977, 121) A further aspect of perception is that previous experiences shape the way we perceive. Perception is not a faculty that allows us to see the world as it is objectively, independent from our history, background beliefs, etc. To put it differently, “experience is not an opening through which a world, existing prior to all experience, shines into a room of consciousness; it is not a mere taking of something alien to consciousness into consciousness” (Husserl 1969, 232). This aspect of perception is closely related to discussions about the theory-ladenness of perception. Obviously, there are many similarities between Husserl’s conception of horizontal intentionality and Giere’s analysis of perception. For Giere, human perception is always incomplete because we only engage with a restricted part of the electromagnetic spectrum. For Husserl, perception is essentially incomplete because it presents its objects only one-sidedly and from a specific perspective. 10 A similar remark can be found in Husserl 1973, 116 f. It is interesting to see that the phenomenologically minded mathematician and physicist Hermann Weyl, father of the gauge principle which is one of the cornerstones of modern physics, basically makes the same claim, intentionally using phenomenological terminology (quoted and discussed in Ryckman 2005, 131). 2 Horizontal intentionality Since it is one of the main problems of current perspectivism to make clear how exactly it differs from standard forms of realism or anti-realism, the present elaboration can contribute to a better understanding of the distinctiveness of perspectivism. We will start with claims shared by most phenomenologists (Sect. 3) and see how these ideas can be radicalized (Sect. 4). This section discusses Husserl’s conception of horizontal intentionality, clarifying to what degree it is in agreement with Giere’s analysis of the partial nature of observation. We will also get a first glimpse of the phenomenological doctrine that there is no view from nowhere available, not even to science. One of Husserl’s main contributions to a proper phenomenological analysis of perceptual experience is his disclosure of what he calls the horizontal structure of experience. Perceptual experiences go beyond what is directly given.9 When you look at the laptop in front of you, your experience presents to you the laptop’s screen, case, keyboard, etc. Your experience has a presentive or in Husserl’s terminology a self- giving character with respect to these aspects of your laptop. But your perceptual experience does not only intend these directly perceived aspects. What is co-given to you within experience is the laptop’s back, that its back has certain properties such as a smooth surface, that it also has a smooth underside, etc. You do not actually see the back or underside of your laptop but they are co-intended aspects of your experience. Furthermore, the laptop is not experienced as an isolated object but as embedded in a surrounding world. Even if your attention is solely directed at the laptop, it is part of your experience that the laptop does not float around in nothingness, but is placed on a desk, which in turn stands on the floor here in your office at the department of your university in your hometown and so on. In Husserl’s terminology, these co-given objects (desk, floor, etc.) are part of the outer horizon of your experience. The hidden but co-intended aspects of the object itself (the laptop) are part of the inner horizon of European Journal for Philosophy of Science (2020) 10: 30 30 Page 8 of 27 your experience (Husserl 1972, 28). Husserl characterizes the perspectival character of perception as follows: your experience (Husserl 1972, 28). 2 Horizontal intentionality What is more, perception is shaped by subjective factors such as one’s history, culture, or previous experiences. It should be noted, however, that there are also crucial differences between Husserl and Giere. Husserl aims at making a general point about perceptual experiences based on European Journal for Philosophy of Science (2020) 10: 30 Page 9 of 27 30 30 descriptive phenomenological reflection. For Husserl, it is essentially (a priori) true that the object of a perceptual experience, i.e., a physical object extended in space, can only be given perspectivally. Giere’s reasoning, on the other hand, is based on results of the empirical sciences, particularly on how human eyes interact with the electromagnetic spectrum. Notwithstanding the differences in scope and methodology, this might be a good example where phenomenological and empirical investigations come to similar conclusions, thereby complementing each other. descriptive phenomenological reflection. For Husserl, it is essentially (a priori) true that the object of a perceptual experience, i.e., a physical object extended in space, can only be given perspectivally. Giere’s reasoning, on the other hand, is based on results of the empirical sciences, particularly on how human eyes interact with the electromagnetic spectrum. Notwithstanding the differences in scope and methodology, this might be a good example where phenomenological and empirical investigations come to similar conclusions, thereby complementing each other. Although perceptual experiences are necessarily perspectival and also shaped by previous experiences and further subjective factors, for Husserl this does not mean that perception cannot be a source of epistemic justification. In fact, for Husserl, perceptual experiences are the prime examples of sources of justification. According to Husserlian phenomenology, epistemology is intrinsically linked to the study of the intentional structures of consciousness. The most fundamental epistemological question turns out to be how subjectivity can be the source of knowledge (cf. Melle in Husserl 1984, page XXXI). Subjectivity is not only at the center of epistemology because justification is always justification for a subject. More importantly, justification is gained through subjective acts: “Subjective acts motivate everything” (Husserl 1984, 121). Here “subjective acts” simply means intentionally directed mental states of a subject. But which acts are justification-conferring? The answer is experiences. More precisely, those mental states that have a presentive character, that present their objects in an intuitive (“anschaulich”) manner. 11 Following Jeffrey Barrett (Barrett 1999, 116), Huggett and Wüthrich recently defined “a theory to be empirically incoherent in case the truth of the theory undermines our empirical justification for believing it to be true” (Huggett & Wüthrich 2013, 277). Hence, if the life-world is the epistemic foundation of mathematical physics, but mathematical physics is interpreted as revealing that the life-world is a mere illusion, this interpretation of physics is empirically incoherent. Note that this is not to say that our ordinary world of tables and chairs is ontologically fundamental. It may well be that physics reveals that our ordinary world is, so to speak, an emergent world, emerging from more fundamental physics such as quantum field theory. For similar discussions in the philosophy of quantum gravity of what it would mean for the very endeavor of physics if it turned out that space and time are not fundamental, cf. Huggett & Wüthrich 2013; Oriti 2014. We must not forget that “[a] central concern of philosophy of science is understanding how the theoretical connects to the empirical, the nature and significance of ‘saving the phenomena’” (Huggett & Wüthrich 2013, 276). 2 Horizontal intentionality For Husserl, this not only includes perceptual expe- riences but also, e.g., introspective intuitions and a priori intuitions (for more details on Husserl’s conception of experiential justification, cf. Berghofer 2018a, 2020). As we have seen above, although Husserl regards experiences as a source of immediate justification, he is well aware that experiences are not windows to the world through which we see how the world is in itself thoroughly objectively. Instead, experiences present their objects in a certain way that at least partly depends on subjective factors such as previous experiences, background beliefs, etc. In this context, Husserl empha- sizes the role of transcendental subjectivity. Every imaginable sense, every imaginable being, whether the latter is called immanent or transcendent, falls within the domain of transcendental subjectivity, as the subjectivity that constitutes sense and being. The attempt to conceive the universe of true being as something lying outside the universe of possible consciousness, possible knowledge, possible evidence, the two being related to one another merely externally by a rigid law, is nonsensical. They belong together essentially; and, as belonging together essentially, they are also concretely one, one in the only absolute concretion: transcendental subjectivity. If transcendental subjectivity is the universe of possible sense, then an outside is precisely – nonsense. (Husserl 1960, 84) Passages like this have led to much controversy in Husserl scholarship, some interpreting this as a methodological claim (as I do), others interpreting it as a commitment to metaphysical idealism. For our purpose, it suffices to note that for Husserl a view from nowhere at the world is in principle impossible. This is true not only for our perceptual encounter with the world but also for our scientific encounter. European Journal for Philosophy of Science (2020) 10: 30 30 Page 10 of 27 “Straightforward experience, in which the life-world is given, is the ultimate foundation of all objective owledge” (Husserl 1970, 226). g ( , ) 13 “But the researcher of nature does not make clear to himself that the constant fundament of his – after all subjective – work of thought is the surrounding life-world; it is always presupposed as the ground, as the field of work upon which alone his questions, his methods of thought, make sense” (Husserl 1970, 295). 14 Giere states that Jakob von Uexkull’s conception of Umwelt “is a more elaborate version of what I am calling a perspective” (Giere 2006, 36). He notes that “this term is borrowed from Husserl, or at least related to Husserl’s use of this term, but I am not in a position to explore this possible connection” (Giere 2006, 123). “But the researcher of nature does not make clear to himself that the constant fundament of his – after all bjective – work of thought is the surrounding life-world; it is always presupposed as the ground, as the field work upon which alone his questions, his methods of thought, make sense” (Husserl 1970, 295). 11 Following Jeffrey Barrett (Barrett 1999, 116), Huggett and Wüthrich recently defined “a theory to be empirically incoherent in case the truth of the theory undermines our empirical justification for believing it to be true” (Huggett & Wüthrich 2013, 277). Hence, if the life-world is the epistemic foundation of mathematical physics, but mathematical physics is interpreted as revealing that the life-world is a mere illusion, this interpretation of physics is empirically incoherent. Note that this is not to say that our ordinary world of tables and chairs is ontologically fundamental. It may well be that physics reveals that our ordinary world is, so to speak, an emergent world, emerging from more fundamental physics such as quantum field theory. For similar discussions in the philosophy of quantum gravity of what it would mean for the very endeavor of physics if it turned out that space and time are not fundamental, cf. Huggett & Wüthrich 2013; Oriti 2014. We must not forget that “[a] central concern of philosophy of science is understanding how the theoretical connects to the empirical, the nature and significance of ‘saving the phenomena’” (Huggett & Wüthrich 2013, 276). 12 “Straightforward experience, in which the life-world is given, is the ultimate foundation of all objective knowledge” (Husserl 1970, 226). 13 “But the researcher of nature does not make clear to himself that the constant fundament of his – after all subjective – work of thought is the surrounding life-world; it is always presupposed as the ground, as the field of work upon which alone his questions, his methods of thought, make sense” (Husserl 1970, 295). 14 Giere states that Jakob von Uexkull’s conception of Umwelt “is a more elaborate version of what I am calling a perspective” (Giere 2006, 36). He notes that “this term is borrowed from Husserl, or at least related to Husserl’s use of this term, but I am not in a position to explore this possible connection” (Giere 2006, 123). 3 The scientific perspective According to Husserl, the scien- tific method is not the only method of gaining knowledge, science does not provide an exhaustive picture of nature, and science is not completely independent from subjective factors. successful in what they are doing. However, he criticizes the conclusions naturalists and objectivist realists draw from the success of science. According to Husserl, the scien- tific method is not the only method of gaining knowledge, science does not provide an exhaustive picture of nature, and science is not completely independent from subjective factors. Edith Stein, one of Husserl’s foremost pupils, put it this way: “What physics [...] reveals pertains to the real nature but it never exhausts nature. And what evades the web of mathematical formulas is not less ‘real’ than what is captured by mathematics” (Stein 2004, 62). Here we find three motifs that are typical for a phenomenology of physics. First, phenomenology does not dispute the success of physics, neither does it object to the implementation of mathematics. Secondly, however, the mathematical picture delivered by physics only reveals one perspective of nature.15 Even if we manage to get a mathematical grip on nature, what we gain from this can never be an exhaustive picture of nature. “We have seen that the methods of the exact natural sciences do not capture reality in its totality, instead they are only concerned with certain sides of nature” (Stein 2004, 73).16 Thirdly, not being mathematizable does not imply not being real. This is not only true for certain aspects of nature but also for totally different entities including values, essences, and consciousness. Mathematics is extremely useful in physics but this does not imply that any science (including philosophy, value theory, etc.) must attempt to mathematize its objects. The idea that scientific knowledge is perspectival and dependent upon the scientist’s experiences and life-world features also prominently in the work of Merleau-Ponty. “Everything that I know about the world, even through science, I know from a perspective that is my own or from an experience of the world without which scientific symbols would be meaningless. The entire universe of science is constructed upon the lived world” (Merleau-Ponty 2012, lxxii). Importantly, such phenomenological reflections on the basic epistemic role of the life-world are supported by recent conceptions of agent-based modeling (Giere 2010) and by experimental results concerning the relationship between perceptual and con- ceptual learning (Landy & Goldstone 2007). 15 “Any understanding of reality is by definition perspectival. Effacing our perspective does not bring us any closer to the world. It merely prevents us from understanding anything about the world at all.” (Zahavi 2019, 28) 16 “Phenomenology is not out to dispute the value of science and is not denying that scientific investigations can lead to new insights and expand our understanding of reality. But phenomenologists do reject the idea that natural science can provide an exhaustive account of reality. Importantly, this does not entail that phenome- nology is, as such, opposed to quantitative methods and studies. The latter are excellent, but only when addressing quantitative questions.” (Zahavi 2019, 52) 3 The scientific perspective Husserl’s main work concerning philosophy of science is his The Crisis of European Sciences and Transcendental Phenomenology. According to Husserl, the success of modern science (beginning with Galileo), i.e., the success of mathematizing and quantizing nature, has led to several misunderstandings. Firstly, it made scientists, as well as philosophers, believe that the methods of science are the only legitimate forms of gaining knowledge. Husserl opposes this methodological naturalism that implies that even philosophy must proceed like a natural science. Secondly, Husserl bemoans that due to the objectivism of modern age it has become commonplace to take the mathematical models and formulae to be the true reality, while the life-world, the world of our everyday experiences, the world of tables and chairs, is demoted to some kind of illusion (cf. particularly Husserl 1970, 48–53).11 Husserl emphasizes that mathematics and geometry are only methods to describe physical reality, they are not the “true” reality lying behind what we can intuitively (“anschaulich”) observe. What he criticizes is that scientists and philosophers seem to have forgotten this and tend to confuse what is a method with what is reality. He stresses that the life-world serves as the epistemic grounding12 and the meaning-foundation of all scientific activities.13 Husserl’s conception of the life-world proved useful in many contexts and his thesis that the life-world remains the epistemic foundation for any scientific theory should be particularly interesting to proponents of perspectivism.14 However, if you wish to abstain from using the terminology of a life-world, the basic idea remains that no matter how abstract your scientific theories are, their justification, ultimately, lies in ordinary experiences, in what is immediately given. In Husserl’s words, “the inductive scientific judging” of the “exact objective sciences that by going beyond the immedi- ately experienced deduces the non-experienced is always dependent on its ultimate legitimizing basis, on the immediate data of experience” (Husserl 1973, 121; my translation). To be sure, Husserl neither criticizes science per se nor its methods. And, of course, he does not dispute its success. The sciences, particularly physics, are extremely European Journal for Philosophy of Science (2020) 10: 30 Page 11 of 27 30 30 successful in what they are doing. However, he criticizes the conclusions naturalists and objectivist realists draw from the success of science. 15 “Any understanding of reality is by definition perspectival. Effacing our perspective does not bring us an closer to the world. It merely prevents us from understanding anything about the world at all.” (Zahavi 201 28) 3 The scientific perspective The upshot is as follows: All we know about the physical world, we know, ultimately, by way of perceptual experiences. But our experiences do not constitute some magical purely objective view from nowhere. Our experiences are shaped by previous experiences as well as by our beliefs and expectations. What is more, perceptual experiences affect our concept formation and symbolic thinking, and the concepts we use, in turn, affect our experiences. Accord- ingly, when we do physics, when we establish a mathematical model that is intention- ally directed at the world, we do not have a purely objective mathematical model on the one hand and objective reality on the other hand. Instead, on both sides we have models influenced by subjective factors. European Journal for Philosophy of Science (2020) 10: 30 30 Page 12 of 27 Referring to a famous example put forward by Arthur Eddington (Eddington 1928, ixf.), I may exemplify the above as follows: Looking at this table in front of me is a perceptual act (precisely but perspectivally) directed at this table in front of me. A physical-mathematical model of this table, describing its atomic structure etc., is also (precisely but perspectivally) directed at this table. Both are legitimate ways of being intentionally directed at the table, each elucidating different aspects of the same object.17 They complement each other and in certain ways they influence each other. They are both legitimate perspectives, one of them being epistemologically more fundamental. It is our everyday experience that is episte- mologically more fundamental because this is the one from which our mathemat- ical model arises and the one the mathematical model, ultimately, must conform to.18 We know from Mirja Hartimo’s analysis of Husserl’s private library that Husserl was not only interested in modern physics but that he studied it in great detail (Hartimo 2018). However, Husserl never practiced philosophy of physics in a narrow sense. He never engaged with, say general relativity, to draw philosophical conclusions. 17 A similar approach to Eddington’s example has been championed by Putnam, who explicitly refers to Husserl in this context (Putnam 1987). Cf. De Caro 2020 for affinities between Putnam and perspectivism. Cf. Zahavi 2004 for affinities between Putnam and the phenomenological tradition. 18 One might argue that while the life-world-perspective is epistemologically more fundamental than the scientific perspective, the scientific perspective reveals ontologically more fundamental aspects in the sense that ordinary objects such as tables and chairs somehow emerge from elementary particles and quantum fields. I see no reason why phenomenologists should deny such a claim. 19 We do know, however, that Husserl was greatly interested in such a project. This is why he supported the work of his pupil Oskar Becker who aimed precisely at a phenomenological grounding of general relativity. In a letter to Weyl, Husserl wrote: “It [Becker’s Habilitation] is nothing less than a synthesis of Einstein’s and yours discoveries with my phenomenological investigations on nature. […] What will Einstein say when it is shown that nature requires a relativity-theoretical structure on the a priori grounds of phenomenology and not on positivistic principles, and that only in this way a completely understandable, and ultimately exact, science is possible.” (Husserl in Mancosu & Ryckman 2005, 160 f.) 20 The classical phenomenologists most explicitly arguing that phenomenology and modern physics can in one way or another complement each other are Oskar Becker and Merleau-Ponty. Hermann Weyl and Fritz London are notable figures in early modern physics making similar claims. For phenomenological motifs in Hermann Weyl’s development of the gauge principle, cf. Ryckman 2005 and Ryckman forthcoming; for phenomenological motifs in Fritz London’s interpretation of quantum mechanics, cf. French 2002 and French forthcoming. 3 The scientific perspective And he never attempted a phenomenological interpretation or grounding of general relativity or quantum mechanics.19 This is a bit surprising since many physicists, as well as many philosophers of physics, believe that the downfall of classical mechanics and the rise of general relativity and particularly quantum mechanics support key ideas of Husserl’s phenomenology of science.20 Concerning general relativity, Merleau-Ponty states: The physics of relativity confirms that absolute and final objectivity is a mere dream by showing how each particular observation is strictly linked to the location of the observer and cannot be abstracted from this particular situation; it also rejects the notion of an absolute observer. We can no longer flatter ourselves with the idea that, in science, the exercise of a pure and unsituated 20 The classical phenomenologists most explicitly arguing that phenomenology and modern physics can in one way or another complement each other are Oskar Becker and Merleau-Ponty. Hermann Weyl and Fritz London are notable figures in early modern physics making similar claims. For phenomenological motifs in Hermann Weyl’s development of the gauge principle, cf. Ryckman 2005 and Ryckman forthcoming; for phenomenological motifs in Fritz London’s interpretation of quantum mechanics, cf. French 2002 and French 21 I am thankful to an anonymous referee of this journal for pressing me on making this clarification. 22 As a side note, it has been argued that general covariance in general relativity (and more precisely gauge invariance in a broader context) “is an indication of the relational character of fundamental observables in physics” (Rovelli 2014, 103). It would be worthwhile to discuss whether general covariance and gauge invariance support perspectivist and phenomenological approaches. European Journal for Philosophy of Science (2020) 10: 30 Page 13 of 27 30 30 intellect can allow us to gain access to an object free of all human traces, just as God would see it. This does not make the need for scientific research any less pressing; in fact, the only thing under attack is the dogmatism of a science that thinks itself capable of absolute and complete knowledge. We are simply doing justice to each of the variety of elements in human experience and, in particular, to sensory perception. (Merleau-Ponty 2004, 44f.) It is to be noted that Merleau-Ponty’s remark is misleading since in the theory of relativity observation is not linked to the location of the observer but to the frame of reference of the observer.21 The principle of relativity implies that there is no privileged frame of reference; the laws of physics are the same in all inertial frames of reference. Special relativity is built upon the principle of relativity (first postulate) and the postulate that in a vacuum the speed of light is constant for all observers. Together, these two postulates have several implications that show that some of the facts that we usually consider to be “objective” are in fact observer- dependent. For instance, special relativity implies the relativity of simultaneity: It depends on the observer’s frame of reference whether two events separated in space occur at the same time. There is no objective or absolute sense in which we could tell that two spatially separate events take place simultaneously. When we turn to general relativity, we see that space and time are not absolute, not a fixed background, but that the geometry of spacetime itself is influenced by what is going on within spacetime, namely by the energy-momentum of matter. This means that there is a reciprocal relationship between spacetime and what it contains (including the embodied observer).22 It is to be noted that Merleau-Ponty’s remark is misleading since in the theory of relativity observation is not linked to the location of the observer but to the frame of reference of the observer.21 The principle of relativity implies that there is no privileged frame of reference; the laws of physics are the same in all inertial frames of reference. 21 I am thankful to an anonymous referee of this journal for pressing me on making this clarification. 22 Special relativity is built upon the principle of relativity (first postulate) and the postulate that in a vacuum the speed of light is constant for all observers. Together, these two postulates have several implications that show that some of the facts that we usually consider to be “objective” are in fact observer- dependent. For instance, special relativity implies the relativity of simultaneity: It depends on the observer’s frame of reference whether two events separated in space occur at the same time. There is no objective or absolute sense in which we could tell that two spatially separate events take place simultaneously. When we turn to general relativity, we see that space and time are not absolute, not a fixed background, but that the geometry of spacetime itself is influenced by what is going on within spacetime, namely by the energy-momentum of matter. This means that there is a reciprocal relationship between spacetime and what it contains (including the embodied observer).22 All this deserves phenomenological reflections on its own, but the theory of relativity is no focus of this paper. Instead, in the following two sections, we will address how perspectivist and phenomenological approaches to science relate to quantum mechanics. This focus on quantum mechanics arises naturally because when Merleau-Ponty spells out his phenomenological perspectivism, he crucially draws on quantum mechanics. Among the most famous of the classical phenomenologists (Husserl, Scheler, Heidegger, Stein, Sartre, Merleau-Ponty), Merleau-Ponty was the one most explicitly engaging with the natural sciences. He believed that modern physics supports phenomenological approaches to science and reality and although he also engaged with the theory of relativity, he believed that quantum mechanics best supports phenomenological approaches. In the following section, we shed light on Merleau-Ponty’s phenomenology of physics, emphasizing how he radicalizes ideas we find in Husserl. In the final section, we shall see that there are interesting systematic similarities between ideas we find in Merleau-Ponty and a current popular interpreta- tion of quantum mechanics. European Journal for Philosophy of Science (2020) 10: 30 30 Page 14 of 27 23 This monograph of London and Bauer basically has two objectives: First, to provide a “concise and simple” (London & Bauer 1983, 219) account of the measurement problem in the spirit of von Neumann’s ground- breaking Mathematische Grundlagen der Quantenmechanik (1932). Providing the axiomatic foundations of quantum mechanics, von Neumann’s book was one of the most influential works of early quantum mechanics. London and Bauer were in broad agreement with von Neumann. They did not understand their monograph as a counter project but as a more accessible version of von Neumann’s highly technical work which was written in German. Secondly, London and Bauer shed more light on the relationship between the observed and the observer, aiming at clarifying the role of consciousness in quantum measurement. In perfect agreement with Merleau-Ponty they hold that modern physics reveals that “the idea of an observable world totally independent of the observer, was a vacuous idea” (London & Bauer 1983, 220). For an analysis of the phenomenological motifs in Fritz London’s approach to quantum mechanics and for how this approach can serve as a starting point for a genuinely phenomenological interpretation of quantum mechanics, cf. French 2002 and particularly French forthcoming. 24 The quoted article by Adam Frank, Marcelo Gleiser, and Evan Thompson is a great example: https://aeon. co/essays/the-blind-spot-of-science-is-the-neglect-of-lived-experience?fbclid=IwAR1QpuiKEPuaE3aH37 xTZA6bhHs9vstxBXt_Znw84NFkSEBdj389BuAirsA. Retrieved on February 13, 2020. 4 Merleau-Ponty’s partial realism European Journal for Philosophy of Science (2020) 10: 30 Page 15 of 27 30 30 Merleau-Ponty considers the “relations between the observer and the observed” to be the “ultimate physical beings” (Merleau-Ponty 1968, 15). Underpinned by the London & Bauer account of quantum mechanics, he doubts “the idea that every object has an individual existence,” and instead refers to physical objects as “generic realities” (Merleau-Ponty 2003, 92). Furthermore, he explicitly addresses one of the most important questions of philos- ophy of science: What do we observe in scientific measurements? Of course, this question is particularly pressing in quantum mechanics. Contrasting the role of the measuring apparatus in classical physics and quantum mechanics, Merleau-Ponty states that while classically “the apparatus is the prolongation of our senses” in quantum mechanics “[t]he apparatus does not present the object to us.” Instead, “[i]t realizes a sampling of this phenomenon as well as a fixation. […] Known nature is artificial nature” (Merleau-Ponty 2003, 93). Unfortunately, Merleau-Ponty does not offer a detailed and sophisticated analysis of what is “artificial” about quantum measurements. In general, in his discussion of quantum mechanics we find much that is inspiring and helpful but his remarks are often vague and in order to advance current debates we would need a more precise analysis of how quantum mechanics supports perspectivist and phenomenological approaches. The next section is intended to make a step towards this goal. This being said, I believe that Merleau-Ponty is right and that there is a fundamental difference between looking through telescopes or microscopes on the one hand and using measuring devices in quantum mechanics on the other hand. In the case of telescopes and microscopes, there is a rather straightforward sense in which we directly observe the object in question. But it would be quite a stretch to say that the same is true when looking at the photographs gained by cloud chambers and bubble chambers that visualize the tracks of charged particles.25 What is more, while cloud chambers and bubble chambers have a photographic readout, the devices that are now common, such as particle colliders like the LHC, have a purely electronic readout. What we gain from LHC experiments is data – big data. “Data pours out of the LHC detectors at a blistering rate. 4 Merleau-Ponty’s partial realism Merleau-Ponty has a reputation for having deeply cared about the natural sciences, particularly psychology. This makes him a promising and popular point of origin for many contemporary phenomenologists working on the interface to psychology and the cognitive sciences. It is less well known, however, that he also explicitly addressed physics, contemplating how philosophy and physics can enrich each other and what a phenomenologically grounded physics may look like. This is particularly true for his “Modern Science and Nature” which is part of the lecture courses published in La Nature. Here, Merleau-Ponty carefully engages with quantum me- chanics, outlining his phenomenological approach to physics. Merleau-Ponty discusses the limits of objectivity and aims at a physics that takes into consideration the physicist who observes and experiments. He believes that modern physics, particularly quantum mechanics, exemplifies or at least leads to a new kind of science that engages in self-criticism, reflectively addressing its relationship to the objects it studies (Merleau-Ponty 2003, 85). In this context, he discusses and draws on the interpretation of measurement in quantum mechanics delivered by London and Bauer (London & Bauer 1939; 1983) that was itself heavily influenced by Husserl’s phenomenology.23 For Merleau-Ponty, physics in its most perfected form abandons the idea of deliv- ering a completely objective picture of the world. Instead, physics needs to put the physicist into physics and account for the fact that the life-world predates all scientific endeavors. In his words: But a physics that has learned to situate the physicist physically, a psychology that has learned to situate the psychologist in the socio-historical world, have lost the illusion of the absolute view from above: they do not only tolerate, they enjoin a radical examination of our belongingness to the world before all science. (Merleau-Ponty 1968, 27) There is an increasing number of philosophers and physicists who insist “that imme- diate experience and the world can never be separated” and agree that science cannot “give us a complete, objective description of cosmic history, distinct from us and our perception of it.”24 Doubting “that the physical object in itself” pre-exists physics, 24 The quoted article by Adam Frank, Marcelo Gleiser, and Evan Thompson is a great example: https://aeon. co/essays/the-blind-spot-of-science-is-the-neglect-of-lived-experience?fbclid=IwAR1QpuiKEPuaE3aH37 xTZA6bhHs9vstxBXt_Znw84NFkSEBdj389BuAirsA. Retrieved on February 13, 2020. 25 For a discussion of unobservable scientific entities from a phenomenological perspective, cf. Wiltsche 2012; Berghofer 2018b. 26 . Retrieved on February 13, 2020. In this context, cf., e.g., Karaca 2017, 344. 27 For discussions of what kind of observation is taking place in LHC experiments, cf., e.g., Beauchemin 2017; Karaca 2017, 2018. Concerning the relationship between theory and observation, Beauchemin argues that reflection on such big data experiments “indicates that the frontier between theory and observation is blurry and that the dichotomy theory-experiment should be revised” (Beauchemin 2017, 275). Although Beauchemin stresses the theory-ladenness of LHC experiments, he does not dispute the objectivity of the corresponding empirical facts. Similarly, Karaca argues that “the exploratory data selection procedure carried out in the ATLAS experiment is theory-laden in the sense that its implementation is crucially dependent on the aforementioned theoretical models that the experiment is aimed to test” (Karaca 2017, 350; cf. also Karaca 2018, 5449), insisting, however, that no vicious circularity is happening there (Karaca 2018, 5450). Werner Callebaut has explicitly argued that Giere’s perspectivism “provides the best resources currently at our disposal to tackle many of the philosophical issues” surrounding big data biology (Callebaut 2012, 69). 25 For a discussion of unobservable scientific entities from a phenomenological perspective, cf. Wiltsche 201 Berghofer 2018b. 26 26 . Retrieved on February 13, 2020. In this context, cf., e.g., Karaca 2017, 344. 27 4 Merleau-Ponty’s partial realism Even after filtering out 99% of it, in 2018 we gathered 88 petabytes of data.”26 I do not say that there is anything genuinely problematic about this process but it is a process that needs careful philosophical-phenomenological reflection and it is a process that is far from delivering a purely objective picture of the world.27 30 Page 16 of 27 European Journal for Philosophy of Science (2020) 10: 30 The point is that similar to Giere, Merleau-Ponty rejects the idea that scientific observation provides us with an objective picture of nature. Of course, they differ in their respective reasons for rejecting this idea. Giere emphasizes that scientific instru- ments only engage with a limited aspect of nature. Merleau-Ponty seems to stress that according to quantum mechanics the act of observing inevitably affects the observed reality (e.g., when electrons, depending on the experimental setup, either behave like particles or like waves). Another important similarity to Giere is that Merleau-Ponty aims at establishing a position that is in between objectivist realism and anti-realism, namely a version of realism that rejects the ideal of providing a complete and purely objective view of the world and that takes into account the subject that is doing science. To get a better grip on Merleau-Ponty’s approach to the scientific realism debate, let us see which positions he rejects. One might think that phenomenologists feel sympa- thetic to instrumentalist accounts (and for some phenomenologists this is certainly true), but Merleau-Ponty clearly opposes such views. Merleau-Ponty does not use the term “instrumentalism,” but he introduces the following position: Physics should not be conceived as a search for the truth, it should give up determining a real physics: it would be only an ensemble of measurements linked to equations, allowing [us] to foresee the result of future measure- ments. Formalist physics receives all freedom, but it loses its ontological content. It signifies no mode of being, no reality. (Merleau-Ponty 2003, 95f.) This position, expressing the core ideas of instrumentalism, is rejected by him without much argument. For Merleau-Ponty, it is clear that physics, correctly interpreted, indeed tells us something significant about the nature of reality. Physics is not a mere tool to make predictions, nor can reality be reduced to what is measured and observed. 28 Unfortunately, in the English translation (Merleau-Ponty 2003, 98), the quotation marks are missing. 29 Merleau-Ponty’s partial realism denies the existence of individual observer-independent physical objects and intrinsic properties. Instead, as we have seen, it regards the “structural relations” that refer “to certain mathematical forms necessary for the description of the relation of the subject to the object” as the fundamental entities. Accordingly, Merleau-Ponty’s partial realism has much common ground with structural realism. In current debates, structural realism is a form of selective realism that focuses on the mathematical structure of scientific theories. A distinction is made between epistemic and ontic structural realism. Broadly speaking, while epistemic structural realism says that structures are all we can know, ontic structural realism (OSR) says that structures are all there is. OSR enjoys much popularity and comes in many flavors, ranging from the claim that objects (at the fundamental level) do not possess intrinsic properties to the eliminativist version, according to which there are no objects but only structures (for more details, cf. Ladyman 2016; Berghofer 2018c). Of course, the difference between Merleau-Ponty and proponents of OSR is that the latter do not regard their structural relations as relations between subject and object. Proponents of OSR presuppose the observer-independence of physical reality. 4 Merleau-Ponty’s partial realism He goes on, drawing on the work of the French physicist and logician Paulette Destouches-Fevrier, to point out that it would also be a mistake to adopt an idealist position. The problem with idealism is that just like standard realism it amounts to a form of objectivism. To be more precise, idealism is an objectivism that “objectifies human representations” (Merleau-Ponty 2003, 96). Instead, Merleau-Ponty is convinced that “[t]he relations between reality and measurement must be conceived outside of the dichotomy of in-itself/representation” (Merleau- Ponty 2003, 96). Acknowledging that “[p]hysics cannot be realist in the classical sense” but “cannot be idealist, either,” Merleau-Ponty chooses to term his position “a ‘partial realism’ or a ‘participationist’ conception” (Merleau-Ponty 2003, 97f.). This terminology, adopted from Paulette Destouches-Fevrier, highlights the inter- relatedness and inseparability of the observer and the observed. The term “partial realism” emphasizes that although this view is not a traditional form of realism, it is supposed to be some form of realism. Returning to our initial question of how to place Merleau-Ponty in the scientific realism debate, we need to ask: which form of realism? What are the fundamental objects of reality according to his partial realism, his participationist conception? European Journal for Philosophy of Science (2020) 10: 30 Page 17 of 27 30 30 In this context, he calls reality a “structural plane,” and continues by quoting a long passage from Destouches-Fevrier.28 Here Destouches-Fevrier says that reality “tran- scends the opposition object-subject.” The focus is on the “structural relations” between subject and object. These structural relations “refer not to an object, but to certain mathematical forms necessary for the description of the relation of the subject to the object.” The ontological significance of the structural relations is highlighted by pointing out that “they are independent of the results of the processes of measurement” and by perhaps misleadingly comparing them “to the Platonic objectivity.” (Destouches-Fevrier quoted in Merleau-Ponty 2003, 98). Unfortunately, Merleau-Ponty does not do much to clarify or go beyond these remarks of Destouches-Fevrier. However, in The Visible and the Invisible he holds that modern physics is obliged “to recognize as ultimate physical beings in full right relations between the observer and the observed” (Merleau-Ponty 1968, 15). In this light, there is little doubt that Merleau-Ponty subscribes to the structuralist view drafted by Destouches-Fevrier. 4 Merleau-Ponty’s partial realism Accordingly, concerning our questions of how to understand the fundamental objects of reality according to Merleau-Ponty’s partial realism, we get the following answer: The fundamental objects are the structural relations between the observer and the observed. These relations can neither be reduced to the objective nor to the subjective. Reality transcends this opposition. Reality can only be understood or even consists in the relations between the observer and the observed.29 Admittedly, all this remains vague. However, in the next section, we shall see that there is a novel interpretation of quantum mechanics that agrees with Merleau-Ponty that quantum mechanics should be understood as revealing the fundamental relatedness between subject and object: QBism. Before turning to a more substantial discussion of quantum mechanics, let me briefly recapitulate and put into perspective what we have achieved so far. In Sect. 1, we have seen that for Giere science is an agent-based endeavor. This idea is fleshed out by Husserl in terms of the life-world serving as the meaning-foundation for all the individual sciences. Merleau-Ponty radicalizes this idea by making the ontological claim that reality is in some sense observer-dependent (perhaps not neces- sarily mind-dependent). Of course, phenomenologists are not committed to this strong claim of observer-dependence, ontologically understood. Husserl argues that subjec- tivity and subjective experiences constitute the source of all knowledge and justification European Journal for Philosophy of Science (2020) 10: 30 30 Page 18 of 27 30 Page 18 of 27 but this is an epistemological claim and Husserl’s transcendental idealism can be interpreted as a methodological-epistemological project free from strong metaphysical implications. According to Husserl’s understanding of the mathematical sciences including theoretical physics, these sciences strive for a maximum of objectivity by looking at the world from the third-person perspective, mathematizing and quantizing their target system. And although full objectivity can never be gained because the scientists’ life-worlds remain the epistemic grounding and meaning-foundations for all scientific theories, for Husserl there is nothing wrong with the individual sciences to proceed in this manner.30 We only need to be careful in how to interpret scientific theories and be aware of their limitations.31 Merleau-Ponty goes beyond such interpretational claims. There is at least one methodological and one ontological claim he adds.32 The methodological claim is that the sciences, particularly physics, must refrain from aiming at a purely objective account of the world. Instead, they must incorporate the first-person perspective into science. 30 However, it should be mentioned that there are some passages in Husserl’s oeuvre that suggest that physics in its most elaborated form would successfully make the physicist and her life-world subjects of investigation, abandoning the idea of a purely objective third-person perspective (Husserl 1970, 295; 2002, 287). It is not entirely clear, however, whether here Husserl wants to say that physics must be phenomenologically clarified so that physics (as it is) can be ultimately grounded and justified. Or whether his claim is to be understood in the Merleau-Pontyan sense that physics can only succeed in its goal of clarifying nature if it succeeds in incorporating the physicist into the physical theories. 31 “There is no pure third-person perspective, just as there is no view from nowhere. This is, of course, not to say that there is no third-person perspective, but merely that such a perspective is, precisely, a perspective from somewhere. It is a view that we can adopt on the world.” (Zahavi 2019, 54) 32 Cf. the beginning of section 2 for a similar distinction we made in the context of perspectivism. 4 Merleau-Ponty’s partial realism Only by doing so can science unveil the most fundamental structures of reality. The ontological claim is that reality is essentially observer-dependent and that the fundamental objects are relations between the observer and the observed. Merleau- Ponty’s position seems to qualify as a version of perspectival realism since he rejects objectivist realism and aims at a position in between objectivist realism and anti- realism. Scientific theories are perspectival in the sense that they need to incorporate the scientists’ relations to the objects. Scientific facts are perspectival in the sense that they depend on, or, perhaps more accurately, consist of the scientists’ relations to the objects. Nevertheless, one might doubt that Merleau-Ponty’s partial realism is a form of realism after all. This is because scientific realism is usually associated with an ontological commitment to the mind-independence of reality. However, we have to note, first, that Merleau-Ponty’s partial realism does not amount to a form of traditional anti-realism (instrumentalism, idealism, constructivism) and, secondly and more im- portantly, that Merleau-Ponty would strongly deny the anti-realist claim that the sciences cannot tell us anything about the nature of reality. In this regard of aiming to be in between standard versions of realism and anti-realism, Merleau-Ponty’s partial realism shares significant similarities with QBism, classified by its chief advocate Christopher Fuchs as a “participatory realism.” Fuchs insists that quantum mechanics must be understood “as being part of a realist program, i.e., as an attempt to say something about what the world is like, how it is put together, and what’s the stuff of it” (Fuchs 2017, 117). Importantly, according to Fuchs, the conclusions he draws from quantum mechanics are not something we need to artificially read into it. Quite the European Journal for Philosophy of Science (2020) 10: 30 Page 19 of 27 30 30 contrary, “[q]uantum theory itself threw these considerations before us!” (Fuchs 2017, 115). This means we need to take quantum mechanics, its features and phenomena, at face value and by doing so we learn something new about reality that was hidden in classical physics. contrary, “[q]uantum theory itself threw these considerations before us!” (Fuchs 2017, 115). This means we need to take quantum mechanics, its features and phenomena, at face value and by doing so we learn something new about reality that was hidden in classical physics. I believe that this perfectly captures Merleau-Ponty’s attitude towards quantum mechanics. 4 Merleau-Ponty’s partial realism Quantum mechanics does not deliver a purely objective view on the world as the objectivist realist would have it; instead, quantum mechanics shows us that a purely objective view on the world is impossible. Quantum mechanics does not represent an observer-independent reality; instead, quantum mechanics calls into ques- tion the idea that there is a purely observer-independent reality behind the phenomena. How quantum mechanics might support perspectivist and phenomenological ap- proaches to science and how QBism relates to Merleau-Ponty’s ideas are the topics of the following section. 5 QBism In quantum mechanics we find many concepts and phenomena that seem to support perspectivist and phenomenological approaches to science and reality, undermining our classical world-view. Determinism is called into question, Heisenberg’s uncertainty principle imposes certain limitations on what we can know about reality, and comple- mentarity and entanglement are often viewed as revealing that acts of measurement necessarily affect observed reality. Particularly in the early days of quantum mechanics, complementarity was considered the key feature of quantum mechanics and was interpreted in a way that seems to support a perspectivist picture. Heisenberg summa- rized the Copenhagen understanding of complementarity as follows: By this term ‘complementarity’ Bohr intended to characterize the fact that the same phenomenon can sometimes be described by very different, possibly even contradictory pictures, which are complementary in the sense that both pictures are necessary if the ‘quantum’ character of the phenomenon shall be made visible. (Heisenberg 1977, 6) Wave-particle duality and Heisenberg’s uncertainty principle are often considered the most prominent manifestations of complementarity. Heisenberg’s uncertainty principle famously says that with respect to complementary variables such as position and momentum the more precisely we determine the one, the less we know about the other. As Heisenberg himself noted this implies that “[e]ven in principle we cannot know the present in all detail” (Heisenberg 1983, 83). Concerning measurements in quantum mechanics, Frescura and Hiley express a common attitude among physicists when they say that the issues surrounding complementarity imply that not all aspects of a system can be viewed simultaneously. By using one particular piece of apparatus only certain features could be made manifest at the expense of others, while with a different piece of apparatus another European Journal for Philosophy of Science (2020) 10: 30 30 Page 20 of 27 complementary aspect could be made manifest in such a way that the original set became non-manifest, that is, the original attributes were no longer well defined (Frescura & Hiley 1984). complementary aspect could be made manifest in such a way that the original set became non-manifest, that is, the original attributes were no longer well defined (Frescura & Hiley 1984). This claim that scientific instruments can only shed light on certain limited aspects of reality is precisely the claim we find in Giere (2006, chapter 3) as discussed in section 1. 5 QBism However, concerning the examples discussed by Giere, one might argue that one could simply add all the perspectives delivered by different instruments to gain one complete picture of reality. Complementarity in quantum mechanics, on the other hand, seems to impose even more rigorous limitations on scientific observations, revealing a genuinely perspectival element of science that disallows gaining one complete picture by adding different perspectives: An increase of information with respect to one set of properties goes hand in hand with a decrease of information with respect to another set of properties. In this picture, quantum mechanics reveals limits to objectivity in the sense that our knowledge of quantum systems is necessarily perspectival, we can always only know certain aspects, never nature in its entirety. Of course, drawing such conclusions from quantum mechanics is a minefield since there is no consensus on its ontological as well as epistemological implications. Interpreting quantum mechanics, offering a solution to the notorious measurement problem, is often considered the main topic of philosophy of physics. As a conse- quence, there exist a number of different interpretations that lead to very different pictures of reality. This is true also for the question of which conclusions to draw from Heisenberg’s uncertainty principle (cf. Hilgevoord & Uffink 2016).33 Many open questions in quantum mechanics concern the wave function and its apparent collapse. Is the wave function something that really exists or is it merely a mathematical tool, useful to make predictions? Why is it that apparently the outcomes of measurements are always definite states? Does the wave function collapse upon measurement? If so, how and why? This is where so-called interpretations of quantum mechanics usually come into play. Some of them, most notably Bohmian mechanics and the many-worlds interpretation, preserve (in some sense and with some costs) the deterministic picture we know from classical mechanics. Proponents of these interpre- tations typically view the wave function as physically real and believe that quantum mechanics provides an objective picture of reality. Other interpretations, such as Rovelli’s relational interpretation (Rovelli 1996), Dieks’ perspectivalism (Dieks 2019a, 2019b), or Healey’s pragmatist approach (Healey 2012) in one way or another contest the idea that physics delivers a purely objective picture of the world. One interpretation in this camp that has gained particular attention recently is QBism. 33 However, it should be mentioned that even in deterministic Bohmian mechanics, “there are unavoidable limitations to our knowledge of particles. In fact, once the wave function is prepared, there is an absolute uncertainty regarding the positions of the particles” (Solé et al. 2016, 22). 5 QBism I shall focus on QBism also because the ideas and the terminology we find here are particu- larly close to what we found in Merleau-Ponty. To be sure, I do not claim that perspectivists or phenomenologists must or should subscribe to QBism. Nor do I argue that they are committed to any “subjective” interpretation of quantum mechanics. However, QBism might be the interpretation that most consistently promotes some of the ideas we find in phenomenology, which is why European Journal for Philosophy of Science (2020) 10: 30 Page 21 of 27 30 30 it is worth considering it in more detail (cf. particularly Bitbol forthcoming and Tremblaye forthcoming). it is worth considering it in more detail (cf. particularly Bitbol forthcoming and Tremblaye forthcoming). In QBism the agent and her experiences and expectations play a central role. The distinctive idea of QBism is to apply a personalist Bayesian account of probability, as it has been developed by Bruno de Finetti, to quantum probabilities. This means that probabilities in quantum mechanics are interpreted not as objective but as subjective probabilities. Accordingly, in QBism quantum states do not represent objective reality but instead represent an agent’s subjective degrees of beliefs about her future experi- ences. Consequently, the wave function is not physically real but a mathematical tool that encodes one’s expectations about one’s future experiences. In short, QBism argues that quantum states do “not represent an element of physical reality but an agent s personal probability assignments, reflecting his subjective degrees of belief about the future content of his experience” (Fuchs & Schack 2015, 1). The first thing to note is that even opponents of any subjective interpretation of quantum mechanics concede that such an account delivers a straightforward solution to the notorious apparent collapse of the wave function. This is because any “approach according to which the wave function is not something real, but represents a subjective information, explains the collapse at quantum measurement perfectly: it is just a process of updating the information the observer has” (Vaidman 2014, 17). To put it differently, according to QBism “[t]he notorious ‘collapse of the wave-function’ is nothing but the updating of an agent’s state assignment on the basis of her experience” (Fuchs et al. 2014, 749). 34 Similarly, QBism dissolves another problem that has been puzzling physicists, namely non-locality (cf. Fuchs et al. 2014; Timpson 2008). 35 This objection has been raised, e.g., by Chen (2019, 6). For how the objection of empirical incoherence emerges from phenomenological reasoning, cf. the beginning of Sect. 2 of the present paper, particularly footnote 11. 5 QBism In this sense, QBism dissolves the measurement problem – the problem does not even show up.34 A further advantage of QBism is that it avoids certain implausible consequences that plague realist interpretations of the wave function. Mathematically speaking, wave functions are vectors in a Hilbert space. This is often expressed by saying that “Wave functions live in Hilbert space” (Griffiths 2018, 94). A Hilbert space is an abstract mathematical concept, namely a complete vector space on which an inner product is defined. But if the wave function is something real, does this mean that mathematical Hilbert space is physically real too? In fact, one can find prominent voices championing Hilbert space realism (e.g. Carroll & Singh 2019) but most consider this an implausible and unwarranted mathematization of nature and it has been pointed out that only “[v]ery few people are willing to defend Hilbert space realism in print” (Wallace 2013, 216). A similar but more subtle form of mathematization takes place in configuration space realism, i.e., the project of reifying the 3 N-dimensional configuration space, N being the number of the particles in the universe. The main proponent of this view is David Albert, who at one point considered our impression that we live in three- dimensional space “somehow flatly illusory” (Albert 1996, 277). Configuration space realism, often referred to as “wave function realism,” has been quite popular and has sparked much controversy. However, if configuration space realism is meant to imply that the physical space of our everyday experiences is demoted to some kind of illusion, then this position is in danger of being empirically incoherent.35 Furthermore, one 30 Page 22 of 27 European Journal for Philosophy of Science (2020) 10: 30 30 Page 22 of 27 might object that configuration space realism “makes the same unmotivated conceptual move as Hilbert space realism: it reifies a mathematical space without any particular justification” (Wallace 2013, 217). Concerning the formal and technical apparatus of the mathematical sciences, Husserl warned us not to be “misled into taking these formulae and their formula-meaning for the true being of nature itself” (Husserl 1970, 44). We see that this problem also arises in quantum mechanics via wave function realism. 5 QBism The most common realist interpretations of quantum mechanics, the many-worlds interpretation, Bohmian mechanics, and GRW theory are all in danger of leading to a mathematization of nature that is not based on physical principles but on mathematical formalism. QBists such as Christopher Fuchs and Blake Stacey have pointed out that in these interpretations “the strategy has been to reify or objectify all the mathematical symbols of the theory and then explore whatever comes of the move” (Fuchs & Stacey 2019, 136). y QBism takes a different approach. Instead of reifying mathematical constructs, the idea is “to reduce the mathematical structure of quantum mechanics to some crisp physical statements” (Fuchs & Stacey 2016, 285). In this respect, QBism is similar to informational approaches to quantum mechanics that seek to reconstruct quantum mechanics based on fundamental physical principles (cf., e.g., Bub 2004, Chiribella et al. 2011, and Goyal 2012). A common idea is that “[i]n quantum mechanics, maximal information is not complete and cannot be completed” (Caves et al. 2002, 3) and according to QBists this result “can be regarded as the greatest triumph of Bayesian reasoning” (Caves et al. 2002, 3). I take it to be a virtue of QBism to resonate well with recent developments in quantum information theory and the insistence that information is necessarily incomplete also resonates well with perspectivism. So far, we have seen that the key move of QBism is to interpret quantum probabilities as personalist Bayesian probabilities and that this move allows QBism to dissolve the mea- surement problem and to avoid implausible mathematizations of nature. But what about more substantive claims about how reality works? Does QBism amount to some sort of instrumentalism according to which quantum mechanics does not teach us anything about reality? Indeed, the charge of instrumentalism is one of the most common objections to QBism. Importantly, proponents of QBism, particularly Christopher Fuchs, vehemently deny such anti-realist interpretations of QBism. Instead, Fuchs argues that, according to QBism, quantum mechanics tells us something very important about reality, namely “that reality is more than any third-person perspective can capture” (Fuchs 2017, 113). In this spirit, one of the objectives of QBism is to put the scientist back into science (Mermin 2014). Fuchs chose the label “participatory realism” for QBism due to the prominent role that the subject and her experiences play in QBism, highlighting the interrelatedness of subject and object (Fuchs 2017). 5 QBism Interpreting probabilities in quantum mechanics as subjective probabilities is only the starting point of the QBist project. The idea is that the fact that quantum mechanics does not deliver a purely objective picture of reality is not a shortcoming of the theory, instead quantum mechanics tells us that reality does not allow being objectively captured at a fundamental level. But what exactly does it mean when Fuchs calls QBism a “participatory realism”? How exactly are subject and object interrelated? Arguably, this concerns the most challenging aspect of QBism and it does not seem that a comprehensive philosophical European Journal for Philosophy of Science (2020) 10: 30 Page 23 of 27 3 30 foundation has been offered to address this question. However, to get a better idea of what QBists have in mind, we turn to the concept of measurement. According to QBism, “[a] measurement does not, as the term unfortunately suggests, reveal a pre- existing state of affairs. It is an action on the world by an agent that results in the creation of an outcome – a new experience for that agent” (Fuchs et al. 2014, 749). We remember that in the context of quantum measurement Merleau-Ponty said that “[t]he apparatus does not present the object to us” but “realizes a sampling of this phenom- enon as well as a fixation. […] Known nature is artificial nature” (Merleau-Ponty 2003, 93). Merleau-Ponty and QBists agree that in quantum mechanics the agent is not an innocent bystander. Instead, measurement is an active, participatory act: “Measurement is not a passive process, but instead a fundamentally participatory one” (Fuchs & Stacey 2019, 163). According to this participatory realism, there is an agent (or a plurality of agents) and an external physical system, and by acting upon the system the agent creates outcomes and these outcomes are the subject matter of quantum mechanics (Fuchs & Stacey 2019, 180). In this sense, “quantum mechanics itself does not deal directly with the objective world; it deals with the experiences of that objective world that belong to whatever particular agent is making use of quantum theory” (Fuchs et al. 2014, 750). Importantly, the agent cannot be reduced to the physical system and the concept of agency is not derivable from quantum mechanics (Fuchs & Stacey 2019, 180). 6 Conclusion Perspectivism is a new and promising approach to the scientific realism debate that aims at a middle way between traditional forms of scientific realism and anti-realism. Perspectivism rejects an objectivist picture, according to which the sciences deliver an exhaustive account of nature that is free from all subjective factors. Instead, perspectivism views science as an agent-based endeavor that delivers a certain perspective on nature. This is not to say that the world does not have the features ascribed to it by science, but to deny that there is nothing to know about the world over and above what is described by our best (possible) scientific theories. Although perspectivism is a novel position in current debates, it does have its forerunners. The present paper discusses to which degree perspectivist ideas can be found in the phenomenological tradition. We have seen that Husserl’s account of horizontal inten- tionality supplements Giere’s analysis of human observation. What is more, Husserl’s claim that all science is epistemically grounded in the life-world and thus can never be entirely free from subjective factors, fits well with Giere’s account of the perspectival character of scientific theories. Then, we have seen how Merleau-Ponty radicalizes this idea by making the further methodological claim that the sciences must incorporate the first-person per- spective into science and the ontological claim that reality is in some sense observer- dependent. Importantly, Merleau-Ponty contends that these are lessons that are motivated by science itself, namely by quantum mechanics. In this respect, Merleau-Ponty’s partial realism is in perfect agreement with Fuchs’ participatory realism. In the final section, we indicated how quantum mechanics could support perspectivist and phenomenological approaches and we shed light on QBism which is an interpretation of quantum mechanics particularly close to the ideas we find in Merleau-Ponty. Acknowledgments I would like to thank Andrea Pace Gainnotta for comments on an earlier version of this paper. Also, many thanks to two anonymous referees for many helpful suggestions. This work was supported by the Austrian Science Fund (FWF) [P 31758]. Funding information Open access funding provided by Austrian Science Fund (FWF). 5 QBism Even more importantly, this is not to be understood in a Kantian sense such that true reality is hidden behind the phenomena. “In a QBist understanding of quantum theory, it is not that nature is hidden from us. It is that it is not all there yet and never will be; nature is being hammered out as we speak” (Fuchs in Schlosshauer 2011, 285). In this picture, “there is no such thing as the universe in any completed and waiting-to-be-discovered sense” (Fuchs in Schlosshauer 2011, 285). Let us summarize how QBism relates to the scientific realism debate. QBism is not realist in the sense that it reifies the mathematical symbols that occur in the formalism of quantum mechanics. Particularly, the wave function is not interpreted as something that exists physically. However, QBism is anti-instrumentalist since it holds that there are very important lessons about reality we can learn from quantum mechanics. We learn that we live in a participatory universe and “that reality is more than any third-person perspective can capture” (Fuchs 2017, 113). QBism is not solipsist since it presupposes the existence of an external system upon which the agent acts, but it remains unclear how exactly to view the relationship between the agent and physical reality. There is general consensus that QBism delivers a consistent interpretation of quantum mechanics that avoids problems surrounding the apparent collapse of the wave function and non-locality. However, there seems to be a lack of a clear philosophical foundation. “Now, as a formal proposal, quantum Bayesianism is relatively clear and well developed. But it is rather less transparent philosophically” (Timpson 2008, 580). Perspectivist and phenome- nological approaches to science and physics might help QBism to find a suitable philo- sophical foundation. What should be clear from this section is that there are many systematically significant parallels between Fuchs’ participatory realism and Merleau-Ponty’s partial realism. Both emphasize the role of the subject (agent) and her experiences and draw attention to the relationship between the observer and the observed (the agent and the external system). Both insist that the physicist is not an innocent bystander, rejecting the idea that European Journal for Philosophy of Science (2020) 10: 30 30 Page 24 of 27 30 Page 24 of 27 scientific observation provides us with an objective picture of nature. 5 QBism Both abstain from mathematizing nature but declare that quantum mechanics quite straightforwardly tells us something very important about the nature and structure of reality. The lessons they draw, of course, are very different from what is claimed by objectivist realists. One lesson, namely the idea that a purely objective third-person perspective on the world is either impossible or cannot capture all of nature, is at the heart of any perspectivist account. It plays an important role in the phenomenological tradition where we find various degrees of perspectivism. The phenomenological tradition offers a rich experience-first epistemology, developing concepts such as horizontal intentionality and life-world that could prove immensely useful for advancing perspectivism. The present paper is supposed to pave the way for further attempts to introduce a phenomenological perspectivism into the philosophy of science. 6 Conclusion Funding information Open access funding provided by Austrian Science Fund (FWF Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and European Journal for Philosophy of Science (2020) 10: 30 Page 25 of 27 30 30 indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. References Albert, D. (1996). Elementary quantum metaphysics. In J. Cushing, A. Fine, & S. Goldstein (Eds.), Bohmian mechanics and quantum theory: An appraisal (pp. 277–284). Dordrecht: Springer. Albertazzi, L. (Ed.). (2013). Handbook of experimental phenomenology. Malden: Wiley-Blackwell. Barrett, J. (1999). The quantum mechanics of minds and worlds. Oxford: Oxford University Press. Beauchemin, P.-H. (2017). Autopsy of measurements with the ATLAS detector at the LHC. Synthese, 19 275–312. Berghofer, P. (2018a). Husserl’s conception of experiential justification: What it is and why it matters. Husserl Studies, 34(2), 145–170. Berghofer, P. (2018b). Transcendental phenomenology and unobservable entities. Perspectives, 7(1), 1–13. Berghofer, P. (2018c). Ontic structural realism and quantum field theory: Are there intrinsic properties at th most fundamental level of reality? Studies in History and Philosophy of Modern Physics, 62, 176–188 Berghofer, P. (2020). Towards a phenomenological conception of experiential justification. Synthese, 19 155–183. Bitbol, M. (forthcoming). A phenomenological ontology for physics: Merleau-Ponty and QBism, forthcoming. In P. Berghofer & H. Wiltsche (Ed), Phenomenological approaches to physics. Synthese Library, Springer. Brown, M. (2009). Models and perspectives on stage: Remarks on Giere’s scientific perspectivism. Studies in History and Philosophy of Science, 40, 213–220. y p y f Bub, J. (2004). Why the quantum? Studies in History and Philosophy of Modern Physics, 35, 241–266. Bub, J. (2004). Why the quantum? Studies in History and Philosophy of Modern Physics, 35, 241–266. Callebaut W (2012) Scientific perspectivism: A philosopher of science’s response to the challenge of big Callebaut, W. (2012). Scientific perspectivism: A philosopher of science’s response to the challenge of b data biology. Studies in History and Philosophy of Biological and Biomedical Sciences, 43, 69–80. Carroll, S., & Singh, A. (2019). Mad-dog everettianism: Quantum mechanics at its most minimal. In A. Aguirre, B. Foster, & Z. Merali (Eds.), What is fundamental? (pp. 95–104). Cham: Springer. g ( ) f (pp ) p g Caves, C., Fuchs, C., & Schack, R. (2002). Quantum probabilities as Bayesian probabilities. Physical Review A, 65, 1–6. Chakravartty, A. (2010). Perspectivism, inconsistent models, and contrastive explanation. Studies in History and Philosophy of Science, 41, 405–412. Chakravartty, Anjan (2017). Scientific realism. Edward N. Zalta (ed.), The Stanford encyclopedia of philosophy (summer 2017 edition). URL = <https://plato.stanford.edu/archives/sum2017/entries/scientific-realism/>. Chakravartty, A. (2018). Realism, antirealism, epistemic stances, and voluntarism. In J. Saatsi (Ed.), The Routledge handbook of scientific realism (pp. 225–236). New York: Routledge. Chen, E. K. (2019) Realism about the wave function. Philosophy Compass, 14(7). https://doi.org/10.11 /phc3.12611. References /phc3.12611. Chiribella, G., D’Ariano, G., & Periotti, P. (2011). Informational derivation of quantum theory. Physical /phc3.12611. Chiribella, G., D’Ariano, G., & Periotti, P. (2011). Informational derivation of quantum theory. Physical R i A 84 1 39 Chiribella, G., D’Ariano, G., & Periotti, P. (2011). Informational derivation of quantum theory. Physic Review A, 84, 1–39. Church, J. (2013). Possibilities of perception. Oxford: Oxford University Press. Cretu, A.-M., & Massimi, M. (Eds.). (2020). Knowledge from a human point of view. Cham: Springer. Cretu, A.-M., & Massimi, M. (Eds.). (2020). Knowledge from a human point o De Caro, M. (2020). Hilary Putnam on perspectivism and naturalism. In A.-M. Cretu & M. Massimi (Eds.), Knowledge from a human point of view (pp. 57–70). Cham: Springer. De Caro, M. (2020). Hilary Putnam on perspectivism and naturalism. In A.- Knowledge from a human point of view (pp. 57–70). Cham: Springer. Dieks, D. (2019a). Quantum reality, perspectivalism and covariance. Foundations of Physics, 49, 629–646 Dieks, D. (2019a). Quantum reality, perspectivalism and covariance. Found Dieks, D. (2019b). Quantum mechanics and perspectivalism. In O. Lombardi, S. Fortin, C. López, & F. Holik (Eds.), Quantum worlds: Perspectives on the ontology of quantum mechanics. Cambridge: Cambridge University Press. (Eds.), Quantum worlds: Perspectives on the ontology of quantum mechanics. Cambridge: Cambridge University Press. Eddington, A. (1928). The nature of the physical world. Cambridge: Cambridge University Press. Fine, A. (1984). The natural ontological attitude. In J. Leplin (Ed.), Scientific realism (pp. 83–107). Berkeley: University of California Press. Forbes, C. (2017). A pragmatic, existentialist approach to the scientific realism debate. Synthese, 194(9), Eddington, A. (1928). The nature of the physical world. Cambridge: Cambridge University Press. Fine A (1984) The natural ontological attitude In J Leplin (Ed ) Scientific realism (pp 83 107) B Eddington, A. (1928). The nature of the physical world. Cambridge: Cambridge University Press. Fine, A. (1984). The natural ontological attitude. In J. Leplin (Ed.), Scientific realism (pp. 83–107). Berkeley: University of California Press. University of California Press. Forbes, C. (2017). A pragmatic, existentialist approach to the scientific realism debate. Synthese, 194(9), 3327–3346. Forbes, C. (2017). A pragmatic, existentialist approach to the scientific realism debate. Synthese, 194(9), 3327–3346. French, S. (2002). A phenomenological solution to the measurement problem? Husserl and the foundations of quantum mechanics. Studies in the History and Philosophy of Modern Physics, 33, 467–491. References H i b W (1977) R k th i i f th l ti f t i t I W P i & S Chi i k (Ed ) 63(4), 729–771. Heisenberg, W. (1977). Remarks on the origin of the relations of uncertainty. In W. Price & S. Chissick (Eds.), Heisenberg, W. (1977). Remarks on the origin of the relations of uncertainty. In W. Price & S. Chissick (E The uncertainty principle and foundations of quantum mechanics (pp. 3–6). New York: Wiley. Heisenberg, W. (1983). The Physical Content of Quantum Kinematics and Mechanics. In J. A. Wheeler & W. H. Zurek (Eds.), Quantum theory and measurement, transl. by Wheeler and Zurek (pp. 62–86). Princeton: Princeton University Press. Hilgevoord, Jan & Uffink, Jos (2016). The uncertainty principle. In Edward N. Zalta (ed.), The Stanford encyclopedia of philosophy (winter 2016 edition), URL = <https://plato.stanford.edu/archives/win2016 /entries/qt-uncertainty/>. Huggett, N., & Wüthrich, C. (2013). Emergent spacetime and empirical (in)coherence. Studies in History and Philosophy of Modern Physics, 44, 276–285. p y f y , , Husserl, E. (1960). Cartesian meditations, transl. by Dorion Cairns. The Hague: Martinus Nijhoff. Husserl, E. (1969). Formal and transcendental logic, transl. by Dorion Cairns. The Hague: Martinus Nijhof ( ) g y g j usserl, E. (1970). The crisis of European sciences and transcendental phenomenology, transl. by David Carr. Evanston: Northwestern University Press. y Husserl, E. (1972). Erfahrung und Urteil. Hamburg: Meiner. Husserl, E. (1973). Zur Phänomenologie der Intersubjektivität. Texte aus dem Nachlass. Erster Teil. Husserliana XIII. Dordrecht: Kluwer. Husserl, E. (1977). Phenomenological psychology, transl. by John Scanlon. The Hague: Martinus Nijhoff. Husserl, E. (1982). Ideas pertaining to a pure phenomenology and to a phenomenological philosophy, first book, transl. by Fred Kersten. The Hague: Martinus Nijhoff. Husserl, E. (1982). Ideas pertaining to a pure phenomenology an book, transl. by Fred Kersten. The Hague: Martinus Nijhoff. Husserl, E. (1984). Einleitung in die Logik und Erkenntnistheorie, Vorlesungen 1906/1907. Husserliana XXIV. Dordrecht: Kluwer. Husserl, E. (2001). Analyses concerning passive and active synthesis, transl. by Anthony Steinbock. Dordrecht: Kluwer. Husserl, E. (2002). Logische Untersuchungen, Ergänzungsband, Erster Teil: Entwürfe zur Umarbeitung der VI. Untersuchung und zur Vorrede für die Neuauflage der Logischen Untersuchungen (Sommer 1913). Husserliana XX/1. Dordrecht: Kluwer. Husserliana XX/1. Dordrecht: Kluwer. Karaca, K. (2017). A case study in experimental exploration: Exploratory data selection at the large hadron collider. Synthese, 194, 333–354. Karaca, K. (2018). References 30 Page 26 of 27 European Journal for Philosophy of Science (2020) 10: 30 French, Steven (forthcoming). From a lost history to a new future: Is a phenomenological approach to quantum physics viable? forthcoming in P. Berghofer & H. Wiltsche, Phenomenological approaches to physics. Synthese Library, Springer. Frescura, F. A. M. & Hiley, Basil (1984). Algebras, quantum theory and pre-space. Revista Brasileira de Fisica, Volume Especial, Julho 1984, Os 70 anos de Mario Schonberg, 49–86. Frost-Arnold, G. (2010). The no-miracles argument for realism: Inference to an unacceptable explanatio Philosophy of Science, 77(1), 35–58. Fuchs, C. (2017). On participatory realism. In I. Durham & D. Rickles (Eds.), Information and interaction (p 113–134). Cham: Springer. Fuchs, C., Mermin, D., & Schack, R. (2014). An introduction to QBism with an application to the locality of quantum mechanics. American Journal of Physics, 82, 749–754. Fuchs, C., & Schack, R. (2015). QBism and the Greeks: Why a quantum state does not represent an element of physical reality. Physica Scripta, 90, 1–6. Fuchs, C., & Stacey, B. (2016). Some negative remarks on operational approaches to quantum theory. In G. Chiribella & R. Spekkens (Eds.), Quantum theory: Informational foundations and foils (pp. 283–305). Dordrecht: Springer. p g Fuchs, C., & Stacey, B. (2019). QBism: Quantum theory as a hero’s handbook. In E. Rasel, W. Schleich, & S. Wölk (Eds.), Foundations of quantum theory (pp. 133–202). Amsterdam: IOS Press. Giere, R. (2006). Scientific perspectivism. Chicago: The University of Chicago Press. Giere, R. (2010). An agent-based conception of models and scientific representation. Synthese, 172, 269–28 R. (2010). An agent-based conception of models and scientific representation. Synthese, 172, 269–281. R. (2013). Kuhn as perspectival realist. Topoi, 32, 53–57. Giere, R. (2013). Kuhn as perspectival realist. Topoi, 32, 53–57. Giere, R. (2013). Kuhn as perspectival realist. Topoi, 32, 53–57. Giere, R. (2016). Feyerabend’s perspectivism. Studies in History and Philosophy of Science, 57, 137–141. Goyal, P. (2012). Information physics – towards a new conception of physical reality. Information, 3, 567–594 Griffiths, D. (2018). Introduction to quantum mechanics (3rd ed.). Cambridge: Cambridge University Pres Hartimo, M. (2018). Husserl’s scientific context 1917–1938, a look into Husserl’s private library. The New Yearbook for Phenomenology and Phenomenological Philosophy, 16, 335–355. Healey, R. (2012). Quantum theory: A pragmatist approach. British Journal for the Philosophy of Scienc 63(4), 729–771. y, ( ) Q y p g pp f p y f , 63(4), 729–771. References (2014). Why gauge? Foundations of Physics, 44, 91–104. Ryckman, T. (2005). The reign of relativity: Philosophy of physics 1915–1925. Oxford: Oxford Universi Press. Ryckam, Thomas (forthcoming). The gauge principle, Hermann Weyl, and symbolic construction from the ‘Purely Infinitesimal’” forthcoming in P. Berghofer & H. Wiltsche. Phenomenological approaches to physics. Synthese Library, Springer. hlosshauer, M. (Ed.). (2011). Elegance and enigma: The quantum interviews. Dordrecht: Springer. llars W (1963) Science perception and reality Atascadero: Ridgeview Publishing Company Schlosshauer, M. (Ed.). (2011). Elegance and enigma: The quantum interviews. Dordrecht: Spring Sellars, W. (1963). Science, perception and reality. Atascadero: Ridgeview Publishing Company. hlosshauer, M. (Ed.). (2011). Elegance and enigma: The quantum interviews. Dordrecht: Spring llars, W. (1963). Science, perception and reality. Atascadero: Ridgeview Publishing Company. Smith, J. (2010). Seeing other people. Philosophy and Phenomenological Research, 81(3), 731–748. Solé, A., Oriols, X., Marian, D., & Zanghi, N. (2016). How does quantum uncertainty emerge fr deterministic Bohmian mechanics? Fluctuation and Noise Letters, 15(3), 1–25. g q deterministic Bohmian mechanics? Fluctuation and Noise Letters, 15(3), 1–25. Stein, E. (2004). Einführung in die Philosophie, Edith Stein Gesamtausgabe Band 8. Freiburg: Herder. Teller, P. (2001). Twilight of the perfect model model. Erkenntnis, 55, 393–415. 001). Twilight of the perfect model model. Erkenntnis, Teller, P. (2011). Two models of truth. Analysis, 71(3), 465–472. ( ) y ( ) Teller, P. (2020). What is perspectivism, and does it count as realism? In M. Massimi & C. McCoy (Eds.), Understanding perspectivism (pp. 49–64). New York: Routledge. Teller, P. (2020). What is perspectivism, and does it count as realism? In M. Massimi & C. McCoy (Eds Teller, P. (2020). What is perspectivism, and does it count as realism? Understanding perspectivism (pp. 49–64). New York: Routledge. Understanding perspectivism (pp. 49–64). New York: Routledge. Understanding perspectivism (pp. 49–64). New York: Routledge. Timpson, C. (2008). Quantum Bayesianism: A study. Studies in History and Philosophy of Modern Physics, 39, 579–609. de La Tremblaye, L. (forthcoming). QBism from a phenomenological point of view: Husserl and QBism, forthcoming in P. Berghofer & H. Wiltsche. Phenomenological approaches to physics. Synthese Library, Springer. Vaidman, L. (2014). Protective measurement of the wave function of a single system. In S. Gao (Ed Protective measurement and quantum reality (pp. 15–27). Cambridge: Cambridge University Press. Wallace, D. (2013). A prolegomenon to the ontology of the Everett interpretation. In A. Ney & D. Albe (Eds.), The wave function (pp. 203–222). New York: Oxford University Press. References Lessons from the large hadron collider for model-based experimentation: The concept of a model of data acquisition and the scope of the hierarchy of models. Synthese, 195, 5431–5452. European Journal for Philosophy of Science (2020) 10: 30 Page 27 of 27 30 30 Ladyman, James (2016). Structural realism. Edward N. Zalta (ed.), The Stanford encyclopedia of philosophy (winter 2016 edition). URL = <https://plato.stanford.edu/archives/win2016/entries/structural-realism/>. Landy, D., & Goldstone, R. (2007). How abstract is symbolic thought? Journal of Experimental Psychology, 33(4), 720–733. London, F., & Bauer, E. (1983). The theory of observation in quantum mechanics. In J. A. Wheeler & W. H. Z k (Ed ) Q t th d t ( 217 259) P i t P i t U i it P London, F., & Bauer, E. (1983). The theory of observation in quantum mechanics. In J. A. Wheeler & W. H. Zurek (Eds.), Quantum theory and measurement (pp. 217–259). Princeton: Princeton University Press. Madary, M. (2017). Visual phenomenology. Cambridge: The MIT Press. Mancosu, P., & Ryckman, T. (2005). Geometry, physics and phenomenology: Four letters of O. Becker to H. Weyl. In V. Peckhaus (Ed.), Oskar Becker und die Philosophie der Mathematik (pp. 153–227). München: Wilhelm Fink Verlag. g Massimi, M. (2012). Scientific perspectivism and its foes. Philosophia, 84, 25–52. Massimi, M. (2012). Scientific perspectivism and its foes. Philosophia, 84, 25–52. Massimi M (2018a) Four kinds of perspectival truth Philosophy and Phenomenological Research 9 Massimi, M. (2018a). Four kinds of perspectival truth. Philosophy and Phenomenological Research, 96(2), 342–359. Massimi, M. (2018b). Perspectivism. In J. Saatsi (Ed.), The Routledge handbook of scientific realism (p 164–175). New York: Routledge. Massimi, M., & McCoy, C. (Eds.). (2020). Understanding perspectivism. New York: Routledge. Merleau-Ponty, M. (1968). The visible and the invisible. Evanston: Northwestern University Pres Merleau-Ponty, M. (2003). Nature: Course notes from the Collège de France. Evanston: Northern Univ Press. Merleau-Ponty, M. (2004). The world of perception, transl. by Oliver Davis. London: Routledge. Merleau-Ponty, M. (2012). Phenomenology of perception, transl. by Donald Landes. New York: Routle Mermin, D. (2014). QBism puts the scientist back into science. Nature, 507, 421–423. Oriti, D. (2014). Disappearance and emergence of space and time in quantum gravity. Studies in History of Philosophy of Modern Physics, 46, 186–199. p y f y ovelli, C. (1996). Relational quantum mechanics. International Journal of Theoretical Physics, 35(8), 1637–1678. Rovelli, C. (1996). Relational quantum mechanics. International Journal of Theoretical Physics, 35(8), 1637–1678. Rovelli, C. References , ( ) g q y, ( ), Zahavi, D. (2004). Natural realism, anti-reductionism, and intentionality. The ‘phenomenology’ of Hilary Putnam. In D. Carr & C. Chan-Fai (Eds.), Space, time, and culture (pp. 235–251). Dordrecht: Springer. Zahavi, D. (2019). Phenomenology: The basics. London: Routledge. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
https://openalex.org/W1591973670	https://repositoriosaludmadrid.es//bitstream/20.500.12530/42818/1/PMC4486748.pdf	English	null	What is Needed to Improve Food Sales in Schools? Food Vendors’ Opinion from El Salvador	Frontiers in public health	2,015	cc-by	2,715	OPINION published: 01 July 2015 doi: 10.3389/fpubh.2015.00168 Keywords: school nutrition, sale regulations, child obesity, food vendors, save the children Save the Children International (SCI) implements a school health and nutrition program in 45 schools in El Salvador. The program’s overall goal is to increase the consumption of protein and micronutrients while reducing intake of sugar and fats among school children. g g g Save the Children International started this program because the Latin America and Caribbean region is at the forefront of the double burden of malnourishment, with rocketing rates of overweight, obesity, and high rates of stunting at the same time (1). Latin Americans tend to decrease healthier foods and consume more fat and sugar when emerging from poverty (2). In the 30-year span from 1990 to 2020, Central America is projected to double its prevalence of overweight and obesity among children under five (3). School children are at a crucial age for setting eating and physical activity habits, which in turn determine their nutritional status and long term health, education, and economic productivity. Overweight and obesity have become a public health problem in El Salvador. According to a recent study in 2014 (4), 30% of 389 young adolescents (7th–11th graders) were overweight and within that group, 8% were obese, comparing to only 2% of children being underweight or very thin. In a previous study in 2012, 23% of 1st and 2nd graders were reported to be overweight or obese (5). Edited by: Lesley Drake, Imperial College London, UK Edited by: Lesley Drake, Imperial College London, UK Reviewed by: Meghana M. Wadnerkar-Kamble, University of East Anglia, UK Daniel Rossignol, Rossignol Medical Center, USA Correspondence: Caroline Hilari chilari@savechildren.org El Salvador has a governmental recommendation regulating the sale of soft drink in schools (5, 6), which has been proposed to become a full law (7, 8). Schools stores also have a written contract, which regulates the vendor’s services, based on a model contract from the Ministry of Education. Reviewed by: Meghana M. Wadnerkar-Kamble, University of East Anglia, UK Daniel Rossignol, Rossignol Medical Center, USA Save the Children International’s school store intervention aims to improve the quality of food that children buy at school. SCI trains and supports student brigades to enforce the current food sale regulation. Thus, monitoring and control lies in the hands of students, which fosters child participation. SCI’s program is specifically known for active child participation in design, implementation, and monitoring (9). In addition, there is great emphasis on food hygiene and safe handling, through training of the food vendors (10). The indicator of success is the proportion of schools with stores selling fruit. Of the 45 schools, we found 38 stores selling fruit in 2012 and 41 in 2013. We were intrigued to learn more about the determinants for healthy food stores on the ground. Therefore, SCI conducted key informant interviews in 2014, in six SCI supported schools, three in the Izalco municipality (Sonsonate Department), and the other three in San Pedro (La Paz Department). In each municipality, we chose one urban and two rural school food vendors. We asked open-ended questions in three thematic areas of interest: Correspondence: Caroline Hilari chilari@savechildren.org Specialty section: This article was submitted to Child Health and Human Development, a section of the journal Frontiers in Public Health Received: 29 January 2015 Accepted: 16 June 2015 Published: 01 July 2015 Specialty section: This article was submitted to Child Health and Human Development, a section of the journal Frontiers in Public Health Received: 29 January 2015 Accepted: 16 June 2015 Published: 01 July 2015 (1) Products most sold and products with greatest profit. (2) Training content and usefulness. (3) Knowledge and opinion about sales regulation and enforcement. What is needed to improve food sales in schools? Food vendors’ opinion from El Salvador Caroline Hilari 1* and Margarita Franco2 1 School Health and Nutrition Unit, Department of Child Education and Protection, Save the Children Federation Inc., La Paz, Bolivia, 2 Save the Children International, El Salvador Country Office, San Salvador, El Salvador Keywords: school nutrition, sale regulations, child obesity, food vendors, save the children Citation: Hilari C and Franco M (2015) What is needed to improve food sales in schools? Food vendors’ opinion from El Salvador. Visits were preannounced to the school director to make sure that schools were open and key informants available, but not to the school vendor. We also observed the type of food on sale during these interviews. Vendors gave informed verbal consent to being interviewed. Confidentiality was ensured through conducting the interview in a private setting, where directors or parents could not overhear the conversation. Front. Public Health 3:168. doi: 10.3389/fpubh.2015.00168 Front. Public Health 3:168. doi: 10.3389/fpubh.2015.00168 Front. Public Health 3:168. doi: 10.3389/fpubh.2015.00168 July 2015 \| Volume 3 \| Article 168 Frontiers in Public Health \| www.frontiersin.org 1 Hilari and Franco El-Salvador’s healthy school food stores armed groups. This was mentioned by several principals and parents. armed groups. This was mentioned by several principals and parents. All six schools visited had fruit and self-made juice (fresh fruit, water, and sugar) on sale. However, two stores sold also soft drinks permanently and two others said they sold soft drinks once a week or occasionally. Those who were selling soft drinks permanently also had unhealthy or “junk” foods such as industrially produced chips (“churros” in Central American Spanish). On the other hand, the stores who were NOT selling soft drinks were selling homemade foods such as sandwiches, cooked or baked traditional foods. The following describes the key motivators and barriers for compliance with the national policy, according to food vendors. The other important feature is the contract with the parent association: The other important feature is the contract with the parent association: I like things to be correct, I like complying. If everything is the same as before (selling chips and soft drinks), then we are not complying, right? Legal is legal. They should give us the contract at the beginning of the year, so that we know the conditions. (Divina Providencia) No Aggressive Marketing Strategies by Soft Drink Producers In one school, Pepsi Cola was counteracting the national policy, by giving support with sport equipment, benches, tables, and other furniture. These supplies are highly valued and would be lost if a “no soft drink” policy were enforced. They were forbidding the soft drinks, but it is impos- sible. Some providers, like Pepsi Cola, support our school centers, they support our schools stores, they give footballs, furniture. You can’t forbid them to sell their product, right? (School director, Las Isletas) Children Staying Inside the School Premises Vendors expressed great concern about children being allowed to go out during school hours or the lack of a fence around school premises. They said that this usually leads to buying junk food in nearby stores on the street and then bring it into school premises. This frustrates the in-school vendors because they lose profit and it counteracts the regulation. Finding a Profitable Healthy Food In this context, homemade fruit juice and fresh fruit apparently give greatest profit margin, mentioned by four vendors. Especially if they take advantage of availability and low prices during harvest time, they can maximize their profit and it is higher than for chips and soft drinks. Continuous Control of the School Store I have a little field, my mom collects the mango, so I don’t spend basically anything (on fruit), it is all for the profit. (Divina Providencia) Fruit juice and fresh fruit gives most profit. Economi- cally, you don’t earn anything out of candy, not a lot of profit. (Las Isletas) In one very healthy food store, the school principal had a strong and clear authority, with a clear commitment to healthy food on school premises: The director here is quite strong, some parents have complained. We even cannot sell the same stuff every day; here we have to say what she tells us, she gives us the menu. The director is the one who reigns (here). (Sega) On the other side, fruit preparation is more work for the vendor, in some cases required hiring additional helpers. For the costs, it’s good for me. But fruit is more work. You have to get up really early to get (harvest or buy) the fruit. That said, you have to work for the fruit. (Divina Providencia) Parent and Children’s Attitude to Healthy Food Several vendors mentioned that children would be less likely to bring junk food to school if taught and motivated by an active student volunteer brigade. They also mentioned parent control in the home, so that buying food becomes more conscious about health impacts. In the future, we propose to expand our school store program intervention so that vendors have a facilitating environment to be able to comply with the national regulation. The student brigade comes, they check that all the food is covered. (San Marcelino) The student brigade comes, they check that all the food is covered. (San Marcelino) h l h d h Author Contributions The student brigade comes, they check that all the food CH provided the initial framework, conducted, translated, and analyzed the interviews. MF coordinated data collection in the field, monitoring, and program interventions. Both worked on the final version of this article. We wish to thank Mario Alvarado, Ma-luschka Colindres, Ludin de Chávez, Mara Ruiz, and several drivers for effective collaboration. There is a Health and Nutrition committee. The parent association should be motivating the children about what is harmful to them. (Las Isletas) Two years ago, we had a very active parent association and student brigade. The Save the Children technician had real leadership, he motivated the kids. Every week References 8. Programa Paquete Escolar [Internet]. Consulta sobre Ley de Alimentación Escola; 2014 [cited 2014 November 17]. Available from https://www.mined.gob.sv/index.php/ints/item/7260-presentan-consulta- ciudadana-de-ley-de-alimentaci%C3%B3n-escolar.html 1. James W. The challenge of childhood obesity. Int J Pediatr Obes (2006) 1:7–10. doi:10.1080/17477160600630404 1. James W. The challenge of childhood obesity. Int J Pediatr Obes (2006) 1:7–10. doi:10.1080/17477160600630404 ciudadana-de-ley-de-alimentaci%C3%B3n-escolar.html 9. Vanner C. We are the assets of the school: Children’s Participation Rights in a School Health and Nutrition Project in El Salvador. Int J Child Right (2014) 22:339–60. doi:10.1163/15718182-02202005 2. Vio F, Albala C. Nutrition policy in the Chilean transition. Public Health Nutr (2000) 3:49–55. doi:10.1017/S1368980000000070 3. De Onis M, Blössner M, Borghi E. Global prevalence and trends of overweight and obesity among preschool children. Am J Clin Nutr (2010) 92:1257–64. doi:10.3945/ajcn.2010.29786 10. Ministerio de Educación, Plan Nacional de Educación, Ministerio de Salud. Tiendas Escolares Saludables. San Salvador: United Nations Organization for Food and Agriculture (FAO) (2007). 4. Centro de Defensa del Consumidor (CDC), Red Salvadoreña por el Derecho a la Educación. Por un futuro sano: Análisis de la situación alimentaria y nutricional en centros educativos públicos. San Salvador: Centro de Defensa del Consumidor (CDC) and Red Salvadoreña por el Derecho a la Educación (2014). Conflict of Interest Statement: This research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. We did not at any time receive payment or services from a third party for any aspect of the submitted work. 5. Unidad de Nutrición. Estudio nacional de yoduria, evaluación del estado nutri- cional y de alimentos fortificados en escolares de primero y segundo grado. San Salvador: Ministerio de Salud (2014). 6. Ministerio de Educación (MINED). Lineamientos Básicos para el Fun- cionamiento de Tiendas Escolares Saludables. El Salvador: Centro de Defensa del Consumidor (CDC). Copyright © 2015 Hilari and Franco. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. 7. Consumers International [Internet]. El Salvador: Ley de Alimentación Salud- able Escolar; 2014 [cited 2014 December 10]. Available from: http://es. More Knowledge About Junk Food Knowledge, modified by personal experience, influences the ven- dors’ attitudes and motivation to limit the sale of junk food. All vendors knew that soft drinks are “forbidden” to sell. Above that, some vendors were conscious of the addictive properties of artificially flavored chips and soft drinks, especially Coca Cola. The main harmful content of junk food is considered to be high content in fat, salt, artificial flavors, and colors. Vendors cite hyperactivity as a main health problem due to junk food consumption: This school doesn’t have a wall, and outside there is everything (all type of food products). I am paying my rent in here, if I stop selling this (junk food), the vendor outside of the school earns more. They would have to put a wall around the school so that kids don’t go out. (San Marcelino) Chips makes something explode inside of the children, they become hyperactive. Kids want chips because it is addictive, my son was addicted to soft drinks, but I took it away because I saw a program on TV and then I took it away from him. I agree with not selling junk food, because it is really bad for everything, the soft drinks are harmful for the bones. (Izalco) The kids just go out and buy chips, the door is open, they buy it from the sale posts in the street, anything they want. But as it is not allowed, I don’t sell (chips). (Talcomunca) They buy the chips from outside (of the school), they buy it on their way. (Izalco) Children leaving school has other implications too in the Sal- vadorian context: there is a danger of road accidents and there is great danger of criminal violence, including recruitment into On the other hand, there is very little consciousness among vendors about the role of sugar in overweight, obesity, and July 2015 \| Volume 3 \| Article 168 Frontiers in Public Health \| www.frontiersin.org 2 El-Salvador’s healthy school food stores Hilari and Franco dental caries. Only one vendor mentioned this effect: dental caries. Only one vendor mentioned this effect: the brigade would control us. The kids couldn’t go out to buy in the street. “You have your store here, you don’t have any reason to go out to the street” they would tell them. And they also had festivals to motivate kids for good nutrition. We were famous for having a healthy school store; I could even sell homemade sweets. (Divina Providencia) The ingredients do harm, they make the sugar go up in the blood. At nine years of age they already have (high) blood sugar, hypertension. Super fat, but not at all well-nourished. (Divina Providencia) During observation, particularly in those stores which did not sell soft drinks and junk food, we saw more homemade foods on sale such as juices or sweets. The very high sugar content points to the fact that homemade products are not necessarily healthier than industrially produced foods. Additionally, food vendors need to be informed about the harmful consequences of high sugar content in children’s diet. We propose to classify the “ecologic” or enabling environmen- tal conditions for healthy food in schools into three levels of action: Level of action Enabling condition for healthy food Vendors Finding a profitable healthy food School administration Keeping children inside the school premises Controlling or enforcing the healthy food regulation Keeping soft drink marketing out of schools Civil society, including NGOs Vendors’ knowledge about healthy food Children’s and parents’ attitude toward healthy food Finally, most vendors had received some training on food hygiene by the public health service, supported by SCI. However, only one of the vendors remembered a message about food quality from the training course: They taught us the importance of selling natural foods. (Divina Providencia) They taught us the importance of selling natural foods. (Divina Providencia) We had initially used “selling fruit” per observation as the main indicator for clarifying schools stores as compliant with regula- tions. However, after analysis of the interviews and observations, we propose to use “not selling soft drinks” as indicator, because the absence of soft drinks was a fairly good differentiator variable for a healthy food store classification. References consumersinternational.org/news-and-media/news/2014/12/cdc_ley/ July 2015 \| Volume 3 \| Article 168 Frontiers in Public Health \| www.frontiersin.org 3
https://openalex.org/W1686789558	https://europepmc.org/articles/pmc4573996?pdf=render	English	null	Differential expression of Mediator complex subunit MED15 in testicular germ cell tumors	Diagnostic pathology	2,015	cc-by	3,848	© 2015 Klümper et al. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Differential expression of Mediator complex subunit MED15 in testicular germ cell tumors Niklas Klümper1†, Isabella Syring1,2†, Anne Offermann1, David Adler1, Wenzel Vogel1, Stefan C. Müller2, Ellinger2, Arne Strauß3, Heinz Joachim Radzun4, Philipp Ströbel4, Johannes Brägelmann1,5, Sven Perner1† and Felix Bremmer 4† Jörg Correspondence: felix.bremmer@med.uni-goettingen.de †Equal contributors 4Institute of Pathology, University Hospital of Göttingen, Robert Koch Str. 40, 37075 Göttingen, Germany Full list of author information is available at the end of the article Abstract Background: Testicular germ cell tumors (TGCT) are the most common cancer entities in young men with increasing incidence observed in the last decades. For therapeutic management it is important, that TGCT are divided into several histological subtypes. MED15 is part of the multiprotein Mediator complex which presents an integrative hub for transcriptional regulation and is known to be deregulated in several malignancies, such as prostate cancer and bladder cancer role, whereas the role of the Mediator complex in TGCT has not been investigated so far. Aim of the study was to investigate the implication of MED15 in TGCT development and its stratification into histological subtypes. Methods: Immunohistochemical staining (IHC) against Mediator complex subunit MED15 was conducted on a TGCT cohort containing tumor-free testis (n = 35), intratubular germ cell neoplasia unclassified (IGCNU, n = 14), seminomas (SEM, n = 107) and non-seminomatous germ cell tumors (NSGCT, n = 42), further subdivided into embryonic carcinomas (EC, n = 30), yolk sac tumors (YST, n = 5), chorionic carcinomas (CC, n = 5) and teratomas (TER, n = 2). Quantification of MED15 protein expression was performed through IHC followed by semi-quantitative image analysis using the Definiens software. Results: In tumor-free seminiferous tubules, MED15 protein expression was absent or only low expressed in spermatogonia. Interestingly, the precursor lesions IGCNU exhibited heterogeneous but partly very strong MED15 expression. SEM weakly express the Mediator complex subunit MED15, whereas NSGCT and especially EC show significantly enhanced expression compared to tumor-free testis. Conclusions: In conclusion, MED15 is differentially expressed in tumor-free testis and TGCT. While MED15 is absent or low in tumor-free testis and SEM, NSGCT highly express MED15, hinting at the diagnostic potential of this marker to distinguish between SEM and NSGCT. Further, the precursor lesion IGCNU showed increased nuclear MED15 expression in the preinvasive precursor cells, which may provide diagnostic value to distinguish between benign and pre-malignant testicular specimen, and may indicate a role for MED15 in carcinogenesis in TGCT. Klümper et al. Diagnostic Pathology (2015) 10:165 DOI 10.1186/s13000-015-0398-6 Klümper et al. Diagnostic Pathology (2015) 10:165 DOI 10.1186/s13000-015-0398-6 Background are further subdivided into embryonic carcinomas (EC), yolk sac tumors (YST), chorionic carcinomas (CC), and teratomas (TER) [3]. Generally, NSGCT have a more aggressive and undifferentiated phenotype than SEM and tend to be metastatic [4]. Therefore, the histological distinction between these tumor subentities plays an im- portant role for the therapeutic management and new bio- markers are needed for higher sensitivity in diagnostics. In young men at the age of 15 to 40 years, testicular germ cell tumors (TGCT) are the most frequent malignant tu- mors [1]. Interestingly, an increasing incidence in TGCT has been observed over the last 40 years [2]. TGCT are his- tologically and clinically grouped into seminomas (SEM) and non-seminomatous germ cell tumors (NSGCT), which MED15 is part of the multiprotein Mediator complex (MED) and serves as a hub for important signaling path- ways, transcriptional co-activators and co-repressors form- ing a bridge between the RNA polymerase II (Pol II) and * Correspondence: felix.bremmer@med.uni-goettingen.de †Equal contributors 4Institute of Pathology, University Hospital of Göttingen, Robert Koch Str. 40, 37075 Göttingen, Germany Full list of author information is available at the end of the article Klümper et al. Diagnostic Pathology (2015) 10:165 Page 2 of 6 manufacturer): anti-MED15 rabbit polyclonal (1:100, 11566-1-AP, Proteintech, Chicago, IL, USA). Antibody dilution was conducted using a Ventana diluent. Signal de- tection was done using the ultraView Universal DAB de- tection kit (Ventana Medical System, Tuscon, AZ, USA). Finally, slides were counterstained with haematoxylin and bluing reagent, dehydrated, and mounted. IHC stainings were validated independently by two pathologists (F.B., S.P). Only cases with at least one assessable core were in- cluded in this analysis. Tumor samples with a lack of tis- sue or absence of carcinoma were excluded. transcriptional factors [5, 6]. The Mediator complex has frequently been described to be differentially expressed or mutated in diverse tumor entities [7]. Interestingly, MED15 belongs to the tail module of the Mediator com- plex, which is known to receive and integrate information from diverse signaling pathways such as the transforming growth factor-β (TGF-β) and sterol regulatory element- binding protein (SREBP) pathways [8–10]. While some cancer entities were shown to be strongly associated with MED deregulation, knowledge about the Mediator com- plex expression profile in TGCT has been lacking so far. Results MED MED15 expression in tumor-free testes, precursor lesions, seminomas and non-seminomatous germ cell tumors MED15 expression in tumor-free testes, precursor lesions, seminomas and non-seminomatous germ cell tumors In tumor-free testes the MED15 protein expression was absent or low. Interestingly, pre-eminently the pluripotent spermatogonia exhibit moderate MED15 expression in the tumor-free testis (Fig. 1a). Intratubular germ cell neoplasia unclassified (IGCNU) showed increased nuclear MED15 expression in the preinvasive precursor cells and an in- creased positive index (Fig. 1b). The TMA was constructed as described previously [11]. Briefly, formalin-fixed paraffin-embedded tissues were cut into 4 μm thick sections and were mounted on slides. After staining with haematoxylin and eosin (H&E), areas of nor- mal tissue and primary tumor were determined and circled by a pathologist. Representative cores of the circled re- gions, measuring 0.6 mm in diameter from each formalin- fixed paraffin-embedded (FFPE) benign tissue and primary tumor, were assembled into tissue microarray blocks (recipient blocks) using a semiautomatic tissue arrayer (Beecher Instruments, Sun Prairie, WI, USA). H&E TMA sections were assessed again to confirm the histology of the selected regions. In SEM, MED15 expression was predominantly low or completely absent (Fig. 1c). In contrast, non-seminomatous germ cell tumors (NSGCT) exhibited a significantly higher MED15 expression and positivity index as compared to tumor-free testes and SEM. Especially, the staining pattern in EC showed strong homogeneous expression of MED15 in the nuclei and cytoplasm (Fig. 2a). Likewise, samples of NSGCT, YST (Fig. 2b) and CC (Fig. 2c) exhibited elevated nuclear and cytoplasmic MED15 expression as compared to tumor-free testes and SEM. TER were found with high diversity in MED15 expression in line with their histologic heterogeneity (data not shown). In conclusion, the NSGCT significantly overexpress MED15 as compared to tumor- free testes and SEM (p < 0.001). Further, IGCNU as a pre- malignant precursor lesion of the testes showed enhanced Quantification of protein expression Quantification of MED15 protein expression was per- formed using the semi-quantitative image analysis pro- gram Definiens (Definiens Inc., Munich, Germany). Hereby, the pathologist chose manually the tumor area within the testicular specimens. Afterwards, the program analyzed the selected regions of interest with respect to overall protein expression using the average staining intensity (mean brown chromogen intensity) and the number of posi- tively stained cells determined through an intensity thresh- old in relation to all analyzed cells in a sample (positive index). The statistical evaluation of the expression intensity was performed using the two-sided Student’s t-test in SPSS Statistics 22 (SPSS Inc., Chicago, IL, USA). Tissue samples of primary TGCT In this study tissue microarrays (TMA) containing speci- mens of testes were used to examine the MED15 protein expression by immunohistochemical analysis. The TGCT cohorts, kindly provided by the Institute for Pathology Göttingen and the Urology Department of the University Hospital Bonn, include the following tissue samples: 35 tumor-free testes, 14 intratubular germ cell neoplasia unclassified (IGCNU), 107 seminomas (SEM) and 42 non-seminomatous germ cell tumors (NSGCT), further subdivided into embryonic carcinomas (EC, n = 30), yolk sac tumors (YST, n = 5), chorionic carcinomas (CC, n = 5) and teratomas (TER, n = 2). Ethical approval for using hu- man material in the present study was obtained from the ethics committee of the University Medical Centre Göttingen and Bonn. Background We therefore evaluated the MED15 expression in tumor- free testis and TGCT subentities by immunohistochemical staining (IHC) on a large tissue microarray (TMA) cohort in order to evaluate a possible diagnostic and therapeutic value for MED15 in TGCT. Immunohistochemical analysis (IHC) h h l Immunohistochemical staining was conducted using the Ventana Benchmark automated staining system (Ventana Medical System, Tuscon, AZ, USA). In brief, slides were in- cubated at room temperature with the primary anti- body according to the manufacturer (dilutions, clones, Page 3 of 6 Klümper et al. Diagnostic Pathology (2015) 10:165 Fig. 1 Representative IHC images for MED15 expression from tissue of tumor-free testis (a), intratubular germ cell neoplasia unclassified (IGCNU) (b) and seminoma (SEM) (c). 10× (upper panel) and 40× (lower panel) objective magnification Fig. 1 Representative IHC images for MED15 expression from tissue of tumor-free testis (a), intratubular germ cell neoplasia unclassified (IGCNU) (b) and seminoma (SEM) (c). 10× (upper panel) and 40× (lower panel) objective magnification MED15 protein expression as compared to benign tis- sue (p < 0.01) (Fig. 3a). Interestingly, NSGCT showed an elevated MED15 positive index indicating a homoge- neous expression in the tumor, whereas tumor-free testes and SEM showed significantly lower positive indexes (p < 0.001) (Fig. 3b). MED15 protein expression as compared to benign tis- sue (p < 0.01) (Fig. 3a). Interestingly, NSGCT showed an elevated MED15 positive index indicating a homoge- neous expression in the tumor, whereas tumor-free testes and SEM showed significantly lower positive indexes (p < 0.001) (Fig. 3b). Discussion In this study, we found MED15 nuclear and cytoplasmic expression to be significantly higher in NSGCT as com- pared to tumor-free testes and SEM. Further, the precur- sor lesion intratubular germ cell neoplasia unclassified (IGCNU) exhibited increased nuclear MED15 expression Fig. 2 Representative IHC images for MED15 expression from tissue of the different non-seminomatous germ cell tumors (NSGCT) embryonic carcinoma (EC) (a), yolk sac tumor (YST) (b) and chorionic carcinoma (CC) (c). 10× (upper panel) and 40× (lower panel) objective magnification Fig. 2 Representative IHC images for MED15 expression from tissue of the different non-seminomatous germ cell tumors (NSGCT) embryonic carcinoma (EC) (a), yolk sac tumor (YST) (b) and chorionic carcinoma (CC) (c). 10× (upper panel) and 40× (lower panel) objective magnification Page 4 of 6 Klümper et al. Diagnostic Pathology (2015) 10:165 Fig. 3 a Nuclear MED15 protein expression profile of the total testicular germ cell tumor (TCGT) cohort. b Positive index of the MED15 immunhistochemical staining on the TGCT cohort. (n.s. = not significant, * = p < 0.05, = p < 0.01, * = p < 0.001) Fig. 3 a Nuclear MED15 protein expression profile of the total testicular germ cell tumor (TCGT) cohort. b Positive index of the MED15 immunhistochemical staining on the TGCT cohort. (n.s. = not significant, * = p < 0.05, = p < 0.01, * = p < 0.001) in the preinvasive precursor cells as compared to benign tissue which may be a hint for MED15 overexpression as a clonal event in TGCT development. Interestingly, MED15 was found to be homogeneously expressed in NSGCT (positive index >0.8) pointing at a possible role for MED15 in selection of precursor tumor cells during carcinogenesis. MED15 is part of the Mediator com- plex, which is an evolutionarily conserved multi-protein as- sembly. It plays a pivotal role in transcriptional regulation forming a bridge between transcriptional factors and the RNA polymerase II (Pol II) [5]. The complex can be subdi- vided into four distinct submodules termed the head, mid- dle, tail and kinase. While the head, middle and tail module form a stable core complex, the kinase module associates reversibly with the core complex thereby influencing the activity of the whole complex [6]. Received: 8 July 2015 Accepted: 28 August 2015 Received: 8 July 2015 Accepted: 28 August 2015 Received: 8 July 2015 Accepted: 28 August 2015 Discussion Generally, the head and middle modules interact directly with Pol II, the main part of the core associated basal transcription machinery, whereas the tail and kinase modules serve as major loci for signal transduction and diverse signaling pathways [5]. Genomic alteration as well as altered expression of dis- tinct Mediator complex subunits (MED) can cause dys- functions and dysregulations of cell signaling and may contribute to malignant properties through directly medi- ating target gene transcription [7]. Diverse MED subunits have been associated with tumor development, progres- sion and the emergence of chemoresistance, which re- mains one of the key problems in oncologic therapy. Page 5 of 6 Klümper et al. Diagnostic Pathology (2015) 10:165 Klümper et al. Diagnostic Pathology (2015) 10:165 Even though modern therapeutical options for TGCT such as radical orchiectomy and chemotherapy achieve high cure rates, a small group of patients nevertheless experiences late relapse, metastatic spread or chemore- sistance which leads to poor prognosis and underlines the importance for unraveling the responsible mecha- nisms [12]. Authors’ contributions SP, FB and DA designed the study. SP, FB, NK, and IS designed experiments. NK and FB drafted the manuscript and performed microscopic and histopathologic investigations. WV, JE, SCM, AS, HJR and PS were responsible for clinical, microscopic, histopathologic elements and participated in pathological investigations. NK, IS, FB, AO, JB and DA were responsible for analysis and interpretation of the data. All authors read and approved the final manuscript. Acknowledgement The study was supported by a grant of the Rudolf Becker-Foundation to SP, the Ferdinand Eisenberger-Fellowship of the German Society of Urology (DGU) to IS (SYI1/FE-13), the Gerok-Fellowship grant to JB and a medical doctoral fellowship grant (BONFOR) of the Medical Faculty of the University of Bonn to AO. FB is supported by the research program, faculty of medicine, Georg August- University Göttingen. Abbreviations TGCT: Testicular germ cell tumors; SEM: Testicular seminomas; NSGCT: Non- seminomatous germ cell tumors; EC: Embryonic carcinomas; YST: Yolk sac tumors; CC: Chorionic carcinomas; TER: Teratomas; MED15: Part of the multiprotein Mediator complex; TGF-β: transforming growth factor-β; SREBP: Sterol regulatory element-binding protein; TAZ: Transcriptional co-activator with PDZ-binding motif; FASN: Fatty acid synthase. Conclusions In conclusion, the differential protein expression of Medi- ator complex subunit MED15 in TGCT may provide valu- able as a diagnostic marker and may assist the selection of therapeutic intervention. MED15 is expressed significantly higher in precursor lesion IGCNU as compared to tumor- free testes and may thus prove useful as a diagnostic feature to distinguish between benign and pre-malignant tissue. Furthermore, MED15 may play a role in the differentiation process between seminomas and non-seminomas. MED15 is part of the tail module and is known to serve as a hub for signaling pathways including transforming growth factor-β (TGF-β) signaling as well as the signaling of sterol regulatory element-binding proteins (SREBP) [8, 9]. The TGF-β signaling pathway is a double-edged sword, with either tumor suppressive or oncogenic fea- tures depending on the stage of disease [10]. For example, in castration-resistant prostate cancer, MED15 is frequently overexpressed and correlates with pSmad3 expression, a downstream target indicating activation of the TGF-β sig- naling cascade [13]. MED15 was further described to retain TAZ (transcriptional co-activator with PDZ-binding motif) in the nucleus, which engages in shuttling Smad complexes and dominantly regulates Smad nuclear accumulation, thereby positively regulating TGF-β target gene expres- sion. In human embryonic stem cells MED15 is required for stem cell maintenance and pluripotency through direct interaction with TAZ leading to Smad nuclear accumula- tion and target gene transcription, whereas loss of TAZ leads to inhibition of TGF-β signaling and differentiation. This is in line with the results presented in the current study, which shows a significantly enhanced MED15 expression in pluripotent EC [14]. Interestingly, treat- ment of the seminoma cell line TCam-2 with recom- binant TGF-β1 induces differentiation into a cell type resembling mixed non-seminoma [15] suggesting MED15 as an integrative hub for this differentiation process. 1. Beyer J, Albers P, Altena R, Aparicio J, Bokemeyer C, Busch J, et al. Maintaining success, reducing treatment burden, focusing on survivorship: highlights from the third European consensus conference on diagnosis and treatment of germ-cell cancer. Ann Oncol. 2013;24(4):878–88. doi:10.1093/annonc/mds579. 2. Chia VM, Quraishi SM, Devesa SS, Purdue MP, Cook MB, McGlynn KA. International trends in the incidence of testicular cancer, 1973–2002. Cancer Epidemiol Biomarkers Prev. 2010;19(5):1151–9. doi:10.1158/1055-9965.EPI-10-0031. 3. Horwich A, Shipley J, Huddart R. Testicular germ-cell cancer. Lancet. 2006;367(9512):754–65. doi:10.1016/S0140-6736(06)68305-0. 4. Hermans BP, Sweeney CJ, Foster RS, Einhorn LE, Donohue JP. Risk of systemic metastases in clinical stage I nonseminoma germ cell testis tumor managed by retroperitoneal lymph node dissection. J Urol. 2000;163(6):1721–4. 5. Malik S, Roeder RG. The metazoan Mediator co-activator complex as Author details 1 1Section for Prostate Cancer Research, Institute of Pathology, Center for Integrated Oncology Cologne/Bonn, University Hospital of Bonn, Bonn, Germany. 2Department of Urology and Pediatric Urology, University Hospital of Bonn, Bonn, Germany. 3Department of Urology, University Medical Center, University of Göttingen, Göttingen, Germany. 4Institute of Pathology, University Hospital of Göttingen, Robert Koch Str. 40, 37075 Göttingen, Germany. 5Department of Hematology/Oncology, University Hospital of Bonn, Bonn, Germany. g p In addition, activated TGF-β signaling leads to induction of epithelial-to-mesenchymal transition, which is known to be implicated in tumor progression, metastatic spread and increased chemoresistance [16, 17, 10]. Thus, positive regu- lation of TGF-β signaling might indicate a role of MED15 in the emerging of an aggressive phenotype. Therefore, en- hanced MED15 protein expression found in TCGT could be a hint for non-responsive and invasive tumors and re- quires further investigation. Previously, the SREBP target gene fatty acid synthase (FASN), which is a key enzyme re- sponsible for the endogenous synthesis of fatty acids and described to be upregulated in different tumors, was de- tected to be differentially expressed in TCGT. Especially, EC frequently overexpress FASN, whereas SEM and tumor-free testes express FASN rarely [18]. Considering the strong overexpression of MED15 in TGCT and EC as pre- sented here, the involvement of the MED15-SREBP-FASN axis in tumor formation and differentiation should be in- vestigated in detail with the aim to potentially develop diag- nostic biomarker and identify novel therapeutic targets. Klümper et al. Diagnostic Pathology (2015) 10:165 Klümper et al. Diagnostic Pathology (2015) 10:165 6. Cai G, Imasaki T, Takagi Y, Asturias FJ. Mediator structural conservation and implications for the regulation mechanism. Structure. 2009;17(4):559–67. doi:10.1016/j.str.2009.01.016. 7. Schiano C, Casamassimi A, Rienzo M, de Nigris F, Sommese L, Napoli C. Involvement of Mediator complex in malignancy. Biochim Biophys Acta. 2014;1845(1):66–83. doi:10.1016/j.bbcan.2013.12.001. j 8. Kato Y, Habas R, Katsuyama Y, Naar AM, He X. A component of the ARC/Mediator complex required for TGF beta/Nodal signalling. Nature. 2002;418(6898):641–6. doi:10.1038/nature00969. 9. Yang F, Vought BW, Satterlee JS, Walker AK, Jim Sun ZY, Watts JL, et al. An ARC/Mediator subunit required for SREBP control of cholesterol and lipid homeostasis. Nature. 2006;442(7103):700–4. doi:10.1038/nature04942. 10. Akhurst RJ, Derynck R. TGF-beta signaling in cancer–a double-edged sword. Trends Cell Biol. 2001;11(11):S44–51. 11. Braun M, Scheble VJ, Menon R, Scharf G, Wilbertz T, Petersen K, et al. Relevance of cohort design for studying the frequency of the ERG rearrangement in prostate cancer. Histopathology. 2011;58(7):1028–36. doi:10.1111/j.1365-2559.2011.03862.x. 12. Bosl GJ, Motzer RJ. Medical progress: testicular germ-cell cancer. N Engl J Med. 1997;337:242–53. 12. Bosl GJ, Motzer RJ. Medical progress: testicular germ-cell cancer. N Engl J Med. 1997;337:242–53. 13. Adler D, Menon R, Braun M, Offermann A, Queisser A, Boehm D, et al. MED15, encoding a subunit of the mediator complex, is overexpressed at high frequency in castration-resistant prostate cancer. Int J Cancer J International du Cancer. 2014;135(1):19–26. doi:10.1002/ijc.28647. 14. Varelas X, Sakuma R, Samavarchi-Tehrani P, Peerani R, Rao BM, Dembowy J, et al. TAZ controls Smad nucleocytoplasmic shuttling and regulates human embryonic stem-cell self-renewal. Nat Cell Biol. 2008;10(7):837–48. doi:10.1038/ncb1748. 15. Nettersheim D, Gillis AJ, Looijenga LH, Schorle H. TGF-beta1, EGF and FGF4 synergistically induce differentiation of the seminoma cell line TCam-2 into a cell type resembling mixed non-seminoma. Int J Androl. 2011;34(4 Pt 2):e189–203. doi:10.1111/j.1365-2605.2011.01172.x. 16. Miyai K, Iwaya K, Asano T, Tamai S, Matsubara O, Tsuda H. Fatty acid synthase overexpression in adult testicular germ cell tumors: potential role in the progression of non-seminomatous germ cell tumors. Virchows Arch. 2014;464(2):221–8. doi:10.1007/s00428-013-1525-y. 17. Derynck R, Akhurst RJ, Balmain A. TGF-beta signaling in tumor suppression and cancer progression. Nat Genet. 2001;29(2):117–29. doi:10.1038/ng1001-117. 18. Singh A, Settleman J. EMT, cancer stem cells and drug resistance: an emerging axis of evil in the war on cancer. Oncogene. 2010;29(34):4741–51. doi:10.1038/onc.2010.215. References 1. Beyer J, Albers P, Altena R, Aparicio J, Bokemeyer C, Busch J, et al. Maintaining success, reducing treatment burden, focusing on survivorship: highlights from the third European consensus conference on diagnosis and treatment of germ-cell cancer. Ann Oncol. 2013;24(4):878–88. doi:10.1093/annonc/mds579. 2. Chia VM, Quraishi SM, Devesa SS, Purdue MP, Cook MB, McGlynn KA. International trends in the incidence of testicular cancer, 1973–2002. Cancer Epidemiol Biomarkers Prev. 2010;19(5):1151–9. doi:10.1158/1055-9965.EPI-10-0031. 3. Horwich A, Shipley J, Huddart R. Testicular germ-cell cancer. Lancet. 2006;367(9512):754–65. doi:10.1016/S0140-6736(06)68305-0. 4. Hermans BP, Sweeney CJ, Foster RS, Einhorn LE, Donohue JP. Risk of systemic metastases in clinical stage I nonseminoma germ cell testis tumor managed by retroperitoneal lymph node dissection. J Urol. 2000;163(6):1721–4. 5 M lik S R d RG Th t M di t ti t l 1. Beyer J, Albers P, Altena R, Aparicio J, Bokemeyer C, Busch J, et al. Maintaining success, reducing treatment burden, focusing on survivorship: highlights from the third European consensus conference on diagnosis and treatment of germ-cell cancer. Ann Oncol. 2013;24(4):878–88. doi:10.1093/annonc/mds579. 1. Beyer J, Albers P, Altena R, Aparicio J, Bokemeyer C, Busch J, et al. Maintaining success, reducing treatment burden, focusing on survivorship: highlights from the third European consensus conference on diagnosis and treatment of germ-cell cancer. Ann Oncol. 2013;24(4):878–88. doi:10.1093/annonc/mds579. 5. Malik S, Roeder RG. The metazoan Mediator co-activator complex as an integrative hub for transcriptional regulation. Nat Rev Genet. 2010;11(11):761–72 doi:10 1038/nrg2901 Page 6 of 6 Page 6 of 6 Klümper et al. Diagnostic Pathology (2015) 10:165 Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission
https://openalex.org/W2104710588	https://journals.plos.org/ploscompbiol/article/file?id=10.1371/journal.pcbi.1003363&type=printable	English	null	Electrostatically Accelerated Encounter and Folding for Facile Recognition of Intrinsically Disordered Proteins	PLOS computational biology/PLoS computational biology	2,013	cc-by	10,901	Abstract This is an open-access article distributed under the terms of the Creative Commons unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This work was supported by the National Science Foundation (MCB 0952514) and the Johnson Center for Basic Cancer Research. Part of the computing for this project was performed on the Beocat Research Cluster at Kansas State University, which is funded in part by NSF grants CNS-1006860, EPS-1006860, and EPS-0919443. This work is contribution number 13-392-J from the Kansas Agricultural Experiment Station. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. Competing Interests: The authors have declared that no competing interests exist. * E-mail: jianhanc@ksu.edu A key recent recognition is that frequent requirement of protein folding for specific recognition of IDPs could lead to kinetic bottlenecks [23–25]. As predicted by the dual-transition-state theory [23], the diffusion-limited encounter rate constant repre- sents the upper bound for that of a coupled binding and folding interaction. Importantly, the upper bound can be achieved only if the IDP readily folds upon encounter, which requires folding rates on the order of 10 ms21 or greater [23]. That is, IDPs need to achieve folding rates beyond the typical ms21 ‘‘speed limit’’ estimated for folding of isolated proteins [26] to maximize association kinetics. Therefore, the putative functional advantages of intrinsic disorder, especially structural plasticity for specific interactions with numerous partners [27], come with a potential cost of slow binding kinetics. Such kinetic bottleneck must be resolved for IDPs to be viable in cellular signaling and regulation. Interestingly, a recent survey of binding kinetic data revealed that IDP binding was not systematically slower than that of globular proteins [28]. The implication is that most IDPs do manage to fold rapidly upon nonspecific binding, and this is apparently consistent with the accumulating observations that IDP coupled binding and folding tends to follow induced folding-like baseline mechanisms (i.e., bind then fold) [16,19]. Several factors could contribute to Debabani Ganguly, Weihong Zhang, Jianhan Chen* Department of Biochemistry and Molecular Biophysics, Kansas State University, Manhattan, Kansas, United States of America Electrostatically Accelerated Encounter and Folding for Facile Recognition of Intrinsically Disordered Proteins Debabani Ganguly, Weihong Zhang, Jianhan Chen* Abstract Achieving facile specific recognition is essential for intrinsically disordered proteins (IDPs) that are involved in cellular signaling and regulation. Consideration of the physical time scales of protein folding and diffusion-limited protein-protein encounter has suggested that the frequent requirement of protein folding for specific IDP recognition could lead to kinetic bottlenecks. How IDPs overcome such potential kinetic bottlenecks to viably function in signaling and regulation in general is poorly understood. Our recent computational and experimental study of cell-cycle regulator p27 (Ganguly et al., J. Mol. Biol. (2012)) demonstrated that long-range electrostatic forces exerted on enriched charges of IDPs could accelerate protein-protein encounter via ‘‘electrostatic steering’’ and at the same time promote ‘‘folding-competent’’ encounter topologies to enhance the efficiency of IDP folding upon encounter. Here, we further investigated the coupled binding and folding mechanisms and the roles of electrostatic forces in the formation of three IDP complexes with more complex folded topologies. The surface electrostatic potentials of these complexes lack prominent features like those observed for the p27/ Cdk2/cyclin A complex to directly suggest the ability of electrostatic forces to facilitate folding upon encounter. Nonetheless, similar electrostatically accelerated encounter and folding mechanisms were consistently predicted for all three complexes using topology-based coarse-grained simulations. Together with our previous analysis of charge distributions in known IDP complexes, our results support a prevalent role of electrostatic interactions in promoting efficient coupled binding and folding for facile specific recognition. These results also suggest that there is likely a co-evolution of IDP folded topology, charge characteristics, and coupled binding and folding mechanisms, driven at least partially by the need to achieve fast association kinetics for cellular signaling and regulation. Citation: Ganguly D, Zhang W, Chen J (2013) Electrostatically Accelerated Encounter and Folding for Facile Recognition of Intrinsically Disordered Proteins. PLoS Comput Biol 9(11): e1003363. doi:10.1371/journal.pcbi.1003363 Editor: Bert L. de Groot, Max Planck Institute for Biophysical Chemistry, Germany Editor: Bert L. de Groot, Max Planck Institute for Biophysical Chemistry, Germany Received July 12, 2013; Accepted October 10, 2013; Published November 21, 2013 Received July 12, 2013; Accepted October 10, 2013; Published November 21, 2013 Copyright: 2013 Ganguly et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Copyright: 2013 Ganguly et al. PLOS Computational Biology \| www.ploscompbiol.org Author Summary Intrinsically disordered proteins (IDPs) are key components of regulatory networks that dictate various aspects of cellular decision-making. They are over-represented in major disease pathways, and are considered novel albeit currently difficult drug targets. Recognition of IDPs has extended the traditional protein structure-function para- digm, and various concepts have been proposed on how intrinsic disorder may confer crucial functional advantages. However, the physical basis of these concepts remains poorly established. In particular, while IDPs alone exist as ensembles of fluctuating structures, they frequently fold upon specific binding. Analysis of the physical timescales of protein folding and protein-protein encounter predicts that the requirement of peptide folding for specific binding could lead to a major kinetic bottleneck. In this work, carefully calibrated topology-based coarse-grained models were applied to directly simulate reversible folding and binding and investigate the recognition mechanisms of three IDP complexes. The results strongly support an electrostatically accelerated encounter and folding mech- anism, where long-range electrostatic forces not only accelerate protein-protein encounter via ‘‘electrostatic steering’’ but also promote ‘‘folding-competent’’ encoun- ter topologies to enhance the efficiency of IDP folding upon encounter. In the present work, we investigated the recognition mecha- nisms and the roles of long-range electrostatic interactions in forming of three IDP complexes, namely, p53-TAD1/TAZ2, HIF-1a/TAZ1, and NCBD/ACTR (Table 1). All these complexes have important biological functions. For example, tumor suppres- sor p53 is considered one of the most important proteins in cancer [51]; NCBD and TAZ1/2 are key regulatory domains of CBP, a key component of the general transcriptional machinery that plays critical roles in cell fate regulation [52]. For understanding IDP recognition, these systems involve more complex folded topologies than that of p27 in the p27/Cdk2/cyclin A complex. As shown in Fig. 1, both HIF-1a/TAZ1 and NCBD/ACTR possess extensive binding interfaces, whereas the binding interface in p53-TAD1/ TAZ2 is more localized. Importantly, while strong charge complementary exists near the binding interface (as expected), the surface electrostatic potentials of the folded substrates do not show prominent features like those observed on Cdk2/cyclin A (e.g., see Fig. 1 of reference [24]) to directly suggest that long-range electrostatic forces could promote native-like (and thus more folding-competent) encounter complexes. The NCBD/ACTR complex involves synergistic folding of two IDPs and thus offers a particularly interesting opportunity to understand whether and how electrostatic interactions may modulate the formation of nontrivial folded topologies. Author Summary Amazingly, all three complexes associate with on-rates in excess of 107 M21s21 (see Table 1), a regime that is typically considered ‘‘diffusion-limited’’ and can only be accessed in the limit of ultrafast conformational transitions [40]. efficient folding of IDPs upon binding, in particular small interacting (and folding) domains and simple folded topologies with low contact orders. There also appears to be a delicate balance between pre-folding and conformational flexibility that allows an IDP to quickly fluctuate among accessible conforma- tional states, especially upon encounter [16,29,30]. Nonetheless, it is not yet clear how in general IDPs may achieve fast folding at rates beyond the traditional ms21 folding ‘‘speed limit’’ upon encountering their specific targets. An important characteristics of IDPs is that they are enriched with charged and polar residues [31]. Electrostatics can thus be expected to play key roles in IDP structure and function. For example, the charge content can modulate compaction and other conformational properties of free IDPs [32,33]; DNA search efficiency is controlled by charge composition and distribution in disordered tails of DNA-binding proteins [34,35]. It has been also observed or speculated in a few cases that electrostatics might be important for fast IDP recognition [36–39]. However, these discussions have been often based on the classic electrostatic steering effects [40], and the actual underlying mechanisms of putative electrostatic acceleration were not known. Our recent computational and experimental study of the p27-Cdk2/cyclin A interaction revealed that long-range electrostatic forces could promote facile IDP recognition via an ‘‘electrostatically accelerated encounter and folding mecha- nism’’ [24]. Specifically, the measured p27/Cdk2/cyclin A association rate constants showed a strong salt-dependence, increased ,12 fold when the ionic strength was reduced from 0.6 to 0.075 M. However, the salt-dependence is poorly described by an approximate Debye-Hu¨ckel relation [41] that mainly captures the electrostatic steering effects. Instead, simulations using a series of topology-based coarse-grained models suggested that long-range electrostatic forces exerted on a large number of charges on p27 did not only accelerate the encounter rate (via the classical electrostatic steering effect [40]), but enhance the efficiency of p27 folding upon encounter by promoting native-like encounter topologies. Electrostatically Accelerated Recognition of IDPs Analysis of surface charges in a set of existing IDP complexes further revealed that the vicinity of IDP binding sites tended to be enriched with charges to complement those on IDPs [24] (even though the IDP binding interface itself is more hydrophobic than the rest of the protein surface as previously observed [42]). Electrostatic forces are known to be a dominant long-range force that can guide protein orientation in protein-DNA interactions [43,44] and/or modulate early stages of protein folding [45–47]. One implication of enriched charges near IDP binding sites is thus that the electrostatically accelerated encounter and folding mechanism observed for p27 may be prevalent in signaling and regulatory IDPs. Nonetheless, the ability for long-range electro- static forces to enhance folding upon binding can be surprising, as nonspecific interactions (electrostatic or van der Waals) have been generally expected to accelerate binding but slow down folding [48,49]. It has also been predicted that, while inter-chain electrostatic interactions facilitate binding of disordered chaperone Chz1 to histone variant H2A.Z-H2B, intra-chain electrostatic interactions could lead to premature collapse of Chz1 under low salt conditions and hinder the overall rate of forming the specific complex [50]. Author Summary Introduction Cellular signaling and regulation are frequently mediated by proteins that, in part or as a whole, lack stable structures under physiological conditions [1–3]. Such intrinsically disordered proteins (IDPs) are highly prevalent in proteomes [4] and over- represented in diseases pathways [5,6]. For example, nearly one- third of eukaryotic proteins have been predicted to contain extended disordered regions [7], and about 25% of disease- associated missense mutations can be mapped into predicted disordered regions [8] (although cancer mutations appear to prefer ordered regions [9]). The prevalence of intrinsic disorder suggests that protein conformational heterogeneity could provide crucial functional advantages, for which many concepts have been proposed [10–14]. Understanding the physical basis of how intrinsic disorder mediates protein function (and how such functional mechanism may fail in human diseases [15]) is of fundamental significance and has attracted intense interests in recent years [16]. Important progresses have been made on characterizing the conformational properties of unbound IDPs and determining how these conformational properties contribute to efficient and reliable interactions [16–22]. November 2013 \| Volume 9 \| Issue 11 \| e1003363 PLOS Computational Biology \| www.ploscompbiol.org 1 Topology-based modeling of IDP coupled binding and folding Series of topology-based coarse-grained models were first derived based on the complex structures to allow direct simulation of reversible binding and folding with tractable computational cost. Topology-based modeling is based on the theoretical framework of minimally frustrated energy landscapes for natural proteins [53], and has been highly successful in predicting essential features of protein folding mechanisms [53–55]. Formation of stable IDP complexes such as those studied in this work should also satisfy minimal frustration, and thus topology-based modeling is applicable. Indeed, it has been successfully applied to several IDP November 2013 \| Volume 9 \| Issue 11 \| e1003363 PLOS Computational Biology \| www.ploscompbiol.org 2 Electrostatically Accelerated Recognition of IDPs Table 1. Key properties of three IDP complexes. Namea Length KD b kon (M21s21) PDB IDP Fold Chargesc p53-TAD1/TAZ2 39/90 2.7 mM [77] ,108d 2k8f helix/loops 8(26), 9(+9), 4(+2) HIF-1a/TAZ1 51/99 7 nM [70] 1.36109 [78] 1l8c helices/loops 11(25), 11(+7), 10 (+5) NCBD/ACTR 59/47 34 nM [79] 36107 [65] 1kbh helices - (both IDPs) aAbbreviations: ACTR: the activation domain of p160 steroid receptor co-activator; HIF-1a: hypoxia-inducible factor 1 a subunit; NCBD: the nuclear-receptor co-activator binding domain of CREB binding protein (CBP); p53-TAD1: the transactivation domain 1 of tumor suppressor p53; TAZ1/2: the TAZ domains of CBP. The sequences of all IDPs involved (highlighted in bond fonts) are provided in the Supporting Information. Text S1. bThe experimental KD values were measured at 308 K for p53-TAD1/TAZ2, 298 K for HIF-1a/TAZ1, and 304 K for NCBD/ACTR. Note that KD only weakly depends on temperature for p53-TAD1/TAZ2 (doubled when the temperature is increased from 288K to 308K [77]). cNumbers of charged residues and the net charges (in parentheses) of the IDP, its binding site, and the vicinity of the binding site. Residues at the IDP binding interface are identified as those with greater than 1.0 A˚ 2 solvent accessible surface area changes upon complex formation. Surface residues are identified as those with .5% solvent accessibility. All surface residues within 15 A˚ Ca-Ca distance from the bound IDP but not directly involved in intermolecular contacts are considered to be within the vicinity of the IDP binding site. dEstimated based on the association rate constant of p53-TAD2/TAZ2 (,1010 M21s21 [38]), assuming that TAD1 and TAD2 have similar off rates. TAD2 binds to the TAZ2 primary site with KD ,32 nM [38], about two orders of magnitude stronger than TAD1. Topology-based modeling of IDP coupled binding and folding doi:10.1371/journal.pcbi.1003363.t001 the vicinity of the IDP binding site. dEstimated based on the association rate constant of p53-TAD2/TAZ2 (,1010 M21s21 [38]), assuming that TAD1 and TAD2 have similar off rates. TAD2 binds to the TAZ2 primary site with KD ,32 nM [38], about two orders of magnitude stronger than TAD1. doi:10 1371/journal pcbi 1003363 t001 Figure 1. Structures and surface electrostatic potentials of three complexes. A) p53-TAD1/TAZ2, B) NCBD/ACTR, and C) HIF-1a/TAZ1. TAZ2, NCBD and TAZ1 are shown in molecular surface and colored based on the surface electrostatic potential calculated using PBEQ module of CHARMM [80,81]. Red indicates negative and blue indicate positive charge. p53-TAD1, ACTR and HIF-1a are shown in cartoons, with charged side chains shown in stick. doi:10.1371/journal.pcbi.1003363.g001 Fig re 1 Str ct res and s rface electrostatic potentials of three comple es A) p53 TAD1/TAZ2 B) NCB Figure 1. Structures and surface electrostatic potentials of three complexes. A) p53-TAD1/TAZ2, B) NCBD/ACTR, and C) HIF-1a/TAZ1. TAZ2, NCBD and TAZ1 are shown in molecular surface and colored based on the surface electrostatic potential calculated using PBEQ module of CHARMM [80,81]. Red indicates negative and blue indicate positive charge. p53-TAD1, ACTR and HIF-1a are shown in cartoons, with charged side chains shown in stick. doi:10.1371/journal.pcbi.1003363.g001 doi:10.1371/journal.pcbi.1003363.g001 November 2013 \| Volume 9 \| Issue 11 \| e1003363 November 2013 \| Volume 9 \| Issue 11 \| e1003363 PLOS Computational Biology \| www.ploscompbiol.org PLOS Computational Biology \| www.ploscompbiol.org PLOS Computational Biology \| www.ploscompbiol.org 3 Electrostatically Accelerated Recognition of IDPs Table 2. Dissociation constants, melting temperatures, average reversible coupled binding and folding transition rates calculated using various coarse-grained models with and without explicit charges and/or 0.05 M salt. Table 2. Dissociation constants, melting temperatures, average reversible coupled binding and folding transition rates calculated using various coarse-grained models with and without explicit charges and/or 0.05 M salt. Table 2. Dissociation constants, melting temperatures, average reversible coupled binding and folding transition rates calculated using various coarse-grained models with and without explicit charges and/or 0.05 M salt. Models Calc. Topology-based modeling of IDP coupled binding and folding 2 compares the free energy surfaces as a function of intra- and inter-molecular native contact factions for all three complexes, calculated using calibrated Go¯-like models with and without explicit charges and/or salt (see Table 2). Both p53-TAD1 and HIF-1a recognitions follow induced folding-like mechanisms, where the peptides only gain structures after forming significant numbers of native intermolecular contacts. For example, Fig. 2A shows that p53-TAD1 does not start to fold until Qinter reaches ,0.5. Free NCBD is a molten globule with folded-like secondary structures [63], and its synergistic folding with ACTR has been previously shown to involve multiple stages of selection and induced folding [25,60], reminiscent of the ‘‘extended conforma- tional selection’’ mechanism [30]. Nonetheless, neither protein gains significant secondary (for ACTR) or tertiary (for NCBD) structures until over 20% of native intermolecular contacts are formed (Fig. 2G and 2J). baseline mechanisms of coupled binding and folding and to dissect the effects of long-range electrostatic forces. In particular, the fractions of native contacts formed have been shown to provide natural reaction coordinates for such mechanistic analysis [62]. Fig. 2 compares the free energy surfaces as a function of intra- and inter-molecular native contact factions for all three complexes, calculated using calibrated Go¯-like models with and without explicit charges and/or salt (see Table 2). Both p53-TAD1 and HIF-1a recognitions follow induced folding-like mechanisms, where the peptides only gain structures after forming significant numbers of native intermolecular contacts. For example, Fig. 2A shows that p53-TAD1 does not start to fold until Qinter reaches ,0.5. Free NCBD is a molten globule with folded-like secondary structures [63], and its synergistic folding with ACTR has been previously shown to involve multiple stages of selection and induced folding [25,60], reminiscent of the ‘‘extended conforma- tional selection’’ mechanism [30]. Nonetheless, neither protein gains significant secondary (for ACTR) or tertiary (for NCBD) structures until over 20% of native intermolecular contacts are formed (Fig. 2G and 2J). complexes [56–60], with many key predictions substantiated by independent experimental studies. Nonetheless, important differ- ences do exist between IDPs and structured proteins in sequence compositions and binding interface characteristics [42]. We have previously demonstrated that traditional topology-based models need to be carefully calibrated to ensure proper balance among competing intramolecular and intermolecular interactions (see Methods for detail on the calibration protocol) [61]. We note that the importance of model calibration was also illustrated in a recent study of the HIF-1a/TAZ1 complex [59]. Topology-based modeling of IDP coupled binding and folding Table 2 summarizes the final calibrated models for all three complexes. The calculated residual helicity distributions of the unbound states are show in Fig. S1. Three independent models were constructed for each complex: one without explicit charges (mimicking high salt concentration with fully screened long-range electrostatic interactions), one with explicit charges (mimicking low salt concentration with unscreened long-range electrostatic inter- actions), and a third one with explicit charges and 0.05 M salt (mimicking physiological conditions). All models reproduce the experimental KD to the same order of magnitude, except that the no charge model for HIF-1a/TAZ1 yields a KD value about one order of magnitude too large. We note that calculated KD values can be very sensitive to small changes of in the scaling of intermolecular interactions during model calibration (see Meth- ods). It is computationally expensive to use REX simulations to systematically search for the parameter space, especially for models without explicit charges due to slower transitions. Nonetheless, by performing production simulations at the corre- sponding melting temperatures, remaining imperfections in the balance of various interactions should be further suppressed, allowing reliable comparative studies of the mechanistic roles of electrostatic interactions in coupled binding and folding. Interestingly, formation of all three complexes involves inter- mediates, even though the intermediate in p53-TAD/TAZ2 interaction only become pronounced in the presence of nonspe- cific electrostatic forces (see Fig. 2A vs 2C). Detailed examination of the simulation trajectories and various free energy surfaces using fractions of native contacts formed by different IDP segments (e.g., see Figs. S2, S3, S4) revealed the existence of multiple parallel pathways for forming HIF-1a/TAZ1 and NCBD/ACTR. While these mechanistic details are not the focus of the current work, they appear to be highly consistent with previous experimental and computational studies. For example, as shown in Fig. S2, both the first and third helices of HIF-1a could initiate recognition, with the pathway initiated by the third helix binding being much more prevalent. Similar observations were also made in a separate computational study [59]. Specific recognition of NCBD/ACTR appears to be primarily initiated by the C-terminal segments of these two peptides (Figs. S3, S4), which forms a key intermediate Topology-based modeling of IDP coupled binding and folding KD Tm (K) kTS (ms21) kcap (ns21) kesc (ns21) kevo (ns21) TAD1/TAZ2 No charge 1.462.0 mM 327 4.361.5 1.4 8.0 0.049 Charged, 0.05M salt 1.661.6 mM 340 14.561.1 3.2 4.1 0.08 Explicit charges 4.963.2 mM 335 27.060.2 32.1 0.10 0.16 HIF-1a/TAZ1 No charge 64664 nM 327 6.160.5 2.8 6.3 0.022 Charged, 0.05M salt 9.469.6 nM 340 10.261.8 3.4 5.6 0.039 Explicit charges 1.361.6 nM 345 29.463.7 5.0 0.69 0.048 NCBD/ACTR No charge 67699 nM 318 0.5360.2 0.13 0.61 0.0043 Charged, 0.05M salt 96692 nM 315 1.760.1 0.29 0.31 0.0074 Explicit charges 39614 nM 322 5.260.7 0.79 0.020 0.012 KD was calculated from REX simulations at 300 K(see Table 1 for the experimental values); kTS was calculated from the production Langevin simulations at the corresponding Tm, as kTS = NTS/ttot, where NTS is the number of reversible binding and folding transitions observed during the total simulation time span ttot. As all simulations were performed at Tm, kTS as defined is half of the binding and unbinding rates. kcap kesc and kevo are defined in Eqns. 1–4. The effective concentrations of these simulations are 1.66 mM, 1.66 mM and 1.43 mM for p53-TAD1/TAZ2, HIF-1a/TAZ1 and NCBD/ACTR, respectively. All uncertainties were estimated as the differences between results calculated from the first and second halves of the data. doi:10.1371/journal.pcbi.1003363.t002 KD was calculated from REX simulations at 300 K(see Table 1 for the experimental values); kTS was calculated from the production Langevin simulations at the corresponding Tm, as kTS = NTS/ttot, where NTS is the number of reversible binding and folding transitions observed during the total simulation time span ttot. As all simulations were performed at Tm, kTS as defined is half of the binding and unbinding rates. kcap kesc and kevo are defined in Eqns. 1–4. The effective concentrations of these simulations are 1.66 mM, 1.66 mM and 1.43 mM for p53-TAD1/TAZ2, HIF-1a/TAZ1 and NCBD/ACTR, respectively. All uncertainties were estimated as the differences between results calculated from the first and second halves of the data. doi:10.1371/journal.pcbi.1003363.t002 baseline mechanisms of coupled binding and folding and to dissect the effects of long-range electrostatic forces. In particular, the fractions of native contacts formed have been shown to provide natural reaction coordinates for such mechanistic analysis [62]. Fig. Baseline mechanisms of coupled binding and folding: Effects of electrostatic forces Free energy surfaces were constructed using various combina- tions of folding and binding order parameters to understand the November 2013 \| Volume 9 \| Issue 11 \| e1003363 PLOS Computational Biology \| www.ploscompbiol.org 4 Electrostatically Accelerated Recognition of IDPs Figure 2. Free-energy surfaces at Tm as a function of the fractions of intra- and intermolecular contacts formed, computed using various Go¯ -like models with and without explicit charges and/or 50 mM salt (see Table 2). Rows A–C, D–F and G–L are for the p53-TAD1/ TAZ2, HIF-1a/TAZ1 and NCBD/ACTR complexes, respectively. Qinter is the fraction of intermolecular contacts formed; Qp53, QHIF-1a and QACTR are the fractions of intramolecular contacts formed by p53-TAD1, HIF-1a and ACTR, respectively; QNCBD-tert is the fraction of tertiary intramolecular contacts formed by NCBD (the helical content of NCBD remain similar during coupled binding and folding). Contours are drawn every kT, where k is Boltzmann constant and T is the absolute temperature. doi:10 1371/journal pcbi 1003363 g002 Figure 2. Free-energy surfaces at Tm as a function of the fractions of intra- and intermolecular contacts formed, computed using various Go¯ -like models with and without explicit charges and/or 50 mM salt (see Table 2). Rows A–C, D–F and G–L are for the p53-TAD1/ TAZ2, HIF-1a/TAZ1 and NCBD/ACTR complexes, respectively. Qinter is the fraction of intermolecular contacts formed; Qp53, QHIF-1a and QACTR are the fractions of intramolecular contacts formed by p53-TAD1, HIF-1a and ACTR, respectively; QNCBD-tert is the fraction of tertiary intramolecular contacts formed by NCBD (the helical content of NCBD remain similar during coupled binding and folding). Contours are drawn every kT, where k is Boltzmann constant and T is the absolute temperature. doi:10.1371/journal.pcbi.1003363.g002 examined. The baseline mechanisms for the formation of all three complexes remain induced folding-like. Furthermore, nonspecific electrostatic interactions do not change the relative prevalence of the parallel pathways that exist. For example, HIF-1a still initiates binding mainly through the third helix (Fig. S2); synergistic folding NCBD and ACTR is still mainly initiated through their C- terminal segments (Figs. S3, S4). The key effect of electrostatic forces appears to be substantial reductions in the free energy barriers that separate various basins. That is, even under the no salt condition, strong nonspecific electrostatic interactions do not appear to add to the ruggedness of coupled binding and folding free energy surfaces. Kinetic effects of long-range and nonspecific electrostatic forces Kinetics of coupled binding and folding was derived directly from production Langevin dynamics simulations performed using the calibrated Go¯-like models at their corresponding Tm. The results, summarized in Table 2, show that long-range electrostatic forces accelerate the reversible binding/unbinding transition rates for all three complexes. The overall electrostatic acceleration, estimated by comparing the average transition rates (kTS) calculated using models with and without explicit charges, ranges from ,5 fold for HIF-1a to 10 fold for NCBD/ACTR. The magnitude of acceleration is similar to what was previously measured for other IDPs including p27 [24] and PUMA [39] (both ,10 fold). The presence of 0.05 M salt significantly attenuates the predicted electrostatic acceleration, to only about two fold. However, the effect of salt screening on electrostatic acceleration is likely over-predicted [24], which is due to the Ca-only model used in this work and may be corrected with more detailed protein models [45]. Consistent with the kinetic analysis, there are significant reductions in the free energy barriers along Qinter (see Fig. 3), which has been shown to be a good binding reaction coordinate [61]. In addition, the magnitude of barrier reduction correlates well with the degree of rate acceleration calculated directly from Langevin dynamics simulations, with the largest barrier reduction observed for NCBD/ACTR and the smallest reduction observed from HIF-1a/TAZ1. The enhanced apparent efficiency of folding upon encounter appears to be frequently achieved at the cost of longer folding times. For example, the MFPTs of transitions from the collision complexes to the bound states increase from 0.26 to 3.94 ns for the p53-TAD1/TAZ2 complex (Table S1) and from 8.14 to 44.56 ns for the NCBD/ACTR complex (Tables S3). The net effects on the kinetics of encounter and folding stages can be quantified by calculating three effective rate constants as defined in Eqns. 2–4 (see Methods) [28]. The results, summarized in Table 2 and plotted in Fig. 4, clearly demonstrate that nonspecific electrostatic interaction enhance the encounter rates and reduce the escape To further analyze the effects of electrostatic interactions on different stages of coupled binding and folding, the recognition process was divided into two generic steps, including an encounter step followed by an evolving (folding) step to final bound and folded state (Eq. 1 in Methods). Electrostatically Accelerated Recognition of IDPs passage times (MFPT) and numbers of transitions (Ntran) among these states were then calculated. The results, summarized in Tables S1, S2, S3, show that long-range electrostatic forces greatly reduce the average encounter time, from 0.72 to 0.03 ns for p53- TAD, from 0.37 to 0.20 ns for HIF-1a, and from 7.71 to 1.26 ns for NCBD. At the same time, long-range electrostatic forces also significantly enhance the efficiency of IDP folding upon encounter, allowing much larger fractions of the encounter complexes to eventually evolve to the bound states. For example, for NCBD/ ACTR, only 16 out ,2300 encounter events evolved to the bound state in absence of long-range electrostatic forces (0.7%); whereas with explicit charges, there was ,37% probability (108 out of 288) of forming the specific complex once the proteins were captured into the collision complex state (Table S3). For the HIF-1a/TAZ1 complex, the percentages of collision to specific complex transition are 0.4% without and 7% with explicit charges (Table S2); for p53-TAD1/TAZ2, the production percentages are 0.6% without and 60% with explicit charges (Table S1). It should be emphasized that nonspecific electrostatic interactions significantly stabilize the collision complexes, due to large and complementary net charges of the interacting proteins (see Table 1). As such, much fewer fully unbinding events were observed during production simulations using the charged models. This effect also led to more reversible transitions between the bound and collision complex states and thus an overestimation of the true folding efficiency of IDPs upon collision as estimated above. We also note that the collision complexes as defined in our analysis were not intended to represent so-called ‘‘encounter complexes’’ that have been often considered key intermediates of protein-protein association [66], although encounter complexes are also believed to be mainly stabilized by nonspecific electrostatic interactions. self-consistency between the charge distribution and folded topology in the bound states, despite a lack of apparent complementary between folding topologies and surface electro- static potentials for these IDP complexes (see Fig. 1). Baseline mechanisms of coupled binding and folding: Effects of electrostatic forces An implication is that there exists a level of that was also suggested by an H/D exchange mass spectrometry study [64]. Kinetic data from a recent stop-flow study of the NCBD/ACTR interaction [65] are consistent with the prediction of induced folding as a baseline mechanism and have further confirmed the existence of parallel pathways and multiple folding intermediates. Representative snapshots along the dominant binding and folding pathways of p53-TAD1/TAZ2 and HIF- 1a/TAZ1 are shown in Figs. S5, S6. Explicit inclusion of charges does not significantly perturb the baseline mechanisms of coupled binding and folding. As shown in Fig. 2 and Figs. S2, S3, S4, long-range electrostatic forces do not lead to fundamental changes in any of the free energy surfaces November 2013 \| Volume 9 \| Issue 11 \| e1003363 PLOS Computational Biology \| www.ploscompbiol.org PLOS Computational Biology \| www.ploscompbiol.org 5 Discussion Inspection of the conformational properties of the collision complexes provides further insights into the molecular basis for enhanced efficiency of IDP folding upon encounter due to long- range electrostatic forces. As shown in Fig. 5, without nonspecific electrostatic interactions (models without explicit charges), the initial contacts between two binding partners are largely random, and the distributions of IDP initial contact points on the substrate surface in the collision complexes are relatively uniform (left column). In contrast, with the inclusion of explicit charges, the probabilities of IDP encountering near the native binding interface are dramatically increased. Coupled with reduced escape rates, this allows much higher efficiency of IDP folding upon encounter to achieve higher overall association rate constants (Table 2). The ability of long-range electrostatic forces to guide the recognition process is also reflected in the free energy surfaces as a function of binding RMSD of the IDP and center of mass separation between two peptides. As shown in Fig. 6, long-range electrostatic forces generate a strong free energy gradient that extends over 10–15 A˚ away from the native bound positions, without creating over- stabilized misfolded states at short separation distances. It is intriguing that, even though both NCBD and ACTR are disordered in the unbound state, nonspecific long-range electro- static forces between complementary charges on these two proteins can still manage to promote native-like topologies in the collision complexes. In particular, there is a much higher probability of NCBD and ACTR initiating contacts via the C-terminal helix of NCBD and the second helix of ACTR (Fig. 5E–F). This is part of a key pathway of synergistic folding inherent to the NCBD/ACTR complex that was predicted by coarse-grained and atomistic simulations [25,60] and later substantiated by H/D exchange mass spectrometry [64]. Therefore, nonspecific electrostatic interactions appear to mainly augment existing folding pathways While fulfilling important functional constraints such as structural plasticity for binding numerous specific targets, protein intrinsic disorder can lead to potential kinetic bottlenecks to be viable in cellular signaling and regulation. Our previous work on the p27/Cdk2/cyclin A complex has revealed a mechanism where nonspecific electrostatic interactions not only enhance the protein- protein encounter kinetics but also promote folding-competent encounter topologies to increase the efficiency of IDP folding upon encounter [24]. Kinetic effects of long-range and nonspecific electrostatic forces Such generic decomposition ignores the details of IDP-specific folding pathways, to allow on to focus on the net effects of electrostatic forces on the overall efficiency of IDP folding upon encounter. For this, three general states were identified during production simulations, including the unbound (U), collision complex (CC), and bound (B) states (see Methods for specific criteria for state assignment). The mean first Figure 3. Free energy as a function intermolecular contact fraction at Tm. These profiles were calculated from the REX simulations using WHAM for: A) TAD1/TAZ2, B) HIF-1a/TAZ1, and C) NCBD/ACTR. doi:10.1371/journal.pcbi.1003363.g003 Figure 3. Free energy as a function intermolecular contact fraction at Tm. These profiles were calculated from the REX simulations using WHAM for: A) TAD1/TAZ2, B) HIF-1a/TAZ1, and C) NCBD/ACTR. doi:10.1371/journal.pcbi.1003363.g003 November 2013 \| Volume 9 \| Issue 11 \| e1003363 PLOS Computational Biology \| www.ploscompbiol.org 6 Electrostatically Accelerated Recognition of IDPs Figure 4. Effective rate constants for transitions between the unbound, collision complex and bound states. The rates, as defined in Methods Eqns. 1–4, were calculated using models with and without explicit charges and/or 50 mM salt for: A) TAD1/TAZ2, B) HIF-1a/TAZ1 and C) NCBD/ACTR. The results demonstrate that long-range electrostatic forces increase both the capture and evolution rates and at the same time reduce the escape rates. doi:10.1371/journal.pcbi.1003363.g004 Figure 4. Effective rate constants for transitions between the unbound, collision complex and bound states. The rates, as defined in Methods Eqns. 1–4, were calculated using models with and without explicit charges and/or 50 mM salt for: A) TAD1/TAZ2, B) HIF-1a/TAZ1 and C) NCBD/ACTR. The results demonstrate that long-range electrostatic forces increase both the capture and evolution rates and at the same time reduce the escape rates. doi:10.1371/journal.pcbi.1003363.g004 rates of the collision complexes. Importantly, the effective evolution rates are always faster, by about three fold, in the presence of long-range electrostatic forces, despite longer MFPTs for the transitions from the collision complexes to the bound state observed for the p53-TAD1/TAZ2 and NCBD/ACTR complex- es. The magnitude of electrostatic acceleration of folding upon encounter is similar to what was previously observed for folding and binding of p27 to the Cdk2/cyclin A complex [24]. inherent to the folded topologies to facilitate efficient folding of IDPs upon encounter. Kinetic effects of long-range and nonspecific electrostatic forces Coupled with the previous observation that the vicinity of the IDP binding site tends to be enriched with charges to complement those on IDPs [24], thee current results suggest that there is likely a co-evolution of IDP folded topology, charge characteristics, and coupled binding and folding mecha- nisms. Furthermore, the co-evolution is likely driven by the important need to achieve facile IDP recognition for cellular signaling and regulation. Mechanism of electrostatically accelerated folding upon encounter Mechanism of electrostatically accelerated folding upon encounter Simulation protocols The complexes were simulated in cubic boxes with periodic boundary conditions imposed in CHARMM [71,72]. The box sizes are 100, 100 and 105 A˚ for p53-TAD1/TAZ2, HIF-1a/ TAZ1 and NCBD/ACTR, respectively. Langevin dynamics was performed with 15 fs time steps and a friction coefficient of 0.1 ps21. SHAKE was used to fix all virtual bond lengths [73]. Non-bonded interactions were cut off at 25 A˚ . Unbound IDPs were simulated at 300 K for 750 ns to calibrate the intramolecular interactions. REX-MD was performed using the MMTSB Toolset [68] for calibration of the intermolecular interactions. For this, eight replicas spanning 270 to 400 K were used. The lengths of REX calibration simulations ranged from 1.05 ms (for p53-TAD1/ TAZ2) up to 10 ms (for NCBD/ACTR), as needed for achieving sufficient convergence. Temperature weighted histogram analysis method (WHAM) [74] was used to compute the heat capacity (CV) curves and generate unbiased probability distributions for free energy and thermodynamic analysis. In particular, the dissociation constants (KD) were calculated from the bound and unbound probabilities at 300 K [61], where the unbound state was defined as the state without any native intermolecular contacts formed. For NCBD/ACTR complex, the 1D free energy profile lack significant barriers between the unbound and partially bound intermediate states (Fig. 3C, red trace). Therefore, the unbound probability was calculated as 1 – Pbound, where Pbound is the bound probability (see below for the specific criteria of state assignments). Once calibrated, production simulations of 30–40 ms in lengths were performed using all models at the corresponding TM’s (see Table 2). The TM value was first identified based on the CV curve and then fine tuned to ensure that similar probabilities of sampling the bound and unbound states were observed in the production simulation. Figure 5. Distributions of IDPs on the substrate surfaces in the collision complexes derived from simulations using models with and without explicit charges. For the p53-TAD1/TAZ2 (A–B) and HIF-1a/TAZ1 (C–D) complexes, TAZ2 and TAZ1 are colored based on the probability of each residue in contact with the IDPs in the collision complex ensembles, and p53-TAD1 and HIF-1a are shown only in the fold and bound conformations (yellow cartoon) for reference. For the NCBD/ACTR complex (E–F), both IDPs are shown in the bound and folded conformations and colored based on the probability of each residue involved (nonspecific) intermolecular contacts in the collision complex ensemble. Calibration of topology-based coarse-grained models with and without explicit charges Ca-only sequence-flavored Go¯-like models [67] were first derived from the complex structures of p53-TAD1/TAZ2, HIF1-a/TAZ1 and NCBD/ACTR (see Table 1) using the Multiscale Modeling Tools for Structural Biology (MMTSB) Go¯- Model Builder (http://www.mmtsb.org) [68]. The 3 zinc ions bound to TAZ1 in the HIF1-a/TAZ1 complex were modeled explicitly with distance restraints to the coordinating residues. All three models were then calibrated to balance the intrinsic folding propensity and the strength of intermolecular interactions using a previously described protocol [61]. Briefly, the strengths of intra- molecular native contact were uniformly scaled to reproduce the Electrostatically Accelerated Recognition of IDPs Electrostatically Accelerated Recognition of IDPs Figure 5. Distributions of IDPs on the substrate surfaces in the collision complexes derived from simulations using models with and without explicit charges. For the p53-TAD1/TAZ2 (A–B) and HIF-1a/TAZ1 (C–D) complexes, TAZ2 and TAZ1 are colored based on the probability of each residue in contact with the IDPs in the collision complex ensembles, and p53-TAD1 and HIF-1a are shown only in the fold and bound conformations (yellow cartoon) for reference. For the NCBD/ACTR complex (E–F), both IDPs are shown in the bound and folded conformations and colored based on the probability of each residue involved (nonspecific) intermolecular contacts in the collision complex ensemble. doi:10.1371/journal.pcbi.1003363.g005 experimentally measured residual helicity of unbound IDPs, which are mainly based on NMR secondary chemical shift and/or circular dichroism analysis (p53-TAD1 [69], NCBD/ACTR [63], and HIF1-a [70]). The residual helicity distributions calculated using the final models listed in Table 2 are provided in Fig. S1. Then, the strengths of intermolecular contacts were adjusted, such that binding affinities calculated from replica exchange molecular dynamics (REX-MD) simulations approximately match the experimental values (see Table 1). Following the previously described procedure [24], the calibrated sequence-flavored Go¯- like models were then further modified by assigning proper explicit charges to all charged residues (Lys, Arg, Glu and Asp) as well as zinc ions in the HIF1-a/TAZ1 complex. The charged models were then re-calibrated to reproduce the experimental residual structure level (Fig. S1) and binding affinity (Table 2). Such calibration is critical to avoid inherent bias for particular types of interactions, e.g., intra- vs. inter-molecular or native vs. nonspe- cific electrostatic. Nonspecific electrostatic interactions were modeled using the Debye-Hu¨ckel potential to account for ionic screening. The dielectric constant was set at 80. Simulation protocols d i 10 1371/j l bi 1003363 005 p doi:10.1371/journal.pcbi.1003363.g005 Discussion Using carefully calibrated topology-based coarse- grained models, we have now further demonstrated that similar electrostatically accelerated encounter and folding mechanisms also underlie the formation of three IDP complexes with more complexed folded structures, namely, p53-TAD1/TAZ2, HIF- 1a/TAZ1, and NCBD/ACTR. Importantly, these complexes lack apparent features on the electrostatic surface potentials to directly suggest the ability of nonspecific long-range electrostatic forces to promote native-like encounter topologies to enhance the IDP folding efficiency upon encounter. Nonetheless, there seems to exist a sufficient level of self-consistency between the charge distributions and folded topologies in the bound state to allow accelerated recognition in presence of nonspecific electrostatic interactions. Therefore, enriched charges on IDPs not only play key roles in modulating the conformational properties of the unbound state, but also likely play general and important roles in regulating efficient interactions of IDPs with specific partners. We note that IDPs are frequently regulated by post-translational modifications that add or remove charges. Improved mechanistic understanding of electrostatic forces in IDP recognition derived from the current work will thus help to dissect the profound impacts of post-translational modifications and disease-related mutations on IDP structure and interaction. November 2013 \| Volume 9 \| Issue 11 \| e1003363 PLOS Computational Biology \| www.ploscompbiol.org 7 Free energy and kinetic analysis All free energy profiles were calculated from the REX simulations and the kinetic analysis was performed based on the production simulations, unless otherwise stated. For calculation of contact fractions, a given native contact was considered as formed if the inter-Ca distance was within 1.0 A˚ of the distance in the native complex. Nonspecific intermolecular contacts are consid- ered as formed when the inter-Ca distance is within 10 A˚ cutoff. Three general conformational states were defined for each complex, including the unbound (U), collision complex (CC) and November 2013 \| Volume 9 \| Issue 11 \| e1003363 PLOS Computational Biology \| www.ploscompbiol.org 8 Electrostatically Accelerated Recognition of IDPs Figure 6. Free-energy surfaces at Tm as a function of binding RMSD of the IDP and center of mass separation between two peptides (RCM), computed using various Go¯ -like models with and without explicit charges and/or 50 mM salt (see Table 2). The binding RMSD (of the IDP) was calculated by first aligning the snapshot with respect to the folded structure using only the folded substrate. For NCBD/ACTR, both proteins are IDPs and the (regular) RMSD was calculated using the whole complex. Rows A–C, D–F and G–I are for the p53-TAD1/TAZ2, HIF-1a/TAZ1 and NCBD/ACTR complexes, respectively. Contours are drawn every kT. doi:10.1371/journal.pcbi.1003363.g006 Figure 6. Free-energy surfaces at Tm as a function of binding RMSD of the IDP and center of mass separation between two peptides (RCM), computed using various Go¯ -like models with and without explicit charges and/or 50 mM salt (see Table 2). The binding RMSD (of the IDP) was calculated by first aligning the snapshot with respect to the folded structure using only the folded substrate. For NCBD/ACTR, both proteins are IDPs and the (regular) RMSD was calculated using the whole complex. Rows A–C, D–F and G–I are for the p53-TAD1/TAZ2, HIF-1a/TAZ1 and NCBD/ACTR complexes, respectively. Contours are drawn every kT. doi:10.1371/journal.pcbi.1003363.g006 bound (B) states, to understand the effects of electrostatic forces on protein-protein encounter and subsequent folding upon encounter. U kcap=kesc CC kevo B ð1Þ kcap~MFPT{1 cap ð2Þ protein-protein encounter and subsequent folding upon encounter. The unbound state includes conformations with no specific or nonspecific contacts formed between IDP and substrate, and the collision complex state includes conformations with at least one nonspecific but no specific intermolecular contact formed. PLOS Computational Biology \| www.ploscompbiol.org g (TIF) QACTR-H1 inter ,QACTR-H2 inter and QACTR-H3 inter are the fractions of native intermolecular contacts formed by the first, second and third helices of ACTR, respectively. Contours are drawn every kT. (TIF) Figure S3 2D free energy surfaces at Tm calculated using models with (panels A, C, and E) and without explicit charges (panels B,D, Text S1 Amino acid sequences of all four IDPs simulated. (DOC) F) (see Table 2 of the main text). QACTR-H1 inter ,QACTR-H2 inter and QACTR-H3 inter are the fractions of native intermolecular contacts formed by the first, second and third helices of ACTR, respectively. Contours are drawn every kT. (TIF) (TIF) Figure S6 Representative snapshots along the binding and folding pathways for HIF-1a/TAZ1 extracted from the produc- tion simulation using the calibration model without explicit charges. (TIF) g (TIF) Figure S1 Residual helicities of (a) p53-TAD1, (b) HIf-1a, and (c) ACTR in the unbound states calculated using different Go¯-like models. The solid traces correspond to models without explicit charges and the dashed traces are from the charged models. The black traces were computed from models with no adjustment of the intramolecular interaction strengths (i.e., scale = 1.0), which signif- icantly over-stabilized the helices. The red traces were calculated using the final calibrated models with optimal scaling of intramo- lecular interactions (see Table 2 of the main text). The residual helicity showed minimal dependence on the salt concentration for all peptides and the corresponding profiles are thus not shown. (TIF) Table S1 MFPTs and numbers of transitions (in parenthesis) between conformational sub-states of the p53-TAD1/TAZ2 complex computed from the production Langevin simulations. (DOC) Table S2 MFPTs and numbers of transitions (in parenthesis) between conformational sub-states of the HIF-1a/TAZ1 complex computed from the production Langevin simulations. (DOCX) Table S3 MFPTs and numbers of transitions (in parenthesis) between conformational sub-states of the NCBD/ACTR complex computed from the production Langevin simulations. (DOCX) Figure S2 2D free energy surfaces at Tm calculated using models with (panels A, C, and E) and without explicit charges (panels B,D, F) (see Table 2 of the main text). QHIF-1aA inter and QHIF-1aC inter are the fractions of native intermolecular contacts formed by the first and third helices of HIF-1a, respectively. RCM is the distance between the centers of mass of HIF-1a and TAZ1. Contours are drawn every kT. (TIF) Figure S2 2D free energy surfaces at Tm calculated using models with (panels A, C, and E) and without explicit charges (panels B,D, F) (see Table 2 of the main text). QHIF-1aA inter and QHIF-1aC inter are the fractions of native intermolecular contacts formed by the first and third helices of HIF-1a, respectively. RCM is the distance between the centers of mass of HIF-1a and TAZ1. Contours are drawn every kT. (TIF) Table S4 Averaged on and off rates (kon and koff), as calculated from the mean residence times in either unbound or bound states during the production Langevin simulations at the corresponding Tm (as estimated from short replica exchange simulations). (DOC) Figure S3 2D free energy surfaces at Tm calculated using models with (panels A, C, and E) and without explicit charges (panels B,D, F) (see Table 2 of the main text). p (TIF) Figure S4 2D free energy surfaces at Tm calculated using models with (panels A, C, and E) and without explicit charges (panels B,D, Free energy and kinetic analysis The bound states are defined as following: 1) for p53-TAD1/TAZ2: Ninter$11; 2) for HIF-1a/TAZ1: Ninter$26 for the no charge model, Ninter$23 for the charged model, and Ninter$24 for the charged model with 0.05 M salt; 3) for ACTR/NCBD: Ninter$30. Ninter is the total number of native intermolecular contacts formed. Note that slightly different criteria were used to define the bound state of HIF-1a/TAZ1 due to small shifts of the bound free energy basins calculated using different models (see Fig. 3). 15-ps running averages were used for assigning states, to avoid including fictitious transitions due to rapid small fluctuations in the calculated contact counts (especially between the U and CC states). The overall on and off rates were calculated directly from the average lifetimes of the bound and unbound states (see Table S4). In addition, MFPTs and numbers of transitions among all three states were derived from the production simulation trajectories, and various rates were calculated as defined in Eqns. 2–4. ð1Þ The unbound state includes conformations with no specific or nonspecific contacts formed between IDP and substrate, and the collision complex state includes conformations with at least one nonspecific but no specific intermolecular contact formed. The bound states are defined as following: 1) for p53-TAD1/TAZ2: Ninter$11; 2) for HIF-1a/TAZ1: Ninter$26 for the no charge model, Ninter$23 for the charged model, and Ninter$24 for the charged model with 0.05 M salt; 3) for ACTR/NCBD: Ninter$30. Ninter is the total number of native intermolecular contacts formed. ð2Þ ð2Þ kesc~ MFPTesc\|NesczMFPTevo\|Nevo ð Þ= NesczNevo ð Þ ½ {1 \| Nesc NesczNevo ð3Þ ð3Þ kevo~ MFPTesc\|NesczMFPTevo\|Nevo ð Þ= NesczNevo ð Þ ½ {1 \| Nevo NesczNevo ð4Þ ð4Þ \| Nevo NesczNevo Here, kcap, kesc, and kevo are the capture, escape (to the unbound state) and evolution (to the bound state) rates of the collision November 2013 \| Volume 9 \| Issue 11 \| e1003363 PLOS Computational Biology \| www.ploscompbiol.org 9 References 1. Dyson HJ, Wright PE (2005) Intrinsically unstructured proteins and their functions. Nature Reviews Molecular Cell Biology 6: 197–208. 1. Dyson HJ, Wright PE (2005) Intrinsically unstructured proteins and their functions. Nature Reviews Molecular Cell Biology 6: 197–208. 9. Pajkos M, Meszaros B, Simon I, Dosztanyi Z (2012) Is there a biological cost of protein disorder? Analysis of cancer-associated mutations. Molecular BioSystems 8: 296–307. 2. Uversky VN, Oldfield CJ, Dunker AK (2005) Showing your ID: intrinsic disorder as an ID for recognition, regulation and cell signaling. Journal of Molecular Recognition 18: 343–384. 10. Tsai CJ, Ma B, Sham YY, Kumar S, Nussinov R (2001) Structured disorder and conformational selection. Proteins 44: 418–427. 11. Tompa P, Szasz C, Buday L (2005) Structural disorder throws new light on moonlighting. Trends in Biochemical Sciences 30: 484–489. 3. Smock RG, Gierasch LM (2009) Sending signals dynamically. Science 324: 198– 203. 4. Dunker AK, Lawson JD, Brown CJ, Williams RM, Romero P, et al. (2001) Intrinsically disordered protein. Journal of Molecular Graphics & Modelling 19: 26–59. 12. Boehr DD, Nussinov R, Wright PE (2009) The role of dynam 12. Boehr DD, Nussinov R, Wright PE (2009) The role of dynamic conformational ensembles in biomolecular recognition. Nat Chem Biol 5: 789–796. ensembles in biomolecular recognition. Nat Chem Biol 5: 789–79 13. Hilser VJ, Thompson EB (2007) Intrinsic disorder as a mechanism to optimize allosteric coupling in proteins. Proceedings of the National Academy of Sciences of the United States of America 104: 8311–8315. 5. Iakoucheva LM, Brown CJ, Lawson JD, Obradovic Z, Dunker AK (2002) Intrinsic disorder in cell-signaling and cancer-associated proteins. Journal of Molecular Biology 323: 573–584. 14. Shoemaker BA, Portman JJ, Wolynes PG (2000) Speeding molecular recognition by using the folding funnel: The fly-casting mechanism. Proceedings of the National Academy of Sciences of the United States of America 97: 8868–8873. 6. Uversky VN, Oldfield CJ, Dunker AK (2008) Intrinsically disordered proteins in human diseases: Introducing the D-2 concept. Annual Review of Biophysics 37: 215–246. 15. Vacic V, Markwick PRL, Oldfield CJ, Zhao X, Haynes C, et al. (2012) Disease- Associated Mutations Disrupt Functionally Important Regions of Intrinsic Protein Disorder. Plos Computational Biology 8: e1002709. 7. Oldfield CJ, Cheng Y, Cortese MS, Brown CJ, Uversky VN, et al. (2005) Comparing and combining predictors of mostly disordered proteins. Biochem- istry 44: 1989–2000. 8. Vacic V, Iakoucheva LM (2012) Disease mutations in disordered regions– exception to the rule? 8. Vacic V, Iakoucheva LM (2012) Disease mutations in disordered regions– exception to the rule? Molecular BioSystems 8: 27–32. Author Contributions Conceived and designed the experiments: DG WZ JC. Performed the experiments: DG WZ. Analyzed the data: DG WZ JC. Contributed reagents/materials/analysis tools: DG WZ JC. Wrote the paper: DG WZ JC. Electrostatically Accelerated Recognition of IDPs F) (see Table 2 of the main text). QNCBD-H1 inter ,QNCBD-H2 inter and QNCBD-H3 inter are the fractions of native intermolecular contacts formed by the first, second and third helices of NCBD, respectively. Contours are drawn every kT. (TIF) complex, respectively; Nesc and Nevo are the numbers of escape and evolution transitions. Note that the MFPTs calculated correspond to the average times spent in an initial state before a transition to the final state. Ideally, the average lifetime of CC should be independent of whether the trajectory ends up in either the U or B state for a true three-state model as shown in Eq. 1. However, the actual transitions between the CC and B states involve several intermediates that are not represented in Eq. 1, and the effective MFPTs as calculated thus depend on both the initial and final states (e.g., see Tables S1, S2, S3). Analytical expressions on similar MFPTs involved in amyloid fibril templating can be found a recent theoretical analysis by Schmit [75]. All molecular visualizations were prepared using VMD [76]. F) Figure S5 Representative snapshots along the binding and folding pathways of p53-TAD1/TAZ2 extracted from the production simulation using the calibration model without explicit charges. Electrostatically Accelerated Recognition of IDPs Genes & Development 14: 1553–1577. 25. Zhang W, Ganguly D, Chen J (2012) Residual structures, conformational fluctuations, and electrostatic interactions in the synergistic folding of two intrinsically disordered proteins. Plos Computational Biology 8: e1002353. p p 53. Wolynes PG (2005) Recent successes of the energy landscape theory of protein folding and function. Quarterly Reviews of Biophysics 38: 405–410. 26. Kubelka J, Hofrichter J, Eaton WA (2004) The protein folding ‘speed limit’. Current Opinion in Structural Biology 14: 76–88. 54. Dill KA, Chan HS (1997) From Levinthal to pathways to funnels. Nature Structural Biology 4: 10–19. 55. Shea JE, Onuchic JN, Brooks CL (1999) Exploring the origins of topological frustration: Design of a minimally frustrated model of fragment B of protein A. Proceedings of the National Academy of Sciences of the United States of America 96: 12512–12517. 27. Oldfield CJ, Meng J, Yang JY, Yang MQ, Uversky VN, et al. (2008) Flexible nets: disorder and induced fit in the associations of p53 and 14-3-3 with their partners. BMC Genomics 9 Suppl 1: S1. p pp 28. Huang Y, Liu Z (2009) Kinetic advantage of intrinsically disordered proteins in coupled folding-binding process: a critical assessment of the ‘‘fly-casting’’ mechanism. Journal of Molecular Biology 393: 1143–1159. 56. Turjanski AG, Gutkind JS, Best RB, Hummer G (2008) Binding-induced folding of a natively unstructured transcription factor. PLoS Comput Biol 4: e1000060. J gy 29. Zhou HX, Pang X, Lu C (2012) Rate constants and mechanisms of intrinsically disordered proteins binding to structured targets. Physical chemistry chemical physics: PCCP 14: 10466–10476. 57. Lu Q, Lu HP, Wang J (2007) Exploring the mechanism of flexible biomolecular recognition with single molecule dynamics. Physical Review Letters 98: 128105. 58. Wang J, Wang Y, Chu X, Hagen SJ, Han W, et al. (2011) Multi-Scaled Explorations of Binding-Induced Folding of Intrinsically Disordered Protein Inhibitor IA3 to its Target Enzyme. Plos Computational Biology 7: e1001118. p y 30. Csermely P, Palotai R, Nussinov R (2010) Induced fit, conformational selection and independent dynamic segments: an extended view of binding events. Trends in Biochemical Sciences 35: 539–546. Inhibitor IA3 to its Target Enzyme. Plos Computational Biology 7: e g y p gy 59. De Sancho D, Best RB (2012) Modulation of an IDP binding mechanism and rates by helix propensity and non-native interactions: association of HIF1alpha with CBP. Molecular BioSystems 8: 256–267. 31. Electrostatically Accelerated Recognition of IDPs Brown AM, Zondlo NJ (2012) A Propensity Scale for Type II Polyproline Helices (PPII): Aromatic Amino Acids in Proline-Rich Sequences Strongly Disfavor PPII Due to Proline-Aromatic Interactions. Biochemistry 51: 5041– 5051. 60. Ganguly D, Zhang W, Chen J (2012) Synergistic folding of two intrinsically disordered proteins: searching for conformational selection. Molecular BioSys- tems 8: 198–209. 32. Mao AH, Crick SL, Vitalis A, Chicoine CL, Pappu RV (2010) Net charge per residue modulates conformational ensembles of intrinsically disordered proteins. Proceedings of the National Academy of Sciences of the United States of America 107: 8183–8188. 61. Ganguly D, Chen J (2011) Topology-based modeling of intrinsically disordered proteins: balancing intrinsic folding and intermolecular interactions. Proteins 79: 1251–1266. 62. Cho SS, Levy Y, Wolynes PG (2006) P versus Q: structural reaction coordinates capture protein folding on smooth landscapes. Proceedings of the National Academy of Sciences of the United States of America 103: 586–591. 33. Muller-Spath S, Soranno A, Hirschfeld V, Hofmann H, Ruegger S, et al. (2010) Charge interactions can dominate the dimensions of intrinsically disordered proteins. Proceedings of the National Academy of Sciences of the United States of America 107: 14609–14614. 63. Ebert MO, Bae SH, Dyson HJ, Wright PE (2008) NMR relaxation study of the complex formed between CBP and the activation domain of the nuclear hormone receptor coactivator ACTR. Biochemistry 47: 1299–1308. 34. Vuzman D, Levy Y (2010) DNA search efficiency is modulated by charge composition and distribution in the intrinsically disordered tail. Proceedings of the National Academy of Sciences of the United States of America 107: 21004– 21009. 64. Keppel TR, Howard BA, Weis DD (2011) Mapping unstructured regions and synergistic folding in intrinsically disordered proteins with amide H/D exchange mass spectrometry. Biochemistry 50: 8722–8732. 35. Levy Y, Onuchic JN, Wolynes PG (2007) Fly-casting in protein-DNA binding: frustration between protein folding and electrostatics facilitates target recogni- tion. Journal of the American Chemical Society 129: 738–739. 65. Dogan J, Schmidt T, Mu X, Engstrom A, Jemth P (2012) Fast association and slow transitions in the interaction between two intrinsically disordered protein domains. The Journal of biological chemistry 287: 34316–34324. 36. Galea CA, Nourse A, Wang Y, Sivakolundu SG, Heller WT, et al. (2008) Role of intrinsic flexibility in signal transduction mediated by the cell cycle regulator, p27(Kip1). Journal of Molecular Biology 376: 827–838. 66. Ubbink M (2009) The courtship of proteins: understanding the encounter complex. FEBS Letters 583: 1060–1066. 67. Electrostatically Accelerated Recognition of IDPs 45. Azia A, Levy Y (2009) Nonnative electrostatic interactions can modulate protein folding: molecular dynamics with a grain of salt. Journal of Molecular Biology 393: 527–542. 17. Mittag T, Forman-Kay JD (2007) Atomic-level characterization of disordered protein ensembles. Curr Opin Struct Biol 17: 3–14. 18. Galea CA, Wang Y, Sivakolundu SG, Kriwacki RW (2008) Regulation of cell division by intrinsically unstructured proteins: Intrinsic flexibility, modularity, and signaling conduits. Biochemistry 47: 7598–7609. 46. Stoycheva AD, Onuchic JN, Brooks CL (2003) Effect of gatekeepers on the early folding kinetics of a model beta-barrel protein. Journal of Chemical Physics 119: 5722–5729. g g y 19. Wright PE, Dyson HJ (2009) Linking folding and binding. Current Opinion in Structural Biology 19: 31–38. 47. Stoycheva AD, Brooks CL, Onuchic JN (2004) Gatekeepers in the ribosomal protein S6: Thermodynamics, kinetics, and folding pathways revealed by a minimalist protein model. Journal of Molecular Biology 340: 571–585. 20. Click TH, Ganguly D, Chen J (2010) Intrinsically Disordered Proteins in a Physics-Based World. International Journal of Molecular Sciences 11: 5292– 5309. minimalist protein model. Journal of Molecular Biology 340: 571– 48. Toth-Petroczy A, Simon I, Fuxreiter M, Levy Y (2009) Disordered tails of homeodomains facilitate DNA recognition by providing a trade-off between folding and specific binding. Journal of the American Chemical Society 131: 15084–15085. 21. Fisher CK, Stultz CM (2011) Constructing ensembles for intrinsically disordered proteins. Current Opinion in Structural Biology 21: 426–431. p p gy 22. Das RK, Mittal A, Pappu RV (2013) How is functional specificity achieved through disordered regions of proteins? BioEssays 35: 17–22. 49. Huang Y, Liu Z (2010) Nonnative interactions in coupled folding and binding processes of intrinsically disordered proteins. PLoS ONE 5: e15375. 23. Zhou HX (2010) From Induced Fit to Conformational Selection: A Continuum of Binding Mechanism Controlled by the Timescale of Conformational Transitions. Biophysical Journal 98: L15–L17. 50. Chu X, Wang Y, Gan L, Bai Y, Han W, et al. (2012) Importance of Electrostatic Interactions in the Association of Intrinsically Disordered Histone Chaperone Chz1 and Histone H2A.Z-H2B. Plos Computational Biology 8: e1002608. 24. Ganguly D, Otieno S, Waddell B, Iconaru L, Kriwacki RW, et al. (2012) Electrostatically Accelerated Coupled Binding and Folding of Intrinsically Disordered Proteins. Journal of Molecular Biology 422: 674–684. p gy 51. Kruse JP, Gu W (2009) Modes of p53 Regulation. Cell 137: 609–62 52. Goodman RH, Smolik S (2000) CBP/p300 in cell growth, transformation, and development. References Molecular BioSystems 8: 27–32. 16. Chen JH (2012) Towards the physical basis of how intrinsic disorder mediates protein function. Archives of Biochemistry and Biophysics 524: 123–131. 8. Vacic V, Iakoucheva LM (2012) Disease mutations in disordered regions– exception to the rule? Molecular BioSystems 8: 27–32. November 2013 \| Volume 9 \| Issue 11 \| e1003363 November 2013 \| Volume 9 \| Issue 11 \| e1003363 10 PLOS Computational Biology \| www.ploscompbiol.org Electrostatically Accelerated Recognition of IDPs Karanicolas J, Brooks CL (2002) The origins of asymmetry in the folding transition states of protein L and protein G. Protein Science 11: 2351–2361. 37. Sivakolundu SG, Nourse A, Moshiach S, Bothner B, Ashley C, et al. (2008) Intrinsically unstructured domains of Arf and Hdm2 form bimolecular oligomeric structures in vitro and in vivo. Journal of Molecular Biology 384: 240–254. 68. Feig M, Karanicolas J, Brooks CL (2004) MMTSB Tool Set: enhanced sampling and multiscale modeling methods for applications in structural biology. Journal of Molecular Graphics & Modelling 22: 377–395. 38. Arai M, Ferreon JC, Wright PE (2012) Quantitative analysis of multisite protein- ligand interactions by NMR: binding of intrinsically disordered p53 transactiva- tion subdomains with the TAZ2 domain of CBP. Journal of the American Chemical Society 134: 3792–3803. p g 69. Lee H, Mok KH, Muhandiram R, Park KH, Suk JE, et al. (2000) Local structural elements in the mostly unstructured transcriptional activation domain of human p53. The Journal of biological chemistry 275: 29426–29432. 39. Rogers JM, Steward A, Clarke J (2013) Folding and Binding of an Intrinsically Disordered Protein: Fast, but Not ‘Diffusion-Limited’. Journal of the American Chemical Society 135: 1415–1422. 70. Dames SA, Martinez-Yamout M, De Guzman RN, Dyson HJ, Wright PE (2002) Structural basis for Hif-1 alpha/CBP recognition in the cellular hypoxic response. Proc Natl Acad Sci U S A 99: 5271–5276. 40. Schreiber G, Haran G, Zhou HX (2009) Fundamental Aspects of Protein- Protein Association Kinetics. Chemical Reviews 109: 839–860. 71. Brooks BR, Bruccoleri RE, Olafson BD, States DJ, Swaminathan S, et al. (1983) Charmm - a Program for Macromolecular Energy, Minimization, and Dynamics Calculations. Journal of Computational Chemistry 4: 187–217. 41. Vijayakumar M, Wong KY, Schreiber G, Fersht AR, Szabo A, et al. (1998) Electrostatic enhancement of diffusion-controlled protein-protein association: comparison of theory and experiment on barnase and barstar. Journal of Molecular Biology 278: 1015–1024. y p y 72. Brooks BR, Brooks CL, Mackerell AD, Nilsson L, Petrella RJ, et al. (2009) CHARMM: The Biomolecular Simulation Program. Journal of Computational Chemistry 30: 1545–1614. 42. Meszaros B, Tompa P, Simon I, Dosztanyi Z (2007) Molecular principles of the interactions of disordered proteins. Journal of Molecular Biology 372: 549–561. 73. Ryckaert JP, Ciccotti G, Berendsen HJC (1977) Numerical-Integration of Cartesian Equations of Motion of a System with Constraints - Molecular- Dynamics of N-Alkanes. Journal of Computational Physics 23: 327–341. 42. 78. Sugase K, Lansing JC, Dyson HJ, Wright PE (2007) Tailoring relaxation dispersion experiments for fast-associating protein complexes. Journal of the American Chemical Society 129: 13406–13407. 80. Jo S, Kim T, Iyer VG, Im W (2008) CHARMM-GUI: a web-based graphical user interface for CHARMM. J Comput Chem 29: 1859–1865. J p 81. Jo S, Vargyas M, Vasko-Szedlar J, Roux B, Im W (2008) PBEQ-Solver for online visualization of electrostatic potential of biomolecules. Nucleic Acids Res 36: W270–275. J p 77. Feng H, Jenkins LM, Durell SR, Hayashi R, Mazur SJ, et al. (2009) Structural basis for p300 Taz2-p53 TAD1 binding and modulation by phosphorylation. Structure 17: 202–210. 79. Demarest SJ, Martinez-Yamout M, Chung J, Chen HW, Xu W, et al. (2002) Mutual synergistic folding in recruitment of CBP/p300 by p160 nuclear receptor coactivators. Nature 415: 549–553. 76. Humphrey W, Dalke A, Schulten K (1996) VMD: Visual molecular dynamics. Journal of Molecular Graphics 14: 33–8, 27–8. Electrostatically Accelerated Recognition of IDPs Electrostatically Accelerated Recognition of IDPs Meszaros B, Tompa P, Simon I, Dosztanyi Z (2007) Molecular principles of the interactions of disordered proteins. Journal of Molecular Biology 372: 549–561. 43. Liu H, Shi Y, Chen XS, Warshel A (2009) Simulating the electrostatic guidance of the vectorial translocations in hexameric helicases and translocases. Proceedings of the National Academy of Sciences 106: 7449–7454. 43. Liu H, Shi Y, Chen XS, Warshel A (2009) Simulating the electrostatic guidance of the vectorial translocations in hexameric helicases and translocases. Proceedings of the National Academy of Sciences 106: 7449–7454. y p y 74. Gallicchio E, Andrec M, Felts AK, Levy RM (2005) Temperature weighted histogram analysis method, replica exchange, and transition paths. Journal of Physical Chemistry B 109: 6722–6731. 44. Marcovitz A, Levy Y (2011) Frustration in protein-DNA binding influences conformational switching and target search kinetics. Proceedings of the National Academy of Sciences 108: 17957–17962. 44. Marcovitz A, Levy Y (2011) Frustration in protein-DNA binding influences conformational switching and target search kinetics. Proceedings of the National Academy of Sciences 108: 17957–17962. y y 75. Schmit J (2013) Kinetic theory of amyloid fibril templating. Journal of Chemical Physics 138: 185102. November 2013 \| Volume 9 \| Issue 11 \| e1003363 PLOS Computational Biology \| www.ploscompbiol.org 11 Electrostatically Accelerated Recognition of IDPs PLOS Computational Biology \| www.ploscompbiol.org November 2013 \| Volume 9 \| Issue 11 \| e1003363 PLOS Computational Biology \| www.ploscompbiol.org 12
https://openalex.org/W1977600154	https://re.public.polimi.it/bitstream/11311/1045376/2/11311-1045376%20Barbieri.PDF	English	null	Estimation of Instantaneous Complex Dynamics through Lyapunov Exponents: A Study on Heartbeat Dynamics	PloS one	2,014	cc-by	12,273	Abstract Measures of nonlinearity and complexity, and in particular the study of Lyapunov exponents, have been increasingly used to characterize dynamical properties of a wide range of biological nonlinear systems, including cardiovascular control. In this work, we present a novel methodology able to effectively estimate the Lyapunov spectrum of a series of stochastic events in an instantaneous fashion. The paradigm relies on a novel point-process high-order nonlinear model of the event series dynamics. The long-term information is taken into account by expanding the linear, quadratic, and cubic Wiener-Volterra kernels with the orthonormal Laguerre basis functions. Applications to synthetic data such as the He´non map and Ro¨ssler attractor, as well as two experimental heartbeat interval datasets (i.e., healthy subjects undergoing postural changes and patients with severe cardiac heart failure), focus on estimation and tracking of the Instantaneous Dominant Lyapunov Exponent (IDLE). The novel cardiovascular assessment demonstrates that our method is able to effectively and instantaneously track the nonlinear autonomic control dynamics, allowing for complexity variability estimations. Citation: Valenza G, Citi L, Barbieri R (2014) Estimation of Instantaneous Complex Dynamics through Lyapunov Exponents: A Study on Heartbeat Dynamics. PLoS ONE 9(8): e105622. doi:10.1371/journal.pone.0105622 Editor: Ramesh Balasubramaniam, University of California, Merced, United States of America Editor: Ramesh Balasubramaniam, University of California, Merced, United States of America Editor: Ramesh Balasubramaniam, University of California, Merced, United States of America Received January 13, 2014; Accepted July 25, 2014; Published August 29, 2014 Received January 13, 2014; Accepted July 25, 2014; Published August 29, 2014 enza et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits tion, and reproduction in any medium, provided the original author and source are credited. Copyright: 2014 Valenza et al. This is an open-access article distributed under unrestricted use, distribution, and reproduction in any medium, provided the original Copyright: 2014 Valenza et al. This is an open-access article distributed under the terms of the Creative Commons Attr unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Copyright: 2014 Valenza et al. This is an open-access article distributed under the terms of the Creative Commons Attribut unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Abstract Funding: The research leading to these results has received partial funding from the Department of Anesthesia, Critical Care & Pain Medicine, Massachusetts General Hospital, and Harvard Medical School, Boston, MA, USA, and European Union Seventh Framework Programme FP7/2007-2013 under Grant No. 601165 of the project ‘‘WEARHAP’’. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * Email: g.valenza@ieee.org oscillation could yield a positive LE if the trajectory has regions with large slopes. In chaotic systems [37], stochasticity does not play a crucial role and their dynamics are highly dependent on the initial condition. Although it is straightforward to consider chaotic mathematical systems in which stochastic inputs are suppressed, in actual applications especially related to physiological systems, it is not possible to eliminate such inputs making the chaos assessment simply unreliable [3]. Stationary aperiodic behavior, in fact, can also arise in linear or nonlinear stochastic systems. In light of these considerations, as this work deals with (instantaneous) LEs estimation with applications on heartbeat dynamics, we do not address the issue related to the chaotic behavior of heart rate variability (HRV). Gaetano Valenza1,2, Luca Citi1,3, Riccardo Barbieri1 1 Neuroscience Statistics Research Laboratory, Department of Anesthesia, Critical Care & Pain Medicine, Harvard Medical School, Massachusetts General Hospital, Boston, Massachusetts, United States of America; and Department of Brain and Cognitive Science, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States of America, 2 Research Center E. Piaggio and Department of Information Engineering, University of Pisa, Pisa, Italy, 3 School of Computer Science and Electronic Engineering, University of Essex, Colchester, United Kingdom August 2014 \| Volume 9 \| Issue 8 \| e105622 Introduction Hearth contractions are regarded by many scientists as the foremost example of a physiological system showing predomi- nantly nonlinear behavior, mainly generated through integration of multiple neural signaling at the level of the sinoatrial node [1]. Accordingly, the fluctuations in the interval between consecutive heartbeats have been widely investigated as output of a nonlinear system revealing and quantifying the complexity of cardiovascular control [2–27]. Among all nonlinearity and complexity measures, Lyapunov exponents (LEs) have been proven to provide an important mathematical tool in characterizing dynamical properties of a nonlinear system [28]. LEs were first defined by Lyapunov [29] in order to study the stability of non-stationary solutions of ordinary differential equations and for more than fifty years they have been extensively studied in many disciplines [30–33]. Specifically, they refer to the average exponential rates of divergence or conver- gence of neighboring trajectories in the system phase space. In fact, for a system whose characteristic equations are known, there is a straightforward technique for computing the whole Lyapunov spectrum [28]. Several methods for a reliable data-driven LEs estimation, even in short time data records, have been also proposed [34,35]. In a deterministic nonlinear system with no stochastic inputs, a positive LE reflects sensitive dependence to initial conditions and can be taken as a definition of a chaotic system [36]. Nevertheless, a small amount of noise in a limit cycle Relying on the approach suggested by Chon et al. [38] and, later, by Armoundas et al. [39], we consider the cardiovascular system both chaotic and stochastic. This concept is in agreement with current physiological knowledge, since healthy HRV dynamics can be considered/modeled as the output of a nonlinear deterministic system (the pacemaker cells of sinus node) being forced by a high-dimensional input (the activity in the nerves innervating the sinus node). Accordingly, we model the heartbeat nonlinear dynamics as a third-order Nonlinear Autoregressive (NAR) model embedded in a point process probabilistic frame- work. Such statistical approach allows us to estimate the LEs in an instantaneous fashion by fitting the model to the observed data and applying the Fast Orthogonal Search (FOS) algorithm [40]. Introduction Point-process theory has been widely recognized as an excellent 1 August 2014 \| Volume 9 \| Issue 8 \| e105622 PLOS ONE \| www.plosone.org August 2014 \| Volume 9 \| Issue 8 \| e105622 Estimating the Instantaneous Lyapunov Spectra mathematical tool to characterize the probabilistic generative mechanism of the heartbeat at each moment in time [41]. In the considered model, the intrinsically discrete, unevenly spaced heartbeat intervals are represented by a physiologically-plausible inverse-gaussian (IG) distribution. Defining the first and second- order moments of the IG distribution as function of the past heartbeat intervals (i.e., the RR intervals), it is possible to obtain an effective prediction of the next heartbeat event together with an accurate assessment of instantaneous indices of cardiovascular control. In previous studies [19,41,42], we demonstrated how to estimate heartbeat dynamics even in short recordings under nonlinear and non-stationary conditions using Wiener-Volterra theory for nonlinear systems identification. f (tjHt,j(t))~ j0(t) 2p(t{uj)3 " #1 2 exp { 1 2 j0(t) t{uj{mRR(t,Ht,j(t)) 2 mRR(t,Ht,j(t))2 (t{uj) ( ) ð1Þ ð1Þ where Ht~(uj,RRj,RRj{1,:::,RRj{Mz1) is the history of the point process, j(t) is the vector of the time-varying parameters, mRR(t,Ht,j(t)) represents the first-moment statistic (mean) of the distribution, and j0(t)~hw0 denotes the shape parameter of the IG distribution (as h=m??, the IG distribution becomes more like a Gaussian distribution). As f (tDHt,j(t)) indicates the probability of having a beat at time t given that a previous beat has occurred at uj, mRR(t,Ht,j(t)) can be interpreted as signifying the average (or expected) waiting time before the next beat. We can also estimate the second-moment statistic (variance) of the IG distribution as s2 RR(t)~m3 RR(t)=h. The use of an IG distribution to characterize the R-R intervals occurrences is motivated by the fact that if the rise of the membrane potential to a threshold initiating the cardiac contraction is modeled as a Wiener process with drift, then the probability density of the times between threshold crossings (the RR intervals) is indeed the inverse Gaussian distribution [41,49]. As a matter of fact, when compared with other distributions, the IG model achieves the overall best fit in terms of smaller KS distance [42]. The instantaneous RR mean, mRR(t,Ht,j(t)), can be modeled as a generic function of the past RR values mRR(t,Ht,j(t))~g(RRj{1,RRj{2,:::,RRj{h), where RRj{k denotes the previous kth R–R interval occurred prior to the present time t. Introduction to the Instantaneous Dominant Lyapunov Exponent Nonlinear Modeling of History Dependence The general Nonlinear Autoregressive (NAR) formulation can be written as: y(n)~F(y(n{1),y(n{2),:::,y(n{M))zE(n) ð2Þ ð2Þ where E(n) are independent, identically distributed (i.i.d.) Gaussian random variables. The expected value of y(n) can be written as a Taylor expansion under the hypothesis of being infinitely differentiable in a neighborhood of the working point: E½y(n)~c0 z X M i~1 c1(i)y(n{i) z X ? K~2 X M i1~1 . . . X M iK ~1 cK(i1, . . . ,iK)P K j~1y(n{ij) ð3Þ Introduction to the Instantaneous Dominant Lyapunov Exponent p In this work, the IG mean is modeled as a third-order nonlinear function of the past RR intervals. In order to perform an effective parameter estimation and retain all the historical information of events, the cubic NAR kernels of the Wiener-Volterra series are expanded using the Laguerre bases [43] leading to the definition of cubic Nonlinear Autoregressive Laguerre (NARL) model. Of note, the NARL definition includes an infinite regression of the past events with a parsimonious use of model parameters. From the NARL point-process model and Fast Orthogonal Search algo- rithm, we are able to estimate the complete LE spectrum at each moment in time, providing novel information concerning the complexity dynamics and its variability. Of note, to the best of our knowledge, complexity variability measures have never been estimated from instantaneous indices of complexity and could open novel perspectives on the assessment of discrete stochastic physiological systems. We present two applications on synthetic datasets (the He´non map and Ro¨ssler attractor) and two experimental applications portraying the crucial role of the instantaneous Lyapunov Exponents in assessing autonomic changes in humans (ten heathy subjects undergoing postural changes, and fourteen patients with severe heart failure), focusing our attention on the Instantaneous Dominant Lyapunov Exponent (IDLE, l), which is the first exponent of the Lyapunov spectrum. Preliminary results of these analyses have been presented in [44– 46]. We start with a detailed, exhaustive presentation of our methodological framework through the following ‘‘Materials and Methods’’ section. Materials and Methods Point-Process Models of Heartbeat Dynamics Point-Process Models of Heartbeat Dynamics ð3Þ y A random point process is a stochastic process whose elements are point patterns specified as a locally finite counting measure [47]. Considering the R-waves detected from the Electrocardio- gram (ECG) as such events, point process theory can be used to characterize their probability of occurrence [41,42,48]. Mathe- matically, in the time domain, a simple 1-dimension point process consists of series of timestamps marking the occurrence times t [ ½0,?) of the random events. Given a set of R-wave events fujgJ j~1, let RRj~uj{uj{1w0 denote the jth R–R interval, or equivalently, the waiting time until the next R-wave event. Assuming history dependence, the probability distribution of the waiting time t{uj until the next R-wave event, where uj denotes the previous R-wave event occurred before time t, follows an inverse Gaussian (IG) model: Due to the autoregressive structure of eq. 3, the system can be identified with only exact knowledge of the output data and with only few assumptions on the input data. In practice, this series will obviously be truncated at some small, finite value. Here, we represent the nonlinear cardiovascular system by taking into account up to the cubic nonlinear terms, i.e. c0, c1(i), c2(i,j), and c3(i,j,k). Thus, the model can be written as: August 2014 \| Volume 9 \| Issue 8 \| e105622 PLOS ONE \| www.plosone.org 2 Estimating the Instantaneous Lyapunov Spectra Estimating the Instantaneous Lyapunov Spectra E½y(n)~c0 z X M i~1 c1(i)y(n{i) z X M i~1 X M j~1 c2(i,j)y(n{i)y(n{j)z X M i~1 X M j~1 X M k~1 c3(i,j,k)y(n{i)y(n{j)y(n{k) ð4Þ c2(i,j)~ X Q r~0 X Q s~0 g2(r,s)wr(i)ws (j) if ð11Þ ð11Þ i=jho dimenticato perche avevamo qst ð4Þ c3(i,j,k)~ X K r~0 X K s~0 X K t~0 g3(r,s,t)wr(i)ws(j)wt(k) if i=j : ð12Þ ð12Þ where the quadratic and the cubic terms, c2(i,j) and c3(i,j,k), are assumed to be permutation invariant. This choice of a third order NAR system further gives robustness against the presence of measurement noise in the data [38]. Here g0, g1(r), g3(r,s) and g3(r,s,t) are the Laguerre coefficients. Using eq. 5 and eqs. 9–12, the model in eq. 4 for the instantaneous RR mean becomes: Laguerre Expansion of the cK terms. An important practical limitation in modeling high-order nonlinearities using the model in eq. 4 is the high number of parameters needed to sufficiently fit the observed data. Point-Process Models of Heartbeat Dynamics An advocated approach to solve such a limitation has been proposed by using the Laguerre functions [43,50,51]. Let us define the jth-order discrete time orthonormal Laguerre function: mRR(t,Ht,j(t))~g0(t)z X P i~0 g1(i,t)li(t)z X Q i~0 X Q j~0 g2(i,j,t) li(t) lj(t)z X K i~0 X K j~0 X K k~0 g3(i,j,k,t) li(t) lj(t) lk(t) : ð13Þ mRR(t,Ht,j(t))~g0(t)z X P i~0 g1(i,t)li(t)z X Q i~0 X Q j~0 g2(i,j,t) li(t) lj(t)z X K i~0 X K j~0 X K k~0 g3(i,j,k,t) li(t) lj(t) lk(t) : ð13Þ X Q i~0 X Q j~0 g2(i,j,t) li(t) lj(t)z ð13Þ ð13Þ X K i~0 X K j~0 X K k~0 g3(i,j,k,t) li(t) lj(t) lk(t) : wj(n)~a n{j 2 (1{a) 1 2 X j i~0 ({1)i n i j i aj{i(1{a)i, hereinafter called Nonlinear Autoregressive with Laguerre expan- sion (NARL) model. Here, the Laguerre filter outputs are: hereinafter called Nonlinear Autoregressive with Laguerre expan- sion (NARL) model. Here, the Laguerre filter outputs are: where a is the discrete-time Laguerre parameter (0vav1) which determines the rate of exponential asymptotic decline of these functions, and n§0. Given the Laguerre function, wj(n), and the signal, y(n), the jth-order Laguerre filter output is: li(t)~ X ~N(t) n~1 wi(n)(RR ~N(t){n{RR ~N(t){n{1) ð14Þ ð14Þ lj(n)~ X ? i~0 wj(i)y(n{i{1) ð5Þ ð5Þ with ~N(t)~ limt?t{ N(t)~ maxfk : ukvtg as a left continuous function. The coefficients g0,fg1(i)g, fg2(i,j)g, and fg3(i,j,k)g corre- spond to the time-varying zero-, first-, second-, and third-order NARL coefficients, respectively. When a~0 the filter output becomes lj(n)~({1)jy(n{j{1) and the NARL model corre- sponds, apart for the sign, to the finite NAR model in eq. 4, whereas for a=0 the instantaneous RR mean in terms of NAR equations is theoretically defined as follows: The computation of the Laguerre Filter output can be simplified by using the following recursive relation [43]: l0(n)~ ﬃﬃﬃa p l0(n{1)z ﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ 1{a p y(n{1) ð6Þ ð6Þ lj(n)~ ﬃﬃﬃa p lj(n{1)z ﬃﬃﬃa p lj{1(n)z ð7Þ mRR(t,Ht,j(t))~RR ~N(t){1zc0z X ? i~1 c1(i,t) DRRiz lj(n)~ ﬃﬃﬃa p lj(n{1)z ﬃﬃﬃa p lj{1(n)z ð7Þ mRR(t,Ht,j(t))~RR ~N(t){1zc0z X ? c1( mRR(t,Ht,j(t))~RR ~N(t){1zc0z X ? i~1 c1(i,t) DRRiz lj(n)~ ﬃﬃﬃa p lj(n{1)z ﬃﬃﬃa p lj{1(n)z lj(n)~ ﬃﬃﬃa p lj(n{1)z ﬃﬃﬃa p lj{1(n)z ð7Þ ð7Þ ﬃﬃﬃa p lj{1(n{1), j§1 ð8Þ X ? i~1 X ? j~1 c2(i,j,t) DRRi DRRjz z X ? i~1 X ? j~1 X ? g3(0,0,0,t),:::,g3(K,K,K,t) : More generally, let us consider an n-dimensional linear system in the form yi~Y(t)pi, where Y(t) is a fundamental solution matrix with Y(0) orthogonal, and fpig is an orthonormal basis of Rn. Then, the sum of the corresponding n Lyapunov Exponents (li) is minimized, and the orthonormal basis fpig is called ‘‘normal’’ [54]. One of the key theoretical tools for determining LEs is the continuous QR factorization: Y(t)~Q (t) R(t) [55,56] where Q(t) is orthogonal and R(t) is upper triangular with positive diagonal elements Rii, i~1 : n. Therefore we obtain [54–56]: Parameter Estimation and Model Goodness-of-fit measures. A local maximum likelihood method [41] is used to estimate the time-varying parameter set j(t). Given a local observation interval (t{l,t of duration l, we consider a subset Um:n of the R-wave events, where m~ minfk : ukwt{lg and n~ maxfk : ukƒtg. At each time t, we find the parameter vector j(t) that maximizes the local log-likelihood, given the R-wave events recorded in the local observation interval: L(j(t) j Um:n)~ X n{1 k~mzP{1 w(t{ukz1) log f ukz1 j mRR(ukz1,Hukz1,j(t)),h(j(t)) h i ð16Þ li ~ lim t?? 1 t log EY(t)piE ~ lim t?? 1 t log ER(t)piE ~ lim t?? 1 t log ERii(t)E, 1 ƒ i ƒ n ð19Þ ð16Þ ð19Þ where w(t)~ exp (^wt) with ^w~0:02 s{1, is an exponential weighting function for the local likelihood. This value has been empirically chosen by considering a range of discrete values (^w~f0:005,0:01,0:015,0:02,0:025,0:030g), and by choosing the optimum according to KS plots goodness-of-fit analysis, as described in [41]. We use a Newton-Raphson procedure to maximize the local log likelihood in eq. 16 and compute the local maximum-likelihood estimate of j(t). Because there is significant overlap between adjacent local likelihood intervals, we start the Newton-Raphson procedure at t with the previous local maximum-likelihood estimate at time t{D in which D define how much the local likelihood time interval is shifted to compute the next parameter update. We determined the optimal orders fP,Q,Kg using the Akaike Information Criterion (AIC) by fitting a subset of the data using local likelihood method [41]) as well as the Kolmogorov-Smirnov (KS) statistic in the post hoc analysis. g3(0,0,0,t),:::,g3(K,K,K,t) : Using the Einstein notation, we obtain: Using the Einstein notation, we obtain: mRR(t,Ht,j(t)) ~ c1(i) DRRizc2(i,j) DRRi DRRj z c3(i,j,k) DRRi DRRjDRRk ð20Þ Therefore, the elements of the Jacobian can be computed as follows: LmRR(t,Ht,j(t)) LDRRp ~c1(i)dpz c2(j,i)DRRidjp z c2(i,j)DRRidipzc3(i,j,k)dipDRRjDRRk z c3(i,j,k)djpDRRiDRRk z c3(i,j,k)dkpDRRiDRRj ð21Þ ð21Þ It is known from point process theory [41] that the Conditional Intensity Function (CIF) b(t) is related to the inter-event probability p(t) with a one-to-one relationship: and considering the symmetry properties of the NAR kernels, we finally obtain: b(t)~ p(t) 1{ Ð t ujp(t)dt ð17Þ ð17Þ LmRR(t,Ht,j(t)) LDRRp ~c1(p)z2 X i c2(i,p)DRRi z 3 X i X j c3(i,j,p)DRRiDRRj ð22Þ The estimated CIF is used to evaluate the goodness-of-fit of the proposed heartbeat interval point process probability model, which is based on the KS test [41]. ð22Þ Point-Process Models of Heartbeat Dynamics k~1 c3(i,j,k,t) DRRi DRRjDRRk ð15Þ ð8Þ ð15Þ Since the fwi(t)g form a complete orthonormal set in functional space L2, we can write [52]: c0~g0 ð9Þ where DRRh~(RR ~N(t){h{RR ~N(t){h{1). The autoregressive model is expressed in terms of the derivative RR series, rather than the original RR series, in order to allow the model to cope with highly non-stationary regimes [53]. where DRRh~(RR ~N(t){h{RR ~N(t){h{1). The autoregressive model is expressed in terms of the derivative RR series, rather than the original RR series, in order to allow the model to cope with highly non-stationary regimes [53]. ð9Þ c0~g0 c1(i)~ X P r~0 g1(r)wr(i) ð10Þ ð10Þ Moreover, as mRR(t,Ht,j(t)) is defined in a continuous-time fashion, we can obtain an instantaneous R–R mean estimate at a very fine timescale (with an arbitrarily small bin size D), which requires no PLOS ONE \| www.plosone.org 3 August 2014 \| PLOS ONE \| www.plosone.org 3 August 2014 \| Volume 9 \| Issue 8 \| e105622 www.plosone.org 3 August 2014 \| Volume 9 \| Is PLOS ONE \| www.plosone.org PLOS ONE \| www.plosone.org August 2014 \| Volume 9 \| Issue 8 \| e105622 Estimating the Instantaneous Lyapunov Spectra interpolation between the arrival times of two beats. Given the proposed parametric model, the nonlinear indices of the HR and HRV will be defined as a time-varying function of the parameters j(t)~½h(t),g0 (t),g1 (0,t),:::,g1 (P,t),g2 (0,0,t),:::,g2 (Q,Q,t), l~ lim sup t?? 1 t log Df (t)D ð Þ ð18Þ ð18Þ g3(0,0,0,t),:::,g3(K,K,K,t) : More in detail, we can compare the AIC scores and choose the parameter setup with the minimum AIC value of AIC~{2Lz2 dim (j) where L is the maximized value of the likelihood function for the estimated model, and dim (j) is the number of parameters in the statistical model. where w(t)~ exp (^wt) with ^w~0:02 s{1, is an exponential weighting function for the local likelihood. This value has been empirically chosen by considering a range of discrete values (^w~f0:005,0:01,0:015,0:02,0:025,0:030g), and by choosing the optimum according to KS plots goodness-of-fit analysis, as described in [41]. We use a Newton-Raphson procedure to maximize the local log likelihood in eq. 16 and compute the local maximum-likelihood estimate of j(t). Because there is significant overlap between adjacent local likelihood intervals, we start the Newton-Raphson procedure at t with the previous local maximum-likelihood estimate at time t{D in which D define how much the local likelihood time interval is shifted to compute the next parameter update. We determined the optimal orders fP,Q,Kg using the Akaike Information Criterion (AIC) by fitting a subset of the data using local likelihood method [41]) as well as the Kolmogorov-Smirnov (KS) statistic in the post hoc analysis. More in detail, we can compare the AIC scores and choose the parameter setup with the minimum AIC value of AIC~{2Lz2 dim (j) where L is the maximized value of the likelihood function for the estimated model, and dim (j) is the number of parameters in the statistical model. In our case, the matrix Y(t) corresponds to the Jacobian of the M-dimensional system of the NARL model parameters, where M is the value of the largest model order [39]. Therefore, given the NARL model reported in eq. 13 and using eqs. 9 to 12 bringing back to the NAR framework, it is possible to obtain an M- dimensional state space canonical representation. Standard and Nonlinear Measures of Heartbeat Dynamics J(n)Q(n{1)~Q(n)R(n) with n~1,2:::,N: In order to perform a comparison analysis with standard and nonlinear estimates of heartbeat dynamics, we also calculated the standard mean of the RR intervals (Mean RR), the root mean square of successive differences of intervals (RMSSD) and the number of successive differences of intervals which differ by more The matrix Y(t) corresponds to the Jacobian of this system [39]: The matrix Y(t) corresponds to the Jacobian of this system [39]: J(n)~ 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 .. . .. . .. . .. . P .. . 0 0 0 0 1 LmRR(t,Ht,j(t)) LDRR1 LmRR(t,Ht,j(t)) LDRR2 LmRR(t,Ht,j(t)) LDRR3 LmRR(t,Ht,j(t)) LDRR4 LmRR(t,Ht,j(t)) LDRRM 0 B B B B B B B @ 1 C C C C C C C A : J(n)~ 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 .. . .. . .. . .. . P .. . 0 0 0 0 1 LmRR(t,Ht,j(t)) LDRR1 LmRR(t,Ht,j(t)) LDRR2 LmRR(t,Ht,j(t)) LDRR3 LmRR(t,Ht,j(t)) LDRR4 LmRR(t,Ht,j(t)) LDRRM 0 B B B B B B B @ 1 C C C C C C C A : than 50 ms (pNN50% expressed as a percentage of the total number of heartbeats analyzed) [57]. Referring to morphological patterns of HRV, the triangular index is obtained as the integral of the histogram (i.e. total number of RR intervals) divided by the height of the histogram which depends on the selected bin width [57]. Moreover, we performed the estimation of the dominant Lyapunov exponent according to the algorithm described by Wolf et al. [34] (LD1) and Rosenstein et al. [58] (LD2). Both algorithms are suitably applied to experimental noisy data. Finally, other nonlinear measures such as the approximate entropy (ApEn) [59], Sample Entropy [60], and the Detrended Fluctuation Analysis (DFA) [61] were evaluated. than 50 ms (pNN50% expressed as a percentage of the total number of heartbeats analyzed) [57]. Referring to morphological patterns of HRV, the triangular index is obtained as the integral of the histogram (i.e. total number of RR intervals) divided by the height of the histogram which depends on the selected bin width [57]. Instantaneous Lyapunov Exponents Estimation Considering N data samples, we evaluate the Jacobian over the time series, and determine the LE by means of the QR decomposition: The Lyapunov Exponent (LE) of a real valued function f (t) defined for tw0 is: PLOS ONE \| www.plosone.org August 2014 \| Volume 9 \| Issue 8 \| e105622 August 2014 \| Volume 9 \| Issue 8 \| e105622 4 Estimating the Instantaneous Lyapunov Spectra Estimating the Instantaneous Lyapunov Spectra Estimating the Instantaneous Lyapunov Spectra stochastic series through the IDLE index, we simulated a modified version of the well-known chaotic He´non Map as suggested in [39]. Such a complex system, in which stochastic terms are also considered, is governed by the following differential equations: Synthetic Data Before reporting the implementation of the synthetic datasets, it is important to clarify some important differences from standard definitions that are introduced by our methodology. In fact, standard nonlinear systems such as the He´non Map and Ro¨ssler Attractor are intrinsically defined in the continuos time domain, whereas our methodology deals with stochastic point processes which are a sequence of events. Moreover, additive noise terms have to be considered as well. Therefore, starting from the canonical equations of each nonlinear system, we slightly modify the system equations by adding a noise term. Then, the output of the system is taken as an input of an integrate-and-fire system. The output of such an integrate-and-fire system constitutes the series modeled by the proposed cubic NARL model within the point- process framework. ynz1 ~ bxn z kE(t); xnz1 ~ 1 { ax2 n z yn z kE(t); ð24Þ ð24Þ The time series yn were taken into account fixing b~0:3. The term E(t) is an independent identically distributed Gaussian random variable with zero mean and standard deviation of 1, which is modulated by the coefficient k. The coefficient a is taken as a time-varying variable from a~1 to a~1:09 with step size 0:03. Note that the value a~1:07, in a purely deterministic domain, corresponds to the transition between the non-chaotic and chaotic regime. A total of 100 realizations the He´non Map series were simulated, each of which was comprised of 4000 data He´non Map. In order to test the efficiency of the proposed cubic NARL model in tracking the complexity of a synthetic Figure 2. Simulation results using the Ro¨ ssler equations. (Top) Synthetic interval series from the Ro¨ssler system, (Middle) the a stepping, an (Bottom) IDLE estimates using NARL model. The dotted vertical line indicates the transition between the non-chaotic and chaotic regime. doi:10.1371/journal.pone.0105622.g002 Figure 2. Simulation results using the Ro¨ ssler equations. (Top) Synthetic interval series from the Ro¨ssler system, (Middle) the a steppin Bottom) IDLE estimates using NARL model. The dotted vertical line indicates the transition between the non-chaotic and chaotic regime. doi:10.1371/journal.pone.0105622.g002 Figure 2. Simulation results using the Ro¨ ssler equations. (Top) Synthetic interval series from the Ro¨ssler system, (Middle) the a stepping, and (Bottom) IDLE estimates using NARL model. The dotted vertical line indicates the transition between the non-chaotic and chaotic regime. doi:10.1371/journal.pone.0105622.g002 Figure 2. Simulation results using the Ro¨ ssler equations. Standard and Nonlinear Measures of Heartbeat Dynamics Moreover, we performed the estimation of the dominant Lyapunov exponent according to the algorithm described by Wolf et al. [34] (LD1) and Rosenstein et al. [58] (LD2). Both algorithms are suitably applied to experimental noisy data. Finally, other nonlinear measures such as the approximate entropy (ApEn) [59], Sample Entropy [60], and the Detrended Fluctuation Analysis (DFA) [61] were evaluated. This decomposition is unique except in the case of zero diagonal elements. Then, the LEs li are given by This decomposition is unique except in the case of zero diagonal elements. Then, the LEs li are given by li~ 1 tN X N{1 j~0 ln R(j)ii ð23Þ ð23Þ where t is the sampling time step. The estimation of the LEs is performed at each time t from the corresponding time-varying vector of parameters, j(t). We define the first LE (l1(t)) as the instantaneous dominant Lyapunov exponent (IDLE). In particu- lar, the median IDLE (lRR) and its median absolute deviation (slRR) were considered as group features. Figure 1. Simulation results using the He´non equations. (Top) Synthetic interval series from the He´non map, (Middle) the a stepping, and (Bottom) IDLE averaged estimate from 100 realizations using NARL model with noise level of k~0:001. doi:10.1371/journal.pone.0105622.g001 Figure 1. Simulation results using the He´non equations. (Top) Synthetic interval series from the He´non map, (Middle) the a stepping, and (Bottom) IDLE averaged estimate from 100 realizations using NARL model with noise level of k~0:001. doi:10.1371/journal.pone.0105622.g001 August 2014 \| Volume 9 \| Issue 8 \| e105622 PLOS ONE \| www.plosone.org 5 Estimating the Instantaneous Lyapunov Spectra Estimating the Instantaneous Lyapunov Spectra points with 1000 samples for each of the four a-values. A realization of the simulated time series is illustrated in Fig. 1 along with the a-values and the corresponding IDLE results. data points were generated with 15000 samples for each of the five a-values. The simulated time series is illustrated in Fig. 2 along with the a-values and the corresponding results on the IDLE. The IDLE transitions to positive values reflect the simulated switch to chaotic behavior. Ro¨ssler Attractor. As a further validation, we simulated a modified version of the well-known chaotic Ro¨ssler time series (see previous work [19]). Such a complex system, in which stochastic terms are also considered, is governed by the following differential equations: Synthetic Data (Top) Synthetic interval series from the Ro¨ssler system, (Middle) the a stepping, and (Bottom) IDLE estimates using NARL model. The dotted vertical line indicates the transition between the non-chaotic and chaotic regime. doi:10.1371/journal.pone.0105622.g002 August 2014 \| Volume 9 \| Issue 8 \| e105622 PLOS ONE \| www.plosone.org 6 Experimental Data In order to validate the proposed algorithms performance as related to actual cardiovascular dynamics, we have considered two experimental datasets. Since the experimental protocols are fully described in [19,41] in this paragraph we provide only briefly descriptions of the two datasets. dx dt ~ {z { y z kE(t); dy dt ~ xzayzkE(t); dz dt ~ b z z(x{c) z kE(t); ð25Þ ð25Þ Postural Changes. In order to validate the proposed algorithms performance as related to actual cardiovascular dynamics, we studied the complexity of RR interval series recorded from 10 healthy subjects for the study of cardiovascular and autonomic regulation undergoing a tilt-table protocol. This choice is motivated by the high presence of nonlinearity and non- stationarity in such time series. In this study, a subject lying horizontally in a supine position is then actively or passively tilted to the vertical position. The rest (supine) and upright sessions alternated six times with three possible modality of transition: active stand-up, slow and fast passive tilt. A single-lead ECG was The time series were implemented with sampling time of 0.01 using the Runge-Kutta integration and fixing b~2, and c~4. The term E(t) is an independent identically distributed Gaussian random variable with zero mean and standard deviation of 0.01. The coefficient a is taken as a time-varying variable from a~0:35 to a~0:45 with step size 0:025. Note that the value a~0:432, in a purely deterministic domain, corresponds to the transition between the non-chaotic and chaotic regime. A total of 75000 Figure 3. Box plots of standard and proposed Lyapunov estimations performed on the He´non map among the a2stepping with noise level of k = 0.001. doi:10.1371/journal.pone.0105622.g003 Figure 3. Box plots of standard and proposed Lyapunov estimations performed on the He´non map among the a2stepping with noise level of k = 0.001. doi:10.1371/journal.pone.0105622.g003 August 2014 \| Volume 9 \| Issue 8 \| e105622 PLOS ONE \| www.plosone.org 7 Estimating the Instantaneous Lyapunov Spectra Table 1. IDLE Results from the He´non map Synthetic Dataset. Experimental Data continuously recorded for each subject during the study, and the RR intervals were extracted using a curve length-based QRS detection algorithm [62]. Further details on the experimental protocol can be found in [19,41]. The study was conducted at the Massachusetts Institute of Technology (MIT) General Clinical Research Center (GCRC) and was approved by the MIT Institutional Review Board and the GCRC Scientific Advisory Committee. Patient records/information was anonymized and de- identified prior to analysis. Congestive Heart Failure. The second heartbeat dataset was constituted from data gathered from Congestive Heart Failure (CHF) and reference healthy subjects on a public source: Physionet (http://www.physionet.org/) [63]. The RR time series were recorded from 14 CHF patients (from BIDMC{CHF Database) as well as 16 healthy subjects (from MIT{BIH Normal Sinus Rhythm Database). Each RR time series was artifact-free (upon human’s visual inspection and artifact rejection) and lasted about 50 min (small segments of the original over longer recordings). These recordings have been taken as landmark for studying complex heartbeat interval dynamics [3,8]. Considering the noise level k~0:001, the Kruskal-Wallis test reveals significant differences (pv10{12) for both the standard Lyapunov estimates and the proposed lRR and complexity variability index slRR. In this case, the Dunn test for multiple comparison, which considers a Tukey-Kramer correction, shows that each group of standard estimates of LD1 and LD2 having coherent a parameter are different with all the other groups (pv10{3), whereas coherent a-values of lRR are different with all the other groups (pv10{3) except for a~f1{1:03g (pw0:05). Considering the noise level k~0:01, the Kruskal-Wallis test reveals significant differences (p,51024) only for the proposed lRR and complexity variability index slRR. In this case, the Dunn test for multiple comparison, which considers a Tukey-Kramer correction, shows that coherent a~1 values are different with all the other groups (pv10{3), with a~f1:03,1:06,1:09g equal between each other (pw0:05). For noise level k~0:1 and k~1, Considering the noise level k~0:001, the Kruskal-Wallis test reveals significant differences (pv10{12) for both the standard Lyapunov estimates and the proposed lRR and complexity variability index slRR. In this case, the Dunn test for multiple comparison, which considers a Tukey-Kramer correction, shows that each group of standard estimates of LD1 and LD2 having coherent a parameter are different with all the other groups (pv10{3), whereas coherent a-values of lRR are different with all the other groups (pv10{3) except for a~f1{1:03g (pw0:05). Experimental Data a-values 1 1.03 1.06 1.09 Noise k = 0.001 p-value LD1 0.1424+0.0035 0.1598+0.0041 0.1814+0.0040 0.1960+0.0030 v10{12 LD2 0.0004+0.0002 0.0064+0.0009 0.0377+0.0019 0.1772+0.0030 v10{12 lRR 0.3679+0.0195 0.3530+0.0179 0.4410+0.0268 0.6885+0.0349 v10{12 slRR 0.0407+0.0151 0.0379+0.0112 0.0636+0.0130 0.0747+0.0170 v10{12 Noise k = 0.01 p-value LD1 2.5488+0.0251 2.5552+0.0387 2.5537+0.0280 2.5573+0.0238 w0:05 LD2 0.3038+0.0070 0.3049+0.0077 0.3037+0.0085 0.3033+0.0083 w0:05 lRR 1.2240+0.3232 1.5006+0.4380 1.4822+0.5034 1.4962+0.5237 ,5*1024 slRR 0.4511+0.1559 0.6884+0.2008 0.5780+0.2476 0.7137+0.2830 2v10{5 Noise k = 0.1 p-value LD1 2.6704+0.0265 2.6710+0.0283 2.6756+0.0306 2.6719+0.0350 w0:05 LD2 0.2069+0.0997 0.2161+0.1221 0.2320+0.1334 0.2334+0.1459 w0:05 lRR 0.1180+0.0517 0.1383+0.0565 0.1319+0.0487 0.1566+0.0468 w0:05 slRR 0.1348+0.0376 0.1398+0.0314 0.1360+0.0378 0.1369+0.0247 w0:05 Noise k = 1 p-value LD1 0.2949+0.0033 0.2953+0.0031 0.2949+0.0031 0.2947+0.0023 w0:05 LD2 0.0023+0.0022 0.0015+0.0015 0.0011+0.0011 0.0017+0.0016 w0:05 lRR 20.1032+0.0104 20.1018+0.0107 20.0998+0.0098 20.0993+0.0119 w0:05 slRR 0.0318+0.0056 0.0311+0.0071 0.0337+0.0066 0.0319+0.0071 w0:05 doi:10.1371/journal.pone.0105622.t001 doi:10.1371/journal.pone.0105622.t001 KS plots goodness-of-fit analysis, according to [41]. The simulated time series along with the resulted IDLE series are shown in Fig. 1, whereas the corresponding box plots are shown in Fig. 3 in terms of IDLE median (lRR) and its median absolute deviation slRR. The proposed IDLE is able to track the complexity variation at each moment in time. As a matter of fact, the IDLE goes increasingly high from the non-chaotic behavior to the chaotic one. Remarkably, the non-chaos–chaos transition is instantaneously detected, although a significant oscillatory dynamics is present in the chaotic region. The related IDLE values are reported in Table 1 along with standard estimates of the dominant Lyapunov exponents. A non-parametric statistical analysis has been per- formed in order to quantify the differences between the considered a-values for each of the considered noise level. KS plots goodness-of-fit analysis, according to [41]. The simulated time series along with the resulted IDLE series are shown in Fig. 1, whereas the corresponding box plots are shown in Fig. 3 in terms of IDLE median (lRR) and its median absolute deviation slRR. The proposed IDLE is able to track the complexity variation at each moment in time. As a matter of fact, the IDLE goes increasingly high from the non-chaotic behavior to the chaotic one. Remarkably, the non-chaos–chaos transition is instantaneously detected, although a significant oscillatory dynamics is present in the chaotic region. The related IDLE values are reported in Table 1 along with standard estimates of the dominant Lyapunov exponents. A non-parametric statistical analysis has been per- formed in order to quantify the differences between the considered a-values for each of the considered noise level. Estimating the Instantaneous Lyapunov Spectra Table 2. Median and MAD of IDLE evaluated in the Tilt-Table Experimental Dataset. Subject P-Value KS dist. Rest Stend-Up Rest Slow Tilt Rest Fast Tilt 1 0.0320 0.0458 0.0264+0.0298 0.0108+0.077 0.0304+0.0466 20.0340+0.0514 0.0518+0.0227 20.1165+0.0326 2 0.0340 0.0603 0.1469+0.1712 0.0120+0.1794 0.2534+0.1007 0.0050+0.1372 0.2226+0.0988 0.0075+0.0515 3 ve{8 0.0355 0.0435+0.0552 0.0207+0.0232 0.0439+0.0745 20.0200+0.0423 20.0222+0.0662 20.0313+0.0574 4 0.0300 0.0227 20.0551+0.0150 20.0550+0.0185 0.0557+0.0679 20.0585+0.0315 0.0649+0.0785 0.0084+0.0365 5 0.0220 0.0451 20.0719+0.0613 20.0578+0.0547 20.0045+0.0441 20.0403+0.0355 20.0137+0.0595 20.0388+0.0226 6 0.0020 0.0409 0.0520+0.0894 20.0665+0.0368 20.0032+0.0692 20.0734+0.0442 20.0007+0.0786 20.0362+0.0303 7 0.0020 0.0458 0.0339+0.0448 20.0346+0.0589 0.0278+0.0343 20.0665+0.0341 0.0969+0.0406 20.0258+0.1049 8 0.0760 0.0408 20.0352+0.0467 20.0103+0.0668 20.0159+0.0571 20.0783+0.0405 0.0093+0.0518 20.0612+0.0375 9 ve{6 0.0571 20.0079+0.048 20.0033+0.0703 20.0217+0.0365 0.0216+0.0391 0.0058+0.0394 20.0042+0.0145 10 ve{8 0.0572 0.6521+0.4771 0.0324+0.0830 0.2316+0.1737 20.0052+0.0650 0.2662+0.1708 0.0384+0.1653 the difference between the a-values are not revealed by standard and proposed dominant Lyapunov estimates. Ro¨ssler System. We performed the IDLE estimation by fitting the NARL model on the x series from the modified stochastic Ro¨ssler time series (see eq. 25). The model orders were set as P~3, Q~1, K~1, and a~0:2 were chosen by preliminary KS plots goodness-of-fit analysis, according to [41]. The simulated time series along with the resulted IDLE series are shown in Fig. 2. Clearly, the proposed IDLE is able to track the complexity variation at each moment in time. As a matter of fact, the IDLE goes increasingly high from the non-chaotic behavior to the chaotic one. Remarkably, the non-chaos–chaos transition is instantaneously detected, although a significant oscillatory dynam- ics is present in the chaotic region. Intervals expressed as median + M.A.D. are as follows: {0:0973+0:0044 for a~0:350, {0:0813 + 0:0065 for a~0:375, {0:0850 + 0:0077 for a~0:400, {0:0375 + 0:0075 for a~0:425, 0:0033 + 0:0092 for a~0:450. A non-parametric statistical analysis has been performed in order to quantify the differences between the considered a-values. The Kruskal-Wallis test reveals significant differences (pv10{6) and the Dunn test for multiple comparison, which considers a Tukey-Kramer correction, shows that each IDLE group having coherent a-values is different with all the other group (pv10{5) except for a~f0:375{0:4g (pw0:05). Results Instantaneous Complex Dynamics on Synthetic Data W f d h IDLE i i b fi i Instantaneous Complex Dynamics on Synthetic Data He´non Map. We performed the IDLE estimation by fitting the NARL model on y series from the modified stochastic He´non Map time series (see eq. 24). The series were generated a hundred times for each of the four considered noise levels k~f 0:001, 0:01, 0:1, 1g. For kw0:001, a further constrain of xn~0:1E(t) was imposed as xnv~{0:1 or xnw~{0:5 in order to prevent the system to become unstable. The model orders were set as P~3, Q~1, K~1, and a~0:4 were chosen by preliminary August 2014 \| Volume 9 \| Issue 8 \| e105622 PLOS ONE \| www.plosone.org 8 Instantaneous Complex Dynamics on Postural Changes Instantaneous Complex Dynamics on Postural Changes Before estimating the IDLE from the experimental datasets, we first considered a specific time-domain method [64] for testing the presence of nonlinearity in the heartbeat intervals. The null hypothesis of the test states that the given time series is linear. In the considered recordings, we restricted the test to short-term dependence by setting the number of laps M~8, and a total of 500 bootstrap replications. Concerning the RR series gathered during postural changes, the nonlinearity test shows that the level of nonlinearity of the considered RR intervals is statistically significant for all the considered subjects but one (see Table 2. As also shown in Table 2, the NARL modeling always gives a good model fit, with KS distance v0:0604 in all cases. Specifically, concerning the three experimental sessions, i.e. stand-up, slow-tilt, and fast-tilt, a decrease of the IDLE with respect to the relative rest condition is shown in 25 out of 30 epochs. In particular, in the fast- tilt condition the decrease happens for all subjects, and is more significant than stand-up and slow-tilt. Group statistics of standard and proposed instantaneous measures are shown in Table 3, whose inter-subject analysis was performed using a non-paramet- ric rank-sum test. Results on the proposed IDLE show a non- significant statistical difference between the stand-up epochs and their relative rest epochs (pw0:05) and a significant difference for the slow-tilt epochs (pv0:05). The highest significance was found comparing the fast-tilt epochs with their relative rest (pv0:001). These trends are confirmed by the standard DLE estimation according to the Rosenstein et al. [58] technique, whereas the one suggested by Wolf et al. [34] did not show such significant differences. IDLE dynamics for one representative subject are shown in Fig. 4, whereas the averaged IDLEs for all 10 subjects are shown in Fig. 5, providing a clear portrayal of how different postural stimuli elicit different changes in the dynamic signatures of complexity. Concerning other standard and instantaneous indices, we report significant differences on three session on mRR, ApEn, and SampEn, whereas RMSSD and pNN50% showed significant differences during the slow and fast tilt sessions. P-values are obtained from the nonlinearity test. doi:10.1371/journal.pone.0105622.t002 Using this dataset, we further evaluate the effect of the Laguerre parameter a on the IDLE estimates. August 2014 \| Volume 9 \| Issue 8 \| e105622 August 2014 \| Volume 9 \| Issue 8 \| e105622 Instantaneous Complex Dynamics on Postural Changes Tracking for values a~f0:1,0:2,0:3,:::,0:8g from a representative subject undergoing August 2014 \| Volume 9 \| Issue 8 \| e105622 August 2014 \| Volume 9 \| Issue 8 \| e105622 9 Estimating the Instantaneous Lyapunov Spectra Table 3. Group Statistics of Standard and Instantaneous Heartbeat Dynamics Measures from the Tilt-Table Experimental Dataset. Feature Rest Stand-up p-value Rest Slow Tilt p-value Rest Fast Tilt p-value Standard and Instantaneous Time Domain Measures of HRV Mean RR (ms) 910.94+123.08 781.92+55.96 v0:003 871.86+74.27 772.82+46.10 v0:0002 860.50+80.47 774.66 +44.34 v0:0004 RMSSD 0.0279+0.0123 0.0202+0.0055 w0:05 0.0325 +0.0127 0.0202 +0.0048 v0:03 0.0296+0.0127 0.0200+0.0042 v0:03 pNN50% 6.4335+6.4335 2.3472+2.3472 w0:05 10.0744 +9.4524 1.6860+1.5708 v0:05 7.6677+7.2816 2.0339+2.0339 v0:05 HRV Triangular Index 4.1644+0.6832 3.9065+0.4754 w0:05 3.3012+0.5142 3.5347+0.3497 w0:05 3.6429+0.9743 4.3333+0.6352 w0:05 mRR(ms) 915.10+122.16 769.77+78.22 v0:02 879.60+74.80 773.23+62.29 v0:001 890.17+95.18 776.92+57.40 v0:005 s2 RR (ms2) 394.15+319.26 233.09+139.09 w0:05 435.36+237.42 208.07+112.84 v0:03 440.76+302.71 220.71+108.79 w0:05 Standard and Instantaneous Nonlinear Measures of HRV ApEn 1.122+0.055 0.944+0.079 v10{3 1.167+0.091 0.927+0.125 v10{3 1.087+0.116 0.964+0.072 v0:004 SampEn 1.501+0.192 1.243+0.245 v0:025 1.495+0.173 0.900+0.247 v10{3 1.320+0.247 1.197+0.233 w0:05 DFA-a1 0.9806+0.1039 0.9968+0.1624 w0:05 0.9892+0.0952 1.2128+0.1126 w0:05 0.9788+0.0990 1.0137+0.1425 w0:05 LD1 0.0128+0.0014 0.0125+0.0013 w0:05 0.0133+0.0017 0.0135+0.0015 w0:05 0.0137+0.0013 0.0128+0.0008 w0:05 LD2 0.0029+0.0005 0.0028+0.0005 w0:05 0.0037+0.0006 0.0024+0.0004 v0:005 0.0033+0.0006 0.0020+0.0007 v0:001 lRR 20.0128+0.0480 20.0390+0.0330 w0:05 0.0205+0.0372 20.0491+0.0236 v0:0005 20.0022+0.0332 20.0404 +0.0232 v0:001 slRR 0.0585 +0.0213 0.0652+0.0178 w0:05 0.0693+0.0098 0.0582+0.0113 w0:05 0.0617+0.0131 0.0500+0.0144 w0:05 August 2014 \| Volume 9 \| Issue 8 \| e105622 10 Figure 4. Instantaneous heartbeat statistics computed from a representative subject (subject 1) from the tilt-table protocol. In the top panel, the estimated mRR(t) is superimposed on the recorded RR series. In the bottom panel, the instantaneous averaged IDLE is superimposed on the original IDLE. doi:10.1371/journal.pone.0105622.g004 Estimating the Instantaneous Lyapunov Spectra Estimating the Instantaneous Lyapunov Spectra Figure 4. Instantaneous heartbeat statistics computed from a representative subject (subject 1) from the tilt-table protocol. In the top panel, the estimated mRR(t) is superimposed on the recorded RR series. In the bottom panel, the instantaneous averaged IDLE is superimposed on the original IDLE. doi:10.1371/journal.pone.0105622.g004 postural changes are shown in Fig. 6. Indeed, the IDLE estimates are affected by the choice of the Laguerre parameter a. However, such a variability is significantly less than the variability of the IDLE dynamics within session. As a matter of fact, quantitative results reported in Table 4 show that these differences are associated to a p-value less than 2e{4 for each experimental session. August 2014 \| Volume 9 \| Issue 8 \| e105622 Instantaneous Complex Dynamics on Postural Changes Instantaneous Complex Dynamics on CHF patients The results of the second experimental dataset (on CHF) are shown in Table 5. According to the nonlinearity test, 15 out of 16 RR time series from the healthy subjects showed significant nonlinearity (pv0:05), whereas in the CHF group, 6 out of 14 RR time series failed to reach significance (pw0:05). The fact that a lower degree of nonlinearity was found in the CHF patients suggests that pathological conditions might reduce the nonlinearity in the heartbeat interval series, which is also consistent with previous finding that a healthy heartbeat presents more pro- nounced nonlinear dynamics [3,6,8,65]. Table 5 also demon- strates that the NARL model well fit both pathological and healthy heartbeat series with KS distance v0:082 in all cases. Results averaged among groups are reported in Table 6. We report that standard and instantaneous time domain measures are able to discern the two groups with high statistical significance (p v 5e{4). On the other hand, comparing the standard and proposed instantaneous complexity measures, only the DFA-a2 and the complexity variability slRR are able to provide significant Finally, we report further results on the nonlinearity test separately performed for each of the experimental session, instead of the whole recordings (see Table 2). As a result, under the null hypothesis of linearity according to the time-domain method described in [64] for testing the presence of nonlinearity in the heartbeat intervals, the 49.12% of the resting state (28/57 sessions) were associated to a significant p-value v0:05, along with the 44.4% of the stand-up (8/18 sessions) protocol, the 20% of the slow-tilt (4/20 sessions) protocol, the 10.53% of the fast-tilt (2/19 sessions) protocol. PLOS ONE \| www.plosone.org August 2014 \| Volume 9 \| Issue 8 \| e105622 11 Estimating the Instantaneous Lyapunov Spectra Figure 5. IDLE dynamics averaged for all 10 subjects. The vertical red lines indicate the transition from the supine to the upright position after stand up (top panel), after slow tilt (middle panel), and after fast tilt (bottom panel). doi:10.1371/journal.pone.0105622.g005 Estimating the Instantaneous Lyapunov Spectra Figure 5. IDLE dynamics averaged for all 10 subjects. The vertical red lines indicate the transition from the supine to the upright position after stand up (top panel), after slow tilt (middle panel), and after fast tilt (bottom panel). doi:10.1371/journal.pone.0105622.g005 Figure 5. IDLE dynamics averaged for all 10 subjects. Instantaneous Complex Dynamics on Postural Changes The vertical red lines indicate the transition from the supine to the upright position after stand up (top panel), after slow tilt (middle panel), and after fast tilt (bottom panel). doi:10.1371/journal.pone.0105622.g005 discrimination capability between the two populations with pv0:05. discrimination capability between the two populations with pv0:05. The effective procedure for the time-varying parameter identification is ensured by the combined use of the discrete-time Laguerre expansions for the Wiener-Volterra terms and local maximum likelihood method. In particular, expanding the Volterra terms with the orthonormal Laguerre bases requires a reduced number of parameters to retain the information of all the past events. The nonlinear regression is further performed on the derivative series to better account for nonstationarity [53]. Importantly, unlike other methods that might require large sample size, our method is potentially useful to perform complexity measures in short recordings of the signals of interest. August 2014 \| Volume 9 \| Issue 8 \| e105622 Discussion and Conclusion i 10 1371/j l 0105622 t004 version of the He´non Map and Ro¨ssler system alongside the obtained IDLE results need to be discussed. We are aware that in purely deterministic Ro¨ssler equations the first LE should be zero in the non-chaotic region, whereas it should be increasingly positive in the chaotic region. However, the IDLE results show slightly negative values in the non-chaotic region. Such a behavior may be ascribed to the stochastic input and to the additional integrate-and-fire step which affect the estimation of all complexity measures, including LEs. To this extent, in order to further investigate the effect of noise, results from the He´non Map equations were gathered as a function of the noise level. We demonstrate that, for small amount of noise (k~0:001), standard and instantaneous estimates of the dominant Lyapunov exponent achieve similar results. However, considering He´non Map dynamics with k~0:01, the IDLE is exclusively able to discern the different behavior of the nonlinear system. Of note, for k§0:1, process observations (e.g. RR intervals), which already produced important nonlinear quantifiers for autonomic assessment, based on second- and third-order statistics (instantaneous spectrum and bispectrum) [19]. Most other nonlinearity indices are derived from non-parametric models, whereas our model is purely parametric and the analytically derived indices can be evaluated in a dynamic and instantaneous fashion. We believe these strengths enable our method as a useful tool for assessing nonlinear dynamics of heartbeat intervals in a non-stationary environment. Discussion and Conclusion Novelties and Impact of the proposed Methodology We presented a novel methodology able to instantaneously characterize the complex nonlinear dynamics of a stochastic series of events by using the LEs. The proposed approach relies on the previous literature for the LEs mathematical definition [38,39] and is embedded in a novel IG-based point-process nonlinear framework defined through a third-order Wiener-Volterra repre- sentation, thus advancing on the previous models [19]. As a consequence, the novel instantaneous LEs definition is able to provide a reliable complexity measure tool to examine the unevenly spaced events at very high temporal resolutions, without resorting to any interpolation method. Moreover, goodness of fit measures such as KS distance and autocorrelation plots quanti- tatively allow to verify the model fit as well as to choose the proper model order, which represents another open issue of current parametric approaches. Importantly, the proposed measures also allows for the study of the complexity variability, i.e., the analysis of complex systems referring to the fluctuations in complexity instead of analysis of central tendency. Within the proposed framework, it is always possible to incorporate physiological covariates (such as respiration or blood pressure measures) and produce further instantaneous indices from their dynamic cross spectrum and cross bispectrum [48]. Unlike other paradigms for estimating nonlinearity indices developed in the literature [7,59,66,67], our method is formulated within a probabilistic framework specifically developed for point- August 2014 \| Volume 9 \| Issue 8 \| e105622 PLOS ONE \| www.plosone.org 12 Estimating the Instantaneous Lyapunov Spectra Table 4. IDLE Variability evaluated through a~f0:1,0:2,0:3,:::,0:8g and within each session of the post Table 4. IDLE Variability evaluated through a~f0:1,0:2,0:3,:::,0:8g and within each session of the postural changes protocol. Table 4. IDLE Variability evaluated through a~f0:1,0:2,0:3,:::,0:8g and within each session of the postural changes protocol. Session Through a-values Within Session p-Value Resting State 0.0117+0.0044 0.0585+0.0213 v5e{6 Stand-Up 0.0125+0.0082 0.0652+0.0178 2e{5 Resting State 0.0170+0.0240 0.0693+0.0098 7e{5 Slow Tilt 0.0119+0.0128 0.0582+0.0113 v2e{6 Resting State 0.0167+0.0192 0.0617+0.0131 2e{4 Fast Tilt 0.0119+0.0128 0.0500+0.0144 2e{6 P-values are obtained from the Mann-Whitney test with null hypothesis of equal medians between the two groups. Values are expressed as X~Median(X)+MAD(X). doi:10.1371/journal.pone.0105622.t004 P-values are obtained from the Mann-Whitney test with null hypothesis of equal medians between the two groups. Values are expressed as X~Median(X)+MAD(X). doi:10.1371/journal.pone.0105622.t004 values are obtained from the Mann-Whitney test with null hypothesis of equal medians between the two groups. Values are expressed as X~Median(X)+MAD(X). Study of the Instantaneous Cardiovascular Complex Dynamics doi:10.1371/journal.pone.0105622.t005 P-values are obtained from the nonlinearity test. doi:10.1371/journal.pone.0105622.t005 Lyapunov Exponents was performed and evaluated during postural changes. During the resting condition the cardiovascular and autonomic nervous system are more sensitive to the initial conditions (positive IDLE), whereas a more regular dynamics (negative IDLE values) appear during the tilt phases (see Fig. 4). These results are in agreement with previous findings that complex vagally-driven dynamics are blunted under sympathetic drive [68] and with more recent reports on loss of complexity during states of arousal [20]). Our instantaneous measures also confirm that the instantaneous complexity reflects instantaneous autonomic ner- vous system (ANS) control on the cardiovascular dynamics. We have shown that tracking ANS complexity on healthy subjects undergoing postural changes not only confirms previous results [69,70], but further improves sympathovagal assessment as elicited by different dynamic gravitational stimuli. the noise has an amplitude comparable with the output of the system, thus destroying the different behaviors among the a-values. Lyapunov Exponents was performed and evaluated during postural changes. During the resting condition the cardiovascular and autonomic nervous system are more sensitive to the initial conditions (positive IDLE), whereas a more regular dynamics (negative IDLE values) appear during the tilt phases (see Fig. 4). These results are in agreement with previous findings that complex vagally-driven dynamics are blunted under sympathetic drive [68] and with more recent reports on loss of complexity during states of arousal [20]). Our instantaneous measures also confirm that the instantaneous complexity reflects instantaneous autonomic ner- vous system (ANS) control on the cardiovascular dynamics. We have shown that tracking ANS complexity on healthy subjects undergoing postural changes not only confirms previous results [69,70], but further improves sympathovagal assessment as elicited by different dynamic gravitational stimuli. system, thus destroying the different behaviors among the a values. The use of the Laguerre expansion of the Wiener-Volterra kernels was also investigated through experimental analysis. As the zero-order Laguerre basis is an exponential function, the IDLE estimates present a mild dependence on the a value of the Laguerre functions. Nevertheless, we demonstrated that the actual information needed to characterize the experimental sessions, i.e., the variability within each session, is significantly higher than the variability among all the a values. Anyway, using the hereby proposed approach we clearly demonstrate the ability of the IDLE in tracking the system complexity in an instantaneous fashion. August 2014 \| Volume 9 \| Issue 8 \| e105622 Study of the Instantaneous Cardiovascular Complex Dynamics The novel IDLE index was evaluated in both synthetic and experimental heartbeat series. Estimations on the synthetic dataset were performed on a stochastic version of the well-known chaotic He´non Map and Ro¨ssler attractor. The use of such a modified Figure 6. IDLE dynamics averaged for a~f0:1,0:2,0:3,:::,0:8g from a representative subject undergoing postural changes. Resting state (R) sessions during supine position alternate with upright session after tilt (T). The bold line and gray area indicate the IDLE median and MAD, respectively. Averaged values for all subjects are shown in Table 4. doi:10.1371/journal.pone.0105622.g006 Figure 6. IDLE dynamics averaged for a~f0:1,0:2,0:3,:::,0:8g from a representative subject undergoing postural changes. Resting state (R) sessions during supine position alternate with upright session after tilt (T). The bold line and gray area indicate the IDLE median and MAD, respectively. Averaged values for all subjects are shown in Table 4. doi:10.1371/journal.pone.0105622.g006 August 2014 \| Volume 9 \| Issue 8 \| e105622 PLOS ONE \| www.plosone.org 13 Estimating the Instantaneous Lyapunov Spectra Table 5. Results from the CHF-Healthy Experimental Dataset. Subject Group mRR(ms) p-value KS dist. lRR slRR 01 CHF 995.4+26.4 w0:05 0.0445 0.2268 0.1123 03 CHF 910.25+28.9 w0:05 0.0552 0.2165 0.1257 04 CHF 603.09+22.7 v0:02 0.0456 0.0676 0.0896 05 CHF 655.6+13.3 w0:05 0.0297 20.0114 0.0507 06 CHF 637.4+15.9 v0:001 0.0802 0.0757 0.0659 07 CHF 778.0+7.3 w0:05 0.0363 0.0762 0.0793 08 CHF 800.1+14.1 v0:01 0.0357 20.0965 0.0327 09 CHF 602.5+5.4 w0:05 0.0305 20.0121 0.0555 10 CHF 486.9+6.9 v0:001 0.0329 20.0622 0.0444 11 CHF 685.1+16.0 v0:02 0.0354 20.0014 0.0578 12 CHF 722.8+27.2 v0:001 0.0326 20.0798 0.0380 13 CHF 619.7+5.1 v0:001 0.0386 0.0550 0.0655 14 CHF 837.7+23.4 w0:05 0.0367 0.0041 0.0613 15 CHF 652.15+20.6 v0:005 0.0265 20.0322 0.0550 16265 healthy 1023.9+38.9 v0:001 0.0527 0.0162 0.0438 16272 healthy 924.7+30.6 v0:001 0.0539 20.0304 0.0557 16273 healthy 1046.2+68.7 v0:001 0.0764 0.0998 0.0543 16420 healthy 849.9+39.2 v0:001 0.0394 0.0107 0.0410 16483 healthy 818.5+24.9 v0:001 0.0336 20.0254 0.0396 16539 healthy 831.5+47.9 v0:001 0.0592 0.0736 0.0485 16773 healthy 1238.9+74.7 v0:001 0.0819 0.0842 0.0613 16786 healthy 945.1+37.0 v0:001 0.0503 0.0226 0.0527 16795 healthy 889.4+61.8 v0:001 0.0442 20.0214 0.0462 17052 healthy 939.8+33.3 w0:05 0.0487 0.0682 0.0819 17453 healthy 816.6+31.9 v0:001 0.0416 0.0310 0.0466 18177 healthy 639.3+25.7 v0:001 0.0267 0.0058 0.0608 18184 healthy 831.2+31.7 v0:001 0.0392 20.0373 0.0438 19090 healthy 993.54+41.0 v0:001 0.0565 20.0101 0.0393 19140 healthy 849.1+49.9 v0:001 0.0353 0.0459 0.0513 19830 healthy 821.2+29.4 v0:001 0.0452 20.0795 0.0332 P-values are obtained from the nonlinearity test. Study of the Instantaneous Cardiovascular Complex Dynamics However, this conclusion cannot be made without speculation at this time as the data used for the comparison between CHF patients and healthy subjects is related to long-term ECG monitoring during unstructured activity, whereas the data from the tilt-table experimental dataset is structured. Therefore, the observed sensitivity of the IDLE measures could be due to different physiological behavior occurring in CHF subjects or to differences related to the kind of experimental protocol. Future works are related to pursue this direction in further investigating the potential of these high-order nonlinear models in producing new real-time measures for the underlying complexity of physiological systems, and to the investigation of the instantaneous complexity, along with the baroreflex sensitivity and respiratory sinus arrhythmia, during postural changes in CHF subjects, thus allowing some conclusions on the complex physiological behavior of the cardiovascular system in CHF subjects. of, for example, changes on the noise properties). Of note, we have previously reported that the standard HRV indices defined in the time and frequency domain are unable to distinguish the three possible modality of transition through different p-values [19]. Moreover, the novel complexity features {lRR; slRR} give important information in the complexity evaluation, also useful in distinguishing heartbeat dynamics coming from patients with CHF and healthy subjects. We found that pathological heartbeat dynamics are associated with increased complexity variability, providing an unique measure of complexity able to discern the CHF and healthy populations. Of note, some of the standard measures defined in the time domain have similar p-value than the point-process measures, as well as DFA-a2 shows similar performances than the IDLE median absolute deviation. Never- theless, we point out that we aimed at showing the performances of novel instantaneous measures of complexity based on Lyapunov exponents, providing novel insights on the complexity character- ization of stochastic time-varying discrete point-process systems. In other words, although other HRV-based measures are able to discern CHF from healthy subjects, no other measures have been proposed to characterize the time-varying complexity behavior occurring in pathological vs. a healthy cardiovascular system. In particular, while confirming the results reported in the current literature (i.e., several measures of complexity are not able to characterize the CHF and healthy subjects groups), we show a novel key complexity behavior through the proposed complexity variability framework. Study of the Instantaneous Cardiovascular Complex Dynamics The IDLE, in fact, becomes higher when the simulated system switches from non-chaotic to chaotic behavior (see Fig. 2). In all applications, an IG probability model was used as a stochastic version of the widely-applied deterministic integrate-and-fire models used to simulate heartbeats. Regarding the experimental datasets, we demonstrated that our approach is useful in characterizing the inherent nonlinearity of the cardiovascular system. For the first time, tracking complexity by instantaneous Our experimental findings on the nonlinearity test performed on each experimental session (resting state, stand-up, slow tilt, and fast tilt) suggest that loss of instantaneous heartbeat complexity as a function of velocity of the postural changes is reasonably due to changes in the nonlinearity of the cardiovascular system (instead August 2014 \| Volume 9 \| Issue 8 \| e105622 PLOS ONE \| www.plosone.org August 2014 \| Volume 9 \| Issue 8 \| e105622 14 Estimating the Instantaneous Lyapunov Spectra Table 6. Group Statistics of Standard and Instantaneous Heartbeat Dynamics Measures from CHF-Healthy Experimental Dataset. Table 6. Group Statistics of Standard and Instantaneous Heartbeat Dynamics Measures from CHF-Healthy Experimental Dataset. Table 6. Group Statistics of Standard and Instantaneous Heartbeat Dynamics Measures from CH Table 6. Group Statistics of Standard and Instantaneous Heartbeat Dynamics Measures from CHF-Healthy Experimental Dataset. CHF (n = 14) Healthy (n = 16) p-value Standard and Instantaneous Time Domain Measures of HRV Mean RR (ms) 669.73+68.66 855.74+56.14 v0:002 RMSSD 0.0121+0.0036 0.0432+0.0145 ,4e{4 pNN50% 0.2357+0.2246 21.5406+15.4908 ,1e{4 HRV Triangular Index 2.9551+0.5769 2.5628+0.3593 w0:05 mRR(ms) 671.55+69.6 864.7+53.3 ,4e{3 sRR(ms) 8.31+2.2 24.7+7.0 ,5e{4 Standard and Instantaneous Nonlinear Measures of HRV ApEn 1.2130+0.1032 1.2177+0.1066 .0.05 SampEn 1.5670+0.2690 1.4092+0.1522 .0.05 DFA-a1 0.8498+0.2191 1.0820+0.1467 .0.05 DFA-a2 1.1552+0.1335 0.9286+0.0544 ,0.05 LD1 0.0167+0.0025 0.0165+0.0012 w0:05 LD2 0.0029+0.0008 0.0033+0.0005 w0:05 lRR 0.0014+0.0649 0.0135+0.0368 w0:05 slRR 0.0595+0.0120 0.0476+0.0066 v0:05 P-values are obtained from the Mann-Whitney test with null hypothesis of equal medians between the CHF and healthy subject groups. Values are expressed as X~Median(X)+MAD(X). doi:10.1371/journal.pone.0105622.t006 P-values are obtained from the Mann-Whitney test with null hypothesis of equal medians between the CHF and healthy subject groups. Values are expressed as X~Median(X)+MAD(X). doi:10.1371/journal.pone.0105622.t006 by anesthetic drugs [46]. Looking at the overall results shown using actual heartbeat dynamics data, it seems that the median IDLE is sensitive to changes in ANS regulation induced by orthostatic stress, while IDLE median absolute deviation is sensitive to changes in ANS regulation induced by CHF. Acknowledgments The authors are grateful to Roger G. Mark and Thomas Heldt (Harvard- MIT Division of Health Sciences and Technology) for kindly providing the tilt-table data analyzed in this study. Early versions of this work have been reported in Proceedings of IEEE-EMBC 2012, San Diego (USA), and IEEE-EMBC 2013, Osaka (Japan). Study of the Instantaneous Cardiovascular Complex Dynamics These findings can be linked to the current literature whereby cardiovascular disorders affect complexity and variability, and may lead to serious pathological events such as heart failure [71]. Analyzed the data: GV LC RB. Wrote the paper: GV LC RB. Conceived and designed the methodology and experiments: GV LC RB. References Lanata A, Valenza G, Mancuso C, Scilingo EP (2011) Robust multiple cardiac arrhythmia detection through bispectrum analysis. Expert Systems with Applications 38: 6798–6804. 38. Chon K, Yip K, Camino B, Marsh D, Holstein-Rathlou NH (2000) Modeling nonlinear determinism in short time series from noise driven discrete and continuous systems. International Journal of Bifurcation and Chaos 10: 2745– 2766. 10. Wu G, Arzeno N, Shen L, Tang D, Zheng D, et al. (2009) Chaotic signatures of heart rate variability and its power spectrum in health, aging and heart failure. PloS one 4: e4323. 39. Armoundas A, Ju K, Iyengar N, Kanters J, Saul P, et al. (2002) A stochastic nonlinear autoregressive algorithm reflects nonlinear dynamics of heart-rate fluctuations. Annals of biomedical engineering 30: 192–201. 11. Valenza G, Lanata A, Scilingo EP (2012) Oscillations of heart rate and respiration synchronize during affective visual stimulation. IEEE Transactions on Information Technology in Biomedicine 16: 683–690. 40. Korenberg M (1989) A robust orthogonal algorithm for system identification and time-series analysis. Biological Cybernetics 60: 267–276. 12. Winfree A (1994) Electrical turbulence in three-dimensional heart muscle. Science 266: 1003–1006. 41. Barbieri R, Matten E, Alabi A, Brown E (2005) A point-process model of human heartbeat intervals: new definitions of heart rate and heart rate variability. American Journal of Physiology-Heart and Circulatory Physiology 288: H424. 13. Valenza G, Nardelli M, Lanata A, Gentili C, Bertschy G, et al. (2014) Wearable monitoring for mood recognition in bipolar disorder based on history-dependent long-term heart rate variability analysis. IEEE Journal of Biomedical and Health Informatics. 42. Chen Z, Brown E, Barbieri R (2009) Assessment of autonomic control and respiratory sinus arrhythmia using point process models of human heart beat dynamics. IEEE Transactions on Biomedical Engineering 56: 1791–1802. 14. Chialvo DR, Gilmour Jr RF, Jalife J (1990) Low dimensional chaos in cardiac tissue. Nature 343: 653–657. 43. Marmarelis V (1993) Identification of nonlinea expansions of kernels. Ann Biomed Eng 21. 43. Marmarelis V (1993) Identification of nonlinear biological system using laguerre expansions of kernels. Ann Biomed Eng 21. 15. Valenza G, Lanata´ A, Scilingo EP (2013) Improving emotion recognition systems by embedding cardiorespiratory coupling. Physiological measurement 34: 449. 44. Valenza G, Citi L, Barbieri R (2013) Instantaneous nonlinear assessment of complex cardiovascular dynamics by laguerre-volterra point process models. In: Engineering in Medicine and Biology Society (EMBC), 2013 35th Annual International Conference of the IEEE. IEEE, pp. 6131–6134. 16. References 29. Lyapunov A (1949) Problem general de la stabilite du mouvement. Ann Math Stud 17. 1. Sunagawa K, Kawada T, Nakahara T (1998) Dynamic nonlinear vago- sympathetic interaction in regulating heart rate. Heart and Vessels 13: 157–174. 30. Keller J, Hense A, Rhodin A (2010) Estimating uncertainty in atmospheric models-application and new approaches of lyapunov vector estimations. In: AGU Fall Meeting Abstracts. volume 1, p. 1319. 2. Atyabi F, Livari M, Kaviani K, Tabar M (2006) Two statistical methods for resolving healthy individuals and those with congestive heart failure based on extended self-similarity and a recursive method. Journal of Biological Physics 32: 489–495. g p 31. Herna´ndez-Carrasco I, Lo´pez C, Herna´ndez-Garcı´a E, Turiel A (2011) How reliable are finite-size lyapunov exponents for the assessment of ocean dynamics? Ocean Modelling 36: 208–218. 3. (2009) Special issues on nonlinearity on heart rate. Chaos 19. 4. Valenza G, Citi L, Lanata´ A, Scilingo EP, Barbieri R (2014) Revealing real-time emotional responses: a personalized assessment based on heartbeat dynamics. Nature Scientific reports 4. g 32. Lindsay DH, Campbell A (2011) A chaos approach to bankruptcy prediction. Journal of Applied Business Research (JABR) 12: 1–9. 5. Valenza G, Citi L, Lanata A, Scilingo EP, Barbieri R (2013) A nonlinear heartbeat dynamics model approach for personalized emotion recognition. In: Engineering in Medicine and Biology Society (EMBC), 2013 35th Annual International Conference of the IEEE. IEEE, pp. 2579–2582. 33. Hu J, Gao J, Tung Ww (2009) Characterizing heart rate variability by scale- dependent lyapunov exponent. Chaos: An Interdisciplinary Journal of Nonlinear Science 19: 028506–028506. 34. Wolf A, Swift JB, Swinney HL, Vastano JA (1985) Determining lyapunov exponents from a time series. Physica D: Nonlinear Phenomena 16: 285–317. 6. Glass L (2001) Synchronization and rhythmic processes in physiology. Nature 410: 277–284. 35. Abarbanel HD, Gollub JP (1996) Analysis of observed chaotic data. Physics Today 49: 86. 7. Goldberger A, Peng C, Lipsitz L (2002) What is physiologic complexity and how does it change with aging and disease? Neurobiology of aging 23: 23–26. 36. Ruelle D (1994) Where can one hope to profitably apply the ideas of chaos? Physics Today 47: 24. 8. Poon C, Merrill C (1997) Decrease of cardiac chaos in congestive heart failure. Nature 389: 492–495. 37. Gao J, Cao Y, Tung Ww, Hu J (2007) Multiscale analysis of complex time series: integration of chaos and random fractal theory, and beyond. John Wiley & Sons. 9. References Kanters JK, Højgaard MV, Agner E, Holstein-Rathlou NH (1996) Short-and long-term variations in non-linear dynamics of heart rate variability. Cardiovascular Research 31: 400–409. g g gy y ( ) International Conference of the IEEE. IEEE, pp. 6131–6134. 45. Citi L, Valenza G, Barbieri R (2012) Instantaneous estimation of high-order nonlinear heartbeat dynamics by lyapunov exponents. In: Proceeding of the IEEE-EMBC. pp. 13–16. 17. Schmidt G, Monfill G (1995) Nonlinear methods for heart rate variability assessment. Heart rate variability Armonk: Futura: 87–98. 46. Citi L, Valenza G, Purdon PL, Brown EN, Barbieri R (2012) Monitoring heartbeat nonlinear dynamics during general anesthesia by using the instantaneous dominant lyapunov exponent. In: Engineering in Medicine and Biology Society (EMBC), 2012 Annual International Conference of the IEEE. IEEE, pp. 3124–3127. 18. Jelinek HF, Khandoker AH, Quintana D, Imam MH, Kemp A (2011) Complex correlation measure as a sensitive indicator of risk for sudden cardiac death in patients with depression. In: Computing in Cardiology, 2011. IEEE, pp. 809– 812. 19. Valenza G, Citi L, Scilingo E, Barbieri R (2013) Point-process nonlinear models with laguerre and volterra expansions: Instantaneous assessment of heartbeat dynamics. IEEE Transactions on Signal Processing 61: 2914–2926. 47. Daley D, Vere-Jones D (2003) An introduction to the theory of point processes, vol. I: Elementary Theory and Methods, volume 2. Springer Verlag. 48. Chen Z, Purdon P, Harrell G, Pierce E, Walsh J, et al. (2011) Dynamic assessment of baroreflex control of heart rate during induction of propofol anesthesia using a point process method. Annals of biomedical engineering: 1– 17. 20. Valenza G, Allegrini P, Lanata` A, Scilingo EP (2012) Dominant lyapunov exponent and approximate entropy in heart rate variability during emotional visual elicitation. Frontiers in neuroengineering 5. 21. Schmitt DT, Stein PK, Ivanov PC (2009) Stratification pattern of static and scale-invariant dynamic measures of heartbeat fluctuations across sleep stages in young and elderly. IEEE Transactions on Biomedical Engineering 56: 1564– 1573. 49. Citi L, Brown E, Barbieri R (2012) A real-time automated point process method for detection and correction of erroneous and ectopic heartbeats. 50. Akay M (2000) Nonlinear Biomedical Signal Processing Vol. II: Dynamic Analysis and Modeling. Wiley-IEEE Press. 22. Valenza G, Citi L, Gentili C, Lanata A, Scilingo E, et al. (2014) Characterization of depressive states in bipolar patients using wearable textile technology and instantaneous heart rate variability assessment. IEEE Journal of Biomedical and Health Informatics. 51. Estimating the Instantaneous Lyapunov Spectra Estimating the Instantaneous Lyapunov Spectra Author Contributions Concerning other preliminary applications, the proposed IDLE methodology has been revealed as a powerful tool also in tracking the instantaneous complexity during loss of consciousness induced Analyzed the data: GV LC RB. Wrote the paper: GV LC RB. Conceived and designed the methodology and experiments: GV LC RB. August 2014 \| Volume 9 \| Issue 8 \| e105622 PLOS ONE \| www.plosone.org PLOS ONE \| www.plosone.org 15 References Korenberg M (1991) Parallel cascade identification and kernel estimation for nonlinear systems. Annals of biomedical engineering 19: 429–455. 52. Schetzen M (1980) The volterra and wiener theories of nonlinear systems. 23. Ba¨r KJ, Boettger MK, Koschke M, Schulz S, Chokka P, et al. (2007) Non-linear complexity measures of heart rate variability in acute schizophrenia. Clinical neurophysiology 118: 2009–2015. 53. Granger C, Joyeux R (1980) An introduction to long-memory time series models and fractional differencing. Journal of time series analysis 1: 15–29. d fractional differencing. Journal of time series analysis 1: 15–29 54. Dieci L, Russell R, Van Vleck E (1997) On the computation of lyapunov exponents for continuous dynamical systems. SIAM journal on numerical analysis: 402–423. 24. Tarvainen MP, Georgiadis S, Laitio T, Lipponen JA, Karjalainen PA, et al. (2012) Heart rate variability dynamics during low-dose propofol and dexmedetomidine anesthesia. Annals of biomedical engineering 40: 1802–1813. y 55. Holzfuss J, Parlitz U (1991) Lyapunov exponents from time series. Lyapunov Exponents: 263–270. 25. Valenza G, Citi L, Scilingo EP, Barbieri R (2014) Inhomogeneous point-process entropy: An instantaneous measure of complexity in discrete systems. Physical Review E 89: 052803. 56. Geist K, Parlitz U, Lauterborn W (1990) Comparison of different methods for computing lyapunov exponents. Prog Theor Phys 83: 875–893. 26. Orini M, Laguna P, Mainardi L, Bailo´n R (2012) Assessment of the dynamic interactions between heart rate and arterial pressure by the cross time–frequency analysis. Physiological measurement 33: 315. 57. Rajendra Acharya U, Paul Joseph K, Kannathal N, Lim C, Suri J (2006) Heart rate variability: a review. Medical and Biological Engineering and Computing 44: 1031–1051. y y g 27. Valenza G, Nardelli M, Bertschy G, Lanata A, Scilingo E (2014) Mood states modulate complexity in heartbeat dynamics: A multiscale entropy analysis. EPL (Europhysics Letters) 107: 18003. 58. Rosenstein MT, Collins JJ, De Luca CJ (1993) A practical method for calculating largest lyapunov exponents from small data sets. Physica D: Nonlinear Phenomena 65: 117–134. 28. Shimada I, Nagashima T (1979) A numerical approach to ergodic problem of dissipative dynamical systems. Progress of Theoretical Physics 61: 1605–1616. 59. Pincus S (1991) Approximate entropy as a measure of system complexity. Proceedings of the National Academy of Sciences 88: 2297. PLOS ONE \| www.plosone.org August 2014 \| Volume 9 \| Issue 8 \| e105622 August 2014 \| Volume 9 \| Issue 8 \| e105622 16 67. Thuraisingham R, Gottwald G (2006) On multiscale entropy analysis for physiological data. Physica A: Statistical Mechanics and its Applications 366: 323–332. Estimating the Instantaneous Lyapunov Spectra 60. Richman J, Moorman J (2000) Physiological time-series analysis using approximate entropy and sample entropy. American Journal of Physiology- Heart and Circulatory Physiology 278: H2039. 67. Thuraisingham R, Gottwald G (2006) On multiscale entropy analysis for physiological data. Physica A: Statistical Mechanics and its Applications 366: 323–332. y y gy 61. Peng C, Buldyrev S, Havlin S, Simons M, Stanley H, et al. (1994) Mosaic organization of dna nucleotides. Physical Review E 49: 1685. 68. Hagerman I, Berglund M, Lorin M, Nowak J, Sylve´n C (1996) Chaos-related deterministic regulation of heart rate variability in time-and frequency domains: effects of autonomic blockade and exercise. Cardiovascular research 31: 410– 418. g y 62. Zong W, Moody G (2003) Wqrs-single-channel qrs detector based on length transform. Physionet http://wwwphysionetorg/physiotools/wag/wqrs-1htm. 63. Goldberger A, Amaral L, Glass L, Hausdorff J, Ivanov P, et al. (2000) Physiobank, physiotoolkit, and physionet: Components of a new research resource for complex physiologic signals. Circulation 101: e215. 69. Porta A, Gnecchi-Ruscone T, Tobaldini E, Guzzetti S, Furlan R, et al. (2007) Progressive decrease of heart period variability entropy-based complexity during graded head-up tilt. Journal of applied physiology 103: 1143–1149. p p y g g 64. Barnett A, Wolff R (2005) A time-domain test for some types of nonlinearity. IEEE Transactions on Signal Processing. 53: 26–33. 70. Tulppo MP, Ma¨kikallio TH, Seppa¨nen T, Shoemaker K, Tutungi E, et al. (2001) Effects of pharmacological adrenergic and vagal modulation on fractal heart rate dynamics. Clinical Physiology 21: 515–523. 65. Baillie R, Cecen A, Erkal C (2009) Normal heartbeat series are nonchaotic, nonlinear, and multifractal: New evidence from semiparametric and parametric tests. Chaos: An Interdisciplinary Journal of Nonlinear Science 19: 028503. 71. Schwartz PJ, De Ferrari GM (2011) Sympathetic–parasympathetic interaction in health and disease: abnormalities and relevance in heart failure. Heart failure reviews 16: 101–107. 66. Costa M, Goldberger A, Peng C (2002) Multiscale entropy analysis of complex physiologic time series. Physical Review Letters 89: 68102. PLOS ONE \| www.plosone.org August 2014 \| Volume 9 \| Issue 8 \| e105622 PLOS ONE \| www.plosone.org 17
https://openalex.org/W3004213451	https://zenodo.org/record/3597395/files/1.pdf	English	null	Clinical Pharmacists in Chronic Care [Part 2]	Archives in biomedical engineering & biotechnology	2,020	cc-by	35,333	Abstract Pharmacy practice has changed significantly lately. The professionals have the chance to contribute straightforwardly to patient consideration so as to lessen morbimortality identified with medication use, promoting wellbeing and preventing diseases. Healthcare organizations worldwide are under substantial pressure from increasing patient demand. Unfortunately, a cure is not always possible particularly in this era of chronic complications, and the role of physicians has become limited to controlling and palliating symptoms. The increasing population of patients with long-term conditions are associated with high levels of morbidity, healthcare costs and GP workloads. Clinical pharmacy took over an aspect of medical care that had been partially abandoned by physicians. Overburdened by patient loads and the explosion of new drugs, physicians turned to pharmacists more and more for drug information, especially within institutional settings. Once relegated to counting and pouring, pharmacists headed institutional reviews of drug utilization and served as consultants to all types of health-care facilities. In addition, when clinical pharmacists are active members of the care team, they enhance proficiency by: Providing critical input on medicine use and dosing. Working with patients to solve problems with their medications and improve compliance. Abstract Keywords: Chronic care; Pharmacy intervention; Diabetes care; CVD prevention; Inflammatory bowel disease Abbreviations: AACP: American Association of Colleges of Pharmacy; ACPE: Accreditation Council for Pharmacy Education; IDF: International Diabetes Federation; HbA1c: Hemoglobin A1c; IHD: Ischemic Heart Disease; MI: Myocardial Infarction; CHD: Coronary Heart Disease; DALY: Disability- Adjusted Life Year; QoL: Quality of Life; DRPs: Drug Related Problems; IBD: Inflammatory bowel disease; HRT: Hormone replacement therapy; BMD: Bone-Mineral Density; COPD: Chronic Obstructive Pulmonary Disease; LDL-C: LDL cholesterol; GERD: Gastroesophageal Reflux Disease; OSA: Obstructive Sleep Apnea; SCH: Subclinical Hypothyroidism; NAMI: National Alliance on Mental Illness; MDD: Major Depressive Disorder; NMHS: National Mental Health Survey; ABS: Australian Bureau of Statistics; NSMHWB: National Survey of Mental Health and Wellbeing; CHD: Coronary Heart Disease; MH: Mental Health; ADT: Antidepressant Drug Treatment; CANMAT: Canadian Network for Mood and Anxiety Treatments; PES: Psychiatric Emergency Services; DALY: Disability-Adjusted Life Year; DRPs: Drug-Related Problems; VLW: Value of Lost Economic Welfare; ALS: Amyotrophic Lateral Sclerosis; SNRIs: Serotonin and Norepinephrine Reuptake Inhibitors; TCAs: Tricyclic Antidepressants; ASPs: Antimicrobial Stewardship Programs; ESRD: End-Stage Renal Disease; CKD: Chronic Kidney Disease; MSM: Men who have Sex with Men; NSCLC: Non-small-cell lung cancer; ELISA: Enzyme-Linked Immunosorbent Assay; LLS: Leukemia & Lymphoma Society; ALL: Acute Lymphoblastic Leukemia; AML: Acute Myeloid Leukemia; CML: Chronic Myeloid Leukemia; NRT: Nicotine Replacement Therapy; ADT: Androgen Deprivation Therapy; PSA: Prostate Specific Antigen; DRE: Digital Rectal Examination; PSA: Prostate Specific Antigen; FOBT: Fecal Occult Blood Testing; GLOBOCAN: Global Cancer Incidence, Mortality and Prevalence Abdul Kader Mohiuddin* Abdul Kader Mohiuddin* Department of Medicine, Nasirullah Memorial Trust, Bangladesh Received Date: November 12, 2019 Published Date: January 03, 2020 Corresponding author: Abdul Kader Mohiuddin, Department of Medicine, Nasirullah Memorial Trust, Tejgaon, Dhaka, Bangladesh. Research Article Copyright © All rights are reserved by Abdul Kader Mohiuddin DOI: 10.33552/ABEB.2019.03.000566 Background hospital wards in Aberdeen [3]. The role of clinical pharmacists underwent significant changes from the 1960s through 1990s as their participation in direct patient care enhanced. In the early 1970s, federal funding assisted with greatly expanding clinical pharmacy faculty in Colleges of Pharmacy [4]. Pharmacy education debated where clinical pharmacy fit within pharmacy training. The AACP spearheaded an effort to examine this issue. Till then, two full generations of pharmacists have been educated and trained after the general adoption of the aims of clinical pharmacy Clinical pharmacology is a professional discipline that combines basic pharmacology and clinical medicine. A clinical pharmacist offers invaluable support in the development of a final prescription with better patient management and enhanced safety [1]. Its development began in the early 1950s, primarily as a result of the efforts of Harry Gold. Pharmacist rounding with inpatient hospital services has been traced to the University of Kentucky in 1957 [1,2]. Drug therapy was becoming much more complex. Graham Calder pioneered a new role for pharmacists on This work is licensed under Creative Commons Attribution 4.0 License ABEB.MS.ID.000566. Page 1 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 [4,5]. ACPE has revised the standards for colleges and schools of pharmacy several times since 2000. ACPE Standards 2016 go into effect July 1, 2016. To some extent, pharmacy took over an aspect of medical care that had been partially abandoned by physicians [6]. Overburdened by patient loads and the explosion of new drugs, physicians turned to pharmacists more and more for drug information, especially within institutional settings. A clinical pharmacist often has a somewhat different approach to the use of drugs and may give valuable supplementary information about for example interactions, during the physician’s decision-making process concerning potential changes of and the follow-up of the medication [7,8]. The concept of pharmaceutical care accentuates the pharmacists’ responsibility to pursue the best possible patient outcomes of therapeutic regimen [9]. They possess in-depth knowledge of medications that is integrated with a foundational understanding of the biomedical, pharmaceutical, socio-behavioral, and clinical sciences [10]. To achieve desired therapeutic goals, the clinical pharmacists follow evidence-based therapeutic guidelines, evolving sciences, emerging technologies, and relevant legal, ethical, social, cultural, economic, and professional precept [11- 13]. Background In accordance, clinical pharmacists assume responsibility and accountability for managing medication therapy in direct patient care settings, whether practicing independently or in consultation or collaboration with other health care professionals [14,15]. Their functions encompass comprehensive medication management (ie, prescribing, monitoring, and adjustment of medications), nonpharmacologic guidance, and coordination of care. Interdisciplinary collaboration allows pharmacists opportunities to provide direct patient care or consultations by telecommunication in many different clinical environments, including disease management, primary care, or specialty care [16-19]. Pharmacists may manage chronic or acute illnesses associated with endocrine, cardiovascular, respiratory, gastrointestinal, or other systems [20]. Clinical pharmacist researchers generate, disseminate, and apply new knowledge that contributes to improved HRQoL [21-24]. Within the system of health care, clinical pharmacists are experts in the therapeutic use of medications. They consistently provide medication therapy evaluations and endorsements to patients and allied health professionals (AHPs) [25,26]. Clinical pharmacists are a primary source of scientifically accurate/logical information and advice regarding the safe, appropriate, and cost-effective use of medications [27,28]. They obtain medical and medication history, check medication errors including prescription, dispensing and administration errors, identify drug interactions, monitor ADR, suggest individualization of dosage regimen, provide patient counseling, etc. [29-35]. They also provide information about the use of drugs and medical devices like inhaler, insulin pen, eye drops, nasal sprays, etc. [36]. Participation of a clinical pharmacist in ward/ICU rounds and clinical discussions helps to identify, prevent or reduce drug interaction and ADR [29], [37-39] (Figure 1). Figure 1: Clinical pharmacy offers chronic care services like asthma, diabetes, nutritional supplement counseling, smoking cessation, weight reduction, geriatric care, hyperlipidaemia, hypertension, naturopathy and wound care. The integration of clinical pharmacists into primary care clinics could have positive effects on the clinical outcomes of patients in glycemic control, blood pressure, lipid profile, in accordance with current guidelines. Figure 1: Clinical pharmacy offers chronic care services like asthma, diabetes, nutritional supplement counseling, smoking cessation, weight reduction, geriatric care, hyperlipidaemia, hypertension, naturopathy and wound care. The integration of clinical pharmacists into primary care clinics could have positive effects on the clinical outcomes of patients in glycemic control, blood pressure, lipid profile, in accordance with current guidelines. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Introduction community or primary care setting, in terms of medication, lifestyle management, and health behavior modification [41-45]. It is typically a multi-component intervention that includes medication therapy review, patient medication education, medication monitoring, immunizations, disease self-care and support, and/or prescribing authority. Patients who take voluminous medications due to chronic disease have a high risk of drug duplication, interaction, or ADRS, which could result in extended hospital stays and higher costs [46]. To increase the safety and effectiveness of therapeutics, these patients must have specific needs met, with Population aging has increased the burden of chronic diseases globally. There are both ethical and practical imperatives to address health inequity issues related to chronic disease management for persons with social complexity, existing programs often do not appropriately address the needs of these individuals. This leads to low levels of participation in programs, suboptimal chronic disease management, and higher health-care utilization [40]. Unlike acute conditions, chronic diseases require consistent care and management outside of the healthcare setting, in the Page 2 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 regards to appropriate medication use [47]. Studies have shown that integrating pharmacists into ambulatory clinics can improve chronic disease management and optimal use of medications [48]. Furthermore, pharmacist involvement in patient care may help to curtail inappropriate drug use, specifically in the elderly. A study in Canada saw the proportion of patients receiving an inappropriate medication drop significantly after medication review and optimization by a team that included a pharmacist [49]. Compared to usual care, pharmacist-led care was associated with similar incidences or rates of office, urgent care or ED visits, and hospitalizations and medication adherence, increased the number or dose of medications received and improved study-selected glycemic, blood pressure, and lipid goal attainment [50]. Another recent study shows telehealth-based chronic disease management program including clinical pharmacy specialists imparted statistically significant improvements in diabetes and hypertension outcomes along with clinically significant improvements in the areas of lipid management and tobacco cessation [51]. and optimal use of medications involvement in patient care may ug use, specifically in the elderly. oportion of patients receiving an p significantly after medication m that included a pharmacist [49]. acist-led care was associated with or dose of medications glycemic, blood pressure recent study shows teleh program including cli statistically significant im outcomes along with cli areas of lipid managemen Diabetes Care China and India collectively are home of nearly 110 million diabetic patients [55]. It is a risk factor for CVD and has been associated with 2- to 4-fold higher mortality [56] and another study says that half of all diabetic death was due to CVD [57]. The number of deaths caused by diabetes in the age range of 60–99 years in 2017 was 3,200,000 [58]. About half of diabetes- related mortality (48%) occurs in people younger than 60 years and it continues to reduce life expectancy by 6–8 years in people diagnosed at the age of 50 years [59]. More than 35% patients did not receive any diabetes education, while 30% diabetic patients were compliant with drug regimens and the non-compliance was higher among the lower socioeconomic groups [60]. Pharmacist may provide a face-to-face counseling regarding knowledge on diabetes, self-monitoring of blood glucose, regular checkup of systolic blood pressure, body weight, and serum cholesterol levels. The pharmacist may also counsel regarding non-pharmacological management strategies such as diet control, exercise therapy, and early identification of symptoms of hypoglycemia (blurred vision, rapid heartbeat, sweating, fatigue, headache, dizziness, trouble thinking, seizures, and coma) and its management. In a satisfaction survey of 24 providers or clinical pharmacists, nearly 90% had favorable responses toward the protocol and its effect on access to and quality of care [61]. Management of DM remains a significant challenge in the US, as estimates indicate that greater than 40% of diabetes patients are uncontrolled with a HbA1c. Diabetic patients who received care from the collaborative team, including a clinical pharmacist, had improvement in most key indicators of diabetes like HbA1c, in both high- and low- income countries and in both urban and rural areas [62-71]. Emphasizing medication adherence, particularly for patients with longer duration of diabetes and those with multiple comorbid diseases should be strongly considered in future diabetes management programs implemented to improve glycemic control in patients with type 2 diabetes [72]. A mobile phone text message can serve as a simple and cost-effective option in improving medication adherence and clinical outcomes by providing information between clinic visits has been reported [73,74]. CVD Prevention Hypertension affects 26.4% of the global adult population and a key driver of global disease burden [77]. It is a major risk factor for CHD, stroke, retinopathies, and renal dysfunction. 18 million people die each year from CVDs, an estimated 31% of all deaths worldwide [78]. Of these deaths, 85% are due to MI and stroke [79]. CVD is currently the leading cause of morbidity and mortality and over 80% occur in LMICs [80]. Nearly 50% patients with chest pain related to exercise have obstructive coronary artery disease [81]. Coronary artery spasm plays an important role in the pathogenesis of IHD, including angina pectoris, MI, and sudden death, occurring most often from midnight to early morning [82]. A study of 1,015 patients with stable coronary artery disease (CAD) showed a 4.4- fold escalation in the risk of stroke and a 3.8-fold escalation in the risk of death among patients who self-reported as non-compliant [83]. IHD has topped the list of causes of years of life lost for more than a decade, highlighting the shift in the global burden of disease from communicable to chronic disease. Risk factors for CVD, including raised blood pressure, hypercholesterolaemia and high BMI, are among the most important contributors to DALYs [84]. CVD claimed death of some 900,000 death in US in 2016 [85]. BP is still uncontrolled in 50% of the US population with hypertension. Additionally, BP can remain poorly controlled despite up to six physician visits per year [86]. Across South Asia, overall hypertension prevalence is estimated to be 27%. Prospective Urban Rural Epidemiology study has shown more than 50% are unaware of it and up to 80% of hypertensive patients have low adherence to medication. Uncontrolled BP was found more than 50% in Bangladesh, 70% in Pakistan and almost 60% in Sri Lanka [87]. The goals of treatment of hypertension are to limit target organ damage, thereby reducing the morbidity and mortality associated with the disease [88]. Many factors including socioeconomic status, belief about medications, comorbidity, availability of medications, access to healthcare, level of health literacy, number of medications, duration of therapy, age, gender, culture, educational status, and knowledge of the disease and treatment have been associated with the rate of adherence. Diabetes Care Diabetes Care Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. Page 3 of 41 Diabetes Care Figure 2: Clinical pharmacy offers chronic care services like asthma, diabetes, nutritional supplement counseling, smoking cessation, weight reduction, geriatric care, hyperlipidaemia, hypertension, naturopathy and wound care. The integration of clinical pharmacists into primary care clinics could have positive effects on the clinical outcomes of patients in glycemic control, blood pressure, lipid profile, in accordance with current guidelines. Figure 2: Clinical pharmacy offers chronic care services like asthma, diabetes, nutritional supplement counseling, smoking cessation, weight reduction, geriatric care, hyperlipidaemia, hypertension, naturopathy and wound care. The integration of clinical pharmacists into primary care clinics could have positive effects on the clinical outcomes of patients in glycemic control, blood pressure, lipid profile, in accordance with current guidelines Figure 2: Clinical pharmacy offers chronic care services like asthma, diabetes, nutritional supplement counseling, smokin reduction, geriatric care, hyperlipidaemia, hypertension, naturopathy and wound care. The integration of clinical pharmacis clinics could have positive effects on the clinical outcomes of patients in glycemic control, blood pressure, lipid profile, current guidelines. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Page 3 of 4 ation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Page 3 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 As the seventh-leading cause of death in the US, diabetes can lead to various health complications such as blindness, kidney disease, amputations, and heart disease. The worldwide existing prevalence of DM is about 425 million people, of whom 279 million are in urban areas and 146 million are in rural zones [52]. The IDF estimates that by 2040, one in 10 adults (642 million) will have diabetes. Around 50% (212.4 million) of patients are unaware of their diabetes. More than 12% of total global health expenditure goes to diabetes, according to IDF [53]. Annual worldwide economic cost of diabetic care was calculated at $727 billion in year 2017 which is predicted to be $776 billion for year 2045 respectively [54]. Currently, more than 230 million Asian individuals are living with diabetes, accounting for approximately 55% of the world’s diabetic population. Diabetes Care A report from the National Diabetes Commission suggested that an inappropriate attitude of health care professionals toward an individualized patient schedule on administration times and dosage of each medication, educating the patient on the importance of medication adherence, dietary adherence and exercise on better glycemic control, giving advice on how to reduce adverse effects of medications, and also by teaching how to take medications in the holy month of Ramadan or other religious fasting, and how to use pill boxes and diary logs to reduce forgetfulness. Literature indicates a number of interventional studies involving pharmacist- based educational interventions, showing clinically significant improvements in the clinical outcomes of the diabetes patients [76] (Figure 2). Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. CVD Prevention Lack of medication availability, low level of awareness about the disease and treatment, inability to afford medicines, mistrust in western medicine, and more trust on traditional and spiritual healers are very common in the rural population [89]. Non-drug therapies have been shown to lower BP, enhance antihypertensive drug efficacy, and decrease Page 4 of 41 Page 4 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 cardiovascular risk. All patients with hypertension and those in the prehypertensive category should be advised to make life-style modifications in addition to any pharmacologic treatment that they receive [90]. Surprisingly, 60%- 80% of the population around the world (according to WHO) are partially or fully dependent upon herbal drugs for primary healthcare [91]. Interactions of some ingredients in supplements with other anti-hypertensive and cardiovascular preparations are well-documented [92]. Green tea showed 85% decrease in plasma concentration of nadolol, for example [93]. The pharmacist may play a relevant role in primary and secondary prevention of CVDs, mainly through patient education and advocacy, drug safety management, medication review (review of both drug-food and drug-drug interaction), monitoring and reconciliation, detection and control of specific cardiovascular risk factors (e.g., blood pressure, blood glucose, serum lipids) and clinical events [94]. factors affecting adherence, improve adherence and patient QoL by reducing BP levels in patients treated with antihypertensive agents, increased referral acceptance [86], [95-99]. Interventions that were most effective included combinations of more convenient care, information, reminders, self-monitoring, reinforcement, advocacy, family therapy, psychotherapy, crisis intercession, regular telephone follow-up, and supportive care [83]. Weight loss has been noted to modify risk factors via improving insulin sensitivity, reducing inflammation, decreasing BP and modifying the lipid profile [100]. It is astonishing that 7–28% of patients with coronary heart disease still smoke, but around half of smokers are planning to quit [101]. The World Bank suggests that around 180 million tobacco related deaths could be prevented between now and 2050 if adult tobacco consumption abate by 50% by 2020 [102]. A clinical pharmacist trained for smoking cessation counselling can play a key role in providing such intercessions, including the assessment of pharmacotherapy interactions with tobacco use [103]. Prescription smoking cessation medications include bupropion and varenicline [104]. A recent Canadian survey shows that pharmacist-led intervention resulted in more than 70% of patients using nicotine replacement therapy for smoking cessation [105] (Figure 3). CVD Prevention Pharmacist intervention can increase patients’ knowledge about their condition in a way that positively modifies their beliefs about medicines, increased medication intensification without significant change in medication adherence, modify Figure 3: Effects of pharmacist’s intervention on humanistic, clinical and economic outcomes in patients with CVD [94]. ovarian polycytosis, sleep apnea syndrome, and some neoplasms [111]. Successful obesity treatment plans incorporate diet, exercise, behavior modification (with or without drug treatment), and/or surgical intervention [112]. Prior to recommending any treatment, the clinician must evaluate the patient for the presence of secondary causes of obesity, such as thyroid dysfunction [113,114]. If secondary causes are suspected, then a more complete diagnostic workup and appropriate therapy is important. The clinician should then evaluate the patient for the presence and severity of other obesity- related diseases, evaluating appropriate lab tests as indicated. Based on the outcome of this medical evaluation, the patient should be counseled on the risks and benefits of available treatment options (along with obesity‐related comorbidities, including T2D prevention, and improvements in dyslipidemia, hyperglycemia, osteoarthritis, stress incontinence, GERD, hypertension, and PCOS Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Obesity Management In every single country in the world, the incidence of obesity is rising continuously with coronary artery disease, hypertension, type 2 diabetes mellitus, respiratory disorders and dyslipidemia [106]. The WHO estimated that in 2016 more than 1.9 billion adults were overweight (nearly 40% of the population) and over 650 million (13% of the population) were people with obesity [107]. Globally, the annual cost of obesity-related diseases has reached $2 trillion according to a recent report by McKinsey Global Institute [108]. Obesity increases cardiovascular risk through risk factors such as increased fasting plasma triglycerides, high LDL cholesterol, low HDL cholesterol, elevated blood glucose and insulin levels and high blood pressure [109,110]. Also, obesity causes cerebral vasculopathy, gallbladder lithiasis, arthropathy, Page 5 of 41 Page 5 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 [115,116]. If obesity is present without other comorbid conditions, then the goal would be absolute weight loss. In the presence of [115,116]. If obesity is present without other comorbid conditions, then the goal would be absolute weight loss. In the presence of comorbid conditions, relatively small reductions in total body weight can have significant effects on comorbidity [117] (Figure 4). comorbid conditions, relatively small reductions in total body weight can have significant effects on comorbidity [117] (Figure 4). Figure 4: 2013 AACE Obesity Treatment Algorithm [118]. Figure 4: 2013 AACE Obesity Treatment Algorithm [118]. of disease burden in 2020 [128]. In 2015 alone, the death toll due to asthma was 383,000 globally [129]. One in five individuals with a diagnosis of COPD, asthma, or both asthma and COPD in primary care settings have asthma-COPD overlap [130]. In COPD, the outpatient therapeutic and management goals are to reduce symptoms and risks from exacerbations, and to maintain drug therapy. In contrast with asthma, COPD is a disease caused by chronic and often daily exposure to noxious particles or gases. The small airways in COPD are gradually destroyed leading to chronic bronchitis and emphysema [131]. Optimal pharmacological treatment including rapid treatment of exacerbations, can improve symptoms, reduce exacerbation frequency, and improve exercise tolerance, while poor medication adherence and suboptimal inhaler technique negatively impact outcomes [132]. Adherence to inhaled medication is poor in the real world and shows great variability, ranging from as low as 20% to over 60% [133]. Obesity Management To improve adherence, the therapeutic decisions should be discussed with the patient and should take into consideration their lifestyle factors, demographic characteristics (age, co-morbidities, physical limitations, psychological and cognitive status), and pharmacological factors (polypharmacy regimens) to choose the best inhaler device for that patient [134]. Pharmacist-led comprehensive therapeutic interchange program of COPD inhalers may provide 30% pharmacy cost savings, improved medication adherence, knowledge of disease, decrease the number of prescriptions for exacerbations for these patients. and reduces 30- day readmission rate [135-138]. A pharmacist-driven spirometry service was associated with quality testing results, identified respiratory disease abnormalities, and helped modifications of pulmonary drug regimens based on evidence-based guidelines [137]. Despite advances in inhaler device technology, estimates of Pharmacists, commonly considered one of the most trustworthy and accessible health care professionals, are ideally situated to provide counseling for weight and lifestyle management. Well trained pharmacists to perform basic physical assessments such as weight, waist circumference, blood glucose monitoring, and pharmacotherapy counseling, while additional training could be easily obtained for services that would encompass dietary counseling, guidance on physical activity, and behavioral counseling [108], [119,120]. As pharmacists currently do not have a well- identified role in obesity management, but study results display that pharmacist intervention was beneficial [121]. According to Canadian Pharmacists Journal, 2016 pharmacist prescribing resulted in >3- fold more patients achieving target LDL-c levels [122]. Pharmacists who identify and treat patients with dyslipidemia, including those with inadequately controlled LDL cholesterol levels, are better than physicians at attaining therapeutic goals [123]. Studies have demonstrated that pharmacist-led interventions, including lipid clinics, can help patients achieve these more aggressive goals [124,125]. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Asthma & COPD Management Asthma and COPD are common chronic conditions that comprise nearly 80% of direct health care costs associated with respiratory diseases in the EU [126]. In the UK alone, 5.4 million patients are currently receiving therapeutics for asthma; of these, 1.1 million are minor. Over three million people expire due to COPD worldwide every year, an estimated 6% of all demises worldwide [127]. It is the second most common reason for emergency hospital admission. According to WHO, COPD has become the fourth leading cause of mortality in the US. It is estimated to become the fifth leading cause Page 6 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 action plans, educating patients, recommending stepping up/down therapy, reviewing inhaler technique and making other relevant recommendations such as device changes (e.g., dry-powder to metered-dose inhaler) [141]. Polypharmacy is burdensome and associated with patients hospitalized with acute exacerbations. It is plausible that repeated pharmacist intervention to ensure optimal pharmacotherapy and minimize adverse effects, with a direct link to a consultant respiratory physician, and the patient’s GP, may lead to improved outcomes [142]. The interventions identified focused on key areas of asthma and COPD management and support including assessment of current symptoms; assessment and rectification of inhaler technique; identification of medication-related problems; medication adherence; provision of written and oral education materials; smoking cessation [143] (Figure 5). those making inhaler errors range up to 90% of patients irrespective of the device type used. Poor inhaler technique accounted for over €750 million in direct and indirect costs in 2015 in the UK, Spain and Sweden [139]. By providing pharmaceutical care to patients with asthma, the pharmacist can help them to achieve treatment goals, e.g. improvement of disease control and reduction of asthma symptoms, exacerbations and medication-related side effects [140]. Step up of therapy comprised increasing or starting corticosteroid/long-acting beta agonist combination inhaler; corticosteroid inhaler; short-acting muscarinic antagonist inhaler; oral corticosteroid; oral montelukast; or long-acting muscarinic antagonist inhaler. Step down of therapy comprised reducing or stopping corticosteroid/long-acting beta agonist combination inhaler; corticosteroid inhaler; or long-acting muscarinic antagonist inhaler. The pharmacist conducted activities such as issuing asthma Figure 5: Unifying model showing key influences on successful delivery of smoking cessation support by pharmacists [144]. Figure 5: Unifying model showing key influences on successful delivery of smoking cessation support by pharmacists el showing key influences on successful delivery of smoking cessation support by pharmacists [144]. Asthma & COPD Management Furthermore, approximately 50–70% of the patients discontinue their osteoporosis medications within the first year of initiation, which results in increased morbidity and mortality [149]. In October 2010, the US FDA issued a safety communication regarding the risks of atypical fractures of the femur, with bisphosphonates drugs, the safety communication appeared to have influenced osteoporosis utilization in Medicaid recipients [150] (Figure 6). Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Osteoporosis However, different types of estrogen or progestogen, as well as different formulations, doses, timing of initiation, durations of therapy, and patient characteristics, may play different roles in the effects of HRT [158]. Pharmacist- physician collaboration is associated with higher treatment rates of osteoporosis [159]. Physicians and pharmacist should invest time to educate patients about the potential side effects and box warnings of estrogen use. Routine women wellness exams should also be focused on the development of any malignancies or adverse effects of hormone replacement therapy given a positive history. The pharmacist can play an important role at multiple levels: supporting patients in treatment, by providing information on the disease, its treatment, proper use of medication, adherence and persistence, as well as raising awareness for the prevention of osteoporosis and identifying patients at risk [160]. The counseling should include educating and assessing the patient for proper use of estrogen medication therapies as they may be prescribed in many various preparations of oral, transdermal, vaginal insert, and topic vaginal creams for positive patient compliance and adherence to therapy [147]. diseases are ulcerative colitis and Crohn’s disease (CD). Crohn’s disease can cause inflammation in any part of the GIT. Ulcerative colitis (UC) is an idiopathic inflammatory condition of the colon which results in diffuse friability and superficial erosions on the colonic wall associated with bleeding [161]. Although these diseases have undetermined etiology, research advances have outlined some of the pathways by which they occur: a) genetic predisposition associated with the environment induces a disruption of the intestinal microbial flora, b) the epithelial cells and the immune system of the intestine itself determine the risk of developing the disease [162]. Treatment of both, IBD and IBD related pain is challenging. The upholder of IBD therapy includes systemic immunosuppressive drugs, such as corticosteroids, anti-tumor TNF antibodies or immunomodulators. Furthermore, the management of an acute flare differs from the strategies for maintenance of remission [163]. A total of about 66,000 US residents with a new IBD diagnosis each year, since 2015 [164]. Direct costs (including consultations, drugs, hospitalization and surgery) of UC amount to $3.4 to $8.6 billion in the US and €5.4 to €12.6 billion in EU [165]. The prevalent populations of patients with CD or UC in the UC in 2016 are expected to incur lifetime total costs of $498 billion and $377 billion, respectively [166]. Osteoporosis Worldwide, it is estimated that 1 in 3 women above the age of 50 will experience osteoporotic fractures, as well as 1 in 5 men [145]. The pervasiveness of osteoporosis is expected to rise in the US from approximately 10 million people to more than 14 million people by 2020 [146]. In 2015, direct medical costs totaled $637.5 million for fatal fall injuries and $31.3 billion for nonfatal fall injuries. During the same year, hospitalizations cost an average of $30,550 per fall admission, totaling $17.8 billion. By 2025, the cost of fractures in the US is expected to exceed $25 billion each year to treat more than three million predicted fractures [147]. Similar to other chronic diseases, osteoporosis has struggled with suboptimal medication adherence, resulting in an increased risk of fractures and all-cause mortality. Two gaps in osteoporosis management are well documented: (a) most patients at high risk for fracture are not identified for treatment, and (b) adherence to osteoporosis pharmacotherapy is suboptimal [148]. Nearly 50% osteoporosis patients are non-adherent to medications. High patient cost and safety concerns are barriers for nearly 60% patients [152]. Improved osteoporosis medication adherence can reduce osteoporosis-related health care costs by preventing fractures. Persistent pharmacotherapy for osteoporosis is necessary to prevent osteoporotic fractures and to reduce osteoporosis-related health care costs [153]. Treatment strategies of osteoporosis include non- pharmacological treatment - diet rich of calcium and vitamin D, healthy lifestyle, proper exercise plan, and pharmacological therapy [154]. BMD monitoring after initiating anti-osteoporosis therapy in the routine clinical practice setting [155]. Page 7 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 Figure 6: Mapping the factors influencing adherence and adherence strategies onto the medication-taking process in patients with osteoporosis [151]. Figure 6: Mapping the factors influencing adherence and adherence strategies onto the medication-taking process in patients with osteoporosis [151]. Figure 6: Mapping the factors influencing adherence and adherence strategies onto the medication-taking process in patient [151] Hypercalciuria, calcium malabsorption, hyperparathyroidism, hyperthyroidism, vitamin D lack, Cushing’s syndrome, and hypocalciuric hypercalcemia attributed to secondary causes to more than 30% women. Disorders of calcium metabolism and hyperparathyroidism contributed to nearly 80% of the secondary causes [156]. Hormone replacement therapy (HRT) is not first- line therapy position for osteoporosis but is best for prevention of osteoporosis [157]. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Osteoporosis IBD is a relapsing–remitting condition that causes inflammation and ulceration in the bowels, affecting approximately 500,000 people in the UK [167] (Figure 7). The increasing incidence of IBD in developing countries parallels the westernization of diet, which includes higher calorie intake, especially from sugar, refined carbohydrates, animal proteins and ultra-processed foods and a lesser intake of fiber and fruits [168,169]. Incidence rate of IBD is stabilizing in some developed countries; however, the incidence rate is increasing in developing countries Inflammatory Bowel Disease (IBD) Inflammatory bowel disease (IBD) is a group of inflammatory conditions of the colon and small intestine. The two most common Page 8 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 such as Asia and Eastern Europe [170]. Anxiety and depression are the most common psychological disorders in patients with IBD [171]. IBD is associated with significantly increased MI compared with non-IBD patients [172]. Patients with IBD are also at risk for asthma or COPD and bronchiectasis [173,174]. Sacroiliitis, an inflammatory arthropathy associated with ankylosing spondylitis, is found in patients with IBD but may go undiagnosed [175]. IBD patients showed increased risk for lymphoma and biliary cancer [176]. The general goals of treatment are to induce and maintain remission, minimize complications and disease manifestations, and improve overall QoL. Personalized IBD pharmacist adherence counselling, based on the Health Beliefs Model of medication perception, may increase medication adherence. Education using pamphlets and ad hoc physician education improved knowledge but not adherence [177]. IBD patients, mainly those having UC, need medications throughout their life with periodic dosing and occasionally, enemas and infusions may also be required. Treatment without adherence is highly regarded as the significant factor for relapse occurrence [178]. Medication nonadherence in IBD can be improved through a single personalized counseling session by IBD pharmacist adherence counselling (IPAC) intervention, and the benefit was durable for 2 years [179]. In addition to the disease, these patients are also managed with potent medications like steroids and biological agents, which have a host of adverse effects. Thus, the importance of the pharmacist who should be alert for any adverse reaction [180]. Pharmacist- led drug monitoring clinics measure thioguanine nucleotides and thiopurine methyltransferase levels four weeks after treatment with thiopurines is started to optimize outcomes [181]. In the event that insurance coverage cannot be obtained for the selected biologic response modifiers, the pharmacist identifies and discusses alternative options with the GI team; some of these options may include switching to another formulary agent or enrolling patients in medication assistance programs. Once insurance coverage of the medication is secured, the pharmacist educates the patient on self- administration, stability and storage requirements, and potential adverse effects. Additionally, the pharmacist highlights the significance of compliance with laboratory monitoring and reviews the importance of communicating with the GI team in the event of potential infection, worsening disease control, or issues obtaining or using the medication. Figure 7: The seven sections of the 2019 IBD Standards [167]. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Inflammatory Bowel Disease (IBD) The IBD pharmacist has a key role in the management of IBD patients contributing not only to medication monitoring, prescribing, and safety but also allowing greater capacity in the physician’s, often highly stretched IBD clinics [182]. Beyond medication therapy coordination from beginning to end, the pharmacist plays an active role in assisting with medication reconciliation and ensuring patients are current on necessary immunizations [183]. 40% of patients with CD do not respond to treatment with biologics, 30% to 50% achieve complete remission after six months and 30% of patients maintain the response for 12 months with continual treatment. Current strategies to overcome loss of response involve increasing the dose, decreasing the interval between administrations or switching to an alternative agent [181]. Figure 7: The seven sections of the 2019 IBD Standards [167]. Thyroid Disorders diseases of the pituitary or hypothalamus. Common symptoms in hypothyroidism include fatigue, weight gain, cold intolerance, bradycardia, constipation, depression, and skin and hair dryness [184]. Hyperthyroidism, on the other hand, presents with symptoms nearly opposite, including weight loss, heat intolerance, Primary hypothyroidism (due to thyroid gland dysfunction) is the most common, with typical causes being Hashimoto’s disease, or iatrogenic (due to exposure to radiation or thyroid surgery). Secondary hypothyroidism occurs as a result of Page 9 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 tachycardias or palpitations, hyper-defecation, nervousness, and hyper-hydrosis [185]. Iodine nutrition is a key determinant of thyroid disease risk; however, other factors, such as ageing, smoking status, genetic susceptibility, ethnicity, endocrine disruptors and the advent of novel therapeutics, including immune checkpoint inhibitors, also influence thyroid disease epidemiology [186]. The global prevalence of hypothyroidism is 4.6%, with prevalence being more common in women and in older individuals [187,188]. In UK, the prevalence is around 3.5–5% and in USA, 0.2-3.7% [187]. The prevalence of thyrotoxicosis is 10-fold higher in women. Elevated diastolic blood pressure is present in ~30% of patients with overt hypothyroidism and heart failure develops in 6–16% of patients with hyperthyroidism [77]. There is a high (>20%) prevalence of hypothyroidism in patients with T2DM, hypertension, and patients having both [189]. Hypothyroidism can directly cause obstructive sleep apnea (OSA) [190,191], Bruyneel et.al, 2019 reported 16% of OSA patients had a thyroid problem and 8% of these were newly diagnosed with subclinical hypothyroidism (SCH) [192], over 50% of the patients studied did not receive any treatment, found in a recent Egyptian study [193]. Both hypothyroidism and hyperthyroidism were strongly associated with erectile and ejaculatory dysfunction: hypothyroidism with delayed ejaculation, hyperthyroidism with pre-mature ejaculation [194]. Yuan et.al, 2019 reported highest prevalence of vitiligo in subclinical hypothyroidism, among 6 types of thyroid disorders [195]. Subclinical hypothyroidism is most often caused by autoimmune (Hashimoto) thyroiditis [196], who are at higher risk of developing audiological abnormalities as compared to the healthy individuals [197]. Hypothyroidism may also cause alveolar hypoventilation, decreased lung volumes, upper airway obstruction, depression in respiratory stimulus, and respiratory failure [198]. Thyroid dysfunction is a common extrapulmonary manifestation in COPD patients [199]. The American College of Physicians and the American Society of Internal Medicine (ASIM) recommend screening women older than 50 years of age for unsuspected but symptomatic thyroid disease [200]. ation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Thyroid Disorders However, the American Thyroid Association recommends that adults begin screening at the age of 35 years, and repeat screening every 5 years thereafter [201] (Figure 8). Figure 8: Managing Primary Hypothyroidism (Adapted from DeGroot LJ, 2016) [202]. Figure 8: Managing Primary Hypothyroidism (Adapted from DeGroot LJ, 2016) [202]. or with diseases causing chronic malabsorption may require higher doses of levothyroxine. Pharmacists play an important role in ensuring patients with hypothyroidism are managed appropriately. This can include: Clinical pharmacist improves treatment outcome in term of knowledge, attitude and practice scores of the patients after advocacy [203]. Thyroid medications demand careful, patient- specific dosing. Once a physician has changed the dosage of medication to achieve the desired levels of thyroid hormone in a patient, it is decisive to maintain that particular dose and medication until otherwise directed. This is an important counseling point and one that many patients may not be aware of the degree of significance. While there are only a few medicines currently used to treat thyroid disorders, pharmacists can provide education on new or existing medications. Patients should have their thyroid function monitored annually once their treatment is established. It should be noted that pregnant women and patients with nephrotic syndrome • Counselling patients about the timing of levothyroxine dosing and administration with other medicines; • Discussing with patients the appropriateness of combined T4/T3 therapies in light of the large amount of debate on patient forums; • Ensuring thyroid function tests are monitored adequately • Ensuring thyroid function tests are monitored adequately — tests should be assessed four to six weeks after a dose change and at least annually when replacement therapy is stabilized; — tests should be assessed four to six weeks after a dose change and at least annually when replacement therapy is stabilized; Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 preponderance of 13.7%. This means that 150 million Indians need active therapeutic intervention [211]. Canadian Mental Health Association estimated that 500,000 Canadians miss work every week due to MH issues, costing the Canadian economy around $51 billion per year, as depicted by Hayes et.al, 2019 [212]. According to Australian Bureau of Statistics (ABS) National Survey of Mental Health and Wellbeing (NSMHWB) 45% of Australians exposed to a mental disorder in their lifetime, with 20% experiencing a mental disorder in the past [213]. Thyroid Disorders Shpigelman et.al, 2019 reported that individuals with silent psychiatric disabilities have lower levels of self-esteem and body image compared to individuals with visible physical disabilities. Gender, family status and the acerbity level of the infirmity were found to be correlated with self-esteem and body image [214]. Mental disorders and suicide resulting from workload or work-related stress have become major occupational health issues worldwide, particularly in Asian countries [215]. Depression and anxiety in more common chronic physical circumstances such as CHD or diabetes can be correlated with increased mortality, as reported by Uhlenbusch et.al, 2019. Also, depression is associated with an escalation of about 50% in costs of chronic medical ailment [216]. Rokach et.al, 2019 revealed that anxiety and depression lead to sexual dysfunction is between 30% and 70% in sexually active men and women in high-income countries [217]. Despite psychiatry’s present status as the sixth largest medical specialty, the availability of clinicians has not kept up with demand [218]. For over 40 years, clinical pharmacists have handout to these care models in capacities ranging from educator to consultant to provider. Medicines are a major treatment modality of management for many mental illnesses and pharmacists are therefore well positioned to reinforce MH services with the potential to reduce the associated burden of mental disorders [219] (Figure 9). thyroid stimulating hormone values change slowly, so frequent testing is unnecessary; • Ensuring adequate titration of levothyroxine dose after test results are known [204]. • Ensuring adequate titration of levothyroxine dose after test results are known [204]. Whether stopping by in person or picking up the phone to call, patients should be inspirited to reach out to their pharmacists— the most accessible healthcare provider—the next time a question arises about any medication, as it is both the duty and the joy of a pharmacist to provide this service. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Neurological Disorders primary care settings [223]. Transitions in care have the aptitude to be destabilizing periods for many patients and is an area where pharmacist-guided medication therapy management (MTM) has been found to be propitious [224]. The pharmacist interim prescriber clinic was associated with a compelling decrease in mean number of patients seen per month in PES [218]. The clinical pharmacist can make an impact by bettering mild-to-moderate MH conditions, promoting interdisciplinary collaboration, and increasing documentation and backlash that align with published treatment guidelines [206]. Chavez et.al, 2019 reported that pharmacists improved rate of patient interest in behavioral health counseling during the consult and recommending counseling directly to the patient or even initiating the referral themselves [225]. Bingham et.al, 2018 highlighted the value of the pharmacist’s involvement, suggesting the potential for improved nutrition, physical activity, and sleep for patients with MH conditions, at least in the short term [226]. Throughout the US, pharmacists have crafted intercessions designed to prevent, identify, and manage opioid misuse and abuse [227]. Also, Eltorki et.al, 2019 reported that physicians and nurses have mostly positive perceptions and confidence from clinical pharmacists at the psychiatric hospital [228]. In Australia, The Pharmaceutical Society’s Mental Healthcare Framework admits pharmacists as primary health care professionals who have a significant role to play within MH care. Globally, the International Pharmaceutical Federation has urged members to include pharmacists as part of their “human resource development policy” so that “an increase by 20% of service coverage for severe mental disorders can be achieved” [229]. Chronic neurological diseases like Alzheimer’s disease (AD), Parkinson’s disease (PD), dystonia, Amyotrophic lateral sclerosis (ALS, Lou Gehrig’s disease), Huntington’s disease, neuromuscular disease, multiple sclerosis (MS) and epilepsy, to mention only a few — afflict millions of people worldwide and account for tremendous morbidity and mortality [230]. Contributing 11·6% of global DALYs and 16·5% of deaths from all causes, neurological disorders remain the leading group cause of DALYs and the second leading group cause of deaths in the world [231]. Direct and indirect costs for healthcare related to AD are estimated at nearly $500 billion annually [232]. PD is the second most prevalent neurodegenerative disease after AD, affecting approximately 4‐10 million people worldwide, and is expected to double in prevalence by 2030 as the population ages [233]. The incurred medical expenses were approximately $14 billion in 2010 [234], which is $52 billion now, in US only [235, 236]. Psychiatric Disorders The pharmacist interim prescriber clinic was associated with a compelling decrease in mean number of patients seen per month in PES [218]. The clinical pharmacist can make an impact by bettering mild-to-moderate MH conditions, promoting interdisciplinary collaboration, and increasing documentation and backlash that align with published treatment guidelines [206]. Chavez et.al, 2019 reported that pharmacists improved rate of patient interest in behavioral health counseling during the consult and recommending counseling directly to the patient or even initiating the referral themselves [225]. Bingham et.al, 2018 highlighted the value of the pharmacist’s involvement, suggesting the potential for improved nutrition, physical activity, and sleep for patients with MH conditions, at least in the short term [226]. Throughout the US, pharmacists have crafted intercessions designed to prevent, identify, and manage opioid misuse and abuse [227]. Also, Eltorki et.al, 2019 reported that physicians and nurses have mostly positive perceptions and confidence from clinical pharmacists at the psychiatric hospital [228]. In Australia, The Pharmaceutical Society’s Mental Healthcare Framework admits pharmacists as primary health care professionals who have a significant role to play within MH care. Globally, the International Pharmaceutical Federation has urged members to include pharmacists as part of their “human resource development policy” so that “an increase by 20% of service coverage for severe mental disorders can be achieved” [229]. Neurological Disorders Approximately 350,000 individuals in the United States and 2.5 million individuals worldwide have multiple sclerosis. Almost 10% of the cases present before the age of 18 [237, 238]. The global MS drug market was valued at US$16.3 billion in 2016, expecting to reach US$27.8 billion by 2025 [239]. Epilepsy is another most common serious brain conditions, affecting over 70 million people worldwide [240], with an estimated cumulative value of lost economic welfare (VLW) $647.37 billion in 2016 [241]. Clinical pharmacist’s activity can enhance drug therapy’s effectiveness and safety through pharmacotherapy interventions (PIs), thus minimizing DRPs and optimizing the allocation of financial resources associated with health care. Psychiatric Disorders Almost 1 in 5 adults (44 million) in the US exposed to psychiatric illness and distress in a given year, according to the NAMI. A nearly 10 million people suffer a debilitating mental illness that substantially hampers with their QoL [205,206]. Mental and addictive disorders afflicted more than 1 billion people round the globe in 2016. They caused 7% of all global burden of disorder as measured in DALYs and 19% of all years lived with disability [207]. Depression was the dominant cause of disability in the world, and suicide was the 10th leading cause of death in 2015 [206]. Major depressive disorder (MDD) is the fourth cause of infirmity around the world and is estimated to be the second dominant cause of infirmity by 2020 [208]. In EU, factors that had the strongest alliance with depression were chronic diseases, pain, circumspection in daily living, grip strength and cognitive deterioration. The gap in MH service use was nearly 80% [209]. The therapeutic gap in developing countries was 76%–85%, according to WHO. According NMHS, it is 83% in India for mental disorder and 86% for alcohol use disorders [210]. Swaminath et.al, 2019 revealed that mental morbidity above the age of 18 years is 10.6% with a lifetime Figure 9: Role of Pharmacists in Mental health Disorders [220]. Figure 9: Role of Pharmacists in Mental health Disorders [220]. 35% to 70% ceased treatment within 6 months, with up to 25% to 40% of patients having ceased their therapy within the first month [221,222]. Also, Holvast et.al, 2019 depicted non-compliance to antidepressants is high among older patients with depression in 35% to 70% ceased treatment within 6 months, with up to 25% to 40% of patients having ceased their therapy within the first month [221,222]. Also, Holvast et.al, 2019 depicted non-compliance to antidepressants is high among older patients with depression in Antidepressant drug treatment (ADT), alone or in association with psychotherapy, is endorsed by the CANMAT for a minimum duration of 8 months. However, a large proportion of individuals show suboptimal attachment to ADT. In previous studies, more than Page 11 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 primary care settings [223]. Transitions in care have the aptitude to be destabilizing periods for many patients and is an area where pharmacist-guided medication therapy management (MTM) has been found to be propitious [224]. Alzheimer’s disease Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Page 12 of 41 Alzheimer s disease Figure 10: Treatment algorithm for Alzheimer’s disease based on severity of symptoms [242]. Abbreviations: AD, Alzheimer’s disease; ChEI, cholinesterase inhibitor; ER, extended release; XR, extended release. nt algorithm for Alzheimer’s disease based on severity of symptoms [242]. Abbreviations: AD, Alzheimer’s disease; ChEI, itor; ER, extended release; XR, extended release. Figure 10: Treatment algorithm for Alzheimer’s disease based on severity of symptoms [242]. Abbreviations: AD, Alzheimer’s disease; ChEI, cholinesterase inhibitor; ER, extended release; XR, extended release. Figure 10: Treatment algorithm for Alzheimer’s disease based on severity of symptoms [242]. Abbreviations: AD, Alzheim cholinesterase inhibitor; ER, extended release; XR, extended release. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Page 12 o ation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Page 12 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 Patients with AD particularly susceptible to risk of anticholinergic side effects with certain medications and should be assisted by a pharmacist in selecting safe formulation such as OTC product. Pharmacists can also counsel patients and their caregivers on the safe use of alternative medicines that high majority of caregivers had requested relaxing plants and vitamins from the pharmacy for anxiety and insomnia [242]. As AD is a progressive condition, in its early stages, individuals may present with MCI and some 40% of individuals with MCI deteriorated to dementia (estimated out-of-pocket caregiver costs more than 10 billion in 2016 in Canada alone). It is estimated that patients with dementia cost the healthcare system over 300% more than their cognitively intact peers in the same age group [243]. Patient and caregivers’ education, monitoring its progression, becoming familiar with screening tools that can be used in pharmacy practice to assess cognitive function and helping to manage medications for patients in different stages of dementia are essential contribution by the pharmacists [244]. In Germany, pharmacists adapt in identifying problems related to drug administration, adherence, and drug interaction among patients with dementia. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Alzheimer’s disease Meanwhile, in the UK, a trial targeting peoples with dementia initiated on anti-psychotics demonstrated that pharmacist-led medication review successfully limited the prescribing of anti-psychotics to people with dementia because of the increased risk of ADRs. In Japan, study involving hospital pharmacists on donepezil deliberation for patients with AD and their caregivers has heightened medication adherence though this drug could cause insomnia and GI disturbance. In Malaysia, while medication reviews for patients with other chronic diseases have been invoked by hospital pharmacists via medication therapy adherence clinics, this has not been done for AD patients. It is crucial to ensure that the pharmacists are equipped with proficient knowledge on AD because poor management in AD can result in side effects, inappropriate dosing, and non-compliance to medications [245] (Figure 10). Parkinson’s disease Epilepsy Medication nonadherence directly contributes to poor seizure control. A lack of emphasis on correcting poor adherence and failures in patient adherence can result in unwarranted alterations to a patient’s drug regimen [252]. Timely recognition and effective early therapy with first- and second-line antiepileptic drugs (AEDs) may avert unnecessary hospitalizations. Seizures should be recognized and addressed like any other symptom that causes discomfort or reduces QoL. Use of alternative AED administration routes (buccal, intranasal, or subcutaneous) may offer possibilities for effective and individualized AED therapy, even during the last days of life. In hospice or home care, however, also IV treatment is possible via vascular access devices for long-term use. Aggressive unlimited ICU treatment of refractory status epilepticus (SE) in palliative patients is mostly not indicated [253]. There are three types of non-compliance: (i) in medication; (ii) in dietary/exercise; and (iii) in an appointment. First, non-adherence in medication defined as a non-adherence which includes failure to have the prescription dispensed or renewed, the omission of doses, errors of dosage, incorrect administration, errors in the time or frequency of administration, and premature discontinuation of the drug regimen. Second, a non-adherence in dietary/exercise occurs if the patient fails to follow the diet and exercise recommendations. Last, a non-adherence in an appointment occurs if the patient fails to come at clinics for the scheduled check-up [254]. Routine assessment of adherence barriers is imperative from toddlerhood to young adulthood given that the prevalence of barriers and their relative influence on important health outcomes vary by developmental stage [255] (Figure 12). Exhibit 1: Pharmacists’ contribution to PD detection and management [251] • Observe or examine patients who present with or complain of parkinsonian symptoms (tremor, rigidity, bradykinesia, postural hypotension). • Review medications to determine possibility of drug- induced PD. • Refer patients who present with PD symptoms to their family physician for an accurate diagnosis and further referral to neurologist. Therapeutic drug monitoring for AEDs is commonly used to help guide and assist clinicians with optimal dosing in patients. Monitoring serum concentrations can allow clinicians to achieve seizure control while minimizing adverse effects. Established drug levels for various AEDs should be primarily viewed as reference ranges and not therapeutic levels. Pharmacists can play a significant role in optimizing therapy for patients with epilepsy. Patient counseling on the potential adverse effects of AEDs is important. Parkinson’s disease Figure 11: Summary of the identification, diagnosis and prognosis in Parkinson’s disease [250]. Figure 11: Summary of the identification, diagnosis and prognosis in Parkinson’s disease [250]. for PD management [246]. Parkinson’s disease psychosis (PDP) may affect up to 60% of patients with PD over the course of their disease, and is associated with poor prognosis, including increased risks of mortality and nursing home placement [247]. Current management strategies aim to provide symptomatic relief and to PD is a progressive, debilitating neurodegenerative disease that often requires complex pharmacologic treatment regimens. Healthy ageing, primarily when a neurodegenerative disease is present, is possible by applying the correct pharmacological therapy, along with diet and food supplementation often are critical factors Page 13 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 Volume 3-Issue 4 slow down the disease progression. However, no pharmacological breakthrough has been made to protect dopaminergic neurons and associated motor circuitry components [248]. To carry out MTM with PD patients, the pharmacist’s expertise needs to transcend the technical knowledge about the PD pharmacological treatment. It has been estimated that the PD patients’ adherence to the treatment is nearly 40%, compromising the benefits of the therapy. MTM aims at optimizing the pharmacological therapy results, so the pharmacist monitors the results of the treatment prescribed by the doctor and elaborates a healthcare plan to guarantee the treatment’s effectiveness, safety, and convenience, and therefore improve the patients’ QoL [249]. During this time, pharmacists can serve as invaluable partners in the care of people with PD by assisting with complex medication schedules, addressing side effects, assisting with different formulations of medications, obtaining approval for medications on emergency release, ensuring appropriate intake to maximize the absorption of medications, suggesting and monitoring diet and supplementation and guiding the choice of medication based on patient preference, other concurrent medications and medical conditions, and affordability [250]. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Epilepsy Pharmacists involved in the review of prescriptions were able to prevent errors in dose and frequency of administration of AEDs. Patients and family members should be educated on expected CNS and cognitive side effects, potential skin reactions, and the risk for suicidal behavior. Patient education should also address the significance of medication adherence. Patient medication profiles should be investigated for possible drug interactions, and dosage adjustments or alternative agents should be endorsed if necessary. Additionally, pharmacists can advise clinicians on appropriate therapeutic drug monitoring. The development of instruments to guide the care of epileptic patients, such as algorithms and protocols, could assist with the exertion of relevant and effective methods of patient assessment and would also encourage the pharmacotherapeutic monitoring of epileptic patients through pharmaceutical care [257-261]. • Provide written and verbal education to both patients and caregivers on the advantages and disadvantages of various anti- parkinsonian medications. • Develop therapeutic goals with patients with PD, caregivers and other members of the patient’s health care team. • Actively assess anti-parkinsonian medications for appropriateness, effectiveness, tolerability, safety and affordability based on clinical and lifestyle characteristics of the patient. • Assess patient’s adherence to anti-parkinsonian medications. • Assist patients, caregivers and other members of the health care team with adjusting doses of anti-parkinsonian medications. • Assist patients with appropriate dosing, administration and timing of anti-parkinsonian medications. • Recognize real and potential drug-related problems re lated to anti-parkinsonian medications and use full scope of practice available in your jurisdiction to facilitate resolution of drug-related problems in collaboration with patients, caregiv ers and other members of the health care team (Figure 11). • Recognize real and potential drug-related problems re lated to anti-parkinsonian medications and use full scope of practice available in your jurisdiction to facilitate resolution of drug-related problems in collaboration with patients, caregiv ers and other members of the health care team (Figure 11). Page 14 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 Figure 12: Practical approach to modern epilepsy care [256]. Risk of seizure recurrence can be significantly reduced by 30–40% when treatment is immediately introduced once epilepsy is diagnosed. The reduction is effective for the first 2 years. However, long-term outcome is not affected, and in the UK study around 75% were seizure-free at 2 years, no matter whether treatment was delayed until after the second or third seizure or not. Epilepsy Overall, if patients’ relapse, 90% do so within the first 2 years. Figure 12: Practical approach to modern epilepsy care [256]. Risk of seizure recurrence can be significantly reduced by 30–40% when treatment is immediately introduced once epilepsy is diagnosed. The reduction is effective for the first 2 years. However, long-term outcome is not affected, and in the UK study around 75% were seizure-free at 2 years, no matter whether treatment was delayed until after the second or third seizure or not. Overall, if patients’ relapse, 90% do so within the first 2 years. ation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Fibromyalgia billion in US [262], [264,265]. Work disability was found to be 35% in US and Australia and 30% reported in Canada due to FM [266]. FM is more common in female compared to male, with a ratio of 2:1 [267] or 3:1 [268], although other studies reveal 85%-90% FM patients are middle aged women [269,270]. 30-50% of FM patients have anxiety and/or depression at the time of investigation [271], while patients have a lifetime history of depression (50-75%) [272] and depressive disorders (13%–63.8%) [273]. Physical complications of FM are indicated in (Figure 13). Fibromyalgia (FM) is an idiopathic chronic condition that causes comprehensive musculoskeletal pain, hyperalgesia and allodynia, afflicting 2.10% (a total of 4% female and 2-5% male) of the world’s population, 2.3% of the European population, 2.4% of the Spanish population [262,263]. In France, work productivity loss accorded almost 90% of the total costs incurred by patients with FM, with an economic cost of 13000 million euros yearly which is around $100 Figure 13: Physical complications of Fibromyalgia (Source: creakyjoints.org). The APS guidelines cite strong evidence supporting the use of reuptake inhibitors (SNRIs), fluoxetine, tramadol, and pregabalin The APS guidelines cite strong evidence supporting the use of TCAs and moderate evidence for serotonin and norepinephrine reuptak [274]. A The APS guidelines cite strong evidence supporting the use of TCAs and moderate evidence for serotonin and norepinephrine reuptake inhibitors (SNRIs), fluoxetine, tramadol, and pregabalin [274]. A nearly 70% German FM patients used thermal baths, reuptake inhibitors (SNRIs), fluoxetine, tramadol, and pregabalin [274]. A nearly 70% German FM patients used thermal baths, The APS guidelines cite strong evidence supporting the use of TCAs and moderate evidence for serotonin and norepinephrine reuptake inhibitors (SNRIs), fluoxetine, tramadol, and pregabalin [274]. A nearly 70% German FM patients used thermal baths, Page 15 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 35.2% use alternative therapeutics such as homeopathy, dietary supplements, and 18.4% use introspective or meditation-based exercises such as yoga or Tai chi [275]. Low FODMAP was found be compelling in QOL, quality of sleep, anxiety and depression and inflammatory biomarkers in FM patients [276-281]. Cognitive behavioral therapy (CBT) interventions may slow or reverse cortical gray matter atrophy, diminishes circulating proinflammatory cytokines (IL-6, IL-8, and TNF-α level) of FM patients, pain symptoms and pain perceptions, helps FM patient having fear of pain, anxiety, depression and sleep disturbances [282-287]. Recurrent Urinary Tract Infections RUTIs without use of antibiotics within the past years, interest probiotics has increased over the years. Reviews by Akgül et.al, 2018 concluded that evidence of probiotic appliance in UTIs is not yet sufficient to recommend use of probiotics [316]. Different vaccines based on the whole cells (killed or live-attenuated vaccines) and antigens (subunits, toxins and conjugated vaccines) have been evaluated against UTIs pathogens, as reported by Asadi et.al, 2019 [317]. Substantial efforts have been expended in development of endogenous antimicrobial peptides (AMPs) as new therapeutic options suitable in the treatment of drug-resistant microbial infections. For example, Wnorowska et.al, 2019 reported that combination of natural peptide LL-37 with synthetic analogs might be a potential solution to treat UTIs caused by drug-resistant bacteria [318]. UTI is one of the most prevalent diseases with diverse etiological agents annually affecting 250 million and causes death of 150 million people overall. Financial burden of UTIs exceeds $3.5 billion in US alone, while over half of the anti-infection agents prescribed for a suspected UTI in older adults being considered unnecessary. Surprisingly, nosocomial UTIs account for nearly 40% of all hospital acquired infections and around half of UTI in children are missed. Sexual intercourse ≥3 times/week was associated with greater frequency of UTIs. Close proximity of the urethral meatus to the anus and shorter urethra, is a likely factor in women (Figure 15). Genital hygiene practices such as urination after and washing genitals after intercourse, male partner washing genitals before intercourse, taking baths, frequent replacing of underwear, controlled frequency of sexual intercourse, and washing genitals from front to back were associated with a reduced frequency of UTIs [319-322](Figure 15). Between 50% and 60% of adult women will face at least one UTI event in their life, and close to 10% of postmenopausal women indicate that they had a UTI in the previous year, according to Medina et.al, 2019 [292]. Recurrent UTIs (RUTIs) are mainly precipitated by reinfection of the same pathogen mostly caused by frequent sexual intercourse, heterosexual anal intercourse (without circumcision and a condom), different sexual partners (Each sex partner shares his/her UGT microbiota with the other), utilization of spermicide and another sexual partner, sexual intercourse with addicted partners, sexual intercourse with sex workers, sexual intercourse with online dating friends, sexual intercourse with another sex partner for 2 months. [293-303]. Fibromyalgia Physiotherapy and acupuncture, both are compelling, not found to be more beneficial than each other, longer post-treatment follow-up may help arbitrate the superior treatment option [288]. Low to moderate intensity endurance and strength training are strongly suggested in FM patients. Strength training alleviates pain, fatigue, number of tender points, depression, and anxiety, with increased functional capacity and QoL. Exercise activates the endogenous opioid and adrenergic systems but does not consistently alleviate pain in FM patients [264], [275]. Pharmacists can participate in ongoing follow-up to monitor patients’ responses to both pharmacologic and nonpharmacologic intercession. The assessment should consider both the impact on symptoms of FM and on the patient’s function. If patients do not achieve a satisfactory treatment response or experience intolerable adverse events, pharmacists can collaborate with other members of the healthcare team to arbitrate whether a trial of another medication would be appropriate. Because emerging evidence indicates FM has many possible root causes, lack of treatment response to one medication does not affect whether another medication will be effective. Pharmacists can assess and triage patients who present to the pharmacy with a history of symptoms that are associated with FM [289,290] (Figure 14). Figure 14: An integrated approach for FM management [291]. Figure 14: An integrated approach for FM management [291]. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Page 16 of 41 Page 16 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Recurrent Urinary Tract Infections Many other factors have been thought to predispose women to RUTIs, such as voiding patterns pre- and post-coitus, wiping technique, wearing tight undergarments, deferred voiding habits and vaginal douching; nevertheless, there has been no proven association [304]. Obesity was found to be associated with RUTIs in premenopausal women [305]. Several other risk factors are associated with cystitis, a prior UTI, vaginal infection, diabetes, and genetic susceptibility [306]. Pathogens responsible most commonly are Escherichia coli, Klebsiella pneumoniae, Proteus mirabilis, Enterococcus faecalis and Staphylococcus saprophyticus [293], [306-310]. The role of dietary pattern in RUTI is also not clear. Increasing fluid consumption is often recommended for patients with UTI [301], [307-314] but there has been no clear clinical evidence to support this recommendation [315]. After a first episode of a UTI, 27% of women have a confirmed recurrence within the next 6 months [292]. Although there have been few studies on the determent of Figure 15: Bacterial spreads to the urinary bladder due to close proximity to anus, causing inflammation (Source: fabHow). Figure 15: Bacterial spreads to the urinary bladder due to close proximity to anus, causing inflammation (Source: fabHow Pharmacist management of uncomplicated UTI is effective, safe, and patient satisfaction appears very high. In Quebec, pharmacists can prescribe for UTI in women if there has been an interpretation of UTI and a resulting prescription to treat it in the past year. In Saskatchewan, prescribing for UTI in women has been suggested, but is not yet approved. And in Alberta, pharmacists who have Additional Prescribing Authorization are able to prescribe for UTI if it is within their scope of practice and if, through their own estimate or collaboratively with another health professional, it is determined that treatment is appropriate. If a patient has been prescribed an antibiotic for a presumed UTI, the pharmacist should also apprise here to confirm the aptness of treatment. Patients should be asked about symptoms such as dysuria, frequency, urgency, suprapubic pain, flank pain or tenderness, fever, or hematuria in non-catheterized patients. In catheterized patients, symptoms suggestive of UTI include fever, rigors, flank pain or Page 17 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 tenderness, acute hematuria, purulent discharge from catheter site and new or deteriorated mental status (in the presence of leukocytosis) with no attributable alternative cause. In cases of complicated UTI or pyelonephritis, a urine culture should always be sent. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Page 18 of 41 Recurrent Urinary Tract Infections In this model of care delivery, a pharmacist can provide crosscutting medication management services by communicating bidirectionally between the dialysis unit team and the patient’s care providers, family, and payers, closing the loop of communication, improving medication list accuracy, and identifying and resolving MRPs. The pharmacist in this model could function like a consultant, providing medication management services to patients in several dialysis units. Kidney disease is a global public health problem, affecting over 750 million persons worldwide [326]. Over 30 million American adults may have CKD [327] and 1.8 million in UK [328]. The global estimated prevalence of CKD is 13.4% (11.7-15.1%), and patients with ESRD needing renal replacement therapy is estimated between 4.902 and 7.083 million [329]. In 1990, renal failure was considered as the 27th mortality factor in the world and reached the 18th rank in 2010. According to the Centers for Disease Control and Prevention, in 2014, more than 20% of people with serious hypertension suffered from chronic kidney disease and were at risk for ESRD. On the other hand, hypertension is present in over 90% of individuals with advanced kidney disease [327], [330]. In all developed countries and in many developing countries, diabetes and hypertension are considered as the main cause of CKD. Dialysis remains the most commonly employed treatment option for patients with ESRD because not all patients are medically suitable for kidney transplantation, and the demand for kidneys far exceeds the supply. The total cost of dialysis is mostly composed of the costs of the treatment itself (including disposables, machines, accommodation, electricity, water and human resources) and the costs of medications, transportation, complications, additional hospital admissions and interventions [331]. Total annual cost of CKD far exceeds $5 billion in Korea [332], $114 billion in US [327] and Canada $40 billion [333]. The prevalence of CKD and ESRD is projected to rise by up to 80% by 2020 due to ageing population and the rising prevalence of diabetes in Australia [334]. Multidisciplinary healthcare teams of physicians, nurses, dieticians, and clinical pharmacists share the goal of disease aversion progression and managing comorbid conditions in CKD and ESRD patients. At the initial appointment, the patient meets with all team members for a need estimate. The pharmacist obtains the patient’s medication and allergy histories. Recurrent Urinary Tract Infections Pharmacists who are unable to order urine cultures should advocate for or make advises to have them done when they are pertinent and should discourage the sending of urine cultures when they are not indicated. Pharmacists should also familiarize themselves with the local antibiogram, as this will assist in the selection of empiric therapy. They should keep in mind, however, that resistance rates portrayed in hospital antibiograms may not be representative of the expected resistance patterns of uncomplicated infections, as these antibiograms are often heavily influenced by patients with complicated and nosocomial infections, which tend to be more resistant in nature. Medically underserved populations pose a unique challenge for providing effective patient education, compared with the general population. Patient adherence to provider recommendations is key to achieving therapeutic success. Also, as there are many modifiable risk factors for developing UTIs, it is important to effectively communicate these factors to patients to prevent recurrence and subsequent readmissions. As pharmacists already play a key role in medication counseling prior to hospital discharge, this presents an opportunity to incorporate patient education on infection management and prevention by pharmacists as part of ASPs [323-325] (Figure 16). infections, as these antibiograms are often heavily influenced by Figure 16: Proposed algorithm for assessment and management of urinary tract infection by Beahm et.al, 2017 [323]. Figure 16: Proposed algorithm for assessment and management of urinary tract infection by Beahm et.al, 2017 [323]. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Page 18 of 41 Page 18 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 Renal Failure Figure 17: Dialysis facility-centered medication management services model [340]. In this model of care delivery, a pharmacist can provide crosscutting medication management services by communicating bidirectionally between the dialysis unit team and the patient’s care providers, family, and payers, closing the loop of communication, improving medication list accuracy, and identifying and resolving MRPs. The pharmacist in this model could function like a consultant, providing medication management services to patients in several dialysis units. Figure 17: Dialysis facility-centered medication management services model [340]. Recurrent Urinary Tract Infections The pharmacist also educates the patient about the importance of medication management in chronic kidney disease, adherence to drug regimens, and the potential risks of nephrotoxic medications. At subsequent visits, the patient is seen by specific team members, as appropriate to the person’s laboratory results or as requested by the patient or other team members. As specialists in pharmacotherapy, clinical pharmacists are routinely involved in patient care and interact with other health care professionals, addressing multiple, often unmet needs for pharmacotherapy optimization. Ideally, this happens through a preventive, rather than a reactive, approach [335]. The pharmacist continually assesses drug therapy for efficacy and adverse effects, using laboratory results, the results of physical examinations performed during clinic visits, and information obtained during phone conversations with the patient. The main areas of focus for the CKD pharmacist are management of anemia, monitoring for hypertension, reduction of cardiovascular risk, adjustment of doses, and recommendations relating to medications that are eliminated renally. Drug coverage and supply issues involve communication with community pharmacists and the office of the provincial health plan. The CKD pharmacist deals with drug information requests from team members and other health care providers. Management of inventory and reporting (to the public Kidney disease is a global public health problem, affecting over 750 million persons worldwide [326]. Over 30 million American adults may have CKD [327] and 1.8 million in UK [328]. The global estimated prevalence of CKD is 13.4% (11.7-15.1%), and patients with ESRD needing renal replacement therapy is estimated between 4.902 and 7.083 million [329]. In 1990, renal failure was considered as the 27th mortality factor in the world and reached the 18th rank in 2010. According to the Centers for Disease Control and Prevention, in 2014, more than 20% of people with serious hypertension suffered from chronic kidney disease and were at risk for ESRD. On the other hand, hypertension is present in over 90% of individuals with advanced kidney disease [327], [330]. In all developed countries and in many developing countries, diabetes and hypertension are considered as the main cause of CKD. Dialysis remains the most commonly employed treatment option for patients with ESRD because not all patients are medically suitable for kidney transplantation, and the demand for kidneys far exceeds the supply. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Pag Recurrent Urinary Tract Infections The total cost of dialysis is mostly composed of the costs of the treatment itself (including disposables, machines, accommodation, electricity, water and human resources) and the costs of medications, transportation, complications, additional hospital admissions and interventions [331]. Total annual cost of CKD far exceeds $5 billion in Korea [332], $114 billion in US [327] and Canada $40 billion [333]. The prevalence of CKD and ESRD is projected to rise by up to 80% by 2020 due to ageing population and the rising prevalence of diabetes in Australia [334]. Multidisciplinary healthcare teams of physicians, nurses, dieticians, Page 19 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 new infections in 2016 and 2017 have remained constant at 1.8 million cases, 10% of whom were children. About 2.1 million new HIV-positive cases were diagnosed in 2015 only [341,342]. South and Southeast Asia placed the third rank of the highest HIV/AIDS incidence after the states of sub-Saharan Africa and North Africa in 2012, as 3.9 million people were living with HIV and 270,000 people becoming newly [343]. Interestingly, HIV prevalence among prisoners has been reported to vary between different countries from 0%–2% in Australia to 2% in America, 11% in Latin American countries, 10% in the Middle East, and 20% in African countries [344]. In China, men who have sex with men (MSM) transmission has surpassed both injection drug use and blood donors and has become the major HIV transmission route, rose from 44.3% in 2008 to 63.5% in 2012 and to 71.3% in 2015 in Nanjing [345]. Another study revealed that married Indian women who experience physical and sexual violence from husbands face a significantly increased risk of HIV infection as compared with women who are not thus abused [346]. However, the HIV epidemic has cost the global economy over half a trillion dollars so far in the 21st century (between 2000 and 2015), according to a new scientific study [347] (Figure 18). health department) of vaccinations performed in the clinic are also the pharmacist’s responsibility. Other members of the team participate in the patient education classes, which are offered throughout the year to provide patients with self-management information [336]. One large cross-sectional study observed associations between uncontrolled hypertension and CKD patients with greater medication nonadherence. HIV/AIDS There are nearly 40 million HIV-positive people in the world, while the developing countries contain 95% of them. It is estimated that 14 thousand individuals are being infected with the HIV each day worldwide and more than 30 million people have lost their lives because of the AIDS, since the first HIV positive patient was identified. Whilst AIDS-related deaths and total new infections have fallen by 34% and 18%, respectively, since 2010, worldwide Figure 18: Diary of key sentinel timeline events from discovery to evolution of therapy of HIV-1. AZT, Zidovudine [348]. Ever since the first report by the New York Times on a mysterious illness in 1981 and the identification of HIV-1 as the cause of this illness in 1983, significant strides have been made in the treatment and management of HIV-1. Since the introduction of combination antiretroviral therapy in the mid- 1990s, there have been >30 agents approved for the treatment of HIV-1–positive individuals. Figure 18: Diary of key sentinel timeline events from discovery to evolution of therapy of HIV-1. AZT, Zidovudine [348]. Ever since the first report by the New York Times on a mysterious illness in 1981 and the identification of HIV-1 as the cause of this illness in 1983, significant strides have been made in the treatment and management of HIV-1. Since the introduction of combination antiretroviral therapy in the mid- 1990s, there have been >30 agents approved for the treatment of HIV-1–positive individuals. sharing) and to refer them for counseling and testing. Needlestick injuries are fairly common occurrences in the health care field. Guidelines are available from the US Public Health Service for the management of occupational exposure to HIV, HBV, and HCV and recommendations for postexposure prophylaxis. These guidelines are updated regularly and include such topics as implementation of a bloodborne pathogen policy, treatment recommendations after needlestick injuries, monitoring for adverse effects, and laboratory testing to monitor for seroconversion. Health care providers should be knowledgeable about the symptoms and signs of acute retroviral syndrome, characterized by fever, malaise, lymphadenopathy, and skin rash, which occur within the first few weeks after HIV infection. This presentation occurs before the antibody test results become positive. Current guidelines suggest that patients with recently acquired HIV infection might benefit from antiretroviral drugs and may be candidates for clinical drug trials. Recurrent Urinary Tract Infections An earlier study reveals that clinical pharmacists’ interventions reduced DRPs, gaps between clinical interventions and hospital admissions, length of hospital stays, number of transplant rejections, improved outcome of renal function and incidence of ESRD or death [337-339] (Figure 17). Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. HIV/AIDS Anyone with an acute Exhibit 2: Guidelines for HIV Prevention & Treatment [485] HIV infection is diagnosed by tests for antibodies to HIV-1 and HIV-2. Antibody testing starts with a sensitive screening test such as enzyme-linked immunosorbent assay (ELISA). Reactive screening tests must be confirmed by a supplemental test, such as Western Blot, or by immunofluorescence assay. If confirmed by a supplemental test, a positive test indicates that a person is infected with HIV and is capable of transmitting the virus to others. HIV is detectable within 3 months after infection in at least 95% of patients. Although a negative antibody test result indicates that a patient is not infected, it cannot exclude the possibility of a recent infection. Patients with a new diagnosis should receive initial behavioral and psychosocial counseling on-site. Providers should be alert for medical or psychosocial conditions that might require immediate attention. Patients should be encouraged to notify their partners (including sex partners and needle Page 20 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 [354] and greater adherence to and persistence with ART adherence (early studies indicated 95% adherence was needed for viral suppression) [355]. In a US Department of Veterans Affairs Medical Center adherence study, the results demonstrated a 10% increase in adherence associated with a viral load decrease [356]. Pharmacist counseling of patients with HIV includes several key aspects: an evaluation to ensure appropriate dosage, patient administration counseling, ART adherence education, medication interactions, and possible adverse effect management. If lab data are available, monitoring of CD4+ cell counts, and HIV RNA viral load could also be reviewed [357]. One common reason patient cite for missing medication doses is forgetfulness. Reminder strategies such as pillboxes, calendars, or other medication planners can help improve adherence [358]. According to an analysis of the retrospective data, more than 50% of HIV-positive patients continued their regimens at home, and the remainder had provider support for not having a home regimen. A pharmacy resident or a student trained in medication reconciliation could be appointed with gathering patient information through insurance claims, outpatient pharmacies, and patient or caregiver interviews [359]. This visit gives pharmacists the opportunity to detect any problems in adherence and to suggest ways of managing adverse effects or other problems to patients before the drugs have to be discontinued (Figure 19). HIV infection should be referred immediately to an appropriate HIV care provider. HIV/AIDS Once detection has been confirmed, this should prompt education efforts to reduce the spread of HIV to others. This includes counseling patients on high-risk behaviors (eg, sharing of intravenous needles, unprotected sexual behavior). The core goals of management remain maximal suppression of viral replication and promotion of immune reconstitution through combination antiretroviral therapy (ART). In both the unadjusted and the adjusted analyses, patients with pharmacist-assisted ART management achieved more rapid viral suppression than patients managed without such assistance [349,350]. Secondary goals of therapy include promoting long-term adherence, avoiding drug interactions, minimizing toxic effects, simplifying treatment regimens, decreasing drug costs, managing comorbid conditions, and preventing transmission of HIV by achieving undetectable viral load. Pharmacists’ involvement in the care of HIV-positive patients has been associated with improved patient outcomes, including enhanced, reduced pill burden and dosing frequency, greater increases in CD4 cell counts, higher rates of viral suppression, and decreases in medication errors [349]. Pharmacist involvement ensured identification, prevention, and solving of DRPs [351], increased CD4+ T-lymphocyte counts [352], reduced cost associated with medicine, doctor/hospital appointments, laboratory tests, and hospitalizations [353], reduced pill burden and dosing frequency Figure 19: Integrated AYA-centered Framework for differentiated HIV prevention and Treatment Services [360]. Abbreviations: adolescents and young adults (AYA); Pre-Exposure Prophylaxis (PrEP). Figure 19: Integrated AYA-centered Framework for differentiated HIV prevention and Treatment Services [360]. Abbreviations: adolescents and young adults (AYA); Pre-Exposure Prophylaxis (PrEP). to more than 50% of cancer patients; with more than half of global cancer-related mortalities occurring in Asia alone [368]. The 3 most prevalent cancers in 2019 are prostate, colorectal and skin melanoma among males, and breast, uterine corpus, and colorectal among females [369]. Overall cancer death rates declined faster in blacks than whites in US, although rates for cancers of the breast, uterine corpus, and pancreas are increasing in black people [370]. Also, black men have a 70% higher prostate cancer and a more than 2-fold higher mortality rate compared with white men [371]. The n: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. 00566. DOI: 10.33552/ABEB.2019.03.000566. Lung cancer American Cancer Society estimated that in 2018 lung and bronchus cancers would be responsible for 234,030 new cases which represent 14% of all new cancer cases and 1.5 million deaths [397]. Also, Global lung cancer deaths were estimated at 1.7 million in 2015, contributing to approximately 20% of all cancer-related deaths. In 2004, lung cancer was associated with highest treatment cost at $4.2 billion [398], which was $ 268 billion in 2010 [399]. The overall economic impact of lung cancer in Europe is substantial, it was found more than €100 billion, when costs related to disability and premature mortality are considered as well [400]. Non-small- cell lung cancer (NSCLC) is the most common type of lung cancer, accounting for about 80%-85% of all cases [401,402]. More than half of the NSCLC cases are diagnosed at an advanced stage (stages III and IV) [403,404]. Smoking causes at least 80% of lung cancer deaths [405]. Lin et.al, 2019 concluded association between lung cancer incidence and increased reliance on coal for energy generation [406]. Other possible reasons are exposure to indoor and outdoor air pollution, exposure to radiation, and occupational exposure to agents such as asbestos, nickel, chromium, and arsenic [407]. India currently produces approximately 730 million kg of dry tobacco. Its use is associated with several types of cancers which are very common in LMICs contributing to 50% of all cancers in men and 20% in women. Globally, 90% of lung cancer deaths in men and 80% in women are attributable to smoking. By 2030, tobacco use is estimated to kill around 10 million people a year. The epicenter of this tobacco epidemic is LMICs with 70% of estimated deaths and 80% of the 1 billion smokers in the world coming from here [372] (Figure 20). Chemotherapy remains the indispensable choice for the vast majority of patients with advanced NSCLC, including primary tumors and lung metastases. Use of the pulmonary route is a promising way to decrease the severe systemic toxicities associated with chemotherapy. Inhalation allows the administration of high drug doses directly to lung tumors without prior distribution in the organism [409]. Pharmacists are expected to assist patients with completion of adjuvant chemotherapy. Cancer Care So, proper and up-to-date knowledge is necessary in using alternative treatment options as patients who received alternative medicines had a 2.5 greater risk of dying compared to those who received conventional cancer treatment [390]. In oncology, a retrospective observational cohort study of a pharmacist-managed oral chemotherapy management clinic that provided services (including education on oral chemotherapy agents, concurrent medications, symptom management, and insurance assistance) to cancer patients for up to three months found that the program led to reductions in rates of adverse effects, non-adherence, drug interactions, and medication errors over time, as well as potential cost avoidance or cost savings [391]. Patients also appear to value pharmacist-led interventions in the oncology setting. Based on a survey of outpatients, 86% felt it important to discuss their initial treatment with a pharmacist, while 76% requested pharmacy follow-up at future visits; patients were interested in visiting a pharmacist regularly while receiving chemotherapy, and may be willing to pay for pharmacy counseling services [392]. However, inpatient clinical pharmacy services positively impact on patient care. An experienced cancer pharmacist possesses the specific knowledge about the medications used in the care of cancer patients and has been shown to obtain more accurate medication histories than doctors or nurses in the hospital setting [393]. In the UK, pharmacist independent prescribers can prescribe for any condition within their clinical competence including systemic anti-cancer therapy [394]. Greater training of pharmacists may lead to greater therapeutic interventions and interactions with patients regarding their treatment regimens. Pharmacist intervention aids completion of planned chemotherapy and management of treatment-related adverse events [395]. Ambulatory care clinical pharmacists are often responsible for managing comorbid conditions of patients with cancer. There are over 100 types of cancers, located in different organs and sub- Cancer Care In 2019, 1,762,450 new cancer cases and 606,880 cancer deaths are projected to occur in the United States [361]. Globally, cancer is responsible for at least 20% of all mortality [362], 18.1 million new cancer, 9.5 million death in 2018 [363,364], the 5- year prevalence of 43.8 million [365], is predicted to rise by 61.4% to 27.5 million in 2040 [366]. Approximately 70% of deaths from cancer occur in LMICs [367]. Asia, Africa, and Latin America are collectively home Page 21 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 cost of delivering cancer treatment is estimated to rise globally with a projected total spending of $458 billion by 2030 [372]. However, the financial burden stems from employment loss, cost of care even when patients don’t require chemotherapy, out of pocket costs’ opportunity costs of informal care time and can continue long after the death of the patient [373,374]. Studies say 46 billion in productivity lost in major emerging economies due to cancer [375] and economic costs of tobacco-related cancers exceed USD 200 billion each year [376]. Also, cancer causes 2.6 times more likely to file for bankruptcy than the non-cancer people [377]. Cancer trends in young adults, reflect recent changes in carcinogenic exposures, which could foreshadow the future overall disease burden [378]. Cancer cachexia (anorexia, weight loss, loss of adipose tissue and skeletal muscle) is reported in 30%-80% cancer patients and causes 20% of all cancer deaths [379]. Worldwide, some 60%- 80% people depend on alternative medicines [380-382], which is also true for nearly 40% to 70% European [383,384], 50% Italian, 40% Korean, 30% British [385] and up to 87% of Australian cancer patients [386]. Use of unapproved/unlabeled/wrong herbal treatment is not uncommon [387,388] and also drug interactions reported phyto-therapeutics in oncology [389]. So, proper and up-to-date knowledge is necessary in using alternative treatment options as patients who received alternative medicines had a 2.5 greater risk of dying compared to those who received conventional cancer treatment [390]. In oncology, a retrospective observational cohort study of a pharmacist-managed oral chemotherapy management clinic that provided services (including education on oral chemotherapy agents, concurrent medications, symptom management, and insurance assistance) to cancer patients for up to three months found that the program led to reductions in rates of adverse effects, non-adherence, drug interactions, and medication errors over time, as well as potential cost avoidance or cost savings [391]. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Cancer Care Patients also appear to value pharmacist-led interventions in the oncology setting. Based on a survey of outpatients, 86% felt it important to discuss their initial treatment with a pharmacist, while 76% requested pharmacy follow-up at future visits; patients were interested in visiting a pharmacist regularly while receiving chemotherapy, and may be willing to pay for pharmacy counseling services [392]. However, inpatient clinical pharmacy services positively impact on patient care. An experienced cancer pharmacist possesses the specific knowledge about the medications used in the care of cancer patients and has been shown to obtain more accurate medication histories than doctors or nurses in the hospital setting [393]. In the UK, pharmacist independent prescribers can prescribe for any condition within their clinical competence including systemic anti-cancer therapy [394]. Greater training of pharmacists may lead to greater therapeutic interventions and interactions with patients regarding their treatment regimens. Pharmacist intervention aids completion of planned chemotherapy tissues and originating from different cell types. Some cancer types (e.g., colon, breast, and non-Hodgkin’s lymphoma) contain even more specific classifications based on their molecular subtypes. Despite this complexity and variability, most types of cancer are treated with the same generic therapies [396]. cost of delivering cancer treatment is estimated to rise globally with a projected total spending of $458 billion by 2030 [372]. However, the financial burden stems from employment loss, cost of care even when patients don’t require chemotherapy, out of pocket costs’ opportunity costs of informal care time and can continue long after the death of the patient [373,374]. Studies say 46 billion in productivity lost in major emerging economies due to cancer [375] and economic costs of tobacco-related cancers exceed USD 200 billion each year [376]. Also, cancer causes 2.6 times more likely to file for bankruptcy than the non-cancer people [377]. Cancer trends in young adults, reflect recent changes in carcinogenic exposures, which could foreshadow the future overall disease burden [378]. Cancer cachexia (anorexia, weight loss, loss of adipose tissue and skeletal muscle) is reported in 30%-80% cancer patients and causes 20% of all cancer deaths [379]. Worldwide, some 60%- 80% people depend on alternative medicines [380-382], which is also true for nearly 40% to 70% European [383,384], 50% Italian, 40% Korean, 30% British [385] and up to 87% of Australian cancer patients [386]. Use of unapproved/unlabeled/wrong herbal treatment is not uncommon [387,388] and also drug interactions reported phyto-therapeutics in oncology [389]. Lung cancer A clinical trial directly comparing preoperative with postoperative chemotherapy for early-stage NSCLC demonstrated an attrition rate of 34% for postoperative chemotherapy within the protocol-defined window of 6 to 7 weeks owing to a wide range of complicating factors [410]. Pharmacist intervention aids completion of planned chemotherapy and management of treatment-related adverse events [395]. More than 80% lung cancer in UK are preventable. One-fifth of lung cancers occur in people who have never smoked. More than 90% of lung cancer cases are symptomatic at diagnosis, with a cough being most common. Pharmacists are ideally placed to detect Page 22 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 early signs and symptoms of lung cancer, offer advice and refer people to a GP when appropriate. Pharmacists and pharmacy staff who have the opportunity to talk to patients about the signs and symptoms of lung cancer. With practical guidance, pharmacists can appropriately use immune checkpoint inhibitors for stage III NSCLC, educate patients and other clinicians about immunotherapy, and monitor and manage immune-mediated adverse events. Adding a specialist cancer pharmacist to the outpatient lung cancer team led to significant improvements in patient medication adherence. Both patients and GPs were highly satisfied with the service. Medication misadventure and clinic attendances were reduced [393], [411,412]. Figure 20: Algorithm for the management of EGFR-mutated NSCLC [408]. Molecular testing is recommended at multiple points along the treatment pathway for this patient population. Additional challenges arise because of the known benefits of treating oligometastatic disease, especially in the brain, and treating beyond classical radiologic progression. Clinical trials should be considered at all steps along the treatment path. (a) If oligo-progression (such as isolated brain metastasis) occurs, consider local therapy and continuation of tyrosine kinase inhibitor. (b) Re-biopsy currently required. Biopsy of growing lesion is recommended if possible. Testing can be performed on histology or cytology cell block. Consider testing plasma free DNA as an alternative if clinically available. (c) Clinical trials are preferred at all treatment steps, if available. EGFR = epidermal growth factor receptor; TKI = tyrosine kinase inhibitor. Figure 20: Algorithm for the management of EGFR-mutated NSCLC [408]. Molecular testing is recommended at multiple points along the treatment pathway for this patient population. Additional challenges arise because of the known benefits of treating oligometastatic disease, especially in the brain, and treating beyond classical radiologic progression. Lung cancer Clinical trials should be considered at all steps along the treatment path. (a) If oligo-progression (such as isolated brain metastasis) occurs, consider local therapy and continuation of tyrosine kinase inhibitor. (b) Re-biopsy currently required. Biopsy of growing lesion is recommended if possible. Testing can be performed on histology or cytology cell block. Consider testing plasma free DNA as an alternative if clinically available. (c) Clinical trials are preferred at all treatment steps, if available. EGFR = epidermal growth factor receptor; TKI = tyrosine kinase inhibitor. patients with CML have substantially improved due to advances in drug development and rational treatment intervention strategies [421]. Allowed costs for leukemia patients averaged almost $157,000 in the year after diagnosis, with costs for AML almost tripling that amount, according to a new report from the Leukemia & Lymphoma Society (LLS) [422]. Malignant blood disorders cost €12 billion ($13 billion) in EU in 2012, with more than 60% of that amount spent on healthcare costs and nearly a third spent on drugs [423] (Figure 21). Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Leukemia There is no standard of care for the treatment of relapsed or refractory AML. A clinical trial is always the preferred option. The above algorithm is based on current clinical practice and will hopefully change in coming years due to improvements. In particular the targeted and immunotherapeutic agents detailed in this review may ultimately have utility in (1) initial therapy; (2) as a bridge to, or as a temporizing measure before, allo-HSCT; and/or (3) as part of consolidative therapy. Achievement of a complete remission (CR) prior to undergoing alloHSCT is associated with best survival and is generally preferred. The survival of patients with residual disease undergoing alloHSCT varies considerably however and this therapy may be a reasonable option in selected patients not in CR. HMA: Hypomethylating agent. LDAC: Low-dose cytosine arabinoside. Allo-HSCT: Allogeneic Hematopoietic Stem Cell Transplant. Figure 21: Treatment algorithm for patients with RR-AML [424]. There is no standard of care for the treatment of relapsed or refractory AML. A clinical trial is always the preferred option. The above algorithm is based on current clinical practice and will hopefully change in coming years due to improvements. In particular the targeted and immunotherapeutic agents detailed in this review may ultimately have utility in (1) initial therapy; (2) as a bridge to, or as a temporizing measure before, allo-HSCT; and/or (3) as part of consolidative therapy. * Achievement of a complete remission (CR) prior to undergoing alloHSCT is associated with best survival and is generally preferred. The survival of patients with residual disease undergoing alloHSCT varies considerably however and this therapy may be a reasonable option in selected patients not in CR. HMA: Hypomethylating agent. LDAC: Low-dose cytosine arabinoside. Allo-HSCT: Allogeneic Hematopoietic Stem Cell Transplant. therapy, and several classifications on the basis of molecular and histological characteristics have been developed. The histological subtypes described here (top right) are the most frequent subtypes of breast cancer; ductal carcinoma (now referred to as ‘no special type’ (NST)) and lobular carcinoma are the invasive lesions; their preinvasive counterparts are ductal carcinoma in situ and lobular carcinoma in situ (or lobular neoplasia), respectively. The intrinsic subtypes are based on a 50-gene expression signature (PAM50). The surrogate intrinsic subtypes are typically used clinically and are based on histology and immunohistochemistry expression of key proteins: estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2) and the proliferation marker Ki67. Leukemia Hematopoietic cancers constitute a diverse group of diseases including leukemias, lymphomas, plasma cell tumors, myelodysplastic syndromes, and mastocytosis. They arise primarily from two categories of immunological cell types, myeloid and lymphoid cells [413]. AML is the most common form of acute leukemia in adults, accounting for over 80% of all diagnosed acute leukemias [414,415]. Globally, between 1990 to 2018, the number of leukemia cases markedly increased from 297,000 to 437, 033 [416], accounting for close to 250,000 annual deaths due to AML worldwide [417]. Optimization of post-remission therapies to maintain complete remission and prevent relapse is a major challenge in treating patients with AML [418]. Children with Down syndrome have a 150-fold increased risk of developing AML and 20-fold increased risk of developing ALL [419]. The incidence of ALL is about 3.3 cases per 100,000 children [420]. Outcomes for The Leukemia/BMT Program’s pharmacists have advanced specialty training in leukemia and bone marrow transplantation. They work directly with patients and the healthcare team, and are responsible for providing unbiased, patient-specific drug prescribing information and for identifying, preventing and resolving DRPs. They provide medication counseling when chemotherapy starts. While patients are on treatment, they check Page 23 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 in periodically during a clinic visit (or hospitalization for inpatient chemotherapy) to answer any new medication-related questions or provide recommendations for symptom management. They can educate and counsel patients on chemotherapy and supportive-care medications both during their visit to the clinic or infusion center and after they head home. This entails reviewing home medications for potential drug-drug interactions, providing written materials detailing treatment regimens, and recommending options for nausea and pain management. Myeloma regimens, as an example of one condition for which multiple drugs are given on differing schedules, can be particularly complex. To ensure that patients are prepared to follow these regimens, pharmacists may work with nurses to create treatment calendars as visual aids [425]. The pharmacy team would ensure the NRT product is suitable for the patient based on their smoking status and lifestyle and continue to ensure the NRT product is suitable on dispensing. Research has indicated that non-adherence is a major contributing factor for treatment failure in patients with CML receiving imatinib, oncology pharmacists had a nearly 90% imatinib adherence rate compared to 65.8% in the usual care group [426]. Figure 21: Treatment algorithm for patients with RR-AML [424]. Leukemia Tumors expressing ER and/or PR are termed ‘hormone receptor-positive’; tumors not expressing ER, PR and HER2 are called ‘triple-negative’. The relative placement of the boxes aligns with the characteristics (for example, proliferation and grade) in green. −, negative; +, positive. GES, gene expression signature. (a) ESR1 mutations induced by aromatase inhibitor targeted therapy. (b) Artefact; expression of normal breast components due to low tumor cellularity. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Breast cancer The most common breast cancer type is the invasive ductal carcinoma accounting for 70-80% of all breast cancers diagnosed [427]. It starts in a milk passage (a duct), breaks through the wall of the duct and invades the tissue of the breast [428]. In US, 232,000 new cases of breast cancer were diagnosed [429] and claimed the lives of 40,290 women [430] in 2015. First-degree relatives of patients with breast cancer have a 2-fold to 3-fold excess risk for development of the disease [431]. BRCA1 and BRCA2 are the 2 most important genes responsible for increased breast cancer susceptibility [432]. Early breast cancer detection programs depend for effectiveness on the participation rate, which is affected by risk factor awareness [433]. Since 1990, between 384,000 and 614,500 breast cancer deaths have been averted due to increased mammography screening and improved treatment [434]. However, more than 25% breast cancer is projected to be increased by 2020 [435] (Figure 22). All breast cancers arise in the terminal duct lobular units (the functional unit of the breast) of the collecting duct. The histological and molecular characteristics have important implications for Page 24 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 Figure 22: Breast Cancer Features [436]. All breast cancers arise in the terminal duct lobular units (the functional unit of the breast) of the collecting duct. The histological and molecular characteristics have important implications for therapy, and several classifications on the basis of molecular and histological characteristics have been developed. The histological subtypes described here (top right) are the most frequent subtypes of breast cancer; ductal carcinoma (now referred to as ‘no special type’ (NST)) and lobular carcinoma are the invasive lesions; their preinvasive counterparts are ductal carcinoma in situ and lobular carcinoma in situ (or lobular neoplasia), respectively. The intrinsic subtypes are based on a 50-gene expression signature (PAM50). The surrogate intrinsic subtypes are typically used clinically and are based on histology and immunohistochemistry expression of key proteins: estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2) and the proliferation marker Ki67. Tumors expressing ER and/or PR are termed ‘hormone receptor-positive’; tumors not expressing ER, PR and HER2 are called ‘triple-negative’. The relative placement of the boxes aligns with the characteristics (for example, proliferation and grade) in green. −, negative; +, positive. GES, gene expression signature. (a) ESR1 mutations induced by aromatase inhibitor targeted therapy. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Breast cancer (b) Artefact; expression of normal breast components due of low tumor cellularity. Women with breast cancer had a higher risk of developing new comorbidities than women without cancer [437]. stressful life [438], urban living, mastectomy [439], lower socioeconomic status [435], [440], genetic predisposition, African-American origin, not having children or breastfeeding, early menstruation/ late menopause, obesity, alcohol abuse, HRT after menopause, benign breast conditions or having breast proliferation, using contraceptives and exposure to diethylstilbestrol [441], age between 40-60, late age first pregnancy, smoking [442], abortion history [443] are the associated factors. Distressingly, the 5-year cumulative mortality remains unacceptably high at 50%, primarily due to a late-stage presentation [316]. Wearing bra is not associated with breast cancer risk [443] but wearing (tight) bras for many hours and having breast implants [442], [445] may have associations. Around 60% of breast cancer mortality occurs in LMICs [446]. Conflicting data exists about the influence of oral contraceptive pills (OCPs) on the development of breast cancer. There are clear benefits to the use of OCPs, including a reduction in ovarian cancer risk by 40% and reduction in endometrial cancer risk by 60%. Due to evidence from the Women’s Health Initiative, use of hormone replacement therapy is not recommended for patients to prevent the occurrence of breast cancer [447]. Yuan et.al, 2019 reported that medical and surgical abortion and less than 20 years IUD use could increase the risk of breast cancer for post-menopausal women [448]. The prevalence costs of breast cancer care in the US in 2010 was $16.5 billion [449,450] and exceeded $39 billion before 2017 [451]. Pharmacists can improve chemotherapy breast cancer patients’ QOL regarding malaise and nausea by providing personal counseling before the medical examinations. In addition to the attending physician and nurses, pharmacists can also partially alleviate malaise through active intervention that involves patient counseling and guidance [452]. It is worth mentioning that clinical pharmacists are well accepted as patient medication educator and psychological consultant. Wang et.al, 2015 reported reduced fatigue, the symptoms of nausea and vomiting, and pain as they followed clinical pharmacists’ advice. Medication education and nonpharmacological intervention (e.g., cognitive behavioral interventions, relaxation interventions, and Page 25 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 music therapy) by clinical pharmacists could effectively improve the quality of sleep and reduced cancer-related constipation and diarrhea of cancer patients [453]. Breast cancer Clinical pharmacist may observe all patients who receive active therapy and monitors the emergence and management of treatment-related toxicities through patient assessment and evaluation of laboratory results. During discussions with the patient, the clinical pharmacist may reinforce the importance of adhering to therapy as prescribed. This is especially important for patients who receive oral chemotherapy, which is self-administered so patients may have fewer clinic visits than those who receive IV treatment. Depending on the patient’s needs, the clinical pharmacist may also work with other staff members, such as financial counselors or social workers, to optimize patient care. Similarly, clinical pharmacists confer with specialty pharmacists to ensure that insurance issues are resolved so that the patient receives timely access to treatment. In some cases, it may be necessary to switch therapies. When this occurs, clinical pharmacists leverage their specialized training to make treatment recommendations based on the patient’s breast cancer subtype, extent of disease, treatment history, and performance status [454]. [455]. It is the second leading cause of death in US, affecting some 135,000 estimated new patients with more than 50000 deaths every year [456-458]. In 2015, there were 376,000 new cases and 191,000 deaths in China [459]. The overall incidence of CRC is decreasing in many high-income countries, although reported significant increase in Denmark, New Zealand, Australia, UK and Canada, mainly driven by increases in distal (left) tumors of the colon and predominant in [460-467]. Lifestyle determines around 50% to 60% incident of CRC irrespective of age [468-471]. Physical activity may prevent approximately 15% of the colon cancers [472]. Fish, poultry, cheese, fruit, vegetables, tea and coffee were not associated with colorectal-cancer risk [473]. Alcohol consumption, red meat/processed meat, junk food, smoking, diabetes and obesity potentiate the same risk [474-477]. In 2018, the estimated national expenditure was $16.6 billion in US, which was $4.5 billion to $9.6 billion in 2009 and projected to be more than $20 in 2020 [478- 480]. There were over 1.8 million new cases in 2018. Hungary, North Korea, Slovakia, Norway, Denmark, Portugal, Japan are in the top- ranking positions [481]. 5-year survival for patients with stage IV CRC is less than 10% [482]. The overall risk of CRC among patients with ulcerative colitis is about ten times higher than that of the general population [483]. Breast cancer CRC patients have unique psychosocial needs (e.g., isolation, embarrassment) related to altered eating and bowel habits and sexual dysfunction that warrant clinical attention [484] (Figure 23). Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Colorectal cancer Colorectal cancer (CRC) is the third most common cancer worldwide and the fourth most common cause of cancer death Figure 23: Algorithm for colorectal cancer screening [485]. +, Either colorectal cancer or adenomatous polyp; , HNPCC = hereditary nonpolyposis colorectal cancer and FAP = familial adenomatous polyposis. Figure 23: Algorithm for colorectal cancer screening [485]. +, Either colorectal cancer or adenomatous polyp; , HN nonpolyposis colorectal cancer and FAP = familial adenomatous polyposis. m for colorectal cancer screening [485]. +, Either colorectal cancer or adenomatous polyp; *, HNPCC = hereditary ctal cancer and FAP = familial adenomatous polyposis. risk of colon cancer may be increased as much as twofold in men who are in the highest quintile of body size. Potential mechanisms to this relationship include the observation that physical activity stimulates bowel peristalsis, resulting in decreased bowel transit time, and the possibility that exercise can alter levels of blood glucose, insulin, and other hormones, which may reduce tumor cell growth. In the Physicians’ Health Study, men with C-peptide in the top vs the bottom quintile had a 2.7-fold significantly higher risk of Multiple risk factors are associated with the development of this malignancy, including genetic susceptibility, environmental, and lifestyle. It has been suggested that diets high in fiber are protective against the development of colorectal cancer. Cho et.al, 2019 revealed that age, sex, family history of colorectal cancer, and education and additionally adjusted for the five modifiable risk factors (i.e., prior BMI, physical activity, dietary inflammatory index, smoking, and alcohol consumption) are CRC promoters [486]. The Page 26 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 colorectal cancer after control for BMI and exercise [487]. Heavy alcohol consumption increases risk of rectal and colon cancer by as much as two to three times, although some studies have found no significant increase in risk [488]. The strongest evidence-based screening recommendations include offering annual fecal occult blood testing (FOBT) or flexible sigmoidoscopy every 5 years. Each of these screening tests has been associated with reductions in mortality. Patients with a positive specimen should be followed up with colonoscopy. Flexible sigmoidoscopy reduced mortality by two-thirds for lesions within reach of the sigmoidoscope. Hypnosis, music and relaxing video reduced anxiety and pain associated with colonoscopy and need for sedation during colon cancer screening [489-497]. Colorectal cancer The pharmacist can help provide educational and financial resources and relieve any anxiety related to the screening process. Pharmacists have the ability to recommend colorectal cancer screening tips and tools to reduce risk. Pharmacists can also refer patients to qualified health care providers for follow-up care and act as a prescriber-patient liaison. Pharmacists can also play an essential role by reviewing the proper instructions for any colorectal screening preparation medications [498]. Pharmacists should be aware that there are differences in the guidelines with regards to method and frequency of colorectal cancer screening, but the accepted starting range for screening all adult patients at average risk is 50 years. Mass media campaigns have increased the awareness of the need for colon cancer screening, and counseling by pharmacists is a good way to help reinforce these recommendations. Additionally, pharmacists should be aware of patients displaying any of the colorectal cancer warning signs (eg, bleeding, changes in bowel habits, weight loss, abdominal pain) so that they can be referred to appropriate medical care and diagnostic work-up. Chemotherapy-induced diarrhea can negatively affect the QoL of patients and treatment process. A chemotherapeutic treatment for colon cancer consisting of fluorouracil and capecitabine is associated with 50%, and irinotecan with 80% occurrence of diarrheal symptoms. Also, the clinical oncology pharmacist has an important role in the identification and resolution of DRPs. Evaluation of symptom-related quality of life is important for the monitoring of patients receiving chemotherapy [499]. testosterone compared to white males and increased androgen receptor activation. Obesity is associated with an increased risk of prostate cancer mortality and recurrence. Low-fat diets and other dietary considerations such as β-carotene, lycopene, and vitamin E may be protective, although these are still unproven [502]. Smoking has not been associated with an increased risk of prostate cancer, but smokers with prostate cancer have an increase in mortality [505,506]. Alcohol consumption does not appear to be associated with the development of prostate cancer [507,508]. Many patients with localized prostate cancer are asymptomatic, while those with more invasive disease develop symptoms of alterations in urinary frequency, hesitancy, and flow, and new- onset impotence. There are several prostate cancer cell lines that form primary tumors but will not metastasize to bone [509]. Some nonspecific signs of more advanced disease include anemia and weight loss. The prostate specific antigen (PSA) test involves taking a simple blood sample and detecting the enzyme levels. Colorectal cancer While it is simple and readily available, it does generate false-positives and false-negatives and cannot be recommended alone as a screening tool. The American Cancer Society recommends digital rectal examination (DRE) and prostate specific antigen (PSA) be offered annually to men beginning at age 50 years with at least a 10- year life expectancy, and to younger men (45 years old) who are considered to be at high risk for prostate cancer development (eg, those with a strong family history, African Americans) [510,511]. If both tests are normal, no further diagnostic work-up is required. If either is abnormal, further work-up by transrectal ultrasound is indicated. However, to transition to patient-centered care, pharmacy services should organize around the understanding of patients’ needs, preferences, and expectations for the clinical judgment and decision-making processes. Pharmacist–patient communication is an important strategy for humanized practice. This allows the pharmacist to see beyond an individual with health problems to a patient being with particularized needs [512]. The initial management of newly diagnosed prostate cancer should consider the extended natural history of this malignancy and the risk of progression to more aggressive disease. Shared decision- making, the patient’s life expectancy, and personal preferences play an increasing role in the choice of appropriate treatment options. Pharmacists may play an important role in consulting and educating patients about the screening tests, treatments, and related benefits and harms. Enhanced risk-classification methods and expanded treatment options allow pharmacists to counsel patients regarding therapy based on cancer prognosis and patient preference. Androgen deprivation therapy (ADT) remains the standard of care for newly diagnosed metastatic disease and an important option in the management of higher risk localized cancer. Pharmacists are well-equipped to provide patient counseling and management of complications of ADT [513]. Patel et.al, 2016 demonstrated maximized oral chemotherapy treatment outcomes with the addition of a formalized monitoring program directed by an oncology pharmacist [514]. Also, pharmacist led adherence program facilitated timely dose adjustments and high patient adherence [515] Pharmacists can play active roles in the Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Prostate cancer Prostate cancer is the second most frequent malignancy (after lung cancer) in men worldwide, counting 1,276,106 new cases and causing 358,989 deaths (3.8% of all deaths caused by cancer in men) in 2018. Based on GLOBOCAN 2018 estimates, 1,276,106 new cases of prostate cancer were reported worldwide in 2018, with higher prevalence in the developed countries [500,501]. In Europe, prostate cancer is now the most common cancer in men, accounting for 23% of all male cancers and 10% of cancer-related deaths in males in 2012 [502]. In the USA, the total estimated expenditure on prostate cancer was $9.9 billion 2006 [503] and expected to be $39 billion to more than $58 billion in 2020 [504]. The incidence rate of prostate cancer increases dramatically after 55 years of age. Americans have a 5-year survival approximately 15% less than whites, perhaps due to the combination of higher levels of Page 27 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 osteoporosis. Common side effects of systemic chemotherapy, such as neutropenia and thrombocytopenia, can lead to dose reductions or treatment delays, if not managed properly. Pharmacists, using a variety of online resources can counsel patients and ensure the maximum cycles of chemotherapy can be administered [516,517] (Figure 24). management of prostate cancer, particularly by assisting in the prevention and management of the side effects of hormone therapy and chemotherapy. Hormone therapy can be associated with various adverse effects, such as osteoporosis and hot flashes. Pharmacists can suggest therapies to manage hot flashes and promote the intake of adequate amounts of calcium and vitamin D to prevent Figure 24: Prostate-specific antigen (PSA) testing and prostate cancer patient pathway [518]. Screening for prostate cancer is a highly debated topic due to limitations in the sensitivity and specificity of the PSA test, and also due to the potential harms of unnecessary investigations. In the UK, there is currently no formal screening program. The Department of Health has developed the Prostate Cancer Risk Management Program (PCRMP) for men aged over 50 years, which includes a pathway shown in this figure. The program aims to help ensure that the GP and the patient make robust, informed decisions regarding investigation for prostate cancer. Figure 24: Prostate-specific antigen (PSA) testing and prostate cancer patient pathway [518]. Prostate cancer Screening for prostate cancer is a highly debated topic due to limitations in the sensitivity and specificity of the PSA test, and also due to the potential harms of unnecessary investigations. In the UK, there is currently no formal screening program. The Department of Health has developed the Prostate Cancer Risk Management Program (PCRMP) for men aged over 50 years, which includes a pathway shown in this figure. The program aims to help ensure that the GP and the patient make robust, informed decisions regarding investigation for prostate cancer. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. References 24. Hammad EA, Qudah RA, Akour AA (2017) The impact of clinical pharmacists in improving Jordanian patients’ health outcomes. Saudi Med J 38(11): 1077-1089. 1. Miller RR (1981) History of clinical pharmacy and clinical pharmacology. J Clin Pharmacol 21(4): 195-197. 1. Miller RR (1981) History of clinical pharmacy and clinical pharmacology. J Clin Pharmacol 21(4): 195-197. 25. Dalton K, Byrne S (2017) Role of the pharmacist in reducing healthcare costs: current insights. Integr Pharm Res Pract 6: 37-46. 2. Somogyi A, Loke YK, Ferro A, Lewis LD, Cohen AF, et al. (2010) Clinical pharmacology: a declaration of intent. Br J Clin Pharmacol 70(1): 1-2. 2. Somogyi A, Loke YK, Ferro A, Lewis LD, Cohen AF, et al. (2010) Clinical pharmacology: a declaration of intent. Br J Clin Pharmacol 70(1): 1-2. 26. Li J, Li Z (2018) Differences and similarities in clinical pharmacy practice in China and the United States: a narrative review. Eur J Hosp Pharm 25(1): 2-5. 3. Calvert RT (1999) Clinical pharmacy—a hospital perspective. Br J Clin Pharmacol 47(3): 231-238. 4. Carter BL (2016) Evolution of Clinical Pharmacy in the USA and Future Directions for Patient Care. Drugs Aging 33(3): 169-77. 27. Tripathi S, Crabtree HM, Fryer KR, Graner KK, Arteaga GM (2015) Impact of Clinical Pharmacist on the Pediatric Intensive Care Practice: An 11- Year Tertiary Center Experience. J Pediatr Pharmacol Ther 20(4): 290- 298. 5. Dhingra S, Kumria R (2014) A case report on the significance of clinical pharmacy services in India. Clin Case Rep 2(3): 86-87. 6. Lebovitz L, Eddington ND (2019) Trends in the Pharmacist Workforce and Pharmacy Education. Am J Pharm Educ 83(1): 7051. 28. Mekonnen AB, Yesuf EA, Odegard PS, Wega SS (2013) Implementing ward based clinical pharmacy services in an Ethiopian University Hospital. Pharm Pract (Granada) 11(1): 51-57. 7. Veggeland T, Dyb S (2008) The contribution of a clinical pharmacist to the improvement of medication at a geriatric hospital unit in Norway. Pharm Pract (Granada) 6(1): 20-24. 29. Francis J, Abraham S (2014) Clinical pharmacists: Bridging the gap between patients and physicians. Saudi Pharm J 22(6): 600-602. 8. Horn E, Jacobi J (2006) The critical care clinical pharmacist: evolution of an essential team member. Crit Care Med 34(3 Suppl): S46-51. 30. Chalasani SH, Ramesh M, Gurumurthy P (2018) Pharmacist-Initiated Medication Error-Reporting and Monitoring Programme in a Developing Country Scenario. Pharmacy (Basel) 6(4) pii: E133. 9. Conclusion actual number of DRPs necessitating hospital admission may be higher than reported because of lack of documentation, further underestimating the problem. Initially, collecting and interpreting relevant patient information, identifying patient health-care needs, and formulating a DRP list may be challenging for the pharmacist. Preventive or chronic care is a challenge that should be undertaken by health care providers in all practice settings. Pharmacists should The pharmacist’s main responsibility is to maximize positive outcomes of drug therapy and minimize drug misadventures. Patient therapy should result in the achievement of definite outcomes that improve the patient’s QoL. To date, numerous studies have found an increased rate of hospital admission rates secondary to medication noncompliance and/or adverse drug reactions. The The pharmacist’s main responsibility is to maximize positive outcomes of drug therapy and minimize drug misadventures. Patient therapy should result in the achievement of definite outcomes that improve the patient’s QoL. To date, numerous studies have found an increased rate of hospital admission rates secondary to medication noncompliance and/or adverse drug reactions. The Page 28 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 “seize the moment” to educate and counsel patients regarding these various topics when the opportunities arise. Clinical pharmacists use population health methods to generate chronic disease management referrals for patients with uncontrolled chronic conditions. Opportunities for pharmacists to help bring about awareness of recommendations and risk factors for the development of disease, and educate patients as to the benefits of prevention, occur daily. It is important for the pharmacists on the “front line” to have a general understanding of current recommendations for screening and disease prevention so that they can provide appropriate counseling and care for their patients. Also, pediatric clinical pharmacists have evolved over the last 2 decades and have proven to be a key player in the multidisciplinary team. Although, there are ample of evidences of positive impact on clinical, humanistic and economic outcomes and the benefits of clinical pharmacists managing chronic conditions have been extensively published, their involvement in the multidisciplinary team providing care to patients with chronic cases, more high- quality research is warranted. 12. McCarthy MW (2009) Clinical Pharmacy Skills. In; Michelle McCarthy and Denise Kockler. Oxford American Handbook of Clinical Pharmacy, published by Oxford University Press. 13. Institute of Medicine (US) Roundtable on Evidence-Based Medicine (2009) Leadership Commitments to Improve Value in Healthcare: Finding Common Ground: Workshop Summary. Conclusion Washington (DC): National Academies Press (US). 14. Knoer SJ, Eck AR, Lucas AJ (2016) A review of American pharmacy: education, training, technology, and practice. J Pharm Health Care Sci 2: 32. 15. Ma CS, Holuby RS, Bucci LL (2010) Physician and pharmacist collaboration: The University of Hawaii at Hilo College of Pharmacy-- JABSOM experience. Hawaii Med J 69(6 Suppl 3): 42-44. 16. Svingen CG (2019) Clinical Pharmacist Credentialing and Privileging: A Process for Ensuring High-Quality Patient Care. Fed Pract 36(4): 155- 157. 17. Ismail S, Osman M, Abulezz R, Alhamdan H, Quadri KHM (2018) Pharmacists as Interprofessional Collaborators and Leaders through Clinical Pathways. Pharmacy (Basel) 6(1): E24. 18. Ascione FJ (2019) Preparing Pharmacists for Collaborative/Integrated Health Settings. Pharmacy (Basel) 7(2): E47. 19. Moreno G, Lonowski S, Fu J, Chon JS, Whitmire N, et al. (2017) Physician experiences with clinical pharmacists in primary care teams. J Am Pharm Assoc (2003) 57(6): 686-691. Acknowledgement 20. Al Taani GM, Al Azzam SI, Alzoubi KH, Aldeyab MA (2018) Which drugs cause treatment-related problems? Analysis of 10,672 problems within the outpatient setting. Ther Clin Risk Manag 14: 2273-2281. I’m thankful to Dr. Christel G. Svingen, Deputy Director of Pharmacy Red Lake Indian Health Service Hospital, Minnesota for his valuable time to audit my paper and for his thoughtful suggestions. I’m also grateful to seminar library of Faculty of Pharmacy, University of Dhaka and BANSDOC Library, Bangladesh for providing me books, journal and newsletters. 21. Mohiuddin AK (2019) Pharmacists in Public Health: Scope in Home and Abroad. SOJ Pharmacy & Pharmaceutical Sciences 6(1): 1–23. 22. Toklu HZ, Hussain A (2013) The changing face of pharmacy practice and the need for a new model of pharmacy education. J Young Pharm 5(2): 38-40. Conflict of Interest 23. Sakeena MHF, Bennett AA, McLachlan AJ (2019) The Need to Strengthen the Role of the Pharmacist in Sri Lanka: Perspectives. Pharmacy (Basel) 7(2): E54. The author declares that he has no competing interests. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. References J Gerontol B Psychol Sci Soc Sci 67(3): 279-88. 65. Wallgren S, Berry Caban CS, Bowers L (2012) Impact of clinical pharmacist intervention on diabetes-related outcomes in a military treatment facility. Ann Pharmacother 46(3): 353-357. 46. Zhang C, Zhang L, Huang L, Luo R, Wen J (2012) Clinical pharmacists on medical care of pediatric inpatients: a single-center randomized controlled trial. PLoS One 7(1): e30856. 66. Ray S, Lokken J, Whyte C, Baumann A, Oldani M (2019) The impact of a pharmacist-driven, collaborative practice on diabetes management in an Urban underserved population: a mixed method assessment. J Interprof Care 5: 1-9. 47. Han N, Han SH, Chu H, Kim J, Rhew KY, et al. (2018) Service design oriented multidisciplinary collaborative team care service model development for resolving drug related problems. PLoS One 13(9): e0201705. 67. Schmidt K, Caudill JA, Hamilton T (2019) Impact of clinical pharmacy specialists on glycemic control in veterans with type 2 diabetes. Am J Health Syst Pharm 76(Supplement_1): S9-S14. 48. Tan EC, Stewart K, Elliott RA, George J (2014) Pharmacist services provided in general practice clinics: a systematic review and meta- analysis. Res Social Adm Pharm 10(4): 608-622. 68. Sharp LK, Tilton JJ, Touchette DR, Xia Y, Mihailescu D, et al. (2018) Community Health Workers Supporting Clinical Pharmacists in Diabetes Management: A Randomized Controlled Trial. Pharmacotherapy 38(1): 58-68. 49. Fletcher J, Hogg W, Farrell B, Woodend K, Dahrouge S, et al. (2012) Effect of nurse practitioner and pharmacist counseling on inappropriate medication use in family practice. Can Fam Physician 58(8): 862-868. 69. Neves CM, Nascimento MMGD, Silva DAM, Ramalho de Oliveira D (2019) Clinical Results of Comprehensive Medication Management Services in Primary Care in Belo Horizonte. Pharmacy (Basel) 7(2) pii: E58. 50. Greer N, Bolduc J, Geurkink E, Rector T, Olson K, et al. (2016) Pharmacist- led Chronic Disease Management: A Systematic Review of Effectiveness and Harms Compared With Usual Care. Ann Intern Med. 70. Jeong S, Lee M, Ji E (2018) Effect of pharmaceutical care interventions on glycemic control in patients with diabetes: a systematic review and meta-analysis. Ther Clin Risk Manag 14: 1813-1829. 51. Litke J, Spoutz L, Ahlstrom D, Perdew C, Llamas W, et al. (2018) Impact of the clinical pharmacy specialist in telehealth primary care. Am J Health Syst Pharm 75(13): 982-986. 71. References Hazen ACM, Antoinette A de Bont, Leendertse AJ, Zwart DLM, Niek J de Wit, et al. (2019) How Clinical Integration of Pharmacists in General Practice has Impact on Medication Therapy Management: A Theory- oriented Evaluation. Int J Integr Care 19(1): 1. 31. Ernawati DK, Lee YP, Hughes JD (2014) Nature and frequency of medication errors in a geriatric ward: an Indonesian experience. Ther Clin Risk Manag 10: 413-421. 10. Islam MA, Talukder RM, Taheri R, Blanchard N (2016) Integration of Basic and Clinical Science Courses in US PharmD Programs. Am J Pharm Educ 80(10): 166. 32. Chamoun NR, Zeenny R, Mansour H (2016) Impact of clinical pharmacy interventions on medication error nodes. Int J Clin Pharm 38(6): 1436- 1444. 11. American College of Clinical Pharmacy (2008) The definition of clinical pharmacy. Pharmacotherapy 28(6): 816-817. 33. Gonzalez D, Rao GG, Bailey SC, Brouwer KLR, Cao Y, et al. (2017) Precision Dosing: Public Health Need, Proposed Framework, and Anticipated Impact. Clin Transl Sci 10(6): 443-454. Page 29 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 34. Sanii Y, Torkamandi H, Gholami K, Hadavand N, Javadi M (2016) Role of pharmacist counseling in pharmacotherapy quality improvement. J Res Pharm Pract 5(2): 132-137. 55. Yang JJ, Yu D, Wen W, Saito E, Rahman S, et al. (2019) Association of Diabetes With All-Cause and Cause-Specific Mortality in Asia: A Pooled Analysis of More Than 1 Million Participants. JAMA network open 2(4): e192696. 35. Ip RNS, Tenney JW, Chu ACK, Chu PLM, Young GWM (2018) Pharmacist Clinical Interventions and Discharge Counseling in Medical Rehabilitation Wards in a Local Hospital: A Prospective Trial. Geriatrics (Basel) 3(3) pii: E53. 56. Raghavan S, Vassy JL, Ho YL, Song RJ, Gagnon DR, et al. (2019) Diabetes Mellitus-Related All-Cause and Cardiovascular Mortality in a National Cohort of Adults. J Am Heart Assoc 8(4): e011295. 36. Seidling HM, Lampert A, Lohmann K, Schiele JT, Send AJ, et al. (2013) Safeguarding the process of drug administration with an emphasis on electronic support tools. Br J Clin Pharmacol 76 Suppl 1: 25-36. 57. Einarson TR, Acs A, Ludwig C, Panton UH (2018) Prevalence of cardiovascular disease in type 2 diabetes: a systematic literature review of scientific evidence from across the world in 2007-2017. Cardiovasc Diabetol 17(1): 83. 37. Bauer SR, Kane Gill SL (2016) Outcome Assessment of Critical Care Pharmacist Services. Hosp Pharm 51(7): 507-513. 58. References Longo M, Bellastella G, Maiorino MI, Meier JJ, Esposito K, et al. (2019) Diabetes and Aging: From Treatment Goals to Pharmacologic Therapy. Front Endocrinol (Lausanne) 10: 45. 38. Viana SSC, Arantes T, Ribeiro SCDC (2017) Interventions of the clinical pharmacist in an Intermediate Care Unit for elderly patients. Einstein (Sao Paulo) 15(3): 283-288. 59. Ares J, Valdes S, Botas P, Sanchez Ragnarsson C, Rodriguez Rodero S, et al. (2019) Mortality risk in adults according to categories of impaired glucose metabolism after 18 years of follow-up in the North of Spain: The Asturias Study. PLoS One 14(1): e0211070. 39. Sjolander M, Gustafsson M, Gallego G (2017) Doctors’ and nurses’ perceptions of a ward-based pharmacist in rural northern Sweden. Int J Clin Pharm 39(4): 953-959. 60. Mohiuddin AK (2019) Patient Compliance: Fact or Fiction?. Innovations in pharmacy 10 (1): 1-13. 40. Van Oostrom SH, Picavet HS, de Bruin SR, Stirbu I, Korevaar JC, et al. (2014) Multimorbidity of chronic diseases and health care utilization in general practice. BMC Fam Pract 15: 61. 61. Al Omar LT, Anderson SL, Cizmic AD, Vlasimsky TB (2019) Implementation of a Pharmacist-Led Diabetes Management Protocol. Am Health Drug Benefits 12(1): 14-20. 41. Yi JY, Kim Y, Cho YM, Kim H (2018) Self-management of Chronic Conditions Using mHealth Interventions in Korea: A Systematic Review. Healthc Inform Res 24(3): 187-197. 62. Gardea J, Papadatos J, Cadle R (2018) Evaluating glycemic control for patient-aligned care team clinical pharmacy specialists at a large Veterans Affairs medical center. Pharm Pract (Granada) 16(2): 1164. 42. Kent K, Johnson JD, Simeon K, Frates EP (2016) Case Series in Lifestyle Medicine: A Team Approach to Behavior Changes. Am J Lifestyle Med 10(6): 388-397. 63. Shane McWhorter L, McAdam Marx C, Lenert L, Petersen M, Woolsey S, et al. (2015) Pharmacist-provided diabetes management and education via a telemonitoring program. J Am Pharm Assoc (2003) 55(5): 516-526. 43. Middleton KR, Anton SD, Perri MG (2013) Long-Term Adherence to Health Behavior Change. Am J Lifestyle Med 7(6): 395-404. 64. Davis CS, Ross LAR, Bloodworth LS (2017) The Impact of Clinical Pharmacist Integration on a Collaborative Interdisciplinary Diabetes Management Team. J Pharm Pract 30(3): 286-290. 44. Young S (2014) Healthy behavior change in practical settings. Perm J 18(4): 89-92. 45. Newsom JT, Huguet N, McCarthy MJ, Ramage Morin P, Kaplan MS, et al. (2012) Health behavior change following chronic illness in middle and later life. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. References Dobrinas M, Blanc AL, Rouiller F, Christen G, Coronado M, et al. (2014) Clinical pharmacist’s role in implementing a smoking cessation intervention in a Swiss regional hospital: an exploratory study. Int J Clin Pharm 36(3): 526-534. 85. Hinton W, McGovern A, Coyle R, Han TS, Sharma P, et al. (2018) Incidence and prevalence of cardiovascular disease in English primary care: a cross-sectional and follow-up study of the Royal College of General Practitioners (RCGP) Research and Surveillance Centre (RSC). BMJ Open 8(8): e020282. 105. El Hajj MS, Awaisu A, Kheir N, Mohamed MHN, Haddad RS, et al. (2019) Evaluation of an intensive education program on the treatment of tobacco-use disorder for pharmacists: a study protocol for a randomized controlled trial. Trials 20(1): 25. 86. Mohiuddin Ak (2019) Natural Foods and Indian Herbs of Cardiovascular Interest. Pharmacy & Pharmacology International Journal 7(2). 106. Li VW, Lam J, Heise P, Reid RD, Mullen KA (2018) Implementation of a Pharmacist-Led Inpatient Tobacco Cessation Intervention in a Rehabilitation Hospital: A Before-and-After Pilot Study. Can J Hosp Pharm 71(3): 180-186. 87. Gums TH, Uribe L, Vander Weg MW, James P, Coffey C, et al. (2015) Pharmacist intervention for blood pressure control: medication intensification and adherence. J Am Soc Hypertens 9(7): 569-578. 107. Klop B, Elte JW, Cabezas MC (2013) Dyslipidemia in obesity: mechanisms and potential targets. Nutrients 5(4): 1218-1240. 88. Mohiuddin AK (2019) Risks and Reasons Associated with Medication Non-Adherence. Journal of Clinical Pharmacy 1(1): 50–53. 108. Cercato C, Fonseca FA (2019) Cardiovascular risk and obesity. Diabetol Metab Syndr 11: 74. 89. Williams B, Mancia G, Spiering W, Agabiti Rosei E, Azizi M, et al. (2018) 2018 ESC/ESH Guidelines for the management of arterial hypertension. Eur Heart J 39(33): 3021-3104. 109. Jordan MA, Harmon J (2015) Pharmacist interventions for obesity: improving treatment adherence and patient outcomes. Integr Pharm Res Pract 4: 79-89. 90. Niriayo YL, Ibrahim S, Kassa TD, Asgedom SW, Atey TM, et al. (2019) Practice and predictors of self-care behaviors among ambulatory patients with hypertension in Ethiopia. PLoS One 14(6): e0218947. 110. Carbone S, Canada JM, Billingsley HE, Siddiqui MS, Elagizi A, et al. (2019) Obesity paradox in cardiovascular disease: where do we stand? Vasc Health Risk Manag 15: 89-100. 91. National High Blood Pressure Education Program. (2004) The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. References Von Muenster SJ, Carter BL, Weber CA, Ernst ME, Milchak JL, et al. (2008) Description of pharmacist interventions during physician-pharmacist co-management of hypertension. Pharm World Sci 30(1): 128-135. 77. Berta E, Lengyel I, Halmi S, Zrinyi M, Erdei A, et al. (2019) Hypertension in Thyroid Disorders. Front Endocrinol (Lausanne). 10: 482. 99. Ramanath K, Balaji D, Nagakishore Ch, Kumar SM, Bhanuprakash M (2012) A study on impact of clinical pharmacist interventions on medication adherence and quality of life in rural hypertensive patients. J Young Pharm 4(2): 95-100. 78. Okubadejo NU, Ozoh OB, Ojo OO, Akinkugbe AO, Odeniyi IA, et al. (2019) Prevalence of hypertension and blood pressure profile amongst urban- dwelling adults in Nigeria: a comparative analysis based on recent guideline recommendations. Clin Hypertens 25: 7. 79. Web WHO. Cardiovascular disease. World Heart Day. Scale up prevention of heart attack and stroke. 100. Bhat S, Kroehl M, Yi WM, Jaeger J, Thompson AM, et al. (2019) Factors influencing the acceptance of referrals for clinical pharmacist managed disease states in primary care. J Am Pharm Assoc (2003) 59(3): 336-342. 80. Web WHO. (2017) News/ Fact sheets/Detail/Cardiovascular diseases (CVDs). 101. Kane JA, Mehmood T, Munir I, Kamran H, Kariyanna PT, et al. (2019) Cardiovascular Risk Reduction Associated with Pharmacological Weight Loss: A Meta-Analysis. Int J Clin Res Trials 4(1): 131. 81. Ofori SN, Obosi J (2019) Prevalence of hypertension among office workers in a multi-national company in the Niger-Delta with the 2017 American College of Cardiology/American Heart Association Blood Pressure Guidelines. Prev Med Rep 15: 100899. 102. Hasegawa K, Komiyama M, Takahashi Y (2019) Obesity and Cardiovascular Risk After Quitting Smoking: The Latest Evidence. Eur Cardiol 14(1): 60-61. 82. Lanza GA (2019) Diagnostic Approach to Patients with Stable Angina and No Obstructive Coronary Arteries. Eur Cardiol 14(2): 97-102. 103. El Hajj MS, Kheir N, Al Mulla AM, Shami R, Fanous N, et al. (2017) Effectiveness of a pharmacist-delivered smoking cessation program in the State of Qatar: a randomized controlled trial. BMC Public Health 17(1): 215. 83. Yasue H, Mizuno Y, Harada E (2019) Coronary artery spasm - Clinical features, pathogenesis and treatment. Proc Jpn Acad Ser B Phys Biol Sci 95(2): 53-66. 84. Ferdinand KC, Senatore FF, Clayton Jeter H, Cryer DR, Lewin JC, et al. (2017) Improving Medication Adherence in Cardiometabolic Disease: Practical and Regulatory Implications. J Am Coll Cardiol 69(4): 437-451. 104. References Wassell K, Sullivan J, Jett BP, Zuber J (2018) Comparison of clinical pharmacy specialists and primary care physicians in treatment of type 2 diabetes mellitus in rural Veterans Affairs facilities. Am J Health Syst Pharm 75(5 Supplement 1): S6-S12. 52. Meo SA, Sheikh SA, Sattar K, Akram A, Hassan A, et al. (2019) Prevalence of Type 2 Diabetes Mellitus Among Men in the Middle East: A Retrospective Study. Am J Mens Health 13(3): 1557988319848577. 72. Kharjul M, Braund R, Green J (2018) The influence of pharmacist-led adherence support on glycaemic control in people with type 2 diabetes. Int J Clin Pharm 40(2): 354-359. 53. Hughes JD, Wibowo Y, Sunderland B, Hoti K (2017) The role of the pharmacist in the management of type 2 diabetes: current insights and future directions. Integr Pharm Res Pract 6: 15-27. 73. Alqudah S, Jarab AS, Alefishat EA, Mayyas F, Khdour M, et al. Factors Associated with Poor Hemoglobin A1c Control in Patients with Type 2 Diabetes. Curr Diabetes Rev 15(2): 164-170. 54. Muc R, Saracen A, Grabska Liberek I (2018) Associations of Diabetic Retinopathy with Retinal Neurodegeneration on the Background of Diabetes Mellitus. Overview of Recent Medical Studies with an Assessment of the Impact on Healthcare systems. Open Med (Wars) 13: 130-136. 74. Peasah SK, Granitz K, Vu M, Jacob B (2019) Effectiveness of a Student Pharmacist-Led Telephone Follow-Up Intervention to Page 30 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 Improve Hemoglobin A(1C) in Diabetic Patients. J Pharm Pract 25: 897190019857409. in the management of cardiovascular diseases. Open Heart 5(1): e000687. Improve Hemoglobin A(1C) in Diabetic Patients. J Pharm Pract 25: 897190019857409. in the management of cardiovascular diseases. Open Heart 5(1): e000687. 75. Goruntla N, Mallela V, Nayakanti D (2019) Impact of Pharmacist-directed Counseling and Message Reminder Services on Medication Adherence and Clinical Outcomes in Type 2 Diabetes Mellitus. J Pharm Bioallied Sci 11(1): 69-76. 96. Morgado M, Rolo S, Castelo Branco M (2011) Pharmacist intervention program to enhance hypertension control: a randomised controlled trial. Int J Clin Pharm 33(1): 132-140. 97. Bex SD, Boldt AS, Needham SB, Bolf SM, Walston CM, et al. (2011) Effectiveness of a hypertension care management program provided by clinical pharmacists for veterans. Pharmacotherapy 31(1): 31-38. 76. Al Haqan AA, Al Taweel DM, Awad A, Wake DJ (2017) Pharmacists’ Attitudes and Role in Diabetes Management in Kuwait. Med Princ Pract 26(3): 273-279. 98. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. References Tamez Pérez HE, Proskauer Peña SL, Hernŕndez Coria MI, Garber AJ (2013) AACE Comprehensive Diabetes Management Algorithm 2013. Endocrine Practice 19(2): 327-336. 140. Kuipers E, Wensing M, Wong Go E, Daemen BJG, De Smet PAGM, et al. (2019) Adherence to guideline recommendations for asthma care in community pharmacies: actual and needed performance. NPJ Prim Care Respir Med 29(1): 26. 120. De Geeter Michelle, Shawn Riser Taylor, Ece Ilkiz Okarlton, Jamie Ellex, Christian Dolder (2018) Results of a Pharmacist Intervention on Weight Parameters and A1c Compared to Standard Patient Care. J Pharm Technol 34(5): 194-198. 141. Deeks LS, Kosari S, Boom K, Peterson GM, Maina A, et al. (2018) The Role of Pharmacists in General Practice in Asthma Management: A Pilot Study. Pharmacy Basel 6(4): E114. 121. Zanni GR (2014) Hyperlipidemia: Updated Guidelines Impact Counseling. Pharmacy Times®. 122. Hetro A, Rossetto J, Bahlawan N, Ryan M (2003) Clinical pharmacists supporting patients with diabetes and/or hyperlipidemia in a military medical home. J Am Pharm Assoc 55(1): 73-76. 142. Hunt V, Anderson D, Lowrie R, Montgomery Sardar C, Ballantyne S, et al. (2018) A non-randomised controlled pilot study of clinical pharmacist collaborative intervention for community dwelling patients with COPD. NPJ Prim Care Respir Med 28(1): 38. 123. Tsuyuki RT, Rosenthal M, Pearson GJ (2016) A randomized trial of a community-based approach to dyslipidemia management: Pharmacist prescribing to achieve cholesterol targets (RxACT Study). Can Pharm J (Ott) 149(5): 283-292. 143. Twigg MJ, Wright DJ (2017) Community pharmacy COPD services: what do researchers and policy makers need to know? Integr Pharm Res Pract 6: 53-59. 124. O Riordan M (2016) Pharmacist Management of High Cholesterol Gets More Patients to Goal, Lowers LDL Levels: RxACT. TCTMD News. 144. Greenhalgh T, Macfarlane F, Steed L, Walton R (2016) What works for whom in pharmacist-led smoking cessation support: realist review. BMC Med 14(1): 209. 125. Joel C Mars (2010) Pharmacy Perspectives in Dyslipidemia Management. US Pharmacist. 145. Kadam NS, Chiplonkar SA, Khadilkar AV, Khadilkar VV (2018) Prevalence of Osteoporosis in Apparently Healthy Adults above 40 Years of Age in Pune City, India. Indian J Endocrinol Metab 22(1): 67- 73. 126. Cassagnol M, Ezzo D, Patel PN (2013) New Therapeutic Alternatives for the Management of Dyslipidemia.” Journal of Pharmacy Practice 26(6): 528-540. 146. Patel D, Worley JR, Volgas DA, Crist BD (2018) The Effectiveness of Osteoporosis Screening and Treatment in the Midwest. Geriatr Orthop Surg Rehabil 9: 2151459318765844. 127. References Lewis A, Torvinen S, Dekhuijzen PN, Chrystyn H, Watson AT, et al. (2016) The economic burden of asthma and chronic obstructive pulmonary disease and the impact of poor inhalation technique with commonly prescribed dry powder inhalers in three European countries. BMC Health Serv Res 16: 251. 147. Tu KN, Lie JD, Wan CKV, Cameron M, Austel AG, et al. (2018) Osteoporosis: A Review of Treatment Options. P T 43(2): 92-104. 128. Xin C, Xia Z, Jiang C, Lin M, Li G (2016) The impact of pharmacist- managed clinic on medication adherence and health-related quality of life in patients with COPD: a randomized controlled study. Patient Prefer Adherence 10: 1197-1203. 148. Elias MN, Burden AM, Cadarette SM (2011) The impact of pharmacist interventions on osteoporosis management: a systematic review. Osteoporos Int 22(10): 2587-2596. 149. Jaleel A, Saag KG, Danila MI (2018) Improving drug adherence in osteoporosis: an update on more recent studies. Ther Adv Musculoskelet Dis 10(7): 141-149. 129. Usmani OS (2019) Choosing the right inhaler for your asthma or COPD patient. Ther Clin Risk Manag 15: 461-472. 130. Mubarak N, Hatah E, Khan TM, Zin CS (2019) A systematic review and meta-analysis of the impact of collaborative practice between community pharmacist and general practitioner on asthma management. J Asthma Allergy 12: 109-153. 150. Balkhi B, Seoane Vazquez E, Rodriguez Monguio R (2018) Changes in the utilization of osteoporosis drugs after the 2010 FDA bisphosphonate drug safety communication. Saudi Pharm J 26(2): 238-243. 151. Lau E, Papaioannou A, Dolovich L, Adachi J, Sawka AM, (2008) Patients’ adherence to osteoporosis therapy: exploring the perceptions of postmenopausal women. Can Fam Physician 54(3): 394-402. 131. Duong TN, Zeki AA, Louie S (2017) Medical Management of Hospitalized Patients with Asthma or Chronic Obstructive Pulmonary Disease. Hosp Med Clin 6(4): 437-455. 152. McGreevy JL, Kane MP, Busch RS, Bakst G, ElDeiry S (2019) A pharmacist-run anabolic osteoporosis clinic: An abaloparatide descriptive report. J Am Pharm Assoc (2003) 59(4): 593-597. 132. Krishnan JA, Nibber A, Chisholm A, Price D, Bateman ED, et al. (2019) Prevalence and Characteristics of Asthma-Chronic Obstructive Pulmonary Disease Overlap in Routine Primary Care Practices. Ann Am Thorac Soc 16(9):1143-1150. 153. Cho H, Byun JH, Song I, Kim HY, Ha YC, et al. (2018) Effect of improved medication adherence on health care costs in osteoporosis patients. Medicine Baltimore 97(30): e11470. 133. References Bethesda (MD): National Heart, Lung, and Blood Institute (US). 111. Koliaki C, Liatis S, Kokkinos A (2019) Obesity and cardiovascular disease: revisiting an old relationship. Metabolism 98-107. 112. De Lorenzo A, Gratteri S, Gualtieri P, Cammarano A, Bertucci P, et al. Why primary obesity is a disease? J Transl Med 17(1): 169. 92. Mohiuddin AK (2019) A Brief Review of Traditional Plants as Sources of Pharmacological Interests. Open J Plant Sci 4(1): 1-8. 113. Nuffer Monika (2019) Integrative Health and Medicine: Dietary Supplements and Modalities for the Treatment of Obesity. Nutrition in the Prevention and Treatment of Abdominal Obesity 393-408. 93. Mohiuddin AK (2019) Traditional System of Medicine and Nutritional Supplementation: Use Vs Regulation. Journal of Clinical Trials, Pathology and Case Studies 4(1): 5–30. 114. Butryn ML, Webb V, Wadden TA (2011) Behavioral treatment of obesity. Psychiatr Clin North Am 34(4): 841-859. 94. Mohiuddin AK (2019) Patient History and Medical Record: Proper Solution from Accurate Problem Identification. Medicine and Medical Sciences 7(7): 82–86. 115. Sanyal D, Raychaudhuri M (2016) Hypothyroidism and obesity: An intriguing link. Indian J Endocrinol Metab 20(4): 554-557. 95. Omboni S, Caserini M (2018) Effectiveness of pharmacist’s intervention Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. uddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. /ABEB 2019 03 000566 Page 31 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 116. Fruh SM (2017) Obesity: Risk factors, complications, and strategies for sustainable long-term weight management. J Am Assoc Nurse Pract 29(S1): S3-S14. 137. Cawley MJ, Warning WJ 2nd (2018) Impact of a Pharmacist-driven Spirometry Clinic Service within a Community Family Health Center: A 5-year Retrospective Review. J Res Pharm Pract 7(2): 88-94. 117. Tsai Adam Gilden, Daniel H Bessesen (2019) Obesity. Annals of Internal Medicine 170(5). 138. Ottenbros S, Teichert M, de Groot R, Griens F, Sodihardjo F, et al. (2014) Pharmacist-led intervention study to improve drug therapy in asthma and COPD patients. Int J Clin Pharm 36(2): 336-344. 118. Tsai AG (2019) Why I Treat Obesity. Perm J 23. 139. Usmani OS, Lavorini F, Marshall J, Dunlop WCN, Heron L, et al. (2018) Critical inhaler errors in asthma and COPD: a systematic review of impact on health outcomes. Respir Res 19(1): 10. 119. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Archives in Biomedical Engineering & Biotechnology Ye BD, Travis S (2019) Improving the quality of care for inflammatory bowel disease. Intest Res 17(1): 45-53. 187. Calsolaro V, Niccolai F, Pasqualetti G, Calabrese AM, Polini A, et al. (2019) Overt and Subclinical Hypothyroidism in the Elderly: When to Treat? Front Endocrinol (Lausanne) 10: 177. 166. Lichtenstein GR, Shahabi A, Seabury SA, Lakdawalla DN, Espinosa OD, et al. (2019) Lifetime Economic Burden of Crohn’s Disease and Ulcerative Colitis by Age at Diagnosis. Clin Gastroenterol Hepatol S1542-S3565(19): 30765-30767. 188. Leng O, Razvi S (2019) Hypothyroidism in the older population. Thyroid Res 12: 2. 167. Barrett K, Glatter J (2019) New standards for IBD aim to improve patient experience and outcomes. Guidelines in Practice, Gastrointestinal, UK. 189. Talwalkar P, Deshmukh V, Bhole M (2019) Prevalence of hypothyroidism in patients with type 2 diabetes mellitus and hypertension in India: a cross-sectional observational study. Diabetes Metab Syndr Obes 12: 369-376. 168. Sood A, Ahuja V, Kedia S, Midha V, Mahajan R, et al. (2019) Diet and inflammatory bowel disease: The Asian Working Group guidelines. Indian J Gastroenterol 38(3): 220-246. 190. Thavaraputta S, Dennis JA, Laoveeravat P, Nugent K, Rivas AM (2019) Hypothyroidism and Its Association With Sleep Apnea Among Adults in the United States: NHANES 2007-2008. J Clin Endocrinol Metab 104(11): 4990-4997. 169. Rizzello F, Spisni E, Giovanardi E, Imbesi V, Salice M, et al. (2019) Implications of the Westernized Diet in the Onset and Progression of IBD. Nutrients 11(5): E1033. 191. Sriphrapradang C, Pinyopodjanard S, Suntornlohanakul O, Nimitphong H, Chirakalwasan N, et al. (2019) Lack of associations between thyroid function and obstructive sleep apnea severity in adults with prediabetes and diabetes mellitus. Sleep Breath 23(3): 963-967. 170. Malekzadeh MM, Sima A, Alatab S, Sadeghi A, Daryani NE, et al. Iranian Registry of Crohn’s and Colitis: study profile of first nation-wide inflammatory bowel disease registry in Middle East. Intest Res 17(3): 330-339. 192. Bruyneel M, Veltri F, Poppe K (2019) Prevalence of newly established thyroid disorders in patients with moderate-to-severe obstructive sleep apnea syndrome. Sleep Breath 23(2): 567-573. 171. Choi K, Chun J, Han K, Park S, Soh H, et al. (2019) Risk of Anxiety and Depression in Patients with Inflammatory Bowel Disease: A Nationwide, Population-Based Study. J Clin Med 8(5): E654. 193. Sweed RA, Hassan S, ElWahab NHA, Aref SR, Mahmoud MI (2019) Comorbidities associated with obstructive sleep apnea: a retrospective Egyptian study on 244 patients. Sleep Breath 23(4): 1079-1085. 172. Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 158. Zuo H, Sun A, Gao L, Xue W, Deng Y, et al. (2019) Effect of Menopausal Hormone Therapy on Bone Mineral Density in Chinese Women: A 2-Year, Prospective, Open-Label, Randomized-Controlled Trial. Med Sci Monit 25: 819-826. 178. Ashok K, Mathew AA, Thomas A, Mohan D, Gopalakrishna R, et al. (2017) Clinical Pharmacist’s Interventions on Medication Adherence and Knowledge of Inflammatory Bowel Disease Patients. J Young Pharm 9(3): 381-385. 159. Bowers BL, Drew AM, Verry C (2018) Impact of Pharmacist-Physician Collaboration on Osteoporosis Treatment Rates. Ann Pharmacother 52(9): 876-883. 179. Tiao DK, Chan W, Jeganathan J, Chan JT, Perry J, et al. (2017) Inflammatory Bowel Disease Pharmacist Adherence Counseling Improves Medication Adherence in Crohn’s Disease and Ulcerative Colitis. Inflamm Bowel Dis 23(8): 1257-1261. 160. Duquet N (2014) Osteoporosis: treatment and pharmaceutical care]. J Pharm Belg (2): 14-24. 180. McDowell C, Haseeb M (2019) Inflammatory Bowel Disease (IBD). StatPearls, Treasure Island (FL): StatPearls Publishing. 161. Uhlig HH (2013) Monogenic diseases associated with intestinal inflammation: implications for the understanding of inflammatory bowel disease. Gut 62(12): 1795-805. 181. Jones ASC (2014) Optimising therapy for inflammatory bowel disease. Clinical Pharmacist. 162. Cury DB, Oliveira R, Cury MS (2019) Inflammatory bowel diseases: time of diagnosis, environmental factors, clinical course, and management - a follow-up study in a private inflammatory bowel disease center (2003-2017). J Inflamm Res 12: 127-135. 182. Alan Steel, Helen Morgan, Wei Tan, Vanessa Marvin, Mahmood Wahed (2018) PTU-140 Does a dedicated inflammatory bowel disease (IBD) pharmacist clinic improve patient safety? Gut 67: A265-A266. 183. Bhat S (2015) The Pharmacist’s Role in Biologic Management for IBD in a Health System—Integrated Practice Model. Pharmacy Times®. 163. Bon L, Scharl S, Vavricka S, Rogler G, Fournier N, et al. (2019) Association of IBD specific treatment and prevalence of pain in the Swiss IBD cohort study. PLoS One 14(4): e0215738. 184. Leung AM (2016) Thyroid Emergencies. J Infus Nurs 39(5): 281-286. 185. De Leo S, Lee SY, Braverman LE (2016) Hyperthyroidism. Lancet 388(10047): 906-918. 164. Aniwan S, Harmsen WS, Tremaine WJ, Loftus EV (2019) Incidence of inflammatory bowel disease by race and ethnicity in a population- based inception cohort from 1970 through 2010. Therap Adv Gastroenterol 12: 1756284819827692. 186. Taylor PN, Albrecht D, Scholz A, Gutierrez Buey G, Lazarus JH, et al. (2018) Global epidemiology of hyperthyroidism and hypothyroidism. Nat Rev Endocrinol 14(5): 301-316. 165. References López Campos JL, Quintana Gallego E, Carrasco Hernández L (2019) Status of and strategies for improving adherence to COPD treatment. Int J Chron Obstruct Pulmon Dis 14: 1503-1515. 154. Ivanova S, Vasileva L (2017) Current and Emerging Strategies in Osteoporosis Management. Curr Pharm Des 23(41): 6279-6287. 134. Jardim JR, Nascimento OA (2019) The Importance of Inhaler Adherence to Prevent COPD Exacerbations. Med Sci (Basel) 7(4): E54. 155. Leslie WD, Morin SN, Martineau P, Bryanton M, Lix LM (2019) Association of Bone Density Monitoring in Routine Clinical Practice with Anti-Osteoporosis Medication Use and Incident Fractures: A Matched Cohort Study. J Bone Miner Res 34(10): 1808-1814. 135. Abdulsalim S, Unnikrishnan MK, Manu MK, Alsahali S, Alrasheedy AA, et al. (2019) Impact of a Clinical Pharmacist Intervention on Medicine Costs in Patients with Chronic Obstructive Pulmonary Disease in India. Pharmacoecon Open. 156. Tit DM, Bungau S, Iovan C, Nistor Cseppento DC, Endres L, et al. (2018) Effects of the Hormone Replacement Therapy and of Soy Isoflavones on Bone Resorption in Postmenopause. J Clin Med 7(10): E297. 136. Mc Gurran MA, Richter LM, Leedahl ND, Leedahl DD (2019) Impact of a Comprehensive COPD Therapeutic Interchange Program on 30-Day Readmission Rates in Hospitalized Patients. P T 44(4): 185-191. 157. Fait T (2019) Menopause hormone therapy: latest developments and clinical practice. Drugs Context 8: 212551. n: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. 00566. DOI: 10.33552/ABEB.2019.03.000566. Page 32 of 41 Archives in Biomedical Engineering & Biotechnology Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Archives in Biomedical Engineering & Biotechnology Faquih AE, Memon RI, Hafeez H, Zeshan M, Naveed S (2019) A Review of Novel Antidepressants: A Guide for Clinicians. Cureus 11(3): e4185. 231. GBD 2016 Neurology Collaborators (2019) Global, regional, and national burden of neurological disorders, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet Neurology vol. 18,5 (2019): 459-480. 209. Horackova K, Kopecek M, Machů V, Kagstrom A, Aarsland D, Motlova LB, Cermakova P (2019) Prevalence of late-life depression and gap in mental health service use across European regions. Eur Psychiatry: 19-25. 232. Weller J, Budson A (2018) Current understanding of Alzheimer’s disease diagnosis and treatment. F1000Res 7. pii: F1000 Faculty Rev-1161. 210. Singh OP (2018) Closing treatment gap of mental disorders in India: Opportunity in new competency-based Medical Council of India curriculum. Indian J Psychiatry 60(4): 375-376. 233. Martinez-Martin P, Macaulay D, Jalundhwala YJ, Mu F et al. (2019) The long-term direct and indirect economic burden among Parkinson’s disease caregivers in the United States. Mov Disord 34(2): 236-245. 211. Swaminath G, Enara A, Rao R, Kumar KVK, Kumar CN (2019) Mental Healthcare Act, 2017 and homeless persons with mental illness in India. Indian J Psychiatry (Suppl 4): S768-S772. 234. Kowal SL, Dall TM, Chakrabarti R, Storm MV, Jain A (2013) The current and projected economic burden of Parkinson’s disease in the United States. Mov Disord 28(3):311-8. 212. Hayes K, Berry P, Ebi KL (2019) Factors Influencing the Mental Health Consequences of Climate Change in Canada. Int J Environ Res Public Health 16(9) pii: E1583. 235. Stephens W (2019) Financial Burden of Parkinson Disease Is $52 Billion, More Than Double Previous Estimates. AJMC News. 213. Cook L (2019) Mental health in Australia: a quick guide. Parliament of Australia 236. Economic Burden and Future Impact of Parkinson’s Disease Final Report. LewinGroup (2019). 214. Shpigelman CN, HaGani N (2019) The impact of disability type and visibility on self-concept and body image: Implications for mental health nursing. J Psychiatr Ment Health Nurs 26(3-4): 77-86. 237. Magyari M, Sorensen PS (2019) The changing course of multiple sclerosis: rising incidence, change in geographic distribution, disease course, and prognosis. Curr Opin Neurol 32(3):320-326. 215. Yamauchi T, Sasaki T, Yoshikawa T, Matsumoto S, Takahashi M (2018) Incidence of overwork-related mental disorders and suicide in Japan. Occup Med (Lond) 68(6):370-377. 238. Díaz C, Zarco LA, Rivera DM (2019) Highly active multiple sclerosis: An update. Mult Scler Relat Disord 30:215-224. 239. Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 200. Rugge B, Balshem H, Sehgal R, Rose Relevo, Paul Gorman, et al. (2011) Screening and Treatment of Subclinical Hypothyroidism or Hyperthyroidism. Rockville (MD): Agency for Healthcare Research and Quality, US. 224. American Pharmacists Association (2008) National Association of Chain Drug Stores Foundation. Medication therapy management in pharmacy practice: core elements of an MTM service model (version 2.0). J Am Pharm Assoc (2003) 48(3):341-53. 201. Bensenor IM, Olmos RD, Lotufo PA (2012) Hypothyroidism in the elderly: diagnosis and management. Clin Interv Aging 7: 97-111. 225. Chavez B, Kosirog E (2019) Impact on an integrated psychiatric pharmacy service in a primary care clinic. Ment Health Clin. 9(4):269-274 202. De Groot LJ (2000) Diagnosis and Treatment of Graves’ Disease. In: Feingold KR, Anawalt B, Boyce A (ed.) Endotext, South Dartmouth, USA. 226. Bingham J, Axon DR, Scovis N, Taylor AM (2018) Evaluating the Effectiveness of Clinical Pharmacy Consultations on Nutrition, Physical Activity, and Sleep in Improving Patient-Reported Psychiatric Outcomes for Individuals with Mental Illnesses. Pharmacy (Basel) 7(1). pii: E2. 203. Maharjan S, Himal Paudel C (2015) Assessment of Interventions by Pharmacist in Improving Knowledge, Attitude and Practice towards Hypothyroidism among the Patients Attending at an Endocrine Clinic in Nepal. Indian Journal of Pharmacy Practice, 8(2): 67-71. 227. Riley TB, Alemagno S (2019) Pharmacist utilization of prescription opioid misuse interventions: Acceptability among pharmacists and patients. Res Social Adm Pharm. 15(8):986-991. 204. Lloyd J, Yerbury P, Ruszala V (2011) Thyroid disorders management. 228. Eltorki Y, Abdallah O, Omar N, Zolezzi M (2019) Perceptions and expectations of health care providers towards clinical pharmacy services in a mental health hospital in Qatar. Asian J Psychiatr 42:62-66. 205. Mueller L (2018) Changing the Conversation About Mental Health. Chicago Health, USA 206. Harms M, Haas M, Larew J, DeJongh B (2018) Impact of a mental health clinical pharmacist on a primary care mental health integration team. Ment Health Clin pp: 101-105. 229. El-Den S, Chen TF, Moles RJ, O’Reilly C (2018) Assessing Mental Health First Aid Skills Using Simulated Patients. Am J Pharm Educ 82(2):6222. 230. Choonara YE, Pillay V, du Toit LC, Modi et al. (2009) G Trends in the molecular pathogenesis and clinical therapeutics of common neurodegenerative disorders. Int J Mol Sci 10(6): 2510-57. 207. Rehm J, Shield KD (2019) Global Burden of Disease and the Impact of Mental and Addictive Disorders. Curr Psychiatry Rep 21(2): 1-10. 208. Archives in Biomedical Engineering & Biotechnology Panhwar MS, Mansoor E, Al Kindi SG, Sinh P, Katz J, et al. (2019) Risk of Myocardial Infarction in Inflammatory Bowel Disease: A Population- based National Study. Inflamm Bowel Dis 25(6): 1080-1087. 194. Gabrielson AT, Sartor RA, Hellstrom WJG (2019) The Impact of Thyroid Disease on Sexual Dysfunction in Men and Women. Sex Med Rev 7(1): 57-70. 173. Kuenzig ME, Bishay K, Leigh R, Kaplan GG, Benchimol EI et al. (2018) Co-occurrence of Asthma and the Inflammatory Bowel Diseases: A Systematic Review and Meta-analysis. Clin Transl Gastroenterol 9(9): 188. 195. Yuan J, Sun C, Jiang S, Lu Y, Zhang Y, et al. (2019) The Prevalence of Thyroid Disorders in Patients With Vitiligo: A Systematic Review and Meta-Analysis. Front Endocrinol (Lausanne) 9: 803. 174. Vutcovici M, Brassard P, Bitton A (2016) Inflammatory bowel disease and airway diseases. World J Gastroenterol 22(34): 7735-7741. 196. Biondi B, Cappola AR, Cooper DS (2019) Subclinical Hypothyroidism: A Review. JAMA 322(2): 153-160. 175. Kelly OB, Li N, Smith M, Chan J, Inman RD, et al. (2019) The Prevalence and Clinical Associations of Subclinical Sacroiliitis in Inflammatory Bowel Disease. Inflamm Bowel Dis 25(6): 1066-1071. 197. Fayyaz B, Upreti S (2018) Autoimmune inner ear disease secondary to Hashimoto’s thyroiditis: a case report. J Community Hosp Intern Med Perspect 8(4): 227-229. 176. Scharl S, Barthel C, Rossel JB, Biedermann L, Misselwitz B, et al. (2019) Malignancies in Inflammatory Bowel Disease: Frequency, Incidence and Risk Factors-Results from the Swiss IBD Cohort Study. Am J Gastroenterol 114(1): 116-126. 198. Chaudhary SC, Ahmad T, Usman K, Sawlani KK, Gupta KK, et al. (2018) Prevalence of thyroid dysfunction in chronic obstructive pulmonary disease patients in a tertiary care center in North India. J Family Med Prim Care 7(3): 584-588. 177. Waters BM, Jensen L, Fedorak RN (2005) Effects of formal education for patients with inflammatory bowel disease: a randomized controlled trial. Can J Gastroenterol 19(4): 235-244. 199. Akpınar EE (2019) An underestimated comorbidity of COPD: Thyroid dysfunction. Tuberk Toraks 67(2): 131-135. Page 33 of 41 Archives in Biomedical Engineering & Biotechnology Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Archives in Biomedical Engineering & Biotechnology pii: E37 278. Wang X, Li P, Pan C, Dai L et al. (2019) The Effect of Mind-Body Therapies on Insomnia: A Systematic Review and Meta-Analysis. Evid Based Complement Alternat Med 2019: 9359807. 255. Gutierrez-Colina AM, Smith AW, Mara CA, Modi AC (2018) Adherence barriers in pediatric epilepsy: From toddlers to young adults. Epilepsy Behav 80: 229-234. 256. Gschwind M, Seeck M (2016) Modern management of seizures and epilepsy. Swiss Med Wkly 146: w14310. 279. Silva AR, Bernardo A, Costa J, Cardoso A et al. (2014) Dietary interventions in fibromyalgia: a systematic review. Ann Med 51(sup1): 2-14. 257. Koshy S (2012) Role of pharmacists in the management of patients with epilepsy. Int J Pharm Pract 20(1):65-8. 280. Marum AP, Moreira C, Saraiva F, Tomas-Carus P, Sousa-Guerreiro C (2016) A low fermentable oligo-di-mon saccharides and polyols (FODMAP) diet reduced pain and improved daily life in fibromyalgia patients. Scand J Pain 13: 166-172. 258. dos Reis TM, Campos MSA, Nagai MM, Pereira LRL (2016) Contributions of Pharmacists in the Treatment of Epilepsy: A Systematic Review. Pharmacy Times®. 281. Marum AP, Moreira C, Tomas-Carus P, Saraiva F, Guerreiro CS (2017) A low fermentable oligo-di-mono-saccharides and polyols (FODMAP) diet is a balanced therapy for fibromyalgia with nutritional and symptomatic benefits. Nutr Hosp 34(3): 667-674. 259. Ha H (2013) Epilepsy: Treatment and Management. US Pharmacists. 260. Chen C, Lee DS, Hie SL (2013) The impact of pharmacist’s counseling on pediatric patients’ caregiver’s knowledge on epilepsy and its treatment in a tertiary hospital. Int J Clin Pharm 35(5):829-34. 282. Lattanzio SM, Imbesi F (2018) Fibromyalgia Syndrome: A Case Report on Controlled Remission of Symptoms by a Dietary Strategy. Front Med (Lausanne) 5: 94. 261. dos Reis TM, Campos MSA, Nagai MM, Pereira LRL (2016) Contributions of Pharmacists in the Treatment of Epilepsy: A Systematic Review. Am J Pharm Benefits 8(3): e55-e60 283. Zabihiyeganeh M, Vafaee Afshar S, Amini Kadijani A, Jafari D et al. (2019) The effect of cognitive behavioral therapy on the circulating proinflammatory cytokines of fibromyalgia patients: A pilot controlled clinical trial. Gen Hosp Psychiatry 57:23-28. 262. Cabo-Meseguer A, Cerdá-Olmedo G, Trillo-Mata JL (2017) Fibromyalgia: Prevalence, epidemiologic profiles and economic costs. Med Clin (Barc) 149(10):441-448. 284. (2018) Cologne, Germany: Institute for Quality and Efficiency in Health Care (IQWiG) 2006-. Fibromyalgia: Multimodal pain management and cognitive behavioral therapy. 263. Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 247. Lyons KE, Pahwa R, Hermanowicz N, Davis T et al. (2019) Changing the treatment paradigm for Parkinson’s disease psychosis with pimavanserin. Expert Rev Clin Pharmacol 12(7): 681-691. 270. Arout CA, Sofuoglu M, Bastian LA, Rosenheck RA (2018) Gender Differences in the Prevalence of Fibromyalgia and in Concomitant Medical and Psychiatric Disorders: A National Veterans Health Administration Study. J Womens Health (Larchmt) 27(8):1035-1044. 248. Raza C, Anjum R, Shakeel NUA (2019) Parkinson’s disease: Mechanisms, translational models and management strategies. Life Sci 226: 77-90. 271. Bhargava J, Hurley JA (2019) Fibromyalgia. [Updated 2019 May 1]. In: StatPearls [Internet] (ed.) Treasure Island (FL): StatPearls Publishing. 249. Foppa AA, Chemello C, Vargas-Peláez CM, Farias MR (2016) Medication Therapy Management Service for Patients with Parkinson’s Disease: A Before-and-After Study. Neurol Ther 5(1): 85-99. 272. Kaltsas G, Tsiveriotis K. Fibromyalgia. [Updated 2017 Oct 2]. In: Feingold KR, Anawalt B, Boyce A (ed.) Endotext, South Dartmouth USA. 250. Patel T, Chang F, Parkinson Society Canada (2014) Parkinson’s disease guidelines for pharmacists. Can Pharm J (Ott147(3): 161-70. 273. Galvez-Sánchez CM, Duschek S, Reyes Del Paso GA (2019) Psychological impact of fibromyalgia: current perspectives. Psychol Res Behav Manag 12: 117-127. 251. Patel T, Chang F (2015) Practice recommendations for Parkinson’s disease: Assessment and management by community pharmacists. Can Pharm J (Ott) 148(3): 142-9. 274. Bellato E, Marini E, Castoldi F, Barbasetti N et al. (2012) Fibromyalgia syndrome: etiology, pathogenesis, diagnosis, and treatment. Pain Res Treat 2012: 426130. 252. Hill AC, Thomson KE, Newell TG, White HS (2019) Correction of medication nonadherence results in better seizure outcomes than dose escalation in a novel preclinical epilepsy model of adherence. Epilepsia 60(3): 475-484. 275. Mahjoub F, Salari R, Noras MR, Yousefi M (2017) Are Traditional Remedies Useful in Management of Fibromyalgia and Chronic Fatigue Syndrome? A Review Study. J Evid Based Complementary Altern Med 22(4): 1011-1016. 253. Kälviäinen R, Reinikainen M (2019) Management of prolonged epileptic seizures and status epilepticus in palliative care patients. Epilepsy Behav pii: S1525-5050(19)30372-5. 276. Abdul Kader Mohiuddin (2019) Comparison of Drug and Non-Drug Treatment Options of Fibromyalgia Glob J Ortho Res. 1(5): 2019. GJOR. MS.ID.000522. 254. Ernawati I, Islamiyah WR, Sumarno (2018) How to Improve Clinical Outcome of Epileptic Seizure Control Based on Medication Adherence? A Literature Review. Open Access Maced J Med Sci 6(6): 1174-1179. 277. Sawynok J , Lynch ME (2017) Qigong and Fibromyalgia circa 2017. Medicines (Basel) 4(2). Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Page Archives in Biomedical Engineering & Biotechnology Rivera VM (2019) Biosimilar Drugs for Multiple Sclerosis: An Unmet International Need or a Regulatory Risk? Neurol Ther 8(2): 177-184. 216. Uhlenbusch N, Löwe B, Härter M, Schramm C, Weiler-Normann C (2019) et al. Depression and anxiety in patients with different rare chronic diseases: A cross-sectional study. PLoS One 14(2): e0211343. 240. Thijs RD, Surges R, O’Brien TJ, Sander JW (2019) Epilepsy in adults. Lancet 393(10172): 689-701. 217. Rokach A (2019) The Effect of Psychological Conditions on Sexuality: A Review. Psychol Psychother Res Stud 2(2): PPRS.000534. 241. O’Donohoe TJ, Choudhury A, Callander E (2019) The global macroeconomic burden of epilepsy and the role for neurosurgery: A modelling study based upon the 2016 Global Burden of Disease data. Eur J Neurol. 218. Gibu M, Clark J, Gold J (2018) Mental health pharmacists as interim prescribers. Ment Health Clin 7(3):111-115. 219. Mohiuddin AK (2019) Psychiatric Pharmacy: New Role of Pharmacists in Mental Health. J Psychiatry Mental Disord 4(1): 1010. 242. Grossberg GT, Tong G, Burke AD, Tariot PN (2019). Present Algorithms and Future Treatments for Alzheimer’s Disease. J Alzheimers Dis 67(4): 1157-1171. 220. Matlala M, Maponya ML, Chigome AK, Meyer H-jc (2018) Overview of mental health: A public health priority. S Afr Pharm J 85(6): 46-53. 243.Terrie YC (2007) The Pharmacist’s Role in the Management of Alzheimer’s Disease. Pharmacy Times®. 221. Guillaumie L, Ndayizigiye A, Beaucage C, Moisan J, Grégoire JP et al. (2018) Patient perspectives on the role of community pharmacists for antidepressant treatment: A qualitative study. Can Pharm J (Ott) 151(2):142-148. 244. Duong S, Patel T, Chang F (2017) Dementia: What pharmacists need to know. Can Pharm J (Ott) 150(2): 118-129. 245. Yılmaz, Nihan Çarçak, et al. (2017) Pharmacist’s role in pharmacotherapeutic management of Alzheimer’s disease. Istanbul J Pharm vol. 47, no. 1, pp. 1–4. 222. Srimongkon P, Aslani P, Chen TF (2018) Consumer-related factors influencing antidepressant adherence in unipolar depression: a qualitative study. Patient Prefer Adherence 12:1863-1873. 246. Ciulla M, Marinelli L, Cacciatore I, Stefano AD (2019) Role of Dietary Supplements in the Management of Parkinson’s Disease. Biomolecules 9(7). pii: E271. 223. Holvast F, Oude Voshaar RC, Wouters H, Hek K, Schellevis F et al. (2019) Non-adherence to antidepressants among older patients with depression: a longitudinal cohort study in primary care. Fam Pract 36(1): 12-20. Page 34 of 41 Archives in Biomedical Engineering & Biotechnology Archives in Biomedical Engineering & Biotechnology Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 291. Kodner C (2015) Common questions about the diagnosis and management of fibromyalgia. Am Fam Physician 91(7): 472-8. 314. Kennelly M, Thiruchelvam N, Averbeck MA, Konstatinidis C, Chartier- Kastler E et al. (2019) Adult Neurogenic Lower Urinary Tract Dysfunction and Intermittent Catheterisation in a Community Setting: Risk Factors Model for Urinary Tract Infections. Adv Urol 2019: 2757862. 292. Medina M, Castillo-Pino E (2019) An introduction to the epidemiology and burden of urinary tract infections. Ther Adv Urol 11: 1756287219832172. 315. Plüddemann A (2019) Can drinking more water prevent urinary tract infections? The evidence says yes. BMJ Evid Based Med. 24(5): 191-192. 293. Al-Badr A, Al-Shaikh G (2013) Recurrent Urinary Tract Infections Management in Women: A review. Sultan Qaboos Univ Med J 13(3): 359-67. 316. Akgül T, Karakan T (2018) The role of probiotics in women with recurrent urinary tract infections. Turk J Urol 44(5): 377-383. 294. Lema VM (2015) Urinary Tract Infection In Young Healthy Women Following Heterosexual Anal Intercourse: Case Reports. Afr J Reprod Health 19(2): 134-9. 317. Asadi Karam MR, Habibi M, Bouzari S (2019) Urinary tract infection: Pathogenicity, antibiotic resistance and development of effective vaccines against Uropathogenic Escherichia coli. Mol Immunol 108: 56-67. 295. Harrington RD, Hooton TM (2000) Urinary tract infection risk factors and gender. J Gend Specif Med 3(8): 27-34. 318. Wnorowska U, Piktel E, Durnaś B, Fiedoruk K, Savage PB et al. (2019) Use of ceragenins as a potential treatment for urinary tract infections. BMC Infect Dis 19(1): 369. 296. Oladeinde BH, Omoregie R, Olley M, Anunibe JA (2011) Urinary tract infection in a rural community of Nigeria. N Am J Med Sci 3(2): 75-7. 319. Abdul Kader Mohiuddin (2019) Alternative Management of Uncomplicated UTIs In Women. J Urol Neph St 2(2)- 2019. JUNS. MS.ID.000133. 297. Behzadi P, Behzadi E, Pawlak-Adamska EA (2019) Urinary tract infections (UTIs) or genital tract infections (GTIs)? It’s the diagnostics that count. GMS Hyg Infect Control 14: Doc14. 320. Abdul Kader Mohiuddin (2019) Lifestyle Issues and Prevention of Recurrent UTIs. Biomed J Sci & Tech Res 21(3)-2019. BJSTR. MS.ID.003618. 298. Arnold JJ, Hehn LE, Klein DA (2016) Common Questions About Recurrent Urinary Tract Infections in Women. Am Fam Physician 93(7): 560-9. 321. Silverman JA, Schreiber HL 4th, Hooton TM, Hultgren SJ (2013) From physiology to pharmacy: developments in the pathogenesis and treatment of recurrent urinary tract infections. Archives in Biomedical Engineering & Biotechnology Curr Urol Rep 14(5): 448-56. 299. Tan CW, Chlebicki MP (2016) Urinary tract infections in adults. Singapore Med J 57(9): 485-90. 322. Mohiuddin AK. (2019). Lifestyle Issues and Prevention of Recurrent UTIs. Mathews J Urol Nephrol 3(1): e003. 300. Hickling DR, Nitti VW (2013) Management of recurrent urinary tract infections in healthy adult women. Rev Urol 15(2): 41-8. 323. Beahm NP, Nicolle LE, Bursey A, Smyth DJ, Tsuyuki RT (2017) The assessment and management of urinary tract infections in adults: Guidelines for pharmacists. Can Pharm J (Ott) 150(5): 298-305. 301. Jhang JF, Kuo HC (2017) Recent advances in recurrent urinary tract infection from pathogenesis and biomarkers to prevention. Ci Ji Yi Xue Za Zhi 29(3): 131-137. 324. Beahm NP, Smyth DJ, Tsuyuki RT (2018) Outcomes of Urinary Tract Infection Management by Pharmacists (R(x)OUTMAP): A study of pharmacist prescribing and care in patients with uncomplicated urinary tract infections in the community. Can Pharm J (Ott) 151(5): 305-314. 302. Gupta K, Trautner BW (2013) Diagnosis and management of recurrent urinary tract infections in non-pregnant women. BMJ 346: f3140. 303. Li R, Leslie SW (2019) Cystitis. In: StatPearls. Treasure Island (FL): StatPearls Publishing (ed.). 325. Minejima E, Lee E, Quach S, Santos N, Lou M et al. (2019) Understanding patientperceptions and attitudes toward urinary tract infections andtreatment in a medically underserved population. J Am Coll ClinPharm 1–7. 304. Scholes D, Hooton TM, Roberts PL, Stapleton AE et al. (2000) Risk factors for recurrent urinary tract infection in young women. J Infect Dis 182(4): 1177-82. 326. Crews DC, Bello AK, Saadi G (2019) 2019 World Kidney Day Editorial - burden, access, and disparities in kidney disease. J Bras Nefrol 41(1):1-9. 305. Nseir W, Farah R, Mahamid M, Sayed-Ahmad H et al. (2015) Obesity and recurrent urinary tract infections in premenopausal women: a retrospective study. Int J Infect Dis 41: 32-5. 327. Saran R, Robinson B, Abbott KC, Agodoa LYC, Bragg-Gresham J, et al. (2019) US Renal Data System 2018 Annual Data Report: Epidemiology of Kidney Disease in the United States. American journal of kidney diseases: the official journal of the National Kidney Foundation vol. 73,3S1 (2019): A7-A8. 306. Flores-Mireles AL, Walker JN, Caparon M, Hultgren SJ (2015) Urinary tract infections: epidemiology, mechanisms of infection and treatment options. Nat Rev Microbiol 13(5): 269-84. 307. Archives in Biomedical Engineering & Biotechnology Gajdács M, Urbán E (2019) Comparative Epidemiology and Resistance Trends of Proteae in Urinary Tract Infections of Inpatients and Outpatients: A 10-Year Retrospective Study. Antibiotics 8(3). pii: E91. 328. Hu MK, Witham MD, Soiza RL (2019) Oral Bicarbonate Therapy in Non- Haemodialysis Dependent Chronic Kidney Disease Patients: A Systematic Review and Meta-Analysis of Randomised Controlled Trials. J Clin Med 8(2) pii: E208. 308. Mickymaray S, Al Aboody MS (2019) In Vitro Antioxidant and Bactericidal Efficacy of 15 Common Spices: Novel Therapeutics for Urinary Tract Infections? Medicina (Kaunas) 55(6). pii: E289. 329. Lv JC, Zhang LX (2019) Prevalence and Disease Burden of Chronic Kidney Disease. Adv Exp Med Biol 1165: 3-15. 309. Liya SJ, Siddique R (2018) Determination of Antimicrobial Activity of Some Commercial Fruit (Apple, Papaya, Lemon and Strawberry) Against Bacteria Causing Urinary Tract Infection. Eur J Microbiol Immunol (Bp) 8(3): 95-99. 330. Motedayen M, Sarokhani D, Ghiasi B, Khatony A, Hasanpour Dehkordi A (2019) Prevalence of Hypertension in Renal Diseases in Iran: Systematic Review and Meta-Analysis. Int J Prev Med 10: 124. 331. Moreno Velásquez I, Tribaldos Causadias M, Valdés R, Gómez B et al. (2019) End-stage renal disease-financial costs and years of life lost in Panama: a cost-analysis study. BMJ Open 9(5): e027229. 310. Albu S, Voidazan S, Bilca D, Badiu M, Truţă A et al. (2018) Bacteriuria and asymptomatic infection in chronic patients with indwelling urinary catheter: The incidence of ESBL bacteria. Medicine (Baltimore) 97(33): e11796. 332. Bikbov B, Global Burden of Disease Study Genitourinary Disease Expert Group (2019) Letter to the Editor: Chronic Kidney Disease - a Neglected Disease in the Economic Conditions Burden Analysis in Korea. J Korean Med Sci 34(32): e220. 311. Frellick M (2017) Drinking More Water Reduces Repeat Urinary Tract Infections. Medscape. 312. Hooton TM, Vecchio M, Iroz A, et al. (2018) Effect of Increased Daily Water Intake in Premenopausal Women With Recurrent Urinary Tract Infections: A Randomized Clinical Trial. JAMA Intern Med 178(11): 1509– 1515. 333. Bello AK, Ronksley PE, Tangri N, Kurzawa J, Osman MA et al. (2019) Prevalence and Demographics of CKD in Canadian Primary Care Practices: A Cross-sectional Study. Kidney Int Rep 4(4): 561-570. 334. Home dialysis as the ‘first’ treatment option for end-stage kidney disease (ESKD). 313. Luo EK (2018) How can you stop a chronic UTI? MedicalNewsToay, Wed. n: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. Archives in Biomedical Engineering & Biotechnology Muraleetharan D, Fadich A, Stephenson C, Garney W (2018) Understanding the Impact of Fibromyalgia on Men: Findings From a Nationwide Survey. Am J Mens Health 12(4):952-960. 285. McCrae CS, Williams J, Roditi D, Anderson R, Mundt JM et al. (2019) Cognitive behavioral treatments for insomnia and pain in adults with comorbid chronic insomnia and fibromyalgia: clinical outcomes from the SPIN randomized controlled trial. Sleep 42(3) pii: zsy234. 264. AK Mohiuddin (2019) Non-drug pain management: opportunities to explore. BiomedGrid LLC, USA ISBN: 978-1-946628-01-5 265. Pappolla MA, Manchikanti L, Andersen CR, Greig NH (2019) Is insulin resistance the cause of fibromyalgia? A preliminary report. PLoS One 14(5): e0216079. 286. Aman MM, Jason Yong R, Kaye AD, Urman RD (2018) Evidence-Based Non-Pharmacological Therapies for Fibromyalgia. Curr Pain Headache Rep 22(5): 33. 266. Palstam A, Mannerkorpi K (2017) Work Ability in Fibromyalgia: An Update in the 21st Century. Curr Rheumatol Rev 13(3):180-187. 287. McCrae CS, Mundt JM, Curtis AF, Craggs JG, O’Shea AM (2018) Gray Matter Changes Following Cognitive Behavioral Therapy for Patients With Comorbid Fibromyalgia and Insomnia: A Pilot Study. J Clin Sleep Med 14(9): 1595-1603. 267. D’Agnelli S, Arendt-Nielsen L, Gerra MC, Zatorri K (2019) Fibromyalgia: Genetics and epigenetics insights may provide the basis for the development of diagnostic biomarkers. Mol Pain 15: 1744806918819944. 288. Karatay S, Okur SC, Uzkeser H, Yildirim K, Akcay F (2018) Effects of Acupuncture Treatment on Fibromyalgia Symptoms, Serotonin, and Substance P Levels: A Randomized Sham and Placebo-Controlled Clinical Trial. Pain Med 19(3): 615-628. 268. Sosa-Reina MD, Nunez-Nagy S, Gallego-Izquierdo T, Pecos-Martín D (2017) Effectiveness of Therapeutic Exercise in Fibromyalgia Syndrome: A Systematic Review and Meta-Analysis of Randomized Clinical Trials. Biomed Res Int 2356346. 289. McPherson ML, Lofton JC (2010) Fibromyalgia: An update for pharmacists. Pharmacy Today 16(1)48–59. 269. Ghavidel-Parsa B, Bidari A, Amir Maafi A, Ghalebaghi B (2015) The Iceberg Nature of Fibromyalgia Burden: The Clinical and Economic Aspects. Korean J Pain 28(3):169-76. 290. Northcott MJ, Guymer EK, Littlejohn GO (2017) Pharmacological treatment options for fibromyalgia. Clinical Pharmacist. Page 35 of 41 Archives in Biomedical Engineering & Biotechnology Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 335. Stemer G, Lemmens-Gruber R (2011) Clinical pharmacy activities in chronic kidney disease and end-stage renal disease patients: a systematic literature review. BMC Nephrol 12: 35. 355. Murphy P, Cocohoba J, Tang A, Pietrandoni G, Hou J et al. (2012) Impact of HIV-specialized pharmacies on adherence and persistence with antiretroviral therapy. AIDS Patient Care STDS 26(9): 526-31. 336. McKinnon A (2010) Practice spotlight: pharmacist in a chronic kidney disease clinic. Can J Hosp Pharm 63(6): 452-3. 356. Grossberg R, Zhang Y, Gross R (2004) A time-to-prescription-refill measure of antiretroviral adherence predicted changes in viral load in HIV. J Clin Epidemiol 57(10): 1107-10. 337. Raymond CB, Wazny LD, Sood AR (2013) Standards of clinical practice for renal pharmacists. Can J Hosp Pharm 66(6): 369-74. 357. Regal GG, Wong W (2019) The Pharmacist’s Role in Maximizing HIV Antiretroviral Therapy. Pharmacy Times®. 338. Cabello-Muriel A, Gascón-Cánovas JJ, Urbieta-Sanz E, Iniesta-Navalón C (2014) Effectiveness of pharmacist intervention in patients with chronic kidney disease. Int J Clin Pharm 36(5): 896-903. 358. Holtzman CW, Brady KA, Yehia BR (2015) Retention in care and medication adherence: current challenges to antiretroviral therapy success. Drugs 75(5): 445-54. 339. Salgado TM, Moles R, Benrimoj SI, Fernandez-Llimos F (2012) Pharmacists’ interventions in the management of patients with chronic kidney disease: a systematic review. Nephrol Dial Transplant 27(1): 276-92. 359. (2018) Pharmacist Intervention and Continued HIV Care: Examining the Connection. Pharmacy Times®. 360. Guilamo-Ramos V, Thimm-Kaiser M, Benzekri A, Futterman D (2019) Shifting the Paradigm in HIV Prevention and Treatment Service Delivery Toward Differentiated Care for Youth. National Academy of medicine. 340. Pai AB, Cardone KE, Manley HJ, St Peter WL, Shaffer R et al. (2013) Medication reconciliation and therapy management in dialysis-dependent patients: need for a systematic approach. Clin J Am Soc Nephrol 8(11): 1988-99. 361. Siegel RL, Miller KD, Jemal A (2019) Cancer statistics, 2019. CA Cancer J Clin 69(1): 7-34. 341. Kempton J, Hill A, Levi JA, Heath K, Pozniak A (2019) Most new HIV infections, vertical transmissions and AIDS-related deaths occur in lower- prevalence countries. J Virus Erad 5(2): 92-101. 362. Shah SC, Kayamba V, Peek RM Jr, Heimburger D (2019) Cancer Control in Low- and Middle-Income Countries: Is It Time to Consider Screening? J Glob Oncol 5: 1-8. 342. Archives in Biomedical Engineering & Biotechnology Eilami O, Nazari A, Dousti M, Sayehmiri F, Ghasemi M (2019) Investigation of HIV/AIDS prevalence and associated risk factors among female sex workers from 2010 to 2017: a meta-analysis study. HIV AIDS (Auckl) 11: 105-117. 363. Ferlay J, Colombet M, Soerjomataram I, Mathers C, Parkin DM et al. (2019) Estimating the global cancer incidence and mortality in 2018: GLOBOCAN sources and methods. Int J Cancer 144(8): 1941-1953. 343. Sulung N, Asyura R (2019) The Analysis of Spirituality of Patients with HIV/AIDS in Taking Lessons and Self-Acceptance. Indian J Palliat Care 25(2): 232-235. 364. Feng RM, Zong YN, Cao SM, Xu RH (2019) Current cancer situation in China: good or bad news from the 2018 Global Cancer Statistics? Cancer Commun (Lond 39(1): 22. 344. Sayyah M, Rahim F, Kayedani GA, Shirbandi K, Saki-Malehi A (2019) Global View of HIV Prevalence in Prisons: A Systematic Review and Meta- Analysis. Iran J Public Health 48(2): 217-226. 365. WHO (2018) Latest global cancer data: Cancer burden rises to 18. 1 million new cases and 9. 6 million cancer deaths in 2018. Presse Release. 366. Schüz J, Espina C, Wild CP (2019) Primary prevention: a need for concerted action. Mol Oncol 13(3): 567-578. 345. Zhu Z, Yan H, Wu S, Xu Y, Xu W et al. (2019) Trends in HIV prevalence and risk behaviours among men who have sex with men from 2013 to 2017 in Nanjing, China: a consecutive cross-sectional survey. BMJ Open 9(1): e021955. 367. WHO (2018) Cancer. Health Topics. 368. World Cancer Day. 346. Silverman JG, Decker MR, Saggurti N, Balaiah D, Raj A (2008) Intimate partner violence and HIV infection among married Indian women. JAMA 300(6): 703-10. 369. Miller KD, Nogueira L, Mariotto AB, Rowland JH, Yabroff KR et al. (2019) Cancer treatment and survivorship statistics, 2019. CA Cancer J Clin. 370. DeSantis CE, Miller KD, Goding Sauer A, Jemal A, Siegel RL (2019) Cancer statistics for African Americans, 2019. CA Cancer J Clin 69(3): 211- 233. 347. Global Burden of Disease Health Financing Collaborator Network (2018) Spending on health and HIV/AIDS: domestic health spending and development assistance in 188 countries, 1995-2015. Lancet 391(10132): 1799-1829. 371. Layne TM, Graubard BI, Ma X, Mayne ST, Albanes D (2019) Prostate cancer risk factors in black and white men in the NIH-AARP Diet and Health Study. Prostate Cancer Prostatic Dis 22(1): 91-100. 348. Archives in Biomedical Engineering & Biotechnology ABEB. 00566. DOI: 10.33552/ABEB.2019.03.000566. uddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. ABEB.2019.03.000566. Page 36 of 41 Archives in Biomedical Engineering & Biotechnology Atta MG, De Seigneux S, Lucas GM (2019) Clinical Pharmacology in HIV Therapy. Clin J Am Soc Nephrol 14(3): 435-444. 372. Ghose S, Radhakrishnan V, Bhattacharya S (2019) Ethics of cancer care: beyond biology and medicine. Ecancermedicalscience 13: 911. 349. Tseng A, Foisy M, Hughes CA, Kelly D, Chan S et al. (2012) Role of the Pharmacist in Caring for Patients with HIV/AIDS: Clinical Practice Guidelines. Can J Hosp Pharm 65(2): 125-45. 373. Allcott N, Dunham L, Levy D, Carr J, Stitzenberg K (2019) Financial burden amongst cancer patients treated with curative intent surgery alone. Am J Surg pii: S0002-9610(18)31611-8. 350. Nevo ON, Lesko CR, Colwell B, Ballard C, Cole SR et al. (2015) Outcomes of pharmacist-assisted management of antiretroviral therapy in patients with HIV infection: A risk-adjusted analysis. Am J Health Syst Pharm 72(17): 1463-70. 374. Coumoundouros C, Ould Brahim L, Lambert SD, McCusker J (2019) The direct and indirect financial costs of informal cancer care: A scoping review. Health Soc Care Community. 351. Molino CGRC, Carnevale RC, Rodrigues AT, Moriel P, Mazzola PG (2017) HIV pharmaceutical care in primary healthcare: Improvement in CD4 count and reduction in drug-related problems. Saudi Pharm J 25(5): 724-733. 375. WHO (2018) $46 billion in productivity lost to cancer in major emerging economies. Presse Release. 352. Molino Cde G, Carnevale RC, Rodrigues AT, Visacri MB, Moriel P et al. (2014) Impact of pharmacist interventions on drug-related problems and laboratory markers in outpatients with human immunodeficiency virus infection. Ther Clin Risk Manag 10: 631-9. 376. World Cancer Day. 377. PDQ Screening and Prevention Editorial Board (2002) Cancer Prevention Overview (PDQ®): Health Professional Version. 2019 Aug 2. In: PDQ Cancer Information Summaries. Bethesda (MD): National Cancer Institute (US). 353. Carnevale RC, de Godoi Rezende Costa Molino C, Visacri MB, Mazzola PG, Moriel P (2015) Cost analysis of pharmaceutical care provided to HIV- infected patients: an ambispective controlled study. Daru 23: 13. 378. Sung H, Siegel RL, Rosenberg PS, Jemal A (2019) Emerging cancer trends among young adults in the USA: analysis of a population-based cancer registry. Lancet Public Health 4(3): e137-e147. 354. Ma A, Chen DM, Chau FM, Saberi P (2010) Improving adherence and clinical outcomes through an HIV pharmacist’s interventions. AIDS Care 22(10): 1189-94. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. uddin. Clinical Pharmacists in Chronic Care [Part 2]. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Archives in Biomedical Engineering & Biotechnology Zappa C, Mousa SA (2016) Non-small cell lung cancer: current treatment and future advances. Transl Lung Cancer Re 5(3): 288-300. 385. Kwon JH, Lee SC, Lee MA, Kim YJ, Kang JH et al. (2019) Behaviors and Attitudes toward the Use of Complementary and Alternative Medicine among Korean Cancer Patients. Cancer Res Treat 51(3): 851-860. 406. Lin CK, Lin RT, Chen T, Zigler C, Wei Y, et al. (2019) A global perspective on coal-fired power plants and burden of lung cancer. Environ Health 18(1): 9. 386. Jones E, Nissen L, McCarthy A, Steadman K, Windsor C (2019) Exploring the Use of Complementary and Alternative Medicine in Cancer Patients. Integr Cancer Ther 18: 1534735419854134. 407. Alberg AJ, Brock MV, Ford JG, Samet JM, Spivack SD (2013) Epidemiology of lung cancer: Diagnosis and management of lung cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines. Chest 143(5 Suppl): e1S-e29S. 387. Chotipanich A, Sooksrisawat C, Jittiworapan B (2019) Association between complementary and alternative medicine use and prolonged time to conventional treatment among Thai cancer patients in a tertiary-care hospital. PeerJ 7: e7159. 408. Cabanero M, Sangha R, Sheffield BS, Sukhai M, Pakkal M, et al. (2017) Management of EGFR-mutated non-small-cell lung cancer: practical implications from a clinical and pathology perspective. Curr Oncol 24(2): 111-119. 388. Ekor M (2014) The growing use of herbal medicines: issues relating to adverse reactions and challenges in monitoring safety. Front Pharmacol 4: 177. 409. Abdul Kader Mohiuddin (2019) Alternative Treatments for Cancer Prevention and Cure [Part 1]. Adv Pharmacol ClinTrials 4(4): 000168. 389. Haefeli WE, Carls A (2014) Drug interactions with phytotherapeutics in oncology. Expert Opin Drug Metab Toxicol 10(3): 359-77. 410. Pisters KM, Vallières E, Crowley JJ, Franklin WA, Bunn PA, et al. (2010) Surgery with or without preoperative paclitaxel and carboplatin in early-stage non-small-cell lung cancer: Southwest Oncology Group Trial S9900, an intergroup, randomized, phase III trial. J Clin Oncol 28(11): 1843- 1849. 390. Kabat G (2018) Why Resorting to Alternative Medicine To Treat Cancer Is A Bad Idea. Forbes. 391. Wong SF, Bounthavong M, Nguyen C, Bechtoldt K, Hernandez E (2014) Implementation and preliminary outcomes of a comprehensive oral chemotherapy management clinic. Am J Health Syst Pharm 71(11): 960-5. 411 (2007) Pharmacists at the front line in early detection of lung cancer. The Pharmaceutical Journal. 392. Archives in Biomedical Engineering & Biotechnology McKee M, Frei BL, Garcia A, Fike D, Soefje SA (2011) Impact of clinical pharmacy services on patients in an outpatient chemotherapy academic clinic. J Oncol Pharm Pract 17(4): 387-94. 412. (2019) Lung cancer: A patient’s perspective. Pharmacy Magazine (UK). 413. Robbins SL, Kumar V, Cotran RS (2010) Robbins and Cotran Pathologic Basis of Disease. Philadelphia, PA: Elsevier. 393. Walter C, Mellor JD, Rice C, Kirsa S, Ball D et al. (2016) Impact of a specialist clinical cancer pharmacist at a multidisciplinary lung cancer clinic. Asia Pac J Clin Oncol 12(3): e367-74. 414. Vakiti A, Mewawalla P. Cancer, Acute Myeloid Leukemia (AML, Erythroid Leukemia, Myelodysplasia-Related Leukemia, BCR-ABL Chronic Leukemia) In: Stat Pearls. Treasure Island (FL): Stat Pearls P. 394. Allison J, Fisher J, Souter C, Bennie M (2019) What patient assessment skills are required by pharmacists prescribing systemic anti-cancer therapy? A consensus study. J Oncol Pharm Pract 25(8): 1933-1944. 415. Naik J, Themeli M, de Jong-Korlaar R, Ruiter RWJ, Poddighe PJ, et al. (2019) CD38 as a therapeutic target for adult acute myeloid leukemia and T-cell acute lymphoblastic leukemia. Haematologica 104(3): e100-e103. 395. Morimoto Y, Takei H, Tachibana K, Nakazato Y, Tanaka R et al. (2018) [Role of Pharmacists in Completion of Adjuvant Cisplatin-Vinorelbine Chemotherapy in Japanese Patients with Non-small Cell Lung Cancer]. Yakugaku Zasshi 138(3): 437-442. 416. Bawazir A, Al-Zamel N, Amen A, Akiel MA, Alhawiti NM, et al. (2019) The burden of leukemia in the Kingdom of Saudi Arabia: 15 years period (1999-2013). BMC Cancer 19(1): 703. 417. Dhall A, Zee BM, Yan F, Blanco MA (2019) Intersection of Epigenetic and Metabolic Regulation of Histone Modifications in Acute Myeloid Leukemia. Front Oncol 9: 432. 396. Abdul Kader Mohiuddin (2019) Cost of Biotech Drug Development and Affordability Issues in LMICs. Arch Biomed Eng & Biotechnol 2(3): 2019. ABEB.MS.ID.000538. 418. Medeiros BC, Chan SM, Daver NG, Jonas BA, Pollyea DA (2019) Optimizing survival outcomes with post-remission therapy in acute myeloid leukemia. Am J Hematol 94(7): 803-811. 397. Kim S-K, Kalimuthu S (2014) Chapter 1. Introduction to Anti-cancer Drugs from Marine Origin. In: Se-Kwon Kim. Handbook of Anticancer Drugs from Marine Origin. Publisher: Springer. 419. Murphy BR, Roth M, Kolb EA, Alonzo T, Gerbing R, et al. (2019) Development of acute lymphoblastic leukemia following treatment for acute myeloid leukemia in children with Down syndrome: A case report and 398. Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 379. Park B, You S, Cho WCS, Choi JY, Lee MS (2019) A systematic review of herbal medicines for the treatment of cancer cachexia in animal models. J Zhejiang Univ Sci B 20(1): 9-22. 399. Rezaei S, Akbari Sari A, Woldemichael A, Soofi M, Kazemi A et al. (2016) Estimating the Economic Burden of Lung Cancer in Iran. Asian Pac J Cancer Prev 17(10): 4729-4733. 400. Wood R, Taylor-Stokes G (2019) Cost burden associated with advanced non-small cell lung cancer in Europe and influence of disease stage. BMC Cancer 19(1): 214. 380. Zhang Q (2015). Traditional and Complementary Medicine in Primary Health Care. In: Medcalf A, Bhattacharya S, Momen H (ed.) Health For All: The Journey of Universal Health Coverage. Hyderabad (IN): Orient Blackswan. Chapter 12 USA. 401. Sun L, Yim WS, Fahey P, Wang S, Zhu X, et al. (2019) Investigation on Advanced Non-Small-Cell Lung Cancer among Elderly Patients Treated with Chinese Herbal Medicine versus Chemotherapy: A Pooled Analysis of Individual Data. Evid Based Complement Alternat Med 2019: 1898345. 381. Mawoza, T, Nhachi, C Magwali, (2019) Prevalence of Traditional Medicine Use during Pregnancy, at Labour and for Postpartum Care in a Rural Area in Zimbabwe. Clinics in mother and child health 16(2),pii: 321. 382. Krupa J, Sureshkumar J, Silambarasan R, Priyadarshini K, Ayyanar M (2019) Integration of traditional herbal medicines among the indigenous communities in Thiruvarur District of Tamil Nadu, India. Journal of Ayurveda and integrative medicine, 10(1), 32–37. 402. Abdelaziz HM, Elzoghby AO, Helmy MW, Samaha MW, Fang JY, et al. (2019) Liquid crystalline assembly for potential combinatorial chemo- herbal drug delivery to lung cancer cells. Int J Nanomedicine 14: 499-517. 403. Blandin Knight S, Crosbie PA, Balata H, Chudziak J, Hussell T, et al. (2017) Progress and prospects of early detection in lung cancer. Open Biol 7(9): 170070. 383. Wode K, Henriksson R, Sharp L, Stoltenberg A, Hök Nordberg J (2019) Cancer patients’ use of complementary and alternative medicine in Sweden: a cross-sectional study. BMC Complement Altern Med 19(1):62. 404. Seung SJ, Hurry M, Hassan S, Walton RN, Evans WK (2019) Cost-of- illness study for non-small-cell lung cancer using real-world data. Curr Oncol 26(2): 102-107. 384. Jermini M, Dubois J, Rodondi PY, Zaman K, Buclin T et al. (2019) Complementary medicine use during cancer treatment and potential herb- drug interactions from a cross-sectional study in an academic centre. Sci Rep 9(1): 5078. 405. Archives in Biomedical Engineering & Biotechnology Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. /ABEB.2019.03.000566. Page 37 of 41 Archives in Biomedical Engineering & Biotechnology Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Archives in Biomedical Engineering & Biotechnology Poehls UG, Hack CC, Wunderle M, Renner SP, Lux MP, et al. (2019) Awareness of breast cancer incidence and risk factors among healthy women in Germany: an update after 10 years. Eur J Cancer Prev. 455. Araghi M, Soerjomataram I, Jenkins M, Brierley J, Morris E, (2019) Global trends in colorectal cancer mortality: projections to the year 2035. Int J Cancer 144(12): 2992-3000. 434. Hendrick RE, Baker JA, Helvie MA (2019) Breast cancer deaths averted over 3 decades. Cancer 125(9): 1482-1488. 456. Recio-Boiles A, Waheed A, Cagir B (2019) Cancer, Colon In: Stat Pearls Treasure Island (FL): Stat Pearls. 435. Vishwakarma G, Ndetan H, Das DN, Gupta G, Suryavanshi M, et al. (2019) Reproductive factors and breast cancer risk: A meta-analysis of case- control studies in Indian women. South Asian J Cancer 8(2): 80-84. 457. Virk GS, Jafri M, Ashley C (2019) Colonoscopy and colorectal cancer rates among octogenarians and nonagenarians: nationwide study of US veterans. Clin Interv Aging 14: 609-614. 436. Harbeck N, Penault-Llorca F, Cortes J, Gnant M, Houssami N, et al. (2019) Breast cancer. Nat Rev Dis Primers 5(1): 66. 458. (2002) PDQ Screening and Prevention Editorial Board. Colorectal Cancer Prevention (PDQ®): Health Professional Version. 2019 Apr 16. In: PDQ Cancer Information Summaries. Bethesda (MD): National Cancer Institute (US). 437. Ng HS, Vitry A, Koczwara B, Roder D, McBride ML (2019) Patterns of comorbidities in women with breast cancer: a Canadian population-based study. Cancer Causes Control 30(9): 931-941. 438. Bahri N, Fathi Najafi T, Homaei Shandiz F, Tohidinik HR, Khajavi A (2019) The relation between stressful life events and breast cancer: a systematic review and meta-analysis of cohort studies. Breast Cancer Res Treat 176(1): 53-61. 459. Chen W, Zheng R, Baade PD, Zhang S, Zeng H, et al. (2016) Cancer statistics in China, 2015. CA Cancer J Clin 66(2): 115-132. 460. Araghi M, Soerjomataram I, Bardot A, Ferlay J, Cabasag CJ, et al. (2019) Changes in colorectal cancer incidence in seven high-income countries: a population-based study. Lancet Gastroenterol Hepatol 4(7): 511-518. 439. Ho-Huynh A, Tran A, Bray G, Abbot S, Elston T, et al. (2019) Factors influencing breast cancer outcomes in Australia: A systematic review. Eur J Cancer Care (Engl) 28(4): e13038. 461. Wray AJD, Minaker LM (2019) Is cancer prevention influenced by the built environment? A multidisciplinary scoping review. Cancer. 125(19): 3299-3311. 440. Feng Y, Spezia M, Huang S, Yuan C, Zeng Z, et al. Archives in Biomedical Engineering & Biotechnology Lam MS, Cheung N (2016) Impact of oncology pharmacist-managed oral anticancer therapy in patients with chronic myelogenous leukemia. J Oncol Pharm Pract 22(6): 741-748. 448. Yuan X, Yi F, Hou C, Lee H, Zhong X, et al. (2019) Induced Abortion, Birth Control Methods, and Breast Cancer Risk: A Case-Control Study in China. J Epidemiol 29(5): 173-179. 427. Erfani N, Nazemosadat Z, Moein M (2015) Cytotoxic activity of ten algae from the Persian Gulf and Oman Sea on human breast cancer cell lines; MDA-MB-231, MCF-7, and T-47D. Pharmacognosy Res 7(2): 133-137. 449. Yabroff KR, Lund J, Kepka D, Mariotto A (2011) Economic burden of cancer in the United States: estimates, projections, and future research. Cancer Epidemiol Biomarkers. Prev Oct 20(10): 2006-2014. 428. Joseph OA (2016) The prospects of medicinal plants in the treatment of breast cancer. European Pharmaceutical Review. 450. Mariotto AB, Yabroff KR, Shao Y, Feuer EJ, Brown ML (2011) Projections of the cost of cancer care in the United States: 2010-2020. J Natl Cancer Inst 103(2): 117-128. 429. Azubuike SO, Muirhead C, Hayes L, McNally R (2018) Rising global burden of breast cancer: the case of sub-Saharan Africa (with emphasis on Nigeria) and implications for regional development: a review. World J Surg Oncol 16(1): 63. 451. Iadeluca L, Mardekian J, Chander P, Hopps M, Makinson GT (2017) The burden of selected cancers in the US: health behaviors and health care resource utilization. Cancer Manag Res 9: 721-730. 430. Blumen H, Fitch K, Polkus V (2016) Comparison of Treatment Costs for Breast Cancer, by Tumor Stage and Type of Service. Am Health Drug Benefits 9(1): 23-32. 452. Tanaka K, Hori A, Tachi T, Osawa T, Nagaya K, et al. (2018) Impact of pharmacist counseling on reducing instances of adverse events that can affect the quality of life of chemotherapy outpatients with breast Cancer. J Pharm Health Care Sci. 4: 9. 431. Alkabban FM, Ferguson T (2019) Cancer, Breast. In: Stat Pearls. Treasure Island (FL): Stat Pearls. 453. Wang Y, Wu H, Xu F (2015) Impact of Clinical Pharmacy Services on KAP and QOL in Cancer Patients: A Single-Center Experience. Biomed Res Int 2015: 502431. 432. Mehrgou A, Akouchekian M (2016) The importance of BRCA1 and BRCA2 genes mutations in breast cancer development. Med J Islam Repub Iran 30: 369. 454. Val R, Allison R (2017) Treatment of Advanced Breast Cancer: A Clinical Pharmacy Perspective. American Health and Drug Benefits. 433. Archives in Biomedical Engineering & Biotechnology Jeon SM, Kwon JW, Choi SH, Park HY (2019) Economic burden of lung cancer: A retrospective cohort study in South Korea, 2002-2015. PLoS One 14(2): e0212878. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Page 38 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 retrospective review of Children’s Oncology Group acute myeloid leukemia trials. Pediatr Blood Cancer 66(8): e27700. 442. (2017) InformedHealth.org. Cologne, Germany: Institute for Quality and Efficiency in Health Care (IQWiG); 2006-. Risk factors for breast cancer. 420. Puckett Y, Chan O (2019) Cancer, Acute Lymphocytic Leukemia. In: Stat Pearls. Treasure Island (FL): Stat Pearls. 443. Alexander A, Kaluve R, Prabhu JS, Korlimarla A, Srinath BS, et al. (2019) The Impact of Breast Cancer on the Patient and the Family in Indian Perspective. Indian J Palliat Care 25(1): 66-72. 421. Branford S, Kim DDH, Apperley JF, Eide CA, Mustjoki S, et al. (2019) International CML Foundation Genomics Alliance. Laying the foundation for genomically-based risk assessment in chronic myeloid leukemia. Leukemia 33(8): 1835-1850. 444. Chen L, Malone KE, Li CI (2014) Bra wearing not associated with breast cancer risk: a population-based case-control study. Cancer Epidemiol Biomarkers Prev 23(10): 2181-2185. 422. Franki R (2018) Report details financial burden of blood cancers. MD edge. 445. Rios SSD, Chen ACR, Chen JR, et al. (2016) Wearing a Tight Bra for many hours a day is associated with increased risk of breast cancer. Adv Oncol Res Treat 1:1–5. 423. Davenport L (2016) Costs of Treating Blood Disorders Explored. Medscape. 446. Virani S, Chindaprasirt J, Wirasorn K, Sookprasert A, Somintara O, et al. (2018) Breast Cancer Incidence Trends and Projections in Northeastern Thailand. J Epidemiol 28(7): 323-330. 424. Ramos NR, Mo CC, Karp JE, Hourigan CS (2015) Current Approaches in the Treatment of Relapsed and Refractory Acute Myeloid Leukemia. J Clin Med 4(4): 665-695. 447. Institute of Medicine (US) Committee on the Relationship Between Oral Contraceptives and Breast Cancer. Oral Contraceptives & Breast Cancer. Washington (DC): National Academies Press (US); 1991. A, Oral Contraceptives and Breast Cancer: A Review of the Epidemiological Evidence with an Emphasis on Younger Women. 425. Schlafer D, Panjic EH, Harvey RD (2017) The Evolving Role of the Hematology/Oncology Pharmacist. American Society of Hematology. 426. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Archives in Biomedical Engineering & Biotechnology Cossu G, Saba L, Minerba L, Mascalchi M (2018) Colorectal Cancer Screening: The Role of Psychological, Social and Background Factors in Decision-making Process. Clin Pract Epidemiol Ment Health14: 63-69. 469. Erdrich J, Zhang X, Giovannucci E, Willett W (2015) Proportion of colon cancer attributable to lifestyle in a cohort of US women. Cancer Causes Control 26(9): 1271-1279. 491. Montgomery GH, Schnur JB, Kravits K (2013) Hypnosis for cancer care: over 200 years young. CA Cancer J Clin 63(1): 31-44. 470. Aleksandrova K, Pischon T, Jenab M, Bueno-de-Mesquita HB, Fedirko V, et al. (2014) Combined impact of healthy lifestyle factors on colorectal cancer: a large European cohort study. BMC Med 12: 168. 492. Miller SJ, Schnur JB, Montgomery GH, Jandorf L (2011) African Americans’ and Latinos’ perceptions of using hypnosis to alleviate distress before a colonoscopy. Contemp Hypn Integr Ther 28(3):196-203. 471. Jeon J, Du M, Schoen RE, Hoffmeister M, Newcomb PA, et al. (2018) Colorectal Transdisciplinary Study and Genetics and Epidemiology of Colorectal Cancer Consortium. Determining Risk of Colorectal Cancer and Starting Age of Screening Based on Lifestyle, Environmental, and Genetic Factors. Gastroenterology 154(8): 2152-2164.e19. 493. Umezawa S, Higurashi T, Uchiyama S, Sakai E, Ohkubo H, et al. (2015) Visual distraction alone for the improvement of colonoscopy-related pain and satisfaction. World J Gastroenterol 21(15): 4707-4714. Starting Age of Screening Based on Lifestyle, Environmental, and Genetic Factors. Gastroenterology 154(8): 2152-2164.e19. 494. Lee DW, Chan AC, Wong SK, Fung TM, Li AC, et al. (2004) Can visual distraction decrease the dose of patient-controlled sedation required during colonoscopy? A prospective randomized controlled trial. Endoscopy 36(3): 197-201. 472. Oruç Z, Kaplan MA (2019) Effect of exercise on colorectal cancer prevention and treatment. World J Gastrointest Oncol 11(5): 348-366. 473. Bradbury KE, Murphy N, Key TJ (2019) Diet and colorectal cancer in UK Biobank: a prospective study. Int J Epidemiol 495. Xiaolian J, Xiaolin L, Lan ZH (2015) Effects of visual and audiovisual distraction on pain and anxiety among patients undergoing colonoscopy. Gastroenterol Nurs 38(1): 55-61. 474. Fedirko V, Tramacere I, Bagnardi V, Rota M, Scotti L, et al. (2011) Alcohol drinking and colorectal cancer risk: an overall and dose-response meta-analysis of published studies. Ann Oncol 22(9): 1958-1972. 496. Leung FW (2008) Methods of reducing discomfort during colonoscopy. Dig Dis Sci 53(6): 1462-1467. 475. Vieira AR, Abar L, Chan DSM, Vingeliene S, Polemiti E, et al. Archives in Biomedical Engineering & Biotechnology Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 485. Shapiro NL (2006) Chapter 126. Preventive Care. In: David B. Troy, Paul Beringer (eds.) Remington: The Science and Practice of Pharmacy, Contributor: Joseph Price Remington, Publisher: Lippincott Williams & Wilkins. 463. Poirier AE, Ruan Y, Hebert LA, Grevers X, Walter SD, et al. (2019) Estimates of the current and future burden of cancer attributable to low fruit and vegetable consumption in Canada. Prev Med 122: 20-30. 464. Feletto E, Yu XQ, Lew JB, St John DJB, Jenkins MA, et al. (2019) Trends in Colon and Rectal Cancer Incidence in Australia from 1982 to 2014: Analysis of Data on Over 375,000 Cases. Cancer Epidemiol Biomarkers Prev 28(1): 83-90. 464. Feletto E, Yu XQ, Lew JB, St John DJB, Jenkins MA, et al. (2019) Trends in Colon and Rectal Cancer Incidence in Australia from 1982 to 2014: Analysis 486. Cho YA, Lee J, Oh JH, Chang HJ, Sohn DK, et al. (2019) Genetic Risk Score, Combined Lifestyle Factors and Risk of Colorectal Cancer. Cancer Res Treat 51(3): 1033-1040. of Data on Over 375,000 Cases. Cancer Epidemiol Biomarkers Prev 28(1): 83-90. 487. Pais R, Silaghi H, Silaghi AC, Rusu ML, Dumitrascu DL (2009) Metabolic syndrome and risk of subsequent colorectal cancer. World J Gastroenterol 15(41): 5141-5148. 465. Jenkins MA, Ait Ouakrim D, Boussioutas A, Hopper JL, Ee HC, et al. (2018) Revised Australian national guidelines for colorectal cancer screening: family history. Med J Aust 209(10): 455-460. 488. Rossi M, Jahanzaib Anwar M, Usman A, Keshavarzian A, et al. (2018) Colorectal Cancer and Alcohol Consumption-Populations to Molecules. Cancers (Basel). 10(2): E38. 466. Glover M, Mansoor E, Panhwar M, Parasa S, Cooper GS (2019) Epidemiology of Colorectal Cancer in Average Risk Adults 20-39 Years of Age: A Population-Based National Study. Dig Dis Sci. 467. Siegel RL, Medhanie GA, Fedewa SA, Jemal A (2019) State variation in early-onset colorectal cancer in the United States, 1995-2015. J Natl Cancer Inst 111(10):1104-1106. 489. Elkins G, White J, Patel P, Marcus J, Perfect MM, Montgomery GH. (2006) Hypnosis to manage anxiety and pain associated with colonoscopy for colorectal cancer screening: Case studies and possible benefits. Int J Clin Exp Hypn 54(4): 416-431. 468. Mannucci A, Zuppardo RA, Rosati R, Leo MD, Perea J, et al. (2019) Colorectal cancer screening from 45 years of age: Thesis, antithesis and synthesis. World J Gastroenterol 25(21): 2565-2580. 490. Archives in Biomedical Engineering & Biotechnology (2017) Foods and beverages and colorectal cancer risk: a systematic review and meta-analysis of cohort studies, an update of the evidence of the WCRF- AICR Continuous Update Project. Ann Oncol 28(8): 1788-1802. 497. De Silva AP, Niriella MA, Nandamuni Y, Nanayakkara SD, Perera KR, et al. (2016) Effect of audio and visual distraction on patients undergoing colonoscopy: a randomized controlled study. Endosc Int Open 4(11): E1211-E1214. 476. Pericleous M, Mandair D, Caplin ME (2013) Diet and supplements and their impact on colorectal cancer. J Gastrointest Oncol 4(4): 409-423. 498. Drury J (2013) Colorectal Cancer Risks and the Community Pharmacist’s Role. Pharmacy Times®. 477. Chen Z, Wang PP, Woodrow J, Zhu Y, Roebothan B, et al. (2015) Dietary patterns and colorectal cancer: results from a Canadian population-based study. Nutr J 14: 8. 499. Tezcan S, Izzettin FV, Sancar M, Turhal NS, Yumuk PF (2018) Role of clinical oncology pharmacist in determination of pharmaceutical care needs in patients with colorectal cancer. Eur J Hosp Pharm. 25(e1): e17-e20. 478. National Cancer Institute. Financial Burden of Cancer Care. 500. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, et al. (2018) Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 68(6): 394-424. 479. American Cancer Society. Colorectal Cancer Facts & Figures 2017- 2019. 501. Rawla P (2019) Epidemiology of Prostate Cancer. World J Oncol 10(2): 63-89. 480. Gellad ZF, Provenzale D (2010) Colorectal cancer: national and international perspective on the burden of disease and public health impact. Gastroenterology 138(6): 2177-2190. 502. Smith-Palmer J, Takizawa C, Valentine W (2019) Literature review of the burden of prostate cancer in Germany, France, the United Kingdom and Canada. BMC Urol 19(1): 19. 481. World Cancer Research Fund International. Colorectal cancer statistics. 482. Rodriguez-Bigas MA, Lin EH, Crane CH (2003) Stage IV Colorectal Cancer. In: Kufe DW, Pollock RE, Weichselbaum RR (ed.) Holland-Frei Cancer Medicine. 6th edition. Hamilton (ON): BC Decker. 503. Roehrborn CG, Black LK (2011) The economic burden of prostate cancer. BJU Int108(6): 806-813. 504. (2008) Economic Cost of Cancer Mortality Is High in U.S., Regardless of How Cost Is Measured, JNCI: Journal of the National Cancer Institute 100(24): 1741. 483. Lee SD, Choe JW, Lee BJ, Kang MH, Joo MK, et al. (2015) Butein effects in colitis and interleukin-6/signal transducer and activator of transcription 3 expression. World J Gastroenterol 21(2): 465-474. 505. Archives in Biomedical Engineering & Biotechnology (2018) Breast cancer development and progression: Risk factors, cancer stem cells, signaling pathways, genomics, and molecular pathogenesis. Genes Dis 5(2): 77-106. 462. Poirier AE, Ruan Y, Walter SD, Franco EL, Villeneuve PJ, et al. (2019) The future burden of cancer in Canada: Long-term cancer incidence projections 2013-2042. Cancer Epidemiol 59: 199-207. 441. Sun YS, Zhao Z, Yang ZN, Xu F, Lu HJ, et al. (2017) Risk Factors and Preventions of Breast Cancer. Int J Biol Sci 13(11): 1387-1397. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Page 39 of 41 Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Citation: Abdul Kader Mohiuddin. Clinical Pharmacists in Chronic Care [Part 2]. Arch Biomed Eng & Biotechnol. 3(4): 2019. ABEB. MS.ID.000566. DOI: 10.33552/ABEB.2019.03.000566. Page 41 of 41 Archives in Biomedical Engineering & Biotechnology Pernar CH, Ebot EM, Wilson KM, Mucci LA The Epidemiology of Prostate Cancer. Cold Spring Harb Perspect Med 8(12): a030361. 484. James S, Aparna JS, Paul AM, Lankadasari MB, Mohammed S, et al. (2017) Cardamonin inhibits colonic neoplasia through modulation of MicroRNA expression. Sci Rep 7(1): 13945. 506. Wilson KM, Markt SC, Fang F, Nordenvall C, Rider JR, et al. (2016) Snus use, smoking and survival among prostate cancer patients. Int J Cancer 139(12): 2753-2759. Page 40 of 41 Archives in Biomedical Engineering & Biotechnology Volume 3-Issue 4 513. Kachur E (2018) Prostate Cancer Review. US Pharm 43(8): HS-7-HS-12. 507. Zuccolo L, Lewis SJ, Donovan JL, Hamdy FC, Neal DE, et al. (2013) Alcohol consumption and PSA-detected prostate cancer risk--a case-control nested in the Protect study Int J Cancer 132(9): 2176-2185. 514. Patel JM, Holle LM, Clement JM, Bunz T, Niemann C, et al. (2016) Impact of a pharmacist-led oral chemotherapy-monitoring program in patients with metastatic castrate-resistant prostate cancer. J Oncol Pharm Pract 22(6): 777-783. 508. Zhao J, Stockwell T, Roemer A, Chikritzhs T (2016) Is alcohol consumption a risk factor for prostate cancer? A systematic review and meta-analysis. BMC Cancer 16(1): 845. 515. McNatty A, Dow E, Bryce AH Impact of pharmacist-led monitoring of olaparib (O) for metastatic castrate resistant prostate cancer (mCRPC). Journal of Clinical Oncology 35(6suppl): e569-e569. 509. Thobe MN, Clark RJ, Bainer RO, Prasad SM, Rinker-Schaeffer CW (2011) From prostate to bone: key players in prostate cancer bone metastasis. Cancers (Basel) 3(1): 478-493. 516. Rivera-Ruiz KM Role of The Pharmacist in Prostate Cancer Prevention and Management. 510. Wallner L, Frencher S, Hsu JW, Loo R, Huang J, et al. (2012) Prostate cancer screening trends in a large, integrated health care system. Perm J 16(3): 4-9. 517. Kletas V, De Lemos ML (2012) Prostate Cancer: Pharmacist’s Role in Prevention & Management. Canadian Healthcare Network.ca (Pharmacy Practice). 511. Allard CB, Dason S, Lusis J, Kapoor A (2012) Prostate cancer screening: Attitudes and practices of family physicians in Ontario. Can Urol Assoc J 6(3): 188-1893. 518. Cunliffe A, Harker N, Nicola S (2017) Top tips: prostate cancer. Guidelines in Practice. 512. Scarabelin A, Santana Dosea A, Aguiar PM, Storpirtis S (2019) Pharmacist-Patient Communication in Prostate Cancer as a Strategy to Humanize Health Care: A Qualitative Study. J Patient Exp 6(2): 150-156. Page 41 of 41
https://openalex.org/W2107551808	https://digitalcommons.unl.edu/cgi/viewcontent.cgi?article=1324&context=usdoepub	English	null	Unique electron polarimeter analyzing power comparison and precision spin-based energy measurement	Physical review special topics. Accelerators and beams	2,004	cc-by	13,499	This Article is brought to you for free and open access by the U.S. Department of Energy at DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in US Department of Energy Publications by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln. University of Nebraska - Lincoln University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln US Department of Energy Publications U.S. Department of Energy 2004 Unique electron polarimeter analyzing power comparison and Unique electron polarimeter analyzing power comparison and precision spin-based energy measurement precision spin-based energy measurement J. Grames Thomas Jefferson National Accelerator Facility, grames@jlab.org C. K. Sinclair Thomas Jefferson National Accelerator Facility J. Mitchell Thomas Jefferson National Accelerator Facility E. Chudakov Thomas Jefferson National Accelerator Facility H. Fenker Thomas Jefferson National Accelerator Facility See next page for additional authors Follow this and additional works at: https://digitalcommons.unl.edu/usdoepub Part of the Bioresource and Agricultural Engineering Commons Grames, J.; Sinclair, C. K.; Mitchell, J.; Chudakov, E.; Fenker, H.; Freyberger, A.; Higinbotham, D.W.; Poelker, M.; Steigerwald, M.; Tiefenback, M.; Cavata, C.; Escoffier, S.; Marie, F.; Pussieux, T.; Vernin, P.; Danagoulian, S.; Dharmawardane, V.; Fatemi, R.; Joo, K.; Zeier, M.; Gorbenko, V.; Nasseripour, R.; Raue, B.; Suleiman, R.; and Zihlmann, B., "Unique electron polarimeter analyzing power comparison and precision spin-based energy measurement" (2004). US Department of Energy Publications. 325. https://digitalcommons.unl.edu/usdoepub/325 University of Nebraska - Lincoln University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln US Department of Energy Publications U.S. Department of Energy 2004 Unique electron polarimeter analyzing power comparison and Unique electron polarimeter analyzing power comparison and precision spin-based energy measurement precision spin-based energy measurement J. Grames Thomas Jefferson National Accelerator Facility, grames@jlab.org C. K. Sinclair Thomas Jefferson National Accelerator Facility J. Mitchell Thomas Jefferson National Accelerator Facility E. Chudakov Thomas Jefferson National Accelerator Facility H. Fenker Thomas Jefferson National Accelerator Facility See next page for additional authors Follow this and additional works at: https://digitalcommons.unl.edu/usdoepub Part of the Bioresource and Agricultural Engineering Commons Grames, J.; Sinclair, C. K.; Mitchell, J.; Chudakov, E.; Fenker, H.; Freyberger, A.; Higinbotham, D.W.; Poelker, M.; Steigerwald, M.; Tiefenback, M.; Cavata, C.; Escoffier, S.; Marie, F.; Pussieux, T.; Vernin, P.; Danagoulian, S.; Dharmawardane, V.; Fatemi, R.; Joo, K.; Zeier, M.; Gorbenko, V.; Nasseripour, R.; Raue, B.; Suleiman, R.; and Zihlmann, B., "Unique electron polarimeter analyzing power comparison and precision spin-based energy measurement" (2004). US Department of Energy Publications. 325. https://digitalcommons.unl.edu/usdoepub/325 This Article is brought to you for free and open access by the U.S. Department of Energy at Di it lC @U i it f N b k Li l It h b t d f i l i i US D Authors Authors J. Grames, C. K. Sinclair, J. Mitchell, E. Chudakov, H. Fenker, A. Freyberger, D.W. Higinbotham, M. Poelker, M. Steigerwald, M. Tiefenback, C. Cavata, S. Escoffier, F. Marie, T. Pussieux, P. Vernin, S. Danagoulian, V. Dharmawardane, R. Fatemi, K. Joo, M. Zeier, V. Gorbenko, R. Nasseripour, B. Raue, R. Suleiman, and B. Zihlmann This article is available at DigitalCommons@University of Nebraska - Lincoln: https://digitalcommons.unl.edu/ usdoepub/325 This article is available at DigitalCommons@University of Nebraska - Lincoln: https://digitalcommons.unl.edu/ usdoepub/325 University of Nebraska - Lincoln University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln University of Nebraska - Lincoln University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln US Department of Energy Publications U.S. Department of Energy 2004 Unique electron polarimeter analyzing power comparison and Unique electron polarimeter analyzing power comparison and precision spin-based energy measurement precision spin-based energy measurement J. Grames Thomas Jefferson National Accelerator Facility, grames@jlab.org C. K. Sinclair Thomas Jefferson National Accelerator Facility J. Mitchell Thomas Jefferson National Accelerator Facility E. Chudakov Thomas Jefferson National Accelerator Facility H. Fenker Thomas Jefferson National Accelerator Facility See next page for additional authors Follow this and additional works at: https://digitalcommons.unl.edu/usdoepub Part of the Bioresource and Agricultural Engineering Commons Grames, J.; Sinclair, C. K.; Mitchell, J.; Chudakov, E.; Fenker, H.; Freyberger, A.; Higinbotham, D.W.; Poelker, M.; Steigerwald, M.; Tiefenback, M.; Cavata, C.; Escoffier, S.; Marie, F.; Pussieux, T.; Vernin, P.; Danagoulian, S.; Dharmawardane, V.; Fatemi, R.; Joo, K.; Zeier, M.; Gorbenko, V.; Nasseripour, R.; Raue, B.; Suleiman, R.; and Zihlmann, B., "Unique electron polarimeter analyzing power comparison and precision spin-based energy measurement" (2004). US Department of Energy Publications. 325. https://digitalcommons.unl.edu/usdoepub/325 This Article is brought to you for free and open access by the U.S. Department of Energy at U.S. Department of Energy U.S. Department of Energy Follow this and additional works at: https://digitalcommons.unl.edu/usdoepub Part of the Bioresource and Agricultural Engineering Commons Grames, J.; Sinclair, C. K.; Mitchell, J.; Chudakov, E.; Fenker, H.; Freyberger, A.; Higinbotham, D.W.; Poelker, M.; Steigerwald, M.; Tiefenback, M.; Cavata, C.; Escoffier, S.; Marie, F.; Pussieux, T.; Vernin, P.; Danagoulian, S.; Dharmawardane, V.; Fatemi, R.; Joo, K.; Zeier, M.; Gorbenko, V.; Nasseripour, R.; Raue, B.; Suleiman, R.; and Zihlmann, B., "Unique electron polarimeter analyzing power comparison and precision spin-based energy measurement" (2004). US Department of Energy Publications. 325. https://digitalcommons.unl.edu/usdoepub/325 Corresponding author.Electronic address: grames@jlab.org †Present address: Wilson Laboratory, Cornell University, Ithaca, NY 14853, USA. ‡Present address: Renaissance Technologies Corporation, 600 Route 25A, East Setauket, NY 11733-2841, USA. xPresent address: Carl Zeiss Lithos GmbH, Carl Zeiss Straße 50, D-73447 Oberkochen, Germany. R. Suleiman R. Suleiman Massachusetts Institute of Technology, Cambridge, Massachusetts 02139-4307, USA B. Zihlmann Vrije Universiteit, 1081 HV, Amsterdam, The Netherlands (Received 17 March 2004; published 19 April 2004) B. Zihlmann Vrije Universiteit, 1081 HV, Amsterdam, The Netherlands (Received 17 March 2004; published 19 April 2004) Precision measurements of the relative analyzing powers of ﬁve electron beam polarimeters, based on Compton, Møller, and Mott scattering, have been performed using the CEBAF accelerator at the Thomas Jefferson National Accelerator Facility (Jefferson Laboratory). A Wien ﬁlter in the 100 keV beam line of the injector was used to vary the electron spin orientation exiting the injector. High statistical precision measurements of the scattering asymmetry as a function of the spin orientation were made with each polarimeter. Since each polarimeter receives beam with the same magnitude of polarization, these asymmetry measurements permit a high statistical precision comparison of the relative analyzing powers of the ﬁve polarimeters. This is the ﬁrst time a precise comparison of the analyzing powers of Compton, Møller, and Mott scattering polarimeters has been made. Statistically signiﬁcant disagreements among the values of the beam polarization calculated from the asymmetry measurements made with each polarimeter reveal either errors in the values of the analyzing power or failure to correctly include all systematic effects. The measurements reported here represent a ﬁrst step toward understanding the systematic effects of these electron polarimeters. Such studies are necessary to realize high absolute accuracy (ca. 1%) electron polarization measurements, as required for some parity violation measurements planned at Jefferson Laboratory. Finally, a comparison of the value of the spin orientation exiting the injector that provides maximum longitudinal polarization in each experimental hall leads to an independent and very precise (better than 104) absolute measurement of the ﬁnal electron beam energy. DOI: 10.1103/PhysRevSTAB.7.042802 Unique electron polarimeter analyzing power comparison and precision spin-based energy measurement J. M. Grames, C. K. Sinclair,† J. Mitchell,‡ E. Chudakov, H. Fenker, A. Freyberger, D.W. Higinbotham, M. Poelker, M. Steigerwald,x and M. Tiefenback Thomas Jefferson National Accelerator Facility, Newport News, Virginia 23606, USA C. Cavata, S. Escofﬁer, F. Marie, T. Pussieux, and P. Vernin CEA Saclay, DSM/DAPNIA/SPHN, 91191 Gif sur Yvette Cedex, France S. Danagoulian North Carolina Agricultural and Technical State University, Greensboro, North Carolina 27411, USA V. Dharmawardane Old Dominion University, Norfolk, Virginia 23529, USA R. Fatemi, K. Joo, and M. Zeier University of Virginia, Charlottesville, Virginia 22904-4714, USA V. Gorbenko Kharkov Institute of Physics and Technology, Kharkov 61108, Ukraine R. Nasseripour and B. Raue Florida International University, Miami, Florida 33199, USA R S l i Unique electron polarimeter analyzing power comparison and precision spin-based energy measurement J. M. Grames,* C. K. Sinclair,† J. Mitchell,‡ E. Chudakov, H. Fenker, A. Freyberger, D.W. Higinbotham, M. Poelker, M. Steigerwald,x and M. Tiefenback Thomas Jefferson National Accelerator Facility, Newport News, Virginia 23606, USA C. Cavata, S. Escofﬁer, F. Marie, T. Pussieux, and P. Vernin CEA Saclay, DSM/DAPNIA/SPHN, 91191 Gif sur Yvette Cedex, France S. Danagoulian North Carolina Agricultural and Technical State University, Greensboro, North Carolina 27411, USA V. Dharmawardane Old Dominion University, Norfolk, Virginia 23529, USA R. Fatemi, K. Joo, and M. Zeier University of Virginia, Charlottesville, Virginia 22904-4714, USA V. Gorbenko Kharkov Institute of Physics and Technology, Kharkov 61108, Ukraine R. Nasseripour and B. Raue Florida International University, Miami, Florida 33199, USA DOI: 10.1103/PhysRevSTAB.7.042802 PACS numbers: 29.25.Bx, 29.27.Hj, 29.30.Dn, 13.60.Fz Authors Authors EVIEW SPECIAL TOPICS - ACCELERATORS AND BEAMS,VOLUME 7, 042802 (2004) PHYSICAL REVIEW SPECIAL TOPICS - ACCELERATORS AND BEAMS,VOLUME Experimental plan and CEBAF accelerator The experimental plans to compare the effective ana- lyzing powers of Compton, Møller, and Mott polarime- ters, using the CEBAF accelerator, are discussed in the remainder of this section. The CEBAF accelerator [3] shown in Fig. 1 is a 6 GeV, 200 A continuous beam electron accelerator in which a beam of highly polarized electrons is recirculated up to 5 times through two super- conducting linear accelerators (linacs). Radio frequency (rf) separators allow beams to be extracted after any recirculation pass and delivered simultaneously to three experimental halls, shown as A, B, and C in Fig. 1. For this experiment, a ﬁve-pass beam was delivered to all three experimental halls. The locations of the ﬁve polarimeters are indicated in Fig. 1, and their operating parameters are given in Table I. All ﬁve polarimeters are discussed in greater detail in the following sections. g The most desirable characteristic of any polarimeter is a large and well-known effective analyzing power. However, precise knowledge of the effective analyzing power is limited because it is not generally a directly measured quantity. Direct measurement of the effective analyzing power of a particular scattering process re- quires very difﬁcult double scattering experiments. Small effective analyzing powers, low cross sections, and discrimination against backgrounds make such ex- periments highly impractical in most cases. For all the polarimeters used in the measurements reported here, and indeed in almost all cases, the effective analyzing power of a polarimeter is determined by computer simulation. These simulations include not only the physics asymme- try of the underlying scattering process, but also the de- tails of the real detector, systematic effects to the extent they are identiﬁed, multiple scattering, and background effects. It is quite possible that the true value of the effective analyzing power of a polarimeter differs from the simulated value by more than the uncertainty as determined by simulation. The importance of the Levchuk effect [1,2] to the effective analyzing power of The accelerator delivers a highly polarized beam from the injector simultaneously to all experimental hall po- larimeters. This offers the advantage that all polarimeters receive the same magnitude of beam polarization. However, since the total precession from the injector to each experimental hall is different, the measurable com- ponents of the beam polarization are generally not equal at the various polarimeters. I. INTRODUCTION The use of beams of polarized electrons in nuclear and high energy physics experiments provides an important degree of freedom for understanding fundamental inter- actions. For example, the spin structure of nucleons can be explored using a beam of longitudinally polarized 2004 The American Physical Society 042802-1 042802-1 042802-1 1098-4402=04=7(4)=042802(18)$22.50 042802-1 1098-4402=04=7(4)=042802(18)$22.50 UNIQUE ELECTRON POLARIMETER ANALYZING POWER . . PRST-AB 7 042802 (2004) Møller polarimeters, which emerged long after Møller polarimeters became a common way of measuring elec- tron beam polarization, is a good example of this reality. electrons in conjunction with either a polarized target or recoil polarimetry or by studying parity violation in the scattering of longitudinally polarized electrons from un- polarized targets. In general, the uncertainty in the knowledge of the electron beam polarization is a signiﬁ- cant contribution to the overall error bar in these mea- surements. While experiments using polarized targets or recoil polarimetry do not generally require the highest precision electron polarimetry, this is not the case with parity violation measurements. For some planned parity violation measurements, absolute knowledge of the elec- tron beam polarization at the 1% level is desired. This is beyond the present state of the art in electron polarimetry. electrons in conjunction with either a polarized target or recoil polarimetry or by studying parity violation in the scattering of longitudinally polarized electrons from un- polarized targets. In general, the uncertainty in the knowledge of the electron beam polarization is a signiﬁ- cant contribution to the overall error bar in these mea- surements. While experiments using polarized targets or recoil polarimetry do not generally require the highest precision electron polarimetry, this is not the case with parity violation measurements. For some planned parity violation measurements, absolute knowledge of the elec- tron beam polarization at the 1% level is desired. This is beyond the present state of the art in electron polarimetry. The presence of ﬁve polarimeters at Jefferson Laboratory, using all three basic electron scattering in- teractions, led to a joint effort by all the polarimeter groups at the laboratory to precisely compare the effec- tive analyzing powers of these polarimeters, by using each to measure the polarization of the same beam. The effective analyzing power for each polarimeter was de- termined through simulations done by the experimental groups that constructed and commissioned these polar- imeters. I. INTRODUCTION The measurements reported here have the aim of revealing any errors in the effective analyzing powers of the ﬁve polarimeters. y p p y All electron beam polarimeters developed to date for electron energies above a few keV rely on the spin depen- dence in one of three electron scattering processes: Mott, Compton, or Møller scattering. The spin dependence of each of these three scattering processes gives the physics analyzing power. The scattering target may be polarized, as in Compton or Møller polarimeters, or unpolarized, as in Mott polarimeters. In the former case, the analyzing power is the product of the physics analyzing power and the target polarization, and in the latter case, it is simply the physics analyzing power. The effective analyzing power of a polarimeter incorporates many additional de- tails, such as detector acceptance and resolution, system- atic effects, and backgrounds. In each case, the spin dependent cross section yields a measurable asymmetry in the scattered ﬂux equal to the product of the beam polarization and the effective analyzing power. In gen- eral, no polarimeter measures the total beam polariza- tion. Rather, the particular components of the beam polarization measured are determined by the polarimeter design. Experimental plan and CEBAF accelerator p Although the nine recirculation arcs and the experi- mental hall beam transport lines lie in different horizon- tal planes, there is, with very high precision, no net vertical bend between the injector (and its Mott polar- imeter) beam line and the beam lines through the various experimental hall polarimeters. As a result, any vertical component of polarization at the exit of the injector appears unaltered at the experimental hall polarimeters (for further explanation, see Sec. IIIB). In the horizontal plane, the beam polarization undergoes a large precession from the injector to the various experimental hall polar- imeters, due to the large net horizontal bend angle and the high beam energy. Thus, in general, the polarization at 042802-2 042802-2 042802-2 042802-2 PRST-AB 7 042802 (2004) J. M. GRAMES et al. North Linac South Linac 5 East Arcs 4 West Arcs Polarized Photocathode Guns A B C Moller Polarimeters Compton Polarimeter Mott Polarimeter Wien filter Beam Switchyard FIG. 1. Schematic of the CEBAF accelerator showing elements of the experiment. Wien filter Polarized Photocathode Guns Mott Polarimeter North Linac Compton Polarimeter 4 West Arcs South Linac Beam Switchyard FIG. 1. Schematic of the CEBAF accelerator showing elements of the experiment. TABLE I. Operating parameters of the ﬁve Jefferson Laboratory electron polarimeters. Polarimeter Reaction Intensity Target Measurement Injector Mott ~e ZA 5 A Gold (1 m) Px; Py Hall A Compton ~e ~ 100 A Photon ( 1064 nm) Pz Hall A Møller ~e ~e 500 nA Supermendur (13 m) Pz; Px Hall B Møller ~e ~e 5 nA Permendur (25 m) Pz; Py Hall C Møller ~e ~e 2 A Iron (4 m) Pz TABLE I. Operating parameters of the ﬁve Jefferson Laboratory electron polarimeters. Polarimeter Reaction Intensity Target Measurement electron beams, stated earlier, means that the experimen- tal hall polarimeters have all been constructed to measure the longitudinal component of the beam polarization, whereas the injector Mott polarimeter can measure only the transverse component of the beam polarization. each experimental hall polarimeter has a longitudinal component parallel to the beam momentum, a horizontal component transverse to the beam momentum (hereafter referred to as ‘‘horizontal’’), and a small vertical compo- nent. Clearly in those cases where the longitudinal com- ponent is small, the horizontal component is large. Experimental plan and CEBAF accelerator Since all four experimental hall polarimeters have been de- signed to measure the longitudinal polarization, they may be subject to signiﬁcant systematic effects arising from the large horizontal component when the longitu- dinal component is small, as in some of the measurements reported here. II. ELECTRON POLARIMETERS AT JEFFERSON LAB The ﬁve electron polarimeters are described in the sections below. The references given in each section pro- vide additional details. p The uncertainty in the total precession between the injector and the experimental hall polarimeters compli- cates precise measurements of the relative analyzing powers. The solution to this problem is to conduct the measurements in a way that does not rely on only one measurement of a single component of the beam polar- ization. This is accomplished by adjusting the orientation of the beam polarization with a spin rotator, in this case a Wien ﬁlter, common to all beams. The beam polarization measured by each polarimeter can then be plotted against the common spin orientation. A ﬁt to these data yields both the magnitude of the measured polarization and the spin orientation at the injector that results in the maxi- mum value of the measured polarization at each polar- imeter. The importance of longitudinally polarized A. Mott polarimeter in the injector To support a reliable, high precision measurement of the beam polarization in the electron injector a 5 MeV Mott scattering polarimeter (see Fig. 2) has been devel- oped [4,5]. The Mott scattering asymmetry arises from the spin-orbit coupling between the electron spin and its orbital angular momentum in the Coulomb potential of the target nucleus.The polarimeter is located in the 5 MeV region of the injector on a dedicated beam line 12.5 from the injector beam line. The polarimeter measures both transverse components of the beam polarization and has been studied [6] over a range of energies (2–8 MeV) with target foils of varying thickness and three atomic num- bers (79 for gold, 47 for silver, and 29 for copper). The 042802-3 042802-3 042802-3 UNIQUE ELECTRON POLARIMETER ANALYZING POWER . . . NIQUE ELECTRON POLARIMETER ANALYZING POWER . . . 042802 (2004) PRST-AB 7 PRST-AB 7 PHOTOMULTIPLIER SCINTILLATOR ELASTIC MOTT ELECTRON BEAM DIRECTION Be RING TARGET LADDER VACUUM WINDOW LINEAR FEEDTHROUGH THIN Al VACUUM WINDOW Cu BAFFLE 1.01m FIG. 2. Schematic of the injector Mott polarimeter as installed in the 5 MeV region of the CEBAF injector. 1.01m PHOTOMULTIPLIER SCINTILLATOR ELASTIC MOTT ELECTRON TARGET LADDER BEAM DIRECTION FIG. 2. Schematic of the injector Mott polarimeter as installed in the 5 MeV region of the CEBAF injector. maximum analyzing power occurs at an electron scatter- ing angle close to 172 for all values of the beam energy and atomic number measured. from any longitudinal polarization component would be parity violating, and are thus negligible. The polarimeter has four electron detector arms, two in the horizontal plane and two in the vertical plane, with each member of a pair separated by 180 in azimuth. Internal collimation deﬁnes the acceptance of the indi- vidual detector arms and assures that each detector views only the central area of the target foil. A combination of thin and thick scintillation counters are used in coinci- dence to discriminate against photons. The energy of the elastically scattered 5 MeVelectron is totally absorbed in the thick detector, and the signal from this detector is required to be above a suitable threshold. The rf time structure on the electron beam allows the use of time- of-ﬂight detection, which discriminates against electrons scattered from material other than the primary target. B. Compton polarimeter in experimental Hall A The ﬁrst of the two electron polarimeters in the Hall A beam line is a Compton polarimeter [7–9] (see Fig. 3). The Compton scattering asymmetry results from the interaction of a longitudinally polarized electron beam and a circularly polarized photon (target) beam. The photon beam circulates in a Fabry-Perot cavity centered within a chicane comprised of four vertical bend dipole magnets. An external laser (250 mW at 1064 nm) is locked to and ﬁlls the cavity with circularly polarized light (Pcirc > 99%). The gain of the cavity results in about 1200 W of circulating optical beam power. A remotely controlled quarter-wave plate external to the Fabry-Perot cavity may be rotated to select either state of circular polarization and is used to reverse the overall sign of the scattering asymmetry for systematic correction. The op- tical transport mirrors are oriented to maintain the opti- cal polarization into the cavity. The cavity maintains the photon beam at a small angle to the incident electron beam. Compton backscattered photons exit through the third chicane dipole and are detected using a PbWO4 crystal calorimeter. A detector for the scattered electron also exists, but was not used in this experiment. The effective Compton analyzing power depends on the beam energy and is a modeled number, rather than one calculated from ﬁrst principles. The effective analyzing power is modeled for each run (electron spin orientation) separately. For this experiment the effective analyzing The analyzing power of the standard 1 m gold target foil used in these measurements was determined by mea- suring the scattering asymmetry for a large number of gold foils with thicknesses ranging between 50 nm and 5 m. These measurements allowed a high precision extrapolation to zero foil thickness. The analyzing power of a zero thickness foil is taken to be that calculated for scattering by a single atom. At the 5 MeV beam energy used in the experiment, the effective analyzing power, or effective Sherman function, for the standard foil is Seff 0:4008 0:0014 0:0040. The ﬁrst uncertainty is in- strumental and the second is the theoretical uncertainty in the calculated Sherman function for single atom scat- tering. The theoretical uncertainty includes the uncer- tainties in the radiative correction and the nuclear size effect. As the target is unpolarized, asymmetries arising 042802-4 042802-4 042802-4 042802-4 042802-4 PRST-AB 7 042802 (2004) J. M. GRAMES et al. B. Compton polarimeter in experimental Hall A Hall A Dipoles Beam chicane Optical cavity Beam direction Photon detector Electron detector 15.35 m FIG. 3. (Color) Schematic of the Hall A Compton polarimeter. Photon detector E or Electron detector Optical cavity FIG. 3. (Color) Schematic of the Hall A Compton polarimeter. electron and recoil target electron), centered about 90 in the center of mass, are detected in coincidence. The targets are 13 m thick and are positioned at angles of 20 horizontally with respect to the beam, giving an effective target thickness of 38 m. The longitudinal component of the beam polarization is determined by using the oppositely oriented foils to subtract the asym- metry arising from the horizontal component of the beam polarization. The targets are cooled by conduction through the target support. At 500 nA beam current, the estimated target temperature rise is several degrees Kelvin and the associated relative change in target polar- ization is estimated to be below 0.1%. The simulated value of the physics analyzing power of the polarimeter is Azz 0:7600 0:0023. The effective analyzing power is Aeff 0:0604 0:0018. power is Aeff 0:024. The systematic uncertainty ranged between 0.6% and 2.6% over the range of spin angles of this experiment. The systematic uncertainty of the laser polarization is 1%. D. Møller polarimeter in experimental Hall B 0:7995 0:0060; the uncertainty arises from the statis- tics of the simulation. The effective analyzing power of the polarimeter is Aeff 0:0640 0:0006. The Hall B Møller polarimeter [11] uses a polarized target similar to that in Hall A, a two quadrupole mag- netic spectrometer, and coincidence detection of the scat- tered and recoil electrons. Two 25 m thick permendur foils (49% Fe, 49% Co, 2% V) are oriented with their planes at 20 vertically with respect to the beam di- rection, giving an effective target thickness of 73 m. The target is polarized to ’ 7:5% along the beam direc- tion by the 120 G longitudinal ﬁeld of a pair of Helmholtz coils. This polarimeter is operated at low beam currents (a few nanoamps average) typical of Hall B experiments. In contrast to the Hall A Møller polarimeter the longitu- dinal component of the beam polarization is determined by subtracting the asymmetry arising from the vertical component of the beam polarization. However, only one target was used for this experiment, and no correction for the vertical component was made. The physics analyzing power is simulated to be Azz 0:7826 0:0062. The effective analyzing power is Aeff 0:0587. III. EXPERIMENTAL SETUP The polarized electron beam is produced by photo- emission from a semiconductor cathode using polarized laser light [14,15]. The beam polarization depends upon the speciﬁc cathode material and the wavelength and degree of polarization of the incident light. The cathode, held at a potential of 100 kV, is a wafer of strained gallium arsenide activated to negative electron afﬁnity. Irradiating the cathode with light of a wavelength slightly shorter than that corresponding to the minimum direct band gap energy of the GaAs produces a beam of highly longitudinally polarized electrons. The longitudinal elec- tron polarization is directly proportional to the circular polarization of the optical beam. It is straightforward to produce optical circular polarization greater than 99%, giving an electron beam longitudinal polarization greater than 70% from such a cathode. It should be noted that the beam from the photocathode has a pure longitudinal polarization, and that any linear polarization of the illu- minating light does not produce transverse electron polarization. C. Møller polarimeter in experimental Hall A Downstream of the Compton polarimeter is a Møller polarimeter [10] (see Fig. 4) consisting of a solid polar- ized target, a magnetic spectrometer (three quadrupoles and one dipole), and lead glass and scintillator detectors. The polarimeter uses either of two iron-alloy targets (supermendur) which are polarized by a weak 240 G longitudinal magnetic ﬁeld created by a pair of Helmholtz coils. The target polarization is measured to be 0:0795 0:0024. The Møller pairs (scattered incident Target Quad1 Quad2 Quad3 Dipole Detector Helmholtz T op View coils 700 cm Beam Side View 9.6cm 48.5 cm FIG. 4. (Color) Schematic of the Hall A Møller polarimeter. Target Quad1 Quad2 Quad3 Dipole Detector Helmholtz T op View coils 9.6cm Dipole Quad2 Quad3 Target Quad1 Detector T op View Helmholtz coils 700 cm Beam Side View 48.5 cm FIG. 4. (Color) Schematic of the Hall A Møller polarimeter. FIG. 4. (Color) Schematic of the Hall A Møller polarimeter. 042802-5 042802-5 042802-5 042802-5 UNIQUE ELECTRON POLARIMETER ANALYZING POWER . . . PRST-AB 7 042802 (2004) E. Møller polarimeter in experimental Hall C The Hall C Møller polarimeter [12,13] (see Fig. 5) consists of a polarized iron target, a two quadrupole magnetic spectrometer, a collimator system, and lead glass and scintillation detectors. The target is a pure iron foil positioned normal to the incident beam within a 3 T longitudinal magnetic ﬁeld created by a pair of superconducting Helmholtz coils. In this target design the out-of-plane magnetization is saturated in the external ﬁeld, yielding a target polarization Pz 0:0800 0:0004. The uncertainty in the target polarization in- cludes an estimate of the effect of target heating by a 2:5 A beam. The Møller scattered and recoil electrons are detected in coincidence. The collimator system passes electrons Møller scattered near 90 in the center of mass, where the analyzing power is a maximum, and discrim- inates against electrons from Mott scattering and other backgrounds. With the target polarization parallel to the beam direction, this polarimeter should be very insensi- tive to any effects arising from transverse polarization components. The physics analyzing power of the polar- imeter determined by Monte Carlo simulation is Azz A. Orienting the beam polarization The spin rotator is a Wien ﬁlter located in the 100 keV beam line following the electron gun. AWien ﬁlter [16] is a static electromagnetic device with electric ~E and magnetic ~B ﬁelds perpendicular to both the particle velocity ~ and each other as shown in Fig. 6. TheWien angle (Wien) is the angle by which the beam polarization is rotated, in the plane of the electric ﬁeld, relative to the beam momentum. The Wien angle is di- rectly proportional to the applied ﬁelds (Wien ~E ~B) and at the injector beam energy (100 keV) is domi- nated by the contribution from the electric ﬁeld integral: ~B ~E a g 2 a 0:17%; (1) (1) where the Lorentz factor 1:1957 at 100 keV, g is the electron gyromagnetic factor, and a g 2 =2 1.0m 7.85m Helmholtz collimator Q1 beam detectors Q2 3.20m target FIG. 5. Schematic of the Hall C Møller polarimeter. 7.85m FIG. 5. Schematic of the Hall C Møller polarimeter. 042802-6 042802-6 042802-6 J. M. GRAMES et al. PRST-AB 7 042802 (2004) X Z P ηWien MAGNETIC FIELD ELECTRIC FIELD Beam FIG. 6. Diagram of a Wien ﬁlter indicating the rotation of the beam polarization relative to the beam direction (Wien) in crossed magnetic and electric ﬁelds ( E B ). In this experiment, the electric ﬁeld and the spin rotation are in the horizontal plane. FIG. 6. Diagram of a Wien ﬁlter indicating the rotation of the beam polarization relative to the beam direction (Wien) in crossed magnetic and electric ﬁelds ( E B ). In this experiment, the electric ﬁeld and the spin rotation are in the horizontal plane. 1:159 652 103. The utility of theWien ﬁlter is that the polarization of a beam passing through the device can be rotated in the plane of the electric ﬁeld without changing the central beam orbit. ~F q ~E ~ ~B 0; (2) (2) requiring that j E B j . Th l li d h l d requiring that j E B j . requiring that j E B j . The voltage applied across the two electrodes was calibrated against the common DAC setting, with the results shown in Fig. 7. A. Orienting the beam polarization The response is modeled by a second-order polynomial giving the power supply offset, gain, and linearity as a function of the DAC setting The transverse electric ﬁeld is produced along the mid- plane of the Wien ﬁlter by two electrodes which span its length. The electrode voltages are set by two opposite polarity 15 kV power supplies controlled by a common digital to analog converter (DAC), so that the potential on axis is zero. A magnetic ﬁeld normal to the electric ﬁeld is applied to balance the Lorentz force on the beam axis, VWien p0 p1 Sdac p2 S2 dac : (3) (3) FIG. 7. Wien voltage with second-order polynomial ﬁt (upper panel) and ﬁt residuals (lower panel) both shown as a function of the DAC set point (Sdac). FIG. 7. Wien voltage with second-order polynomial ﬁt (upper panel) and ﬁt residuals (lower panel) both shown as a function of the DAC set point (Sdac). 042802-7 042802-7 UNIQUE ELECTRON POLARIMETER ANALYZING POWER . . . 042802 (2004) PRST-AB 7 through the arc, and the ﬁnal polarization value for each orbit measured. The small precessions of the spin in the arc quadrupoles add when the spin tune and the vertical betatron tune are similar, leading to a small but measur- able net effect. Detailed simulations of the polarization difference for different orbit pairs, incorporating mea- sured values of the orbits, agreed well with the polariza- tion measurements. For the conditions of the experiment, the longitudinal polarization difference was about 0.5% per millimeter of absolute orbit displacement. These re- sults show clearly that spin transport through the CEBAF accelerator is very well understood, and that for normal operation of the accelerator, betatron motion in the arcs does not lead to any signiﬁcant vertical polarization component. While there is a clear systematic variation in the ﬁt residuals, the magnitude of this effect is much too small to inﬂuence any of the results presented here. The overall sign of the spin rotation is set by a high voltage relay that determines which power supply contacts each Wien electrode. The Wien ﬁlter magnetic ﬁeld is produced with a win- dow-frame dipole magnet. The ends of this magnet are terminated with nickel plates having 2 cm diameter cir- cular beam apertures. C. Beam requirements and beam delivery g g g Following the Wien ﬁlter, the beam has in general a horizontal polarization component. This component is rotated about the beam axis by any net longitudinal magnetic ﬁeld integral. Magnetic solenoids are used for beam focusing in several meters of beam line following the Wien ﬁlter. Each of these solenoids is comprised of two nominally identical segments oriented to provide equal and opposite longitudinal ﬁeld integrals, thus giv- ing no net rotation out of the plane to any horizontal polarization component. Any vertical component of po- larization exiting the injector arises from imperfect can- cellation of the longitudinal ﬁeld integral of the focusing solenoids acting on the horizontal polarization compo- nent exiting the Wien ﬁlter and is small. The accelerator was conﬁgured for ﬁve-pass recircula- tion and a nominal ﬁnal beam energy of 5.645 GeV. The ﬁve-pass setup allows delivery of the same energy beam simultaneously to the three end stations. The high beam energy was chosen to minimize the systematic uncertain- ties for the Møller and Compton polarimeters and to provide the highest analyzing power for the Compton polarimeter. The polarimeter beam intensity requirements deter- mined how the electron injector was operated. The operational beam intensity required for the ﬁve polar- imeters varies by 4 orders of magnitude (see Table I). The electron beam produced by a dc gun must ultimately have a bunch structure compatible with the fundamental fre- quency (1497 MHz) of the accelerator rf. This require- ment is assured by an rf chopping system in the 100 keV region of the injector. Three conditions for beam delivery were needed: A. Orienting the beam polarization The magnetic ﬁeld on the axis of the Wien ﬁlter, with the nickel plates installed, was care- fully measured with a precision Hall probe. The elec- trodes were shaped to create an electric ﬁeld proﬁle on the Wien ﬁlter axis, as calculated with the code Poisson [17], which very accurately matched the measured mag- netic ﬁeld proﬁle. There was no detectable saturation of the magnet over the full range of ﬁeld strengths used. B. Spin dynamics were analyzed by each respective polarimeter group. The measured components of the beam polarization re- ported by each group are given in Table II. To address this concern the photoinjector was operated in two different modes. In the ﬁrst of these, a true dc beam was delivered from the photoemission gun by ap- plying a dc current to the laser. This beam was chopped and bunched in the normal manner, with beam intensity adjustment provided by the chopping apertures.With a dc beam from the polarized gun, the polarization is inde- pendent of the time within the bunch. Both Mott and Møller measurements were made with this beam. We report the measured longitudinal polarization for the Møller and Compton polarimeters, and the measured transverse polarization in the horizontal plane for the Mott polarimeter. The measured polarization values are proportional to the product of the longitudinal beam polarization from the photocathode and the cosine of the Wien angle plus the total precession angle. The measured polarizations from each polarimeter are modeled as The second operating mode was used for the Compton measurements, using the rf driven laser and fully open chopping apertures to reach the higher beam intensity. With the rf driven laser illuminating the cathode, and the fully open chopping apertures, the chopping system passes essentially all of the beam, so there is no issue of beam polarization dependence on time within the bunch. The same laser and optical polarization elements were used for both the dc and the rf cases. Only the electrical drive to the laser was changed between the dc and the rf cases. Pmeas Pi cosi VWien i g ; (4) (4) where Pi, i, and i are the ﬁt parameters. Pi and i are the magnitude of the measured beam polarization and the total spin precession between the Wien ﬁlter and the particular polarimeter. Although the ﬁt function is peri- odic, the total beam precession modulo 2 is known to much better than one full revolution. Accordingly, the seed value of i for the ﬁtting routine is set to an appropriate value for each experimental hall. The value g a! 0:01 is a correction for a single dipole magnet (! 15) located between the electron source and the Wien ﬁlter. B. Spin dynamics The spin dynamics of the CEBAF accelerator have been extensively studied. Detailed calculations and mea- surements have demonstrated that the polarization trans- port is very well understood, and that any loss of beam polarization through the full ﬁve-pass accelerator is com- pletely negligible [18]. (1) solely to the Mott polarimeter at 2 A, (2) solely to the Compton polarimeter above 70 A, and (2) solely to the Compton polarimeter above 70 A, and (3) simultaneously to the three Møller polarimeters between 5 nA and 2 A. The magnetic recirculation system, where the vast ma- jority of the spin precession occurs, has no horizontal bends between the equal and opposite vertical bends that take the beam between the two linacs and the various horizontal planes of the recirculation arcs. All bends in each recirculation arc are set in a common horizontal plane by precision survey techniques. This geometry naturally eliminates the difﬁculties associated with the noncommutation of ﬁnite spin rotations which have caused problems in some more complex beam transport systems. These conditions were obtained by a combination of choices which deserves some discussion. More than 80% of a dc electron beam from the polarized source is lost on the apertures of the rf chopping system. A substantial improvement, resulting in a more efﬁcient use of the electrons, has been made by using an rf driven diode laser [20] providing short optical pulse widths (50 psec FWHM) synchronous with the accelerator rf. Using this laser the emitted electrons have a time structure giving high transmission through the chopping apertures. The sensitivity of the ﬁnal beam polarization to the beam orbit through a single recirculation arc was care- fully studied in a dedicated experiment [19]. For this, the spin tune (s 6:4) and the vertical betatron tune (y 6:0) of the arc were close in value. Large equal and opposite vertical betatron oscillations were excited There is evidence that the electron polarization de- pends somewhat on its time within the bunch [21]. This led to the concern that the low beam intensity polar- imeters, which used a small fraction of the bunch as deﬁned by the chopping apertures, would receive a beam polarization somewhat different from the high 042802-8 042802-8 042802-8 042802-8 J. M. GRAMES et al. PRST-AB 7 042802 (2004) beam intensity Compton polarimeter, which used essen- tially all ( > 98%) of the bunch. B. Spin dynamics The coefﬁcient i is the constant relating the spin rotation of the Wien ﬁlter to the Wien ﬁlter high voltage. Beam was delivered to the Mott polarimeter alone, using magnetic deﬂection in the injector, simultaneously to the three Møller polarimeters using rf separation of the high energy beam [22] and to the Compton polarimeter alone by magnetically deﬂecting the high energy beam. The probable values of the ﬁt parameters and their uncertainties are determined from the polarimeter data of Table II using a nonlinear least squares ﬁtting routine [23,24]. Because we are investigating the relative analyzing power of the polarimeters without presuming IV. POLARIMETER MEASUREMENTS AND EXPERIMENTAL RESULTS Polarimeter data were collected at 12 different Wien angles spanning jWienj < 110. The polarimeter data TABLE II. Summary of polarization measurements, using the effective analyzing power quoted for each polarimeter. All measurements were made using the dc laser mode except for the Compton measurements which all used the rf laser mode. The uncertainties below are statistical only. Wien Mott Compton A Møller A Møller B Møller C (deg) (%) (%) (%) (%) (%) 10:54 71:05 0:90 74:65 0:17 10:54 71:22 0:85 57.07 58:90 0:30 36:42 0:18 32:65 1:02 73:62 0:17 108.79 69:00 0:30 32:09 0:14 70:64 1:02 43:56 0:26 93.27 72:80 0:30 14:41 1:30 9:45 0:10 64:31 1:61 59:80 0:27 77.75 70:60 0:30 10:82 0:61 12:44 0:16 51:97 0:91 69:50 0:18 77.75 10:81 0:74 36.38 59:77 0:16 8:65 0:88 67:27 0:18 10.54 10:70 0:40 73:02 0:14 24:05 0:89 49:37 0:20 10:54 13:90 0:40 71:52 1:87 74:73 0:14 43:40 0:99 28:65 0:26 10:54 78:95 1:73 41:55 48:00 0:40 58:52 0:66 41:55 56:71 0:79 60:16 62:90 0:40 45:38 0:13 69:11 1:02 30:66 0:26 84:99 72:20 0:20 12:39 0:17 61:95 1:05 58:06 0:22 108:79 68:00 0:50 24:02 0:14 45:19 0:99 72:59 0:23 042802-9 042802-9 042802-9 UNIQUE ELECTRON POLARIMETER ANALYZING POWER . . . 042802 (2004) PRST-AB 7 TABLE III. i determined by Eq. (4) for each polarimeter and the weighted average, . TABLE III. i determined by Eq. (4) for each polarimeter and the weighted average, . data measured at identical Wien angle settings. Consequently, the analysis was segmented into two stages. In the ﬁrst stage all three ﬁt parameters are esti- mated and the weighted average () of the set of i (see Table III) is computed. In the second stage each data set was ﬁt again using Eq. (4) with i replaced by the weighted average value . The ﬁrst stage analysis of the data gives a i that varies by 4.0% between the extremes of the polarimeter data sets. This variation is attributed to an unmeasured systematic effect and is included in the uncertainty of the ﬁt. IV. POLARIMETER MEASUREMENTS AND EXPERIMENTAL RESULTS Although unexpected, this variation has little impact on the ﬁnal determination of the beam polarization and spin phase, as demonstrated below. Polarimeter i ( deg=kV) Mott 5:553 0:038 Møller A 5:666 0:043 Møller B 5:704 0:047 Møller C 5:615 0:052 Compton 5:779 0:169 (weighted average) 5:630 0:022 systematic uncertainties, care must be taken to estimate the probable values of the ﬁt parameters and their un- certainties. In doing so we cannot use only the statistical errors reported in Table II because this would underesti- mate the true variance in a polarimeter data set. The approach taken [24,25] is to assume that we do not know the individual measurement errors and to assign to each measurement in a polarimeter data set an equal unit weight error. Consequently, the standard error of estimate for the ﬁt parameters from each data set is given by the root mean square statistical errors of estimate of the covariance matrix, weighted by the reduced chi square of the ﬁt "2=N M p . Here "2 is the least squares estimate evaluated at the most probable ﬁt pa- rameters, N is the number of measurements in a data set, and M is the number of ﬁt parameters. This approach depends upon the following assumptions: The ﬁnal ﬁt results and residuals are shown in Fig. 8. In Table IV we compare the results obtained for Pi and i for each polarimeter using both the individual i for that polarimeter and using . In all cases the differences in the ﬁt results using either i or are small compared to the standard error from the least squares ﬁtting routine. It is important to understand that the systematic un- certainties associated with each polarimeter are not in- cluded in the polarization results given in Table IV and Fig. 8. A motivation for this experiment is to ﬁnd evi- dence of undetected systematic effects or other problems with the quoted effective analyzing powers of the various polarimeters by comparing the polarization determined by each polarimeter when all measure the same polarized beam. When systematic effects are included, the uncer- tainties on the polarization values reported in Table IV and Fig. 8 will be larger. (i) we are using the correct model to describe the data [see Eq. IV. POLARIMETER MEASUREMENTS AND EXPERIMENTAL RESULTS (4)], A comparison of the beam polarizations determined from each polarimeter data set is made in Fig. 9 by plotting the ratio of the measured polarization values (Pi) with respect to a reference. In this case, the Mott polarimeter was chosen as the reference because it had the smallest relative uncertainty in the ﬁt parameter Pi. The ratio of the effective analyzing power of any polarimeter relative to the Mott polarimeter is the inverse of the ratio shown in Fig. 9. (ii) the systematic errors do not change within a data set, and (ii) the systematic errors do not change within a data set, and (iii) the total measurement error of each data point within a data set is approximately equal, requiring that each measurement has a very similar statistical precision. While the ﬁrst and third assumptions are valid, it is important to note that the systematic uncertainty may depend on the degree of horizontal polarization for some polarimeters, so the second assumption may not be met in all cases. We have not attempted to analyze how this might affect our results. As noted earlier, when the longitudinal polarization component is small, the horizontal component is large. Some polarimeters may have systematic effects that de- pend on the horizontal component. An indication of the To compare the polarimeters one must additionally constrain the parameter i to be equal for all polarimeter TABLE IV. Polarization and phase results for polarimeter data using 5:630=kV; italicized text show the values obtained using the individual i. The uncertainties are the total standard error in the ﬁt parameters using the least squares ﬁtting routine. Polarization (%) Phase (deg) Polarimeter i i Mott 72:10 0:28 72.190.24 88:88 0:32 88.820.27 Møller A 76:76 1:04 76.791.06 10 983:84 0:63 10 983.750.64 Møller B 69:84 0:64 69.810.59 10 500:48 0:56 10 500.300.53 Møller C 73:30 0:62 73.290.66 10 023:39 0:61 10 023.400:65 Compton 73:33 1:42 74.251.84 10 984:82 1:33 10 983.072.36 042802-10 042802-10 TABLE IV. Polarization and phase results for polarimeter data using 5:630=kV; italicized text show the values obtained using the individual i. The uncertainties are the total standard error in the ﬁt parameters using the least squares ﬁtting routine. TABLE IV. Polarization and phase results for polarimeter data using 5:630=kV; italicized text show the values obtained using the individual i. The uncertainties are the total standard error in the ﬁt parameters using the least squares ﬁtting routine. IV. POLARIMETER MEASUREMENTS AND EXPERIMENTAL RESULTS Fit residuals (lower plot) are shown using only statistical uncertainties from polarimeter data. Note: the lower plot legend applies to the upper plot. FIG. 10. (Color) Polarimeter data and ﬁt (upper plot) using and data set limited to be within 25% of the maximum polarization. Fit residuals (lower plot) are shown using only statistical uncertainties from polarimeter data. Note: the lower plot legend applies to the upper plot. horizontal component of polarization may be an impor- tant source of systematic effects for the Hall A Møller polarimeter. proton scattering methods. An additional beneﬁt of the polarimeter intercomparison is that the spin precession data can be used for a precision determination of the beam energy, due to the linear relationship between the spin precession and the beam energy. One of two independent spin-based approaches, described below, is shown to be capable of yielding an extremely precise absolute mea- surement <104 of the beam energy. IV. POLARIMETER MEASUREMENTS AND EXPERIMENTAL RESULTS Polarization (%) Phase (deg) Polarimeter i i Mott 72:10 0:28 72.190.24 88:88 0:32 88.820.27 Møller A 76:76 1:04 76.791.06 10 983:84 0:63 10 983.750.64 Møller B 69:84 0:64 69.810.59 10 500:48 0:56 10 500.300.53 Møller C 73:30 0:62 73.290.66 10 023:39 0:61 10 023.400:65 Compton 73:33 1:42 74.251.84 10 984:82 1:33 10 983.072.36 042802-10 042802-10 042802-10 PRST-AB 7 J. M. GRAMES et al. 042802 (2004) FIG. 8. (Color) Measured polarizations and ﬁts (upper plot) using and all data. Fit residuals (lower plot), with only statistical uncertainties from the ﬁts shown. In general, the uncertainties from the ﬁts are smaller than the symbol sizes used in plotting the data. Note: the lower plot legend applies to the upper plot. FIG. 8. (Color) Measured polarizations and ﬁts (upper plot) using and all data. Fit residuals (lower plot), with only statistical uncertainties from the ﬁts shown. In general, the uncertainties from the ﬁts are smaller than the symbol sizes used in plotting the data. Note: the lower plot legend applies to the upper plot. importance of this effect may be had by ﬁtting Eq. (4) to data sets containing only large longitudinal polarization values. Figure 10 and TableVshow the results of ﬁts where the data ﬁtted have been restricted to be within 25% of the maximum polarization value. Such ﬁts do not give as good values for i, but still give good values for Pi. The ratio of these polarization values to the Mott polarization value is given in Fig. 11. These results indicate that the FIG. 9. (Color) The relative analyzing powers for the ﬁve Jefferson Laboratory electron beam polarimeters, normalized to the Mott polarimeter for comparison. FIG. 9. (Color) The relative analyzing powers for the ﬁve Jefferson Laboratory electron beam polarimeters, normalized to the Mott polarimeter for comparison. 042802-11 042802-11 042802-11 UNIQUE ELECTRON POLARIMETER ANALYZING POWER . . . 042802 (2004 PRST-AB 7 FIG. 10. (Color) Polarimeter data and ﬁt (upper plot) using and data set limited to be within 25% of the maximum polarization. Fit residuals (lower plot) are shown using only statistical uncertainties from polarimeter data. Note: the lower plot legend applies to the upper plot. FIG. 10. (Color) Polarimeter data and ﬁt (upper plot) using and data set limited to be within 25% of the maximum polarization. V. BEAM ENERGY MEASUREMENTS The beam energy is usually measured by either high precision magnetic spectrometer or elastic electron- TABLE V. Polarization and phase results for polarimeter data restricted to be within 25% of the maximum polarization measured; italicized text show results using all data (see Table IV) for comparison. The uncertainties are the total standard error in the ﬁt parameters using the least squares ﬁtting routine. Results presented in this table use 5:630=kV. Polarization (%) Phase (deg) Polarimeter 25% restriction All data 25% restriction All data Mott 72:27 0:23 72.100.28 88:59 0:50 88.880.32 Møller A 75:40 0:33 76.761.04 10 981:60 0:68 10 983.840.63 Møller B 70:01 1:80 69.840.64 10 501:07 4:54 10 500.480.56 Møller C 73:37 0:42 73.300.62 10 021:70 0:74 10 023.390.61 Compton 73:85 2:15 73.331.42 10 982:86 4:10 10 984.821.33 042802-12 042802-12 TABLE V. Polarization and phase results for polarimeter data restricted to be within 25% of the maximum polarization measured; italicized text show results using all data (see Table IV) for comparison. The uncertainties are the total standard error in the ﬁt parameters using the least squares ﬁtting routine. Results presented in this table use 5:630=kV. Polarization (%) Phase (deg) Polarimeter 25% restriction All data 25% restriction All data Mott 72:27 0:23 72.100.28 88:59 0:50 88.880.32 Møller A 75:40 0:33 76.761.04 10 981:60 0:68 10 983.840.63 Møller B 70:01 1:80 69.840.64 10 501:07 4:54 10 500.480.56 Møller C 73:37 0:42 73.300.62 10 021:70 0:74 10 023.390.61 Compton 73:85 2:15 73.331.42 10 982:86 4:10 10 984.821.33 042802-12 042802-12 042802-12 042802-12 PRST-AB 7 J. M. GRAMES et al. 042802 (2004) FIG. 11. (Color) The relative analyzing powers for the ﬁve Jefferson Laboratory electron beam polarimeters, normalized to the Mott polarimeter for comparison. The solid symbol markers represent the results for the data set limited to be within 25% of the maximum measured polarization. The open symbol markers are the results shown in Fig. 9. FIG. 11. (Color) The relative analyzing powers for the ﬁve Jefferson Laboratory electron beam polarimeters, normalized to the Mott polarimeter for comparison. The solid symbol markers represent the results for the data set limited to be within 25% of the maximum measured polarization. The open symbol markers are the results shown in Fig. 9. A. Method No. 1: Beam energy measured by spin precession between the injector and the experimental halls After manipulation, the ﬁnal beam energy is written in terms of these accelerator parameters and the total spin precession determined from the polarimeter measure- ments as The electron beam gains an initial energy in the in- jector. After injection into the main accelerator the elec- tron beam successively gains energy in each linac during each recirculation pass (see Fig. 1). The dipole magnets in the recirculation and experimental hall transport arcs precess the beam polarization. The total spin precession between the injector and any experimental hall, as mea- sured by the Mott polarimeter and the corresponding experimental hall polarimeter, can be exactly calculated. For n recirculations through the accelerator (n 5 for this experiment) the total precession, n, can be summed and written, after some algebraic manipulation, as E 4men g2 E0!t !h nE12!t!hn1 !12 2n1 !t !h : (6) (6) The main advantage of this method is that at the high- est CEBAF energies one can take advantage of the very large total precession ( > 10 000) to reach an absolute measurement of the beam energy to better than 104. To do so requires precise knowledge of the accelerator pa- rameters in Eq. (6). The sensitivity of the beam energy to these parameters is given in TableVII and is described in more detail below. n g 2 2me fn!1 n 1 !2E0 n 2 n 1 !1 n 1 !2E1 nn 1 2 !1 !2 E2 E0 nE1 E2 !hg; Uncertainty in the injector beam energy (E0) is a signiﬁcant contribution to the total uncertainty because this fraction of the beam energy is precessed by each TABLE VI. Transformations to practical accelerator parameters. Quantity Transformation Final beam energy E0 nE1 E2 ! E Linac imbalance E1 E2 ! E12 Total bend angle n!1 n 1 !2 !h ! !t Linac skewness !1 !2 ! !12 TABLE VI. Transformations to practical accelerator parameters. (5) where E0, E1, and E2 are the energy gains of the injector, north linac and south linac, !1 and !2 are the bend angles of the east and west recirculation arcs, and !h is the bend angle of the respective experimental hall transport arc (h 2 fA;B;Cg). Note that Eq. V. BEAM ENERGY MEASUREMENTS For comparison, the beam energy was measured using the magnetic spectrometer method [26]. Two pairs of beam proﬁle monitors measured the beam direction be- fore and after a string of eight well-measured dipole magnets leading into Hall A to determine the resulting beam deﬂection. This measurement gave a ﬁve-pass beam energy of 5646:5 3:0 MeV. practice, this is assured by measuring and correcting the total path length of each recirculation pass. The system developed to do this allows the path length on each pass to be set with a 2( precision of better than 0.25 rf degree, leading to negligible differences in the energy gain on each pass [27]. It is useful to transform Eq. (5) to parameters more practical (see Table VI) for evaluating the beam energy. A. Method No. 1: Beam energy measured by spin precession between the injector and the experimental halls Consider the case in which the energy gained in one full recirculation pass is ﬁxed. The west recirculation arcs follow both linacs, so the precession in these arcs does not depend on the equality of the linac energy gains. However, the east recirculation arcs follow only the north linac, and thus the precession in these arcs depends on E12. In particular, precession in the east arcs is greater when E12 > 0. The precision with which the bend angles of the recir- culation arcs !1; !2 and hall transport arcs !h is known also impacts the energy measurement. These angles have been determined by high precision survey measurements of beam line elements in the linacs and at the beginning and end of the transport beam lines into experimental Halls A and C. The survey measurements are made with respect to the accelerator site reference grid, which in turn is established within a network of survey monuments spanning the accelerator site. E12 was not measured at the time of the precession measurements reported here. Instead, this measurement was made soon afterward, when the energy gain in each linac had been increased a nominal 7%. To measure E12 a single pass beam was sent to Hall A, and its energy was determined by the standard method of measuring the total deﬂection through the eight Hall A transport di- poles. The beam energy was measured with both linacs powered, and with only the north linac powered and the beam simply drifted through the unpowered south linac. At the time of the measurements reported here, gyro- theodolite measurements [29] were made at a number of locations throughout the accelerator site to conﬁrm the overall shape and stability of the accelerator site reference grid. A gyrotheodolite is an instrument that measures, with very high precision (ca. 3 arc sec), the horizontal direction of a line with respect to true north—the Earth’s axis of rotation. The gyrotheodolite was used to transfer a common high precision directional reference to locations throughout the accelerator tunnel complex. One can then compare the azimuths as determined from the gyrotheo- dolite measurements to azimuths calculated from preci- sion survey measurements. A. Method No. 1: Beam energy measured by spin precession between the injector and the experimental halls (5) assumes that the energy gain on each pass through each linac is the same. In 042802-13 042802-13 042802-13 042802-13 042802 (2004) PRST-AB 7 PRST-AB 7 NIQUE ELECTRON POLARIMETER ANALYZING POWER . . . 042802 (2004) UNIQUE ELECTRON POLARIMETER ANALYZING POWER . . . TABLE VII. The derivative of Eq. (6) with respect to each of the dependent parameters is shown alongside the corresponding the numerical value calculated for Hall B. The numerical values for Halls A and C are within 5% of the values for Hall B. TABLE VII. The derivative of Eq. (6) with respect to each of the dependent parameters is shown alongside the corresponding the numerical value calculated for Hall B. The numerical values for Halls A and C are within 5% of the values for Hall B. Derivative Numerical value @E=@n 4me=g 2 =!t !h 0:544 MeV= deg @E=@E0 !t !h =!t !h 1:00 MeV=MeV @E=@!12 nn 1 E12=2n 1 =!t !h 0:00713 MeV= deg @E=@E12 nf!t !h n 1 !12=2n 1 !t !h g 0:556 MeV=MeV @E=@!t E E0 nE12=2n 1 =!t !h 3:527 MeV= deg @E=@!h E E0 nE12=2n 1 =!t !h 3:449 MeV= deg Derivative Numerical value @E=@n 4me=g 2 =!t !h 0:544 MeV= deg @E=@E0 !t !h =!t !h 1:00 MeV=MeV @E=@!12 nn 1 E12=2n 1 =!t !h 0:00713 MeV= deg @E=@E12 nf!t !h n 1 !12=2n 1 !t !h g 0:556 MeV=MeV @E=@!t E E0 nE12=2n 1 =!t !h 3:527 MeV= deg @E=@!h E E0 nE12=2n 1 =!t !h 3:449 MeV= deg use of precision survey measurements to determine the angles necessary for the energy determination. The an- gles used are given in Table VIII. dipole magnet of the accelerator. The injector beam mo- mentum was measured immediately prior to the experi- ment using a recent high precision calibration of the injector spectrometer [28]. This gives a value for the injector beam energy of E0 62:89 0:06 MeV. Finally, the polarization precession depends on the linac energy imbalance, E12, which is the difference in the energy gain between the north and the south linacs. B. Method No. 2: Beam energy measured by relative spin precession differences between experimental halls The alternative to comparing the total precession be- tween the injector and an experimental hall polarimeter is simply to use the precession difference between each experimental hall. The advantage is that prior to extrac- tion, the beam to each hall undergoes identical precession. In this way the injector energy, linac imbalance, and recirculation arc bend angles are eliminated from the equation and the precession between any two experimen- tal hall polarimeters is given simply by The results are reported in Table IX. These measurements gave a value of 4:13 0:66 MeV for E12 at this energy. The results are reported in Table IX. These measurements gave a value of 4:13 0:66 MeV for E12 at this energy. g 12 gy The value of E12 at the energy used for the precession measurements was determined by scaling to the lower energy, using accelerator parameters measured at each energy. Two different methods were used to do this scal- ing. In the ﬁrst of these, the sum of the accelerating gradients of the 160 cavities in each linac was compared at the two beam energies. In the second, detailed mea- surements of the beam orbit through the ﬁrst and second recirculation arcs, coupled with measurements of the dipole bus current in these arcs, were used. These two scaling methods gave values for E12 of 4:75 0:76 MeV and 5:69 0:95 MeV, respectively. The weighted mean of these two values is 5:17 0:59 MeV. For the analysis presented here, we use the value E12 5:2 1:0 MeV. The enlarged uncertainty in the value of E12 is chosen to account for any undetected systematic effects associated with determining E12 at a different time and with differ- ent linac energy gains. $ g 2 2 E mec2 $%; (7) (7) where $ and $% are the measured differences in the precession and bend angle between two respective hall polarimeters. E is the ﬁnal beam energy common to both. In this case, the uncertainty in the bend angle is the main contribution to the energy uncertainty. The disadvantage of this method is that the precession difference between different experimental hall polarimeters is much smaller than the precession through the full ﬁve-pass accelerator. Thus, even at the maximum CEBAF energy, the statisti- cal precision of this energy measurement is of order 103. A. Method No. 1: Beam energy measured by spin precession between the injector and the experimental halls There are three sources of measurement uncertainty in determining these angles: (a) the 3 arc sec uncertainty of the gyrotheodolite; (b) an azimuthal uncertainty resulting from the transverse position measurement uncertainty of 0.25 mm and the separation of the beam line elements; and (c) an azimu- thal uncertainty resulting from the transverse component of the ﬁducialization uncertainty of 0.05 mm and the separation of the beam line elements. Using normal error propagation, these differences in azimuth were less than 1.5 arc sec in the linacs, approximately 10 arc sec in the beam switchyard, less than 5 arc sec in the Hall A trans- port line, and less than 10 arc sec in the Hall C transport line. Overall, the gyrotheodolite measurements verify the accelerator site reference grid at the level of the 3 arc sec gyrotheodolite measurement uncertainty and justify the TABLE VIII. The gyrotheodolite survey measurements are shown along with the resulting calculated parameters needed for the spin precession results. The angle of the beam line to Hall B was not surveyed with the gyrotheodolite and is included as an assumed value. Quantity Acquired Angle (deg) !hA Measured 37:4913 0:0020 !hB Assumed 0:0000 0:0100 !hC Measured 37:4774 0:0057 !1 Measured 180:0000 0:0020 !2 Measured 180:0000 0:0020 !12 Calculated 0:0000 0:0020 !tA Calculated 1657:4913 0:0024 !tB Calculated 1620:0000 0:0101 !tC Calculated 1582:5226 0:0058 042802-14 042802-14 042802-14 042802-14 J. M. GRAMES et al. PRST-AB 7 PRST-AB 7 042802 (2004) TABLE IX. Summary of measurements to determine the linac energy imbalance. TABLE IX. Summary of measurements to determine the linac energy imbalance. cession for each polarimeter is that listed in Table IV, corrected by the polarization orientation as measured in the injector by the Mott polarimeter (1:12 0:32). cession for each polarimeter is that listed in Table IV, corrected by the polarization orientation as measured in the injector by the Mott polarimeter (1:12 0:32). linac energy imbalance. Quantity Acquired Beam energy (MeV) Injector Measured 67:89 0:07 North linac only Measured 671:14 0:28 North and south linac Measured 1270:26 0:52 E1 Calculated 603:25 0:29 E2 Calculated 599:12 0:59 B. Method No. 2: Beam energy measured by relative spin precession differences between experimental halls The beam energies obtained from the precession differ- ences between the four experimental hall polarimeters are given in Table XI. With the above value for E12, we can determine four values for the ﬁnal beam energy from the measurements of the total precession to the four hall polarimeters. These values are presented in Table X. The total ﬁve-pass pre- C. Final energy determination another and consistent with the much more precise values determined by the total precession method. However, there are large and statistically untenable discrepancies in the energies determined by using the precession dif- ferences between the Hall B polarimeter and the Hall A and C polarimeters.When comparing either the Hall A or Hall C polarimeter to the Hall B polarimeter these dis- crepancies, about 37 MeV (see Table XI), clearly indicate that the difference is associated with whether the com- parison polarimeter is located in Hall A (higher energy) or Hall C (lower energy). Inspection of Eq. (7) shows that the discrepancies associated with the Hall B polarimeter can be resolved by either increasing the total precession from the injector to the Hall B polarimeter, , by 3:4 or an angular shift of the Hall B beam line, !hB, by 0:26 further from Hall A and closer to Hall C. We make the following observations with respect to our results. (i) Polarimeters: It is clear from these results that the ﬁve Jefferson Lab polarimeters do not all agree on the measured beam polarization at the level of a few percent. The effective analyzing power of each of these polar- imeters is a simulated number. Polarimeter comparisons of this type help to reveal undetected systematic effects and other sources of error in the simulated values of the effective analyzing powers. A thorough understanding of polarimeter systematic effects will be necessary to achieve absolute electron polarimetery at the 1% level, as is required for some proposed parity violation mea- surements. While precision comparisons of polarimeter effective analyzing powers can never guarantee that the effective analyzing powers used are correct, agreement among the values of the beam polarization as measured by very different polarimeters provides strong circum- stantial evidence that no large systematic effects have been overlooked. In general, our results indicate how difﬁcult it will be to establish a convincing case that a polarimeter is capable of 1% absolute electron polariza- tion measurements. Inserting these two possibilities into Eq. (6) for the energy determined from the total spin precession method yields a beam energy of 5649:05 0:66 for the modiﬁed precession or 5648:10 0:66 for the modiﬁed beam line. In either case, these new values agree better with the energy as determined by the total precession to the Halls A and C polarimeters. C. Final energy determination Both the injector to experimental hall and experimen- tal hall to experimental hall spin-based energy measure- ment methods indicate discrepancies associated with the Hall B polarimeter. The energy as determined by the precession from the injector to the Hall B polarimeter is somewhat in disagreement ( > 2() with the energy determined with the three other polarimeters, all of which are themselves in good agreement. The energies determined from the precession difference between the Hall A and Hall C polarimeters are consistent with one TABLE X. Final beam energies measured by total precession evaluated at E12 5:2 1:0 MeV. Polarimeters (deg) E (MeV) Mott-Compton 10 985:94 1:37 5649:21 0:89 Mott-Møller A 10 984:96 0:71 5648:70 0:65 Mott-Møller B 10 501:60 0:64 5647:20 0:66 Mott-Møller C 10 024:51 0:69 5649:03 0:71 TABLE X. Final beam energies measured by total precession evaluated at E12 5:2 1:0 MeV. TABLE XI. Summary of energy measurement results comparing only end-station polar- imeters by the relative spin precession method. Polarimeters $ (deg) $% (deg) E (MeV) (E E (%) Møller A-Møller B 483:36 0:84 37:4913 0:0102 5681:10 10:03 0.176 Møller A-Møller C 960:45 0:88 74:9687 0:0060 5645:30 5:17 0.092 Compton A-Møller B 484:34 1:44 37:4913 0:0102 5692:62 17:03 0.299 Compton A-Møller C 961:43 1:46 74:9687 0:0060 5651:07 8:61 0.152 Møller B-Møller C 477:09 0:83 37:4774 0:0115 5609:49 9:89 0.176 TABLE XI. Summary of energy measurement results comparing only end-station polar imeters by the relative spin precession method. 042802-15 042802-15 042802-15 UNIQUE ELECTRON POLARIMETER ANALYZING POWER . . PRST-AB 7 042802 (2004) Asymmetry measurements as a function of this spin orientation were conducted with each polarimeter. Since each polarimeter measures a beam with the same magnitude of polarization, a determination of the relative analyzing power of the various polarimeters is straight- forward. Furthermore, a measurement of the spin orien- tation, as set by the Wien ﬁlter, that produces maximum longitudinal polarization at each polarimeter leads to a very high precision energy measurement. At the accelera- tor energies employed in this experiment, measurement of the total precession from the injector to a single polar- imeter leads to a determination of the beam energy with a precision of about 104. C. Final energy determination It is not possible to determine the source of the dis- crepancy associated with the Hall B polarimeter energy determination from our measurements. The required shift in precession of about 3:4 is very large compared to the standard deviation of the ﬁt value for the Hall B total precession (0:64), making a precession error this large unlikely. The angle of the beam line into Hall B was not determined as part of the precision gyrotheodolite sur- vey, leaving open the possibility that there is a small angular misalignment of the polarimeter beam line. Further measurements are required to resolve the present discrepancy. At this point, we choose to ignore all energy measure- ments involving the Hall B polarimeter. We can deter- mine a best value for the full ﬁve-pass accelerator energy from the three independent measurements of the total precession from the injector to the two Hall A and one Hall C polarimeters. The statistically weighted energy as determined from these three measurements is 5648:94 0:42 MeV. This value is within less than 1 standard de- viation of the value of 5646:5 3:0 MeV as determined by the magnetic spectrometer method. The polarimeter comparison presented here is interest- ing in that we have reported measurements of the longi- tudinal polarization component only. In all cases, these measurements were made with a highly polarized beam. Thus, when the measured longitudinal component is small there is necessarily a large transverse component in the horizontal plane for these measurements. The pres- ence of a large transverse spin component may result in systematic effects in the effective longitudinal analyzing power of a particular polarimeter. Our measurements indicate that large transverse spin components may be an important source of systematic effects for conven- tional Møller polarimeters. VI. CONCLUSIONS A careful comparison of the relative analyzing powers of the ﬁve Jefferson Laboratory electron polarimeters, based on Mott, Møller, and Compton scattering, has been performed. This was accomplished by using a Wien ﬁlter spin rotator located in the low energy region of the injector to orient the electron spin between 110 relative to the beam direction in the horizontal plane. Finally, it is worth noting that the beam polarizations determined by the injector Mott polarimeter, the Hall A Compton polarimeter, and the Hall C Møller polarimeter are in reasonable statistical agreement with a single value. Note that the errors on the values given in Table IV are from the ﬁt only and do not include either the measure- ment statistics or the systematic uncertainty estimated for 042802-16 042802-16 042802-16 042802-16 J. M. GRAMES et al. PRST-AB 7 042802 (2004) each polarimeter. The actual uncertainties in the mea- sured polarization values are larger than those shown in Table IV. This result implies that the analyzing powers of these three very different polarimeters are reasonably well understood. value of the accelerator energy with exceptional precision. While spin-based energy measurements are too cumber- some for routine use, a program of such measurements could provide useful high precision cross calibrations of more easily applied energy measurement techniques. (ii) Wien ﬁlter: The value of the Wien angle coefﬁcient , as determined from ﬁts of the longitudinal polarization measured with each polarimeter as a function of theWien angle, depends on the particular polarimeter used. While this effect is not understood, the resulting uncertainty in the measured analyzing power and precession angle due to the uncertainty in the value of is small. Two data sets, taken with the injector Mott polarimeter about one month apart, gave ﬁt values for differing by about 1%. Precision measurement of the gap high voltage and the magnet current associated with each measurement, as well as the magnet hysteresis history, is clearly war- ranted. Operationally, the beam orbit through the Wien ﬁlter may be slightly different with different Wien ﬁlter settings, particularly as the Wien angle becomes large. While this systematic effect does not affect the conclu- sions from the measurements reported here, it should be better understood for future polarimeter comparison measurements. ACKNOWLEDGMENTS This work is the result of contributions by many scien- tists dedicated to producing and measuring the polarized electron beam at Jefferson Laboratory. C. Curtis and the alignment group conducted the high precision gyrotheo- dolite measurements necessary to attain the spin-based energy measurement results. This work was supported by the U.S. DOE Contract No. DE-AC05-84-ER40150. [1] L. G. Levchuk, Nucl. Instrum. Methods Phys. Res., Sect. A 345, 496 (1994). [2] M. Swartz et al., Nucl. Instrum. Methods Phys. Res., Sect. A 363, 526 (1995). [3] C.W. Leemann, D. R. Douglas, and G. A. Krafft, Annu. Rev. Nucl. Part. Sci. 51, 413 (2001). (iii) Energy measurements: The total precession energy measurement using only the Hall A and Hall C polar- imeters yields a statistically weighted mean, from three independent measurements, of 5648:94 0:42 MeV, or a precision of 0:74 104. The value is well within 1 stan- dard deviation of the magnetic arc spectrometer measure- ment of 5646:5 3:0 MeV. The systematic uncertainties of the magnetic spectrometer measurement are very dif- ferent than the systematic uncertainties of the total pre- cession measurements. The statistically weighted mean of these two total energy measurements is 5648:89 0:42 MeV. A major source of uncertainty in the present total precession measurements is the linac energy imbal- ance. The value used in the above measurements includes an enlarged uncertainty to account for the fact that the energy imbalance was not measured at the same time, or at the exact energy, of the spin-based energy measure- ments. This source of uncertainty could be greatly re- duced in a future spin-based energy measurement. There is clear evidence of problems with the energy measure- ments based on the Hall B polarimeter, and some evi- dence of a misalignment of the beam to the Hall B polarimeter. [4] J. S. Price et al., in Proceedings of the 12th International Symposium on High-Energy Spin Physics (SPIN 96), Amsterdam, Netherlands, 1996 (World Scientiﬁc, Singapore, 1996), pp. 727–729. [5] J. S. Price et al., in Proceedings of the 7th International Workshop on Polarized Gas Targets and Polarized Beams, Urbana, IL, 1997 (AIP, Woodbury, NY, 1997), pp. 446–450. [6] M. Steigerwald, in Proceedings of the International Workshop on Polarized Sources and Targets, Erlangen, Germany, 1999 (University of Erlangen–Nurnberg, Erlangen, Germany, 1999), pp. 258–261. [7] N. Falletto et al., Nucl. Instrum. Methods Phys. Res., Sect. A 459, 412 (2001). [8] M. ACKNOWLEDGMENTS Baylac et al., Phys. Lett. B 539, 8 (2002). [9] HAPPEX, K. A. Aniol et al., Phys. Lett. B 509, 211 (2001). [10] A.V. Glamazdin, et al., Fizika B 8, 91 1999. [11] R. Nasseripour, B. Raue, L. Kramer, and F. Fortier, Bull. Am. Phys. Soc. 45, 91 (2000). [12] M. Hauger et al., Nucl. Instrum. Methods Phys. Res., Sect. A 462, 382 (2001). [13] L. J. deBever, M. Loppacher, J. Zhao, W. A. Tobias, and B. Zihlmann, in Proceedings of the 12th International Symposium on High-Energy Spin Physics (SPIN 96), Amsterdam, Netherlands, 1996 (Ref. [4]), pp. 768–770. Jefferson Lab is developing high precision electron polarimetry to meet the needs of future experiments. A comparison of the beam polarization measured by differ- ent polarimeters observing the same beam is an impor- tant way to gain conﬁdence that the effective analyzing powers of these polarimeters have been correctly deter- mined. The results presented here clearly demonstrate that measurements of the total precession through the CEBAF accelerator can be used to determine the absolute [14] C. K. Sinclair, in Proceedings of the 1999 IEEE Particle Accelerator Conference (PAC 99), New York (IEEE, Piscataway, NJ, 1999), pp. 65–69. [15] M. Poelker et al., in Proceedings of the 14th International Spin Physics Symposium (SPIN 2000), Osaka, Japan, 2000 (AIP, Melville, NY, 2000), pp. 943–948. [16] M. Salomaa and H. Enge, Nucl. Instrum. Methods 145, 279 (1977). 042802-17 042802-17 042802-17 042802-17 042802-17 UNIQUE ELECTRON POLARIMETER ANALYZING POWER . . . 042802 (2004) PRST-AB 7 ed. 1998, http://www.triumf.info:8081/physica/html/ homepage.html [17] K. Halbach, Lawrence Livermore National Laboratory Technical Report No. UCRL-17436, 1967. [24] W. Press, S. Teukolsky, W. Vetterling, and B. Flannery, Numerical Recipes in Fortran 77 (Cambridge University Press, New York, 1992), 2nd ed. [18] J. M. Grames, in Proceedings of the International Workshop on Polarized Sources and Targets, Erlangen, Germany, 1999 (Ref. [6]), pp. 266–269. [19] J. M. Grames, Ph.D. thesis, University of Illinois at Urbana–Champaign, 2000, ftp://ftp.jlab.org/pub/ grames/thesis.pdf [25] P. Bevington, Data Reduction and Error Analysis for the Physical Sciences (McGraw-Hill, New York, 1969). [26] J. Berthot and P. Vernin, Nucl. Phys. News 9N4, 12 (1999). g p [20] M. Poelker, Appl. Phys. Lett. 67, 2762 (1995). [27] D. Hardy, J. Tang, L. R. M. Tiefenback, M. Crofford, and G. A. Krafft, in Proceedings of the 1997 IEEE Particle Accelerator Conference (PAC 97), Vancouver, B.C., Canada (IEEE, Piscataway, NJ, 1997), pp. 2265– 2267. ACKNOWLEDGMENTS [21] J. Schuler et al., in Proceedings of the Low Energy Polarized Electron Workshop (LE98), St. Petersburg, Russia, 1998 (SPES Lab Publishing, St. Petersburg, Russia, 1998). [22] A. Krycuk, J. Fugitt, A. Johnson, R. Kazimi, and L. Turlington, in Proceedings of the 1993 IEEE Particle Accelerator Conference (PAC 93), Washington, DC (IEEE, Piscataway, NJ, 1993), pp. 939–940. [28] R. Kazimi (private communication). [29] C. Curtis, Jefferson Laboratory Technical Report, 2000, survey and alignment data transmittals No. L625, No. L634, and No. L643. [23] J. Chuma, Physica Reference Manual v1.3, Vancouver, Canada, triumf computing document tri-cd-93-01 042802-18 042802-18
W2162205373.txt	https://www.scielo.br/j/anaismp/a/JVtcpBBmtpgtbsST4mBxg7P/?lang=pt&format=pdf	pt		Tecido urbano e mercado imobiliário em São Paulo: metodologia de estudo com base na Décima Urbana de 1809	Anais do Museu Paulista	2,005	cc-by	13,961	Tecido urbano e mercado imobiliário em São Paulo: metodologia de estudo com base na Décima Urbana de 1809 Beatriz Piccolotto Siqueira Bueno Faculdade de Arquitetura e Urbanismo da USP RESUMO: Este artigo apresenta uma metodologia inédita de espacialização da Décima Urbana, primeiro imposto predial estabelecido para as cidades brasileiras. Focaliza o caso de São Paulo, em 1809. Os dados recolhidos na documentação textual foram processados em banco de dados e cartografados na primeira planta cadastral da cidade, elaborada pelo engenheiro Carlos Bresser, entre 1844-1847, e confrontados com a documentação iconográfica dos viajantes e de Militão Augusto de Azevedo, de modo a precisar as informações obtidas. A Décima Urbana de 1809 contém informações sobre a localização dos imóveis, seus proprietários, inquilinos, tipologias (casas térreas, sobrados, lojas), finalidades (uso próprio, aluguel), usos (residencial, comercial, misto) e valor, que hoje nos permitem reconstituir hipoteticamente o velho tecido urbano da cidade de São Paulo e aspectos da dinâmica do seu mercado imobiliário em fins do período colonial. PALAVRAS-CHAVE: São Paulo. Décima Urbana. Tecido Urbano. Mercado imobiliário rentista. Período Colonial. Décima Urbana, the first property tax established in Brazilian cities. The case of São Paulo in 1809 is studied. The data gathered from textual documentation was processed in a database and cartographed on the first official city plan, elaborated by engineer Carlos Bresser between 1844-1847, and then confronted with the iconographic documentation produced by visiting travellers and by photographer Militão Augusto de Azevedo, so as to cross-reference the information obtained. The Décima Urbana of 1809 contains data about the siting of buildings, their proprietors, tenants, typology (single and two-storey houses, shops), finalities (own use, rent), uses (residential, commercial, mixed) and value. This allows for the present day hypothetical reconstruction of old São Paulo’s urban mesh and of aspects of the real estate market dynamics at the end of the colonial period. KEYWORDS: São Paulo. Décima urbana. Urban mesh. Property Rent Market. Colonial Period. ABSTRACT: This article presents a new spatialisation methodology for the Anais do Museu Paulista. São Paulo. N. Sér. v.13. n.1. p. 59-97. jan. - jun. 2005. 59 1. Esta pesquisa é parte de um Projeto Temático financiado pela Fapesp, intitulado Urbanização dispersa e mudanças no tecido urbano.Estudo de caso: Estado de São Paulo, que está sendo desenvolvido no Laboratório de Estudos sobre Urbanização,Arquitetura e Preservação (LAP), da Faculdade de Arquitetura e Urbanismo da USP, sob a coordenação geral do Prof. Dr. Nestor Goulart Reis Filho.Dentro dos objetivos do projeto temático, sob nossa coordenação, subtema específico contempla o velho tecido urbano da cidade de São Paulo em paralelo ao estudo da dinâmica do seu mercado imobiliário. Se propõe a analisar a questão numa perspectiva histórica, de longa duração, enfocando de 1809 a 1950. Por não se tratar de um período homogêneo, dividimos a pesquisa em três módulos: 1809 a 1870; 1870 a 1930; 1930 a 1950. Neste artigo,apresentaremos algumas conclusões referentes ao período colonial, com base em documentação pouco utilizada pelos historiadores, geógrafos, arquitetos, urbanistas e economistas, de extrema relevância para pesquisas dessa natureza. 2. Foi critério da autora manter a grafia de nomes dos proprietários como são citados na Décima Urbana de 1809. 60 Reconstituir hipoteticamente a velha tessitura da cidade de São Paulo e os agentes sociais envolvidos na sua produção, em fins do período colonial, foi um dos desafios assumidos neste artigo, parte de uma pesquisa em andamento1. Ao espacializarmos uma documentação inédita, a Décima Urbana de 1809, pudemos constatar aspectos surpreendentes da cidade de São Paulo, jamais enfocados pela historiografia. Verificamos que boa parte do tecido urbano era produto da iniciativa privada e 50% das casas destinadas à renda de aluguel. Constatamos a existência de um mercado imobiliário “rentista” na São Paulo colonial e considerável concentração de imóveis nas mãos das ordens religiosas, em especial dos beneditinos, detentores de 61 prédios. No entanto, se individualmente foram as ordens religiosas os principais agentes detentores do patrimônio imobiliário urbano paulistano, no conjunto, 81% dos prédios eram patrimônio laico, envolvendo nomes conhecidos como os dos Coronéis Luiz Antônio de Souza e Jozé Arouche de Toledo, bem como outros menos famosos de comerciantes e negociantes.2 A elite tinha muitas vezes dupla morada, em geral gozando de chácaras nos arredores da cidade e de amplos sobrados nas ruas centrais. No entanto, em 1809, os célebres sobrados dos Quatro Cantos pertencentes aos Prados e Jordãos ainda não haviam sido construídos. Também o sobrado de ângulo, na esquina das ruas do Ouvidor e São Bento, pertencente ao futuro Brigadeiro Luiz Antônio não estava lá; correspondia ainda a uma simples loja e um lanço – a Casa Souza – onde o proprietário praticava um dos seus importantes negócios. Não foi tarefa fácil espacializar a Décima, sendo fundamental a eleição de um mapa-base contemporâneo à documentação manuscrita – a planta cadastral de Carlos Bresser (1844 -1847) –, sendo igualmente árduo simular o percurso do fiscal inventariante – responsável pela atribuição do imposto –, por vezes pouco objetivo. A metodologia de pesquisa consistiu portanto de quatro procedimentos associados: 1) compilação dos dados da documentação manuscrita localizada no Arquivo do Estado de São Paulo; 2) aplicação das informações obtidas na Décima, lote a lote, no mapa-base escolhido; 3) confrontação dos dados com a iconografia dos viajantes contemporâneos ao período – Ender (1817), Pallière (1821), Edmund Pink (1823), Burchell (1827) – e com as fotos de Militão Augusto de Azevedo; 4) verificação de alguns imóveis sabidamente pertencentes a certas famílias, via fontes secundárias. Nesse sentido, a tese de doutorado de Paulo Garcez Marins foi de extrema valia, além das fichas do Fundo Aguirra do Museu Paulista, que ainda merecem estudo mais detido. A documentação iconográfica de viajantes para a cidade de São Paulo é escassa e parcial, focalizando especialmente os Largos da Sé, Misericórdia e do Colégio. Ao contrário, as fotos de Militão Augusto de Azevedo são muito mais fartas e espacialmente abrangentes. Embora correspondam à segunda metade do século XIX, tais fotografias nos permitem entrever as velhas Anais do Museu Paulista. v. 13. n.1. jan.- jun. 2005. casas térreas e sobrados de beirais largos, vergas retas ou de arco abatido, com rótulas e muxarabis do período colonial, em meio às edificações alteadas, maquiadas de platibandas, ferragens e vidraças. A interpretação da documentação iconográfica precisa, portanto, a volumetria sugerida pela Décima de 1809. Possivelmente ocorreram imprecisões, mas alguns aspectos de conjunto podem ser inferidos desde já. Trata-se de uma cidade concentrada na colina entre os rios Anhangabaú e Tamanduateí, predominantemente térrea (86%) e residencial (86%), com sobrados localizados em determinadas ruas de uso misto, correspondendo às áreas mais valorizadas da cidade. Ao descortinarmos os proprietários e inquilinos das casas e lojas, surgiu o perfil da população urbana e o cenário ganhou vida, permitindo elucidar aspectos obscuros da historiografia, inferir outros e vislumbrar futuras linhas de pesquisa ainda inexploradas sobre a velha São Paulo. A historiografia Divulgada nos estudos de Raquel Glezer, mas ainda inexplorada desse ponto de vista, a Décima Urbana foi o primeiro imposto predial, estabelecido para a Corte e principais vilas, cidades e lugares notáveis da faixa litorânea, pago à Fazenda Real, correspondendo a 10% do rendimento líquido de todos os bens de raiz, incidindo sobre proprietários e inquilinos, com exceção daqueles pertencentes às Santas Casas de Misericórdia. Ampliado em sua área de abrangência, por Alvará emitido ainda em 3/6/1809, o tributo foi estendido a todas as povoações, para além da faixa costeira, mantendo as mesmas isenções do texto anterior e reforçando a exigência de pronto pagamento. Para fins de tributação, resultou no arrolamento, em livro específico, de todos os prédios circunscritos no perímetro urbano, então definido para tanto, bem como na primeira numeração dos edifícios da cidade. Nesse sentido, contém informações preciosas sobre a localização dos imóveis, seus proprietários, inquilinos (em caso de imóvel de aluguel), tipologias, finalidades, usos e valor (do prédio e do aluguel), que hoje nos permitem reconstituir o velho tecido urbano do atual centro histórico da cidade de São Paulo, em 1809, e entender aspectos do seu mercado imobiliário. Convém salientar que a Décima Urbana teve longa vida, apesar dos percalços iniciais para sua cobrança. Sua escala de abrangência e alíquota foram alteradas em 1867: [...] cobrança de12% na décima adicional dos prédios das corporações de mão morta, os prédios dos bancos, companhias e sociedades anônimas, e associações pias, beneficentes ou religiosas, incluindo no valor locativo o do terreno anexo (GLEZER, 1992, p.102-103). Annals of Museu Paulista. v. 13. n.1. Jan.- Jun. 2005. 61 A Décima Urbana tornou-se sinônimo de área urbana. Por ser sinônimo de área urbana, presta-se de maneira exemplar à reconstituição da sua tessitura. Embora muito tenha sido escrito sobre a cidade de São Paulo, escassa é a literatura sobre essas questões. Do ponto de vista do traçado planimétrico e volumétrico, os estudos recentes de Nestor Goulart Reis Filho (2004) e Benedito Lima de Toledo (2004) muito acrescentaram ao já conhecido. Do ponto de vista dos agentes sociais, produção e vivência do espaço urbano, no que diz respeito à casa comum, à exceção do capítulo “Sociabilidades paulistanas”, ainda inédito, da tese de Paulo Garcez Marins (1999), com base nos inventários post-mortem e na iconografia, raros foram os estudos de conjunto a descortinar essas questões. Pioneiro na localização e estudo do perfil social dos moradores das famosas casas térreas e sobrados de rótulas estampados nas fotos de Militão Augusto de Azevedo, o autor desenvolveu minuciosa metodologia de pesquisa, envolvendo o confronto de fontes documentais diversas – inventários, cartografia, iconografia, antigas numerações das ruas da cidade de São Paulo e a Décima Urbana de 1872. Na mesma linha, mais recentemente, o doutorado de Maria Luiza Ferreira de Oliveira (2003) contemplou aspectos das relações sociais, do cotidiano dos setores médios e do mercado imobiliário “rentista” na cidade de São Paulo, entre 1870-1900, também com base nos inventários de família. Em relação a pesquisas sobre a história do mercado imobiliário na cidade, acreditava-se que o mesmo fosse inexistente no período colonial. À exceção dos trabalhos pioneiros de Fania Fridman (1999) e Nireu Cavalcanti (2004), circunscritos ao caso carioca, nada foi escrito sobre patrimônio e mercado imobiliário urbano em tempos tão recuados. Os estudos de João Luís Fragoso (1998) apontaram para a questão ao demonstrar que a maioria das propriedades imobiliárias do centro do Rio estavam concentradas nas mãos de ricos comerciantes. No entanto, apenas Fania Fridman – com base na documentação oficial dos Arquivos da Cúria Metropolitana do Rio de Janeiro – e Nireu Cavalcanti – com base nos Livros da Décima Urbana da Corte de 1809 a 1812 – caracterizaram a sua existência, natureza e dinâmica específica. Outros estudos foram desenvolvidos sobre a matéria, enfocando no entanto períodos mais recentes. Convém destacar as dissertações de mestrado do historiador Reinéro Lérias (1988) e da geógrafa Mônica Silveira Brito (2000) – ambas com foco em São Paulo na Primeira República – e do arquiteto Walter Pires (2003) – com foco específico nas estruturas fundiárias dos atuais bairros do Cambuci, Jardim da Glória e Chácara Klabin, entre 1876 e 1904. Para as questões relacionadas às transformações urbanísticas na área central da cidade de São Paulo, correspondente à antiga Freguesia da Sé, na Primeira República (1889-1930), é fundamental a tese de doutorado de Heloisa Barbuy (2001) que enfoca as mudanças nos hábitos de consumo do paulistano em paralelo às mudanças nos usos e tipologias dos edifícios, bem como os 62 Anais do Museu Paulista. v. 13. n.1. jan.- jun. 2005. agentes sociais envolvidos. Igualmente importantes são as teses de Cândido Malta Campos Neto (2002) e José Geraldo Simões Jr (1985), focalizando as intervenções urbanísticas induzidas ou patrocinadas pelas administrações de Antônio Prado, Raimundo Duprat e demais prefeitos na zona do triângulo histórico (formado pelas ruas 15 de Novembro, São Bento e Direita), na Rua Líbero Badaró e no Vale do Anhangabaú. A metodologia de pesquisa Para obtermos alguns produtos da Décima Urbana de 1809, elaboramos um banco de dados específico para a pesquisa, contendo os seguintes campos: • Cidade. • Ano do inventário. • Freguesia. • Logradouro, número. • Nome do proprietário (indicação se leigo ou religioso, particular, institucional, ordem religiosa, irmandade, padre secular). • Nome dos inquilinos. • Tipologia do imóvel (sobrado, casa térrea, assobradado). • Subtipologia do imóvel (casa térrea de um a quatro lanços, loja, sobrado de um a três lanços, com ou sem loja, de um, dois ou três andares). • Uso (residencial, comercial, misto). • Finalidade (aluguel, uso próprio, cedido, em obras, outros). • Valor do imóvel. • Valor do aluguel/mês. • Valor da Décima Urbana. Alimentamos o banco de dados com todas as informações coletadas no Livro de 1809 e obtivemos alguns relatórios parciais, que nos permitiram tecer considerações sobre os principais proprietários de imóveis urbanos, localização dos prédios, montante investido em patrimônio imobiliário urbano e natureza específica do mercado imobiliário vigente em fins do período colonial, a saber: a) relatório geral contendo a quantificação dos imóveis catalogados por tipologia, Annals of Museu Paulista. v. 13. n.1. Jan.- Jun. 2005. 63 3. Fundo pertencente ao Serviço de Documentação Textual e Iconográfica do Museu Paulista-USP, desde 1962, contém uma preciosa coleção de fichas, mapas, cadastros, livros e fotografias, organizadas por João Baptista de Campos Aguirra, em mais de 20 anos de pesquisa nos cartórios da cidade, envolvendo informações sobre as transações imobiliárias realizadas do período colonial à Primeira República, rua a rua. 4. Em parceria com Paulo Garcez Marins. 64 tipo de proprietário, usos e finalidades; b) lista da quantidade de imóveis por proprietário e valor total do montante imobilizado em patrimônio imobiliário urbano por eles; c) lista de proprietários contendo os endereços de todos os seus imóveis; d) lista decrescente do valor dos imóveis e respectivos endereços. Para a reconstituição de aspectos concretos do tecido urbano de São Paulo, em 1809, cartografamos todas as informações arroladas no inventário da Décima na primeira planta cadastral da cidade – Mappa da Cidade de São Paulo offerecido a sua Magestade... – elaborada pelo engenheiro Carlos Bresser, entre 1844-1847, eleita mapa-base para essa etapa da pesquisa. As informações nele cartografadas, resultaram em quatro pranchas inéditas: Prancha I – tipologias predominantes (casas térreas ou sobrados); Prancha II – finalidade dos imóveis (uso próprio ou aluguel); Prancha III – usos (residencial, comercial ou misto); Prancha IV – principais agentes detentores de patrimônio imobiliário urbano (leigos e religiosos). Convém ressaltar que as informações disponibilizadas no inventário de 1809 foram confrontadas com outros três mapas de São Paulo datados de 1810, a saber: Planta da Cidade de S. Paulo [...] levantada em 1810 pelo Engenheiro Rufino Felizardo e Costa (REIS FILHO, 2004: p. 86); Planta da Imperial Cidade de São Paulo, levantada em 1810 pelo Engenheiro Rufino Felizardo e Costa e copiada em 1841 com todas as alterações (REIS FILHO, 2004, p. 119) e Carta da capital de São Paulo que o Exmo. Snr. Barão de Caxias mandou executar pelo Engenheiro da Columna José Jacques da Costa Ourique, Fortificador da capital, 1842 (REIS FILHO, 2004, p. 121). Na preparação dos mapas, em Adobe Photoshop, contamos com a colaboração de Marcos Fernandes Calixto Rios, bolsista de iniciação científica do Projeto Temático, que sob nossa orientação está desenvolvendo pesquisa específica sobre a Rua 15 de Novembro, de 1809 a 1954, aprofundando questões referentes às transações imobiliárias e mudanças arquitetônicas e de uso na área central da cidade. Em paralelo, também com o auxílio gráfico do aluno Marcos Rios, preparamos as pranchas focalizando aspectos de algumas ruas e largos da cidade de São Paulo no período, com base na Décima de 1809 e na iconografia disponível, nomeando os proprietários das casas e dando vida ao cenário. Apresentaremos aqui o caso do Largo da Sé, confrontando imagens de Thomas Ender (1817), Edmund Pink (1821-1833), William John Burchell (1827) e Militão Augusto de Azevedo com os dados fornecidos pela Décima. Para tanto, foi essencial o recurso ao Fundo Aguirra3 que possibilitou a checagem da localização dos imóveis. O exame do livro da Décima Urbana de 1809 e o mapeamento das informações – em planimetria e volumetria – nos permitiu reconstituir aspectos do espaço intra-urbano da cidade de São Paulo, bem como esboçar um quadro do seu patrimônio e mercado imobiliário rentista, predominante em todas as cidades coloniais e imperiais brasileiras, ao menos até os anos 70 do século XIX, dando subsídios para comparações futuras com os períodos subseqüentes e casos afins. Essa metodologia piloto será aplicada aos demais registros da Décima Urbana disponíveis sobre São Paulo, 1829 e 18724, e sobre Santos (1814, Anais do Museu Paulista. v. 13. n.1. jan.- jun. 2005. 1835-1836, 1836-1837 e 1837-1838), viabilizando o aprofundamento de questões aqui esboçadas. Algumas constatações Quando comparamos os dados estatísticos relativos ao velho tecido urbano da cidade de São Paulo e seu mercado imobiliário no início do século XIX com os dados referentes à megalópole do século XXI, surpreende-nos a dinâmica do seu processo de urbanização. Hoje falamos em “dispersão” para caracterizar as novas configurações atuais. No entanto, a mancha urbanizada, composta pela Região Metropolitana de São Paulo e seus 39 municípios, correspondia aos limites (o Termo) do município nas suas origens. O que hoje chamamos de Centro Histórico correspondeu, do século XVI ao último quartel do XIX, à area efetivamente urbanizada da cidade. A antiga Freguesia da Sé oscilava em torno de 7.000 habitantes em princípios do Oitocentos e o município em torno de 24.000 habitantes. Hoje, a área municipal abriga um total de 11 milhões de pessoas, e a Região Metropolitana gira em torno dos 17 milhões. Nossa memória não alcança aspectos significativos do espaço intraurbano nas suas origens. Recuamos no tempo, portanto, para caracterizá-lo. São bastante conhecidas as descrições dos viajantes, como as de Saint-Hilaire, que apontam aspectos de São Paulo em princípios do século XIX (1819 e 1822): Eu não saberia dizer qual é o número de casas da cidade de São Paulo, mas Spix e Martius nos informam que, em 1815, quando o distrito de que a cidade era sede ainda compreendia 12 paróquias, havia ali um total de 4142 famílias; segundo Eschwege, haveria em 1820 nesse distrito, já reduzido a 11 paróquias, 4017 famílias, finalmente, de acordo com Daniel Pedro Müller, as 9 paróquias e a sucursal de que se compunha o distrito contariam, em 1839, com um total de 4168. Eschwege, supondo que as cifras relativas a todo o distrito se referem unicamente à cidade de São Paulo, afirma existirem nessa cidade 6 pessoas por casa. [...] Funcionários de todas as categorias, artesãos de variadas profissões, numerosos negociantes, proprietários de casas urbanas e de terras – os quais, ao contrário de Minas Gerais, não moram nas suas fazendas – compõem a população da cidade de São Paulo [...] A cidade fica situada [...] numa elevação [...]. Seu contorno é bastante irregular e de forma ligeiramente alongada, situando-se ela no delta formado pelos córregos Anhangabaú e Tamanduateí, os quais [...] vão desaguar no Tietê” [...] As casas, feitas de taipa e bastante sólidas, são todas caiadas e cobertas de telhas. Nenhuma delas sugere opulência, mas vê-se um grande número de sobrados, que chamam atenção por seu aspecto vistoso e limpo. Os telhados não se projetam muito para fora, apenas o suficiente para darem sombra e protegerem as paredes da chuva, e as janelas não são tão juntas umas das outras, como se vê comumente no Rio de Janeiro. As dos sobrados são quase todas envidraçadas, com postigos pintados de verde e com sacadas. As casas de um só pavimento têm gelosias que se abrem de baixo para cima e são feitas de paus cruzados em diagonal (SAINT-HILAIRE, 1976, p.126 -128). Annals of Museu Paulista. v. 13. n.1. Jan.- Jun. 2005. 65 Se os dados sobre a população de São Paulo são pouco precisos na fala dos viajantes, com base no Recenseamento de 1776 e nas informações fornecidas por Afonso de Freitas, em 1822, sabemos que a cidade passou de 2.026 habitantes e 574 prédios para 6.920 habitantes apenas no distrito da Sé e 24.000 no município (REIS FILHO, 2004, p. 70/80). Os dados arrolados na Décima Urbana de 1809 nos permitem precisar o quadro esboçado acima, recorrentemente invocado pela historiografia. Podemos afirmar que o perímetro urbano da capital paulista se constrangeu de 1554 a c.1870 na estreita colina formada pelo curso dos rios Anhangabaú e Tamaduateí, com pouquíssimas ramificações na várzea, compondo-se de, segundo a Décima de 1809, um total de 34 ruas (Boa Vista, S. Bento, Jogo da Bola, da Bua (sic) , do Comércio, Cadeia, Carmo, Cachoeira, Cemitério, Direita, Esperança, S. Ifigênia, da Freira, das Flores, São Francisco ao Jogo da Bola, S. Gonçalo, Guacû, do Hospital, de S. José, S. João, Lorena, Luz, Ouvidor, Príncipe, Pombal, Piques, Quartéis, Rosário a Boa Vista, Rego, Sé, Sé para S. Teresa, Tabatinguera, S. Teresa, Tanque), 13 travessas (do Bexiga, da Boa Vista, Casinhas, Comércio, Cemitério, do Colégio, S. Ifigênia, Fundição, da Lapa, do Príncipe, do Pombal, dos Quartéis e de S. Teresa), um beco (Beco do Barbas) e quatro largos (Largo da Sé, Largo do Colégio, Largo de São Gonçalo e Largo do Bexiga). Nesse perímetro urbano foram inventariados 1.281 imóveis para fins de tributação. Desses, 81,64% estavam concentrados nas mãos de leigos e 18,27% nas mãos das ordens religiosas, irmandades e padres seculares. São Paulo era uma cidade predominantemente térrea. Dos 1.211 imóveis cadastrados no item tipologia, apenas 161 eram sobrados (159 de um andar; um de dois andares e um de três andares), totalizando 13,29% do conjunto do casario. As 1.047 edificações restantes eram casas térreas, correspondendo a 86,45% do tecido urbano (Figura 1). Entre as casas térreas havia subtipologias, tais como: térreas simplesmente; térrea de uma loja; térrea de um lanço; térrea de dois lanços; térrea de três lanços; térrea de um lanço e uma loja; térrea de dois lanços e uma loja; térrea de uma loja e um corredor; térrea de um lanço e um corredor; térrea de um lanço e um “sotio”; térrea de dois lanços e um mirante. Entre os sobrados também, a saber: sobrado de um andar; sobrado de um lanço e um andar; sobrado de três lanços e um andar; sobrado de uma loja e um andar; sobrado de duas lojas e um andar; sobrado de três lojas e um andar; sobrado de quatro lojas e um andar; sobrado de cinco lojas e um andar; sobrado de seis lojas e um andar; sobrado de uma loja, um lanço e um andar; sobrado de um lanço, três lojas e um andar; sobrado de uma loja, dois corredores e um andar; sobrado de uma loja de primeiro e segundo andares; sobrado de três andares. No conjunto foram identificadas duas casas assobradadas, que supomos tratar-se de sobrado de um lado e térrea do outro, decorrente do desnível do terreno. Cabe explicar que segundo Antenor Nascentes, no seu Dicionário da língua portuguesa, a expressão “lanço” significava “extensão do pano de um muro, de uma parede, de uma fachada”. Com base na iconografia, interpretamos a expressão “casa de um lanço” como uma moradia, térrea ou sobrado, de um único 66 Anais do Museu Paulista. v. 13. n.1. jan.- jun. 2005. Figura 1 – Prancha I – Tipologias. Autora: Beatriz P. Siqueira Bueno. Mapa-base: “Mappa da Cidade de São Paulo offerecido a Sua Magestade, o Imperador pelo Presidente da Província Manoel da Fonseca Lima e Silva (1844-1847)”, do Engenheiro Carlos Bresser, datável de 1844-1847. Seção de Cartografia da Fundação Biblioteca Nacional, Rio de Janeiro. Annals of Museu Paulista. v. 13. n.1. Jan.- Jun. 2005. 67 5. Em caso de imóvel de uso misto ou exclusivamente comercial,com lojas na frente,em lugar das janelas observamos a presença de duas a três portas. 6. A “Junta da Décima da Cidade de São Paulo”,em 1809,era composta pelo Escrivão da Executoria Antonio Xavier Ferreira; pelo Fiscal e Bacharel Manuel Joaquim de Ornellas; pelo Louvado Nobre o Tenente Coronel Francisco Alvares Ferreira do Amaral; pelo Louvado do Povo Lourenço da Silva Barros; pelos Carpinteiros José Ferreira e José Joaquim de Carvalho e pelo Pedreiro Manuel Roiz. Em 15/11/ 1809, o Louvado do Povo foi substituído por Thomé Manuel de Jezus Varella. Em 13/12/1809, o Fiscal e Bacharel Manuel Joaquim de Ornellas foi nomeado Superintendente da Décima, em lugar do Desembargador Miguel Antonio Azevedo da Veiga – Ouvidor Geral Corregedor e Juiz Executor da Real Fazenda. Desde então, o posto de Fiscal da Décima foi encabeçado pelo Bacharel Miguel Carlos Aires de Carvalho, provável proprietário da famosa chácara de mesmo nome,nas imediações da cidade. O lançamento do imposto incidiu sobre todos os prédios urbanos, e o inventário teve início na Rua Direita (Freguesia da Sé), em 10/11/1809, e conclusão na Rua do Tanque (Tanque do Zuniga, Freguesia de S. Ifigênia), em 30/12/1809. A soma do imposto recolhido, em 1809, totalizou 1:302 $241rs (um conto, trezentos e dois mil, duzentos e quarenta e um réis). Segundo Maria Lucília Viveiros de Araújo, 1% dos 68 cômodo frontal, com janela-porta-janela, uma porta e duas janelas ou, simplesmente, porta e janela5. Já uma casa de dois lanços seria composta de dois cômodos frontais e assim sucessivamente, cada lanço resultaria no acréscimo de um cômodo ou extensão de fachada correspondente. O lanço seria portanto o módulo que orientava a atribuição de valor aos imóveis, com base na medição das testadas. Sendo a São Paulo colonial uma cidade predominantemente de taipa de pilão, talvez o lanço equivalesse a dois taipais, totalizando 4,40 m, uma vez que o taipal media uma braça (i.e. 2,20 m). O critério de avaliação dos imóveis urbanos envolvia também outras variáveis, tais como o número de pavimentos. Não por acaso, a medição dos imóveis para fins de taxação era atribuição de dois mestres carpinteiros e um mestre-pedreiro, funcionários da Junta da Décima6. RELATÓRIO GERAL QUANTIDADE PERCENTUAL Freguesias Sé e Santa Ifigênia 1.281 100% Tipos de proprietários Leigos Religiosos Total 1.041 233 1.275 81,64% 18,27% Tipologia Assobradada Sobrado Térrea Terreno Total 2 161 1.047 1 1.211 0,16% 13,29% 86,45% 8,25% Finalidade Auguel Cedido Em obras Fechado Outros Próprio Total 638 44 41 68 15 462 1.269 50,27% 3,46% 3,23% 5,35% 1,18% 36,40% Uso Comercial Misto Residencial Total 26 132 1.051 1.211 2,14% 10,90% 86,78% Fonte: Livro da Décima de 1809. Arquivo do Estado de São Paulo. Quanto à destinação, dos 1.269 imóveis arrolados nesse item, 638 (50,27%) eram de aluguel; 462 (36,40%) de uso próprio; 44 (3,46%) encontravam-se cedidos de favor; 68 (5,35%) estavam fechados; 41 (3,23%) em obras; e 15 (1,18%) não foram taxados por serem considerados “insignificantes”, “arruinados” ou por estarem “em conserto” (Figura 2). Anais do Museu Paulista. v. 13. n.1. jan.- jun. 2005. Figura 2 – Prancha II – Finalidade. Autora: Beatriz P. Siqueira Bueno. Mapa-base: “Mappa da Cidade de São Paulo offerecido a Sua Magestade, o Imperador pelo Presidente da Província Manoel da Fonseca Lima e Silva (1844-1847)”, do Engenheiro Carlos Bresser, datável de 1844-1847. Seção de Cartografia da Fundação Biblioteca Nacional, Rio de Janeiro. Annals of Museu Paulista. v. 13. n.1. Jan.- Jun. 2005. 69 10% do imposto era embolsado pelo Coronel Luiz Antonio de Souza, encarregado de supervisionar a sua cobrança em São Paulo. Muitos imóveis de aluguel apresentavam vários inquilinos, por exemplo quando se tratava de uma casa térrea de vários lanços (um inquilino por lanço) ou sobrados de uso misto com mais de uma loja. Esses dados contrariam a visão corrente da historiografia que considera o investimento em casas de aluguel uma tônica apenas do último quartel século XIX. Segundo Raquel Rolnik: A construção de salas e casas para alugar cresceu no começo da década de 1870, ainda que muito antes, junto com os grandes sobrados, já marcassem a paisagem do Triângulo. Em 1822, o viajante Auguste de Saint-Hilaire comentava sobre a existência de casas e salas para alugar: pequenas construções muito baixas de barro e paredes de sarrafo, com tetos cadentes, chãos de pisos sujos nos trechos mais pobres das ruas São Francisco, Rosário ou Boa Vista. Muitas ficavam ao lado dos grandes sobrados; é o caso da esquina da rua do Rosário com a Travessa do Colégio e da Senador Vergueiro com a rua Direita, no coração do triângulo central. Evidentemente a dimensão desse tipo de investimento era bastante restrita, considerando sobretudo o número irrisório de consumidores potenciais – assalariados não proprietários – em um contexto onde imperava o trabalho escravo. Estes, desprovidos de recursos para adquirir um abrigo próprio, mesmo sem vínculos compulsórios com as casas senhoriais, moravam muitas vezes “de favor”, ou sem pagar nada em casas ou cubículos de propriedade das famílias abastadas. Maria Odila Dias relata grande número de casos em que ex-escravos herdaram pequenas casas de morar de seus senhores, ou de homens e mulheres livres e pobres vinculados às casas senhoriais por redes de compadrio ou serviço que habitavam não no interior das casas grandes, mas em casinhas cedidas em suas proximidades” (2003, p. 102). Subestimando o número de casas de aluguel no tecido urbano, Raquel Rolnik insinua que muitos viviam de favor. Ao contrário do que disse a autora, contabilizamos apenas 3% de moradias cedidas nessa condição em oposição a 50% de casas alugadas. Observamos que as verificações de Maria Odila da Silva Dias são verdadeiras, já que boa parte das casas cedidas de favor o eram a parentes próximos ou ex-escravos. No entanto, ao contrário do que afirma Rolnik, numa sociedade escravista, havia sim setores médios de não proprietários capazes de alimentar um mercado locatício bastante intenso. As tabelas seguintes, referentes à distribuição da população por grupos ocupacionais, em 1818, e aos profissionais recenseados na cidade de São Paulo, em 1593, 1793, 1836, podem nos dar uma idéia das camadas médias existentes no cenário urbano da cidade de São Paulo nesse período. 70 Anais do Museu Paulista. v. 13. n.1. jan.- jun. 2005. Cidade de São Paulo: distribuição da população por grupos ocupacionais em 1818 Grupos ocupacionais Corpo militar Magistrado Números 566 1 Clero secular 81 Clero regular 14 Religiosos recolhidos 56 Agricultores 1.640 Negociantes 220 Artistas 277 Jornaleiros 98 Condutores 52 Madeireiros 18 Mineiros Mendigos1 Total 1 152 3.176 Fonte: MORSE, R. Formação histórica de São Paulo. Apud MEMÓRIA URBANA, 2001. (1) Incluindo mendigos mas não escravos. Annals of Museu Paulista. v. 13. n.1. Jan.- Jun. 2005. 71 Cidade de São Paulo: profissionais recenseados Atividades profissionais 1593 1793 1836 Carpintaria 2 10 77 Sapataria (oficina) 1 16 69 Ferraria 1 4 54 Alfaiataria 2 21 49 Olaria 1 4 38 Ourivesaria – – 30 Marcenaria – – 28 Selaria – 4 13 Pintura – – 10 Lataria – – 7 Construção – 4 7 Padaria – – 7 Botica – – 7 Const. de violas – – 6 Barbearia – 6 6 Entalhe – – 3 Tecelagem 2 – 3 Tanoaria – – 2 Serraria – – 2 Caldeiraria – – 2 Chapelaria (oficina) – – 2 Fábrica de foguetes – – 2 Sirga – – 2 Relojoaria – – 1 Total 9 69 427 Fonte: Aspectos da Metrópole Paulista. In: AZEVEDO, A. (org.). A cidade de São Paulo – estudos de geografia urbana. Apud MEMÓRIA URBANA, 2001. 72 Anais do Museu Paulista. v. 13. n.1. jan.- jun. 2005. Alguns outros dados saltam à vista quando analisamos o Livro da Décima Urbana de 1809. Por exemplo, o valor total atribuído ao imóvel número 1 da Rua Direita – 71$440rs – é praticamente igual ao do aluguel anual do imóvel vizinho – 50$000rs –, tipologicamente idêntico (um sobrado de uma loja e um primeiro andar): Freguezia da Sé – Rua Direita Lado Esquerdo No. 1 Propriedade de cazas do Guarda-mor Vicente Ferreira de Oliveira, que consta de huma logea, e hum primeiro andar, q foi avaliada em setenta e hum mil quatrocentos e quarenta, de q abatidos dez por cento vem para a Décima sinco mil quinhentos e vinte nove reis, com que sahe 5$529. No. 2 Propriedade do Reverendo Bartholomeu Pereira Mendes, que consta de huma logea e hum primeiro andar, e da qual he Inquilino o Capitão Antonio Jozé Brito, que jurou ter alugado annualmente pela quantia de cincoenta mil reis, de que vem para a décima quatro mil e quinhentos reis com que sae 4$500. Nesse aspecto, provavelmente era o potencial locatício anual o orientador do valor atribuído ao imóvel, em caso de venda, num mercado imobiliário de base rentista como esse. Quanto aos usos, 86,78% do tecido urbano eram compostos de imóveis residenciais; 2,14% exclusivamente comerciais e 10,90% de uso misto (Figura 3). Ou seja, de um total de 1.211 imóveis cadastrados nessa categoria, 1.051 eram residenciais, 26 comerciais e 132 de uso misto. São Paulo, em 1809, era portanto uma cidade predominantemente térrea, residencial e com boa parte dos seus habitantes vivendo em casas alugadas. Tratava-se de uma cidade concentrada e com espaços pouco especializados, na qual as principais funções urbanas – residência, comércio, serviços, instituições civis e religiosas – mesclavam-se numa mesma área. Também era uma cidade pouco complexa na medida em que seu tecido urbano se definia a partir da relação entre edifícios privados (casas, lojas, igrejas de irmandades laicas e ordens terceiras – restritas aos seus membros –, conventos, mosteiro) e públicos oficiais (Palácio dos Governadores, Casa de Câmara e Cadeia, Quartel das Tropas de Militares, Casa de Fundição, etc.), em meio a espaços públicos de uso coletivo como ruas e largos. Em termos de infra-estrutura, os esforços da administração pública concentravam-se nas pontes que davam acesso à colina, no calçamento de certas vias e largos e no abastecimento de água por meio de fontes e chafarizes. Embora constrangida na colina histórica, com espaços pouco especializados, públicos de uso coletivo ou privados, em 1809, apresentava áreas mais e menos valorizadas em termos imobiliários. Espécie de esquema centrípeto, as zonas mais caras eram aquelas junto aos Largos da Sé, Largo do Annals of Museu Paulista. v. 13. n.1. Jan.- Jun. 2005. 73 Figura 3 – Prancha III – Usos. Autora: Beatriz P. Siqueira Bueno. Mapa-base: “Mappa da Cidade de São Paulo offerecido a Sua Magestade, o Imperador pelo Presidente da Província Manoel da Fonseca Lima e Silva (1844-1847)”, do Engenheiro Carlos Bresser, datável de 1844-1847. Seção de Cartografia da Fundação Biblioteca Nacional, Rio de Janeiro. 74 Anais do Museu Paulista. v. 13. n.1. jan.- jun. 2005. Palácio (atual Pátio do Colégio) e ruas de uso misto a eles contíguas, concentrando o comércio da cidade. Segundo listagem fornecida pelo banco de dados, os imóveis mais caros encontravam-se situados nas Ruas do Carmo, do Comércio, na Travessa das Casinhas, Rua do Rosário, Rua Direita, Rua de S. Bento, Rua do Ouvidor, Rua da Boa Vista e Rua de S. Teresa (Figura 4). Em contrapartida, os imóveis mais baratos ficavam junto das várzeas ou além rios. Na Rua do Piques oscilavam entre 1$440rs e 1$200rs; na Rua de Santa Ifigênia também; na Rua da Luz 1$200rs; nas Ruas do Rego, da Cachoeira e do Tanque (junto ao Tanque do Zuniga, na Freguesia de S. Ifigênia) variavam de $120rs a 1$000rs. Figura 4 – A cidade de São Paulo vista da várzea do Carmo – parte Norte. Arnaud Julien Pallière, 1821. Coleção Beatriz e Mário Pimenta Camargo. Reprodução fotográfica a partir de REIS FILHO, 2004, p. 92-94. Observar os quintais murados e as fachadas posteriores dos sobrados da antiga R. de S. Teresa (atual Roberto Simonsen). À direita, destacam-se as arcadas do futuro sobrado da Marquesa dos Santos, em 1809, ainda pertencente ao Coronel Joaquim Jozé Pinto de Moraes. Annals of Museu Paulista. v. 13. n.1. Jan.- Jun. 2005. 75 Os dez imóveis mais caros de São Paulo em 1809 Logradouro Rua do Carmo, 49 Rua do Comércio, 25 Travessa das Casinhas, 8 Rua do Rosário a Boa Vista, 13 Rua do Comércio, 7 Travessa das Casinhas, 2 Rua Direita, 1 Rua de São Bento, 85 Rua do Ouvidor, 37 Rua Direita, 6 Rua de São Bento, 28 Rua Boa Vista, 37 Rua Direita, 2 Imóvel Valor (Réis) Sobrado de dois lanços e um andar – propriedade do testamenteiro de D. Francisca de Mattos 102$400 Sobrado de cinco lojas e um andar – propriedade de D. Anna de Almeida 96$480 Sobrado de duas lojas e um andar – propriedade de Jozé Pinto Tavares 71$400 Sobrado de cinco lojas e um andar – propriedade de Joaquim Jozé da Silva 65$280 Sobrado de quatro lojas e um andar – propriedade de Luiz Gonzaga de Araújo 63$900 Sobrado de três lojas e um andar – propriedade de Jozé Amaro de Camargo 63$360 Sobrado de uma loja e um andar – propriedade do Guarda-mor Vicente Ferreira de Oliveira 61$440 Sobrado de um lanço, três lojas e 1 andar – propriedade de Felisberto Pedrozo Siqueira 60$000 Sobrado de dois lanços e um andar – propriedade de Antonio Jozé Barboza – Inquilino:Câmara) 57$600 Sobrado de uma loja e um andar – propriedade do Mosteiro de São Bento) 57$600 Sobrado de seis lojas e um andar – propriedade do Reverendo Manuel Francisco de Andrade 51$920 Casa térrea de dois lanços – propriedade de João Soares de Figueiredo Cardozo Barbas 51$200 Sobrado de uma loja e um andar – propriedade do Reverendo Bartholomeu Pereira Mendes 50$000 Fonte: Livro da Décima de 1809. Arquivo do Estado de São Paulo. A casa mais valorizada, em 1809, era 853 vezes mais cara que a menos valorizada; a primeira situando-se numa das principais ruas residenciais – a Rua do Carmo –, junto da Rua de S. Tereza (atual Roberto Simonsen); a outra nas bordas do perímetro urbano. Seria o equivalente a compararmos a residência de Josef Safra, aproximadamente avaliada em torno de R$ 30 milhões, com a 76 Anais do Museu Paulista. v. 13. n.1. jan.- jun. 2005. casa de um mutirão, avaliada em torno de R$ 10 mil; hoje a diferença oscilaria em 3.000 vezes. Nas devidas proporções, com dinâmicas urbanas extremamente diversas, a cidade de 1809 e a de 2005 parecem esboçar uma tônica recorrente da história da urbanização brasileira, a da segregação social nos espaços urbanos. Malgrado pouco especializado, o espaço intra-urbano da cidade colonial, embora aparentemente mais democrático, configurava no seu tecido uma nítida segregação social e espacial dos menos favorecidos. Havia uma concentração de sobrados nas Ruas Direita, do Ouvidor, do Comércio, do Rosário, de São Bento, de Santa Tereza, do Carmo, no Largo da Sé e no Largo do Colégio, tratando-se da área nobre da cidade. À exceção das Ruas do Carmo e de Santa Tereza, predominantemente residenciais, as demais supracitadas caracterizavam-se por um conjunto significativo de imóveis de uso misto ou exclusivamente comerciais. Nas extremidades, distribuía-se um casario térreo e residencial (Figuras 1 e 3). Ao contrário da concentração verificada quanto às tipologias e usos, observamos que se misturavam no tecido urbano os imóveis de uso próprio e aqueles de aluguel (Figura 2). De um total de 1.281 imóveis cadastrados: • 1.041 eram propriedades de leigos; • 233 distribuíam-se nas mãos das ordens religiosas; • 41 pertenciam às irmandades; • 192 aos padres seculares. Dos 748 proprietários registrados, o Mosteiro de São Bento possuía 61 imóveis; o Dr. Antonio Soares Calheiros, 24; o Convento do Carmo, 22; o Coronel Jozé Arouche de Toledo, 18; o Recolhimento de Santa Tereza, 15; a Irmandade de São Gonçalo, 14; o Capitão Manuel da Luz Tralhão e o Capitão Antonio Alvarez dos Reis, 13; o Capitão Gabriel Jozé Roiz ,11; a Câmara de São Paulo, D. Mariana Fortes e Jozé Antonio da Silva Paulista, 10; a Irmandade do Rosário dos Pretos, nove; o Coronel Anastácio de Freitas Trancozo, oito; o Coronel Luiz Antonio de Souza, o Coronel Jozé Vaz de Carvalho, Dionizio Ereopagita e o Reverendo Ignácio Francisco de Moraes, sete. Dos demais proprietários, seis possuíam seis imóveis; dez dispunham de cinco imóveis; 19 detinham quatro imóveis; 34 eram proprietários de três imóveis e 86 de até dois imóveis. Portanto, dos 748 proprietários arrolados, apenas 24 detinham cerca de 1/4 (302 imóveis) dos 1.281 inventariados, configurando uma enorme concentração de prédios urbanos nas mãos de poucos; algo muito semelhante aos dias atuais. Um total de 173 proprietários detinham mais de dois imóveis e 575 apenas um imóvel. Nem sempre quantidade significava qualidade. Os 20 proprietários com mais capital investido em imóveis na cidade de São Paulo, em 1809, nem sempre eram os detentores da maior quantidade de unidades. A localização mais central, junto às ruas comerciais ou à Sé, garantia melhores preços aos imóveis (Figura 5). Annals of Museu Paulista. v. 13. n.1. Jan.- Jun. 2005. 77 Figura 5 – Espacialização dos imóveis pertencentes aos 20 proprietários detentores de maior patrimônio imobiliário urbano em São Paulo, em 1809. Autora: Beatriz P. Siqueira Bueno. Mapa-base: “Mappa da Cidade de São Paulo offerecido a Sua Magestade, o Imperador pelo Presidente da Província Manoel da Fonseca Lima e Silva (1844-1847)”, do Engenheiro Carlos Bresser, datável de 1844-1847. Seção de Cartografia da Fundação Biblioteca Nacional, Rio de Janeiro. 78 Anais do Museu Paulista. v. 13. n.1. jan.- jun. 2005. Os 20 proprietários detentores de maior patrimônio imobiliário urbano em São Paulo em 1809 (percentual de propriedades sobre o total de imóveis da cidade) PROPRIETÁRIO QUANTIDADE PERCENTUAL VALOR (RÉIS) Mosteiro de São Bento 61 4,77% 721$600 Convento do Carmo 22 1,72% 355$360 Câmara de São Paulo 10 0,78% 240$000 Recolhimento de Santa Thereza 15 1,17% 226$480 Coronel-inspetor José Arouche de Toledo 18 1,41% 224$240 7 0,55% 184$542 José Antonio da Siva Paulista 10 0,78% 173$920 Capitão Manuel da Luz Tralhão 13 1,02% 172$080 D. Anna de Almeida 4 0,31% 163$200 Coronel Joze Vaz de Carvalho 7 0,54% 159$200 24 1,88% 145$920 D. Josefa Maria do Esprito Santo 4 0,32% 127$560 Joaquim Jozé da Silva 3 0,23% 126$720 Capitão João Lopes França 5 0,39% 125$600 11 0,86% 117$240 Irmandade do Rozario dos Pretos 9 0,64% 115$280 Tenente Coronel Anastácio de Freitas Trancozo 8 0,63% 102$880 11 0,86% 93$440 D. Ursula Maria Luiza das Virgens 6 0,47% 78$400 Coronel Joaquim Jozé dos Santos 5 0,39% 77$599 Coronel Luiz Antonio de Souza Dr. Antonio Soares Calheiros Capitão Gabriel José Roiz D. Marianna Fortes (pai do futuro Barão de Itapetininga) Annals of Museu Paulista. v. 13. n.1. Jan.- Jun. 2005. 79 Listagem dos principais proprietários urbanos (percentual de propriedades sobre o total de imóveis da cidade) PROPRIETARIO QUANTIDADE PERCENTUAL VALOR (RÉIS) LEIGOS Câmara de São Paulo 10 0,78% 240$000 Coronel/Inspetctor Jozé Arouche de Toledo 18 1,41% 224$240 7 0,55% 184$542 Jozé Antonio da Silva Paulista 10 0,78% 173$920 Capitão Manuel da Luz Tralhão 13 1,02% 172$080 4 0,31% 163$200 24 1,88% 145$920 Joaquim Jozé da Silva 3 0,23% 126$720 Capitão João Lopes França 5 0,39% 125$600 Capitão Gabriel Jozé Roiz 11 0,86% 117$240 Tenente-Coronel Anastácio de Freitas Trancozo 8 0,63% 102$880 11 0,86% 93$440 Coronel Francisco Xavier dos Santos 6 0,47% 92$160 D. Josefa Maria do Espirito Santo 2 0,16% 89$160 Coronel Joze Vaz de Carvalho 3 0,23% 83$360 D. Ursula Maria Luiza das Virgens 6 0,47% 78$400 Coronel Joaquim Jozé dos Santos 5 0,39% 77$599 Coronel Jozé Vaz de Carvalho 4 0,31% 75$840 Manuel Lopes Coimbra 3 0,23% 73$600 Joaquim Jozé de Oliveira 6 0,47% 72$960 13 1,02% 69$430 Dionizio Ereopagita 7 0,55% 69$200 Clara de Souza 6 0,47% 66$540 D. Anna Leoniza de Abelho Fortes 6 0,47% 66$300 Luiz Gonzaga de Araújo 1 0,08% 63$900 Jozé Amaro de Camargo 1 0,08% 63$360 Coronel-Engenheiro João da Costa Ferreira 2 0,16% 62$400 Coronel Luiz Antonio de Souza D. Anna de Almeida Dr. Antonio Soares Calheiros D. Marianna Fortes Capitão Antonio Alvarez dos Reis 80 Anais do Museu Paulista. v. 13. n.1. jan.- jun. 2005. Guarda-Mor Vicente Ferreira de Oliveira 1 0,08% 61$440 Felisberto Pedroso Siqueira 1 0,08% 60$000 Herdeiros do Coronel Joaquim Manuel da Silva 4 0,31% 59$520 D. Caetana de Toledo 4 0,31% 55$320 D. Caetana de Toledo e suas irmãs 3 0,23% 53$960 Dr. Miguel Carlos (Novo Fiscal da Décima) 2 0,16% 30$720 Dr. Nicolao Pereira de Campos Vergueiro 1 0,08% 30$720 Jaime da Silva Telles 4 0,31% 28$160 Mosteiro de São Bento 61 4,77% 721$600 Convento do Carmo 22 1,72% 355$360 Recolhimento de Santa Thereza 15 1,17% 226$480 Irmandade do Rozario dos Pretos 8 0,63% 105$680 Irmandade de São Pedro 3 0,23% 82$880 Reverendo Ignácio Francisco de Moraes 7 0,55% 79$360 14 1,09% 68$360 Reverendo Bartholomeu Pereira Mendes 3 0,23% 65$360 Reverendo Manuel Francisco de Andrade 2 0,16% 61$520 Irmandade das Almas 3 0,23% 56$320 Irmandade do Santíssimo Sacramento 4 0,31% 47$040 Irmandade do Santíssimo Sacramento - Vila de Santos 1 0,08% 31$000 Irmandade do Sr. Jezus de Nazareth 1 0,08% 25$200 Irmandade do Rozario dos Brancos 3 0,23% 21$120 Irmandade dos Remédios 1 0,08% 20$480 Mosteiro de São Bento de Parnahiba 3 0,23% 19$800 Mosteiro de São Bento de Sorocaba 1 0,08% 12$000 Irmandade dos Passos do Carmo 1 0,08% 8$000 Irmandade de Santa Efigênia 1 0,08% 7$680 Ordem Terceira do Carmo 1 0,08% 3$840 Ordem Terceira de São Francisco 1 0,08% 2$400 RELIGIOSOS Irmandade de São Gonçalo Fonte: Livro da Décima de 1809. Arquivo do Estado de São Paulo. Annals of Museu Paulista. v. 13. n.1. Jan.- Jun. 2005. 81 Assim como no Rio de Janeiro, os beneditinos ocupavam posição de destaque, dispondo de significativo patrimônio imobiliário urbano, na sua maioria casas térreas de um lanço, construídas para renda de aluguel, que em média eram alugadas por 12$000rs, chegando a atingir a cifra dos 19$000rs nas melhores localizações. Também se destacaram como empresários urbanos no período colonial, construindo conjuntos de casas de aluguel na área envoltória aos mosteiros, envolvendo inclusive projetos aprovados pelos superiores hierárquicos da Ordem, sediada no Mosteiro de Tibães, em Braga. Em São Paulo, em 1787, foi construído um conjunto de casas de aluguel pelos monges no trecho inicial da atual Rua Florêncio de Abreu, indicado na época como “Rua Nova de São Bento, chamada Rua da Alegria” (REIS FILHO, 2004, p. 78-79). O projeto, hoje, encontra-se localizado no Arquivo Distrital de Braga (Figura 6). Sabemos que as casinhas de aluguel foram construídas, pois figuram na imagem de Pallière, referente a São Paulo, em 1821. Na lateral do mosteiro, observase o portão de acesso à várzea do Tamanduateí e os quintais murados de algumas das casas (Figura 7). No Rio de Janeiro, os beneditinos foram responsáveis por conjuntos de moradias de aluguel de natureza semelhante. Na também recém-aberta Rua Nova de São Bento (atual Rua do Quartel de Bragança, hoje, Conselheiro Figura 6 – Projeto de casas de aluguel construídas pelos beneditinos nas imediações do Mosteiro. Autor: desconhecido, data: 1787. Arquivo Distrital de Braga/ Universidade do Minho – Portugal. Reprodução fotográfica a partir de REIS FILHO, 2004, p. 78. 82 Anais do Museu Paulista. v. 13. n.1. jan.- jun. 2005. Figura 7 – A cidade de São Paulo vista da várzea do Carmo – parte Norte. Arnaud Julien Pallière, 1821. Coleção Beatriz e Mário Pimenta Camargo. Observar, à esquerda e à direita do portão de acesso à várzea, os fundos das casas de aluguel dos beneditinos, projetadas em 1787. Saraiva), os beneditinos construíram um conjunto de casas de aluguel, de ambos os lados, contratando para tanto os mestres carpinteiros Bento Coelho e Francisco Rabelo de Almeida, que seguiram o “risco” assinado pelos monges. Como menciona Nireu Cavalcanti, tratava-se de um comitente qualificado, espécie de empresário urbano do período colonial, com longa experiência em construção, possuidor de pedreiras e olarias que abasteceriam o canteiro de obras. Tratavase também de uma obra complexa, abrangendo a abertura de um logradouro e construção, em ambos os lados, de prédios térreos e sobrados, num período previsto de quatro anos, cabendo aos mestres carpinteiros a edificação de “todas as moradas de casas”. Sem dúvida, para a época, correspondia a um ousado empreendimento imobiliário realizado por uma instituição particular, com recursos próprios, voltado para o mercado locatício. Os prédios e a rua, não por acaso alcunhada de Nova de São Bento, foram projetados portanto pelos próprios monges beneditinos. A rua tinha 928 palmos de comprimento (204 m) e 30 palmos de largura (6,6 m). À exceção dos prédios de esquina que não faziam parte do contrato, todos os outros foram construídos com tipologias idênticas, de acordo com o “risco”. Segundo menciona Nireu Cavalcanti, com base nas especificações do contrato, cada unidade apresentava um programa e uma planta-tipo – no pavimento térreo “uma sala, uma camarinha, armazém e varanda”, com “as portas que lhe forem necessárias para os cômodos”; no fundo do quintal, “encostado ao muro, sua cozinha e chaminé e despensa, com um cano para saírem as águas para uma e outra parte da cerca do Mosteiro”. Sobre parte do térreo, a partir da sala, levantava-se um sótão, chamado no Annals of Museu Paulista. v. 13. n.1. Jan.- Jun. 2005. 83 contrato de “sobrado”, dividido em sala, dotado de duas janelas abertas sobre a varanda e dois quartos “fechados com portas para a sala”. No fundo de cada prédio deveria ser construído um muro de 15 palmos (3,3 m) de altura, delimitando os quintais com os terrenos da horta e a sede do mosteiro. Também foram especificados os materiais e processos construtivos, nos mínimos detalhes. Longe de sermos anacrônicos, nas devidas proporções, tratava-se de um negócio imobiliário destinado ao mercado locatício, com projeto detalhado, especificações técnicas, empreiteiros especializados, nos moldes dos modernos empreendimentos dos nossos dias (CAVALCANTI, 2004, p. 343-344). No que diz respeito às irmandades laicas, surpreende o fato de a Ordem Terceira do Carmo e da Ordem Terceira de São Francisco disporem apenas de um imóvel cada, em São Paulo, alugados por 3$840rs e 2$400rs anuais, respectivamente, quando os estudos de Fania Fridman as apontavam entre os mais ricos proprietários do Rio de Janeiro e ativos empreendedores imobiliários. De qualquer forma, as ordens religiosas – beneditinos, carmelitas e Recolhimento de Santa Teresa – e a Irmandade do Rosário dos Pretos figuravam entre os mais ricos proprietários urbanos de São Paulo em 1809. As demais Irmandades – das Almas (três prédios = 56$320rs), de Santa Ifigênia (um prédio = 7$680rs), do Rosário dos Brancos (três prédios = 21$120rs), do Santíssimo Sacramento (quatro prédios = 47$040rs), do Senhor Jesus de Nazareth (um prédio = 25$200rs), dos Passos do Carmo (um prédio = 8000rs), dos Remédios (um prédio = 20$480rs) e do Santíssimo Sacramento da Vila de Santos (um prédio 31$000rs), não chegavam a totalizar montantes muito expressivos, à exceção das Irmandades de São Gonçalo (14 prédios = 63$360rs) e de São Pedro (três prédios = 82$880rs). Entre os leigos institucionais, destacava-se exclusivamente a Câmara, senhora de 10 “cazinhas” na famosa travessa de mesmo nome – especializada no abastecimento da cidade –, arrematadas por José Mendes, cada uma por 24$000rs, totalizando um montante de 240$000rs anuais. Entre os particulares laicos destacam-se o Coronel Arouche de Toledo e o Coronel Luiz Antonio de Souza – futuro Brigadeiro Luiz Antonio –, sem dúvida, o homem mais rico de São Paulo e uma das maiores fortunas da Colônia no período (ARAÚJO, 2003). Na análise dos dados fornecidos no Livro da Décima Urbana sobressaem algumas curiosidades como a de que o engenheiro-militar João da Costa Ferreira possuía dois valorizadíssimos imóveis, um sobrado de dois lanços e um andar na Rua de São Gonçalo, n. 34 (atrás da Sé, alugado por 38$400rs anuais) e uma “logea” na Travessa das Casinhas, n. 3, alugada a 24$000rs anuais. Também o célebre mestre pedreiro Joaquim Thebas era possuidor de um imóvel, situado à Rua do Rego, n. 40 (próxima à Ponte do Lorena), que por estar em obras o obrigava a residir de aluguel na casa térrea vizinha, pagando 9$600rs anuais. Outra curiosidade é que o Bispo Diocesano (Mateus de Abreu Pereira) habitava num sobrado de dois lanços e um andar – o mais caro de São Paulo, 84 Anais do Museu Paulista. v. 13. n.1. jan.- jun. 2005. situado na Rua do Carmo, n. 49 –, pagando 102$400rs de aluguel anual ao testamenteiro de D. Francisca Maria de Mattos. Não contente com isso, alugava também idêntico sobrado vizinho, n. 48, pertencente ao Convento do Carmo, pagando mais 40$000rs anuais. Sabemos que também era proprietário da Chácara da Glória, correspondente aos atuais bairros Vila Deodoro e Cambuci. É interessante anotar que os homens mais ricos da cidade, em geral, possuíam chácaras no perímetro urbano, que nessa altura englobava o antigo “rossio”6, tal como nos mostra a Planta da Cidade de São Paulo, levantada em 1810, pelo engenheiro Rufino José Felizardo e Costa (Figura 8) . Esse é o caso do Coronel Luiz Antonio de Souza – proprietário de uma chácara junto da Rua da Consolação (esquina com a atual Avenida S. Luís) – , do Coronel Francisco Xavier dos Santos – vizinho do anterior –, de D. Marianna Fortes, do Coronel Arouche e do Coronel Gavião (futuro Brigadeiro Bernardo José Pinto Gavião Peixoto, pai do Capitão José Maria Gavião Peixoto). Todos parecem desfrutar das suas chácaras por puro conforto, como uma espécie de segunda residência, na medida em que acreditamos ainda não se tratar de um investimento para especulação imobiliária, como o será mais tarde. 6. Rossio = área envoltória às vilas pertencente à Câmara Municipal. Patrimônio público, administrado pelas Câmaras, destinava-se à pastagem de animais, coleta de madeira e lenha, expansão da cidade e obtenção de renda com a concessão de terras.Cf.MARX,1991. Figura 8 – Planta da Cidade de S. Paulo [...] levantada em 1810 pelo Engenheiro Rufino José Felisardo e Costa”. Coleção Museu Paulista – USP. Observar as chácaras existentes na então Freguesia de S. Ifigênia: a) Chácara do Gavião; b) Chácara do Dr. Francisco Xavier (depois legada ao primo, o Barão de Itapetininga); c) Terras do Coronel Luiz Antônio de Souza; d) Chácara do Secretário Velho; e) Chácara de D. Marianna Fortes; f) Chácara do Coronel Arouche. Annals of Museu Paulista. v. 13. n.1. Jan.- Jun. 2005. 85 No entanto, em geral, a maioria desses mesmos personagens dispõe também de moradias no perímetro urbano. A Décima de 1809 nos fornece os endereços residenciais urbanos de alguns deles. Por exemplo, o Coronel-Inspector Jozé Arouche de Toledo possuía um sobrado residencial na R. de São Bento, n. 21 (de dois lanços e um andar) e D. Marianna Fortes possuía um sobrado (de dois lanços e um andar) na Rua de Trás da Sé, n. 6. Já o Coronel Luiz Antonio parece residir na chácara, pois no local do seu futuro sobrado (esquina das Ruas do Ouvidor e São Bento) possuía apenas uma “loja e um lanço”, onde estabelecera a “Casa Souza”, e seus outros seis imóveis estavam alugados. À exceção de D. Ursula Maria Luiza das Virgens (tia de Francisco de Paula Xavier de Toledo) que morava na Travessa do Colégio, n. 10, num sobrado de um andar exclusivamente residencial, os outros grandes detentores de patrimônio imobiliário na cidade de São Paulo, em 1809, apresentavam endereços que atestavam suas atividades predominantes. Jozé Antonio da Silva Paulista residia na Rua do Rosário, n. 46, num sobrado de uma loja e dois lanços; o Capitão Manuel da Luz Tralhão, na Rua do Ouvidor n. 6, num sobrado de uma loja e um andar; o Coronel Joze Vaz de Carvalho, na Rua Direita n. 31, num sobrado de uma loja e um andar; o Dr. Antonio Soares Calheiros, no Largo do Bexiga n. 1, num sobrado de uma loja e um andar; D. Josefa Maria do Espirito Santo, na Rua do Comércio n. 1, num sobrado de duas lojas e um andar; Joaquim Jozé da Silva, na Rua do Rosário n. 13, num sobrado de cinco lojas e um andar; o Capitão João Lopes França, na Rua do Rosário n. 59, num sobrado de uma loja e um andar; o Capitão Gabriel Jozé Roiz, no Largo da Sé n. 4, num sobrado de uma loja e um andar; D. Anna de Almeida, na Rua do Comércio n. 25, num sobrado de cinco lojas e um andar (quatro alugadas e quiçá uma, mais o andar, de uso próprio). A localização e tipologia dessas residências de uso misto, é indicativa das suas atividades como importantes negociantes e comerciantes da cidade. Ao contrário do que afirma Raquel Rolnik (2003, p. 102), a propriedade imobiliária não era irrelevante do ponto de vista da composição da riqueza. Os estudos de Zélia Cardoso de Mello (1985) e Maria Lucília Viveiros de Araújo (2003) – para o caso de São Paulo – e de Fania Fridman (1999), João Luís Fragoso (1998) e Nireu Cavalcanti (2004) – para o caso do Rio de Janeiro – desmentem tais considerações feitas sem base empírica, calcadas, em exemplos genéricos: A propriedade imobiliária até a década de 1870 era muito pouco relevante do ponto de vista da composição da riqueza. Lembremo-nos que quando o barão de Iguape faleceu, em 1875, sua neta Ana Brandina da Silva Prado, casada com Antonio Pereira Pinto Jr. a contragosto da família, foi deserdada e recebeu como herança a casa do avô, velho sobrado de taipa situado nos Quatro Cantos, isto é, na rua de São Bento, esquina com a rua Direita, um dos vértices do Triângulo. Sua irmã Anésia, neta predileta, foi contemplada com uma cômoda!. Um inventário de 1868 também demonstra a insignificância das propriedades 86 Anais do Museu Paulista. v. 13. n.1. jan.- jun. 2005. imobiliárias em relação a outras formas de riqueza: 10 mil metros quadrados de terreno perto da cidade (atual rua dos Guaianazes) – 100$000 réis; um sobrado de taipa de pilão na rua Boa Vista, no coração da cidade – 1000$000 réis, Chácara Pacaembú (incluindo os atuais bairros de Perdizes, Pacaembú e parte da Barra Funda, Lapa e Várzea do Tietê) – 2400$000 réis; escravos que iam de Maria, 60 anos, 40$000 réis, a Faustino, 35 anos, mulato, alfaiate, 600$000 réis; piano 100$000 réis; bacia de cobre 60$800, etc. Um piano valia tanto quanto um terreno de 10 mil metros nos arredores da cidade; um escravo jovem e com ofícios valia quase tanto como um grande sobrado no centro da cidade. Como vimos no capítulo 1, essa situação se alterou no final do século, resultado sobretudo do deslocamento do capital imobilizado no escravo para a terra e da possibilidade aberta pelos estabelecimentos bancários de lastrear empréstimos para lavoura e outros negócios através de hipotecas. Era possível também levantar outras hipóteses de constituição de um mercado imobiliário na cidade: por um lado, a quebra do Banco Mauá, uma das mais sólidas casas bancárias do Império, teria gerado receio entre os capitalistas de guardar dinheiro em estabelecimentos bancários. Por outro lado, Raffard apontava a espetacular conversão de imóveis urbanos em investimentos altamente valorizados dos fazendeiros, antes empregados na construção de ferrovias [...]. Os fazendeiros também temiam a depreciação de suas propriedades agrícolas ou quaisquer outros títulos, em conseqüência da abolição da escravatura e da proclamação da República. De qualquer forma, o crescimento demográfico, a imigração e a presença na cidade de contingentes cada vez maiores de assalariados, artesãos e comerciantes, aliada à disponibilidade de capitais para investimentos, tornavam o mercado de imóveis não só possível como altamente rentável (ROLNIK, 2003, p. 102). Certamente o fenômeno assume outras proporções após a abolição da escravidão e a Proclamação da República, mas não era inexistente no período anterior. A sua natureza é que muda – de “rentista” à “capitalista” –, bem como sua área de abrangência – produzindo a primeira expansão urbana além rios Tamanduateí e Anhangabaú. No que diz respeito ao período 1809 a 1870, o estudo de Maria Lucília Viveiros Araújo (2003) dos inventários post-mortem dos membros das famílias mais ricas da cidade de São Paulo nos permite aferir o peso da propriedade imobiliária (rural e urbana) nas suas fortunas: • Coronel – futuro Brigadeiro – Luiz Antonio de Souza (negociante e fazendeiro português) = 12% imóveis rurais e urbanos; 39% bens profissionais; 29% dívidas ativas; 12% escravos (613). • Coronel Francisco Pinto Ferraz (negociante e fazendeiro português), casado com Ana Francisca Novaes de Magalhães = 23% imóveis Annals of Museu Paulista. v. 13. n.1. Jan.- Jun. 2005. 87 rurais e urbanos; 63% dívidas ativas (empréstimo de dinheiro a juros); 2% escravos. • Francisco Inácio de Souza (português, sobrinho e genro do Brigadeiro Luís Antônio; negociante e fazendeiro) = 40% imóveis rurais e urbanos (26); 39% escravos (229); 5% animais. • Coronel – futuro Brigadeiro – Manoel Rodrigues Jordão (negociante e fazendeiro) = 17 % imóveis rurais e urbanos (17); 28% escravos (281); 27% dívidas ativas. • Coronel Joaquim José dos Santos (negociante de escravos, fazendeiro e pai do Barão de Itapetininga) = 53% imóveis rurais e urbanos (12 imóveis, dentre os quais 10 casas na capital); 35% escravos (87 escravos = 32 urbanos e 45 rurais). • Dr. Rodrigo Antônio Monteiro de Barros (Ouvidor e Desembargador, mineiro; negociante e fazendeiro) = 15% imóveis rurais e urbanos; 43% dívidas ativas; 23% rendas diversas; 10% escravos. • Coronel Anastácio de Freitas Trancoso (negociante e fazendeiro) = 61% imóveis; 6% bens profissionais; 17% escravos (26); 7% animais. • Marechal de Campo Manoel de Oliveira Cardoso (negociante de fazendas secas) = 59% bens imóveis; 34% dívidas ativas. • José Pinto Tavares (negociante português) = 18 propriedades (7 casas na capital). • Capitão Manoel da Luz Tralhão (negociante; pardo, solteiro e natural de Cuiabá) = 35% (15 propriedades urbanas; maior quantidade de casas alugadas na cidade); 2% escravos (5). À exceção dos três últimos, os demais eram todos negociantes e fazendeiros. Ao grupo somava-se José Vaz de Carvalho, sogro do Dr. Francisco José de Sampaio Peixoto, que encabeçava junto do genro o négocio de muares, sendo arrematante do contrato de Curitiba entre 1799 e 1805. São, portanto, homens cuja fortuna advém sobretudo de atividades urbanas – negócios, comércio, empréstimo de dinheiro a juros e renda de aluguel. Os estudos realizados para o caso do Rio de Janeiro, por Fragoso e Cavalcanti, ao menos no que diz respeito aos homens essencialmente urbanos (os ricos comerciantes de “grosso trato”, voltados para o comércio internacional), apontam que tais fortunas se ancoravam em grande parte no patrimônio imobiliário “rentista” urbano. Com base nos inventários post-mortem, Zélia Cardoso de Mello (1985, p. 126) caracterizou o perfil empresarial de alguns dos ricos proprietários de imóveis urbanos integrantes da lista dos 20 mais ricos de São Paulo. Por exemplo, o Capitão João Lopes França possuía além de casas de aluguel em São Paulo, uma chácara, um sítio na Freguesia do Juqueri, 15 escravos e uma sociedade 88 Anais do Museu Paulista. v. 13. n.1. jan.- jun. 2005. com negócio de açúcar, bestas e escravos. Seus rendimentos provinham dos aluguéis, empréstimos e negócios realizados. Também o futuro Barão de Itapetininga (1877), homônimo e filho do Coronel Joaquim Jozé dos Santos (negociante de escravos), figurava entre os “empresários” paulistas mais prósperos da segunda metade do Oitocentos. Embora não apresentasse fortuna de origem propriamente agrária, possuía valorizados bens dessa natureza. Anotado em todos os almanaques do período como “proprietário e capitalista”, fazia jus à qualificação, chegando a possuir 32 casas de aluguel em São Paulo, terrenos e chácaras. Além disso foi proprietário de três fazendas, somando 2.000 alqueires com 398.000 pés de café, casa de máquinas, engenho e mais de 200 escravos. Suas receitas provinham de aluguéis, café e empréstimo de dinheiro a juros. Entre os bens recebidos por herança relacionam-se 14 casas em São Paulo, “terras no Chá” (herdadas do primo Francisco Xavier dos Santos) e escravos. Por sua vez, o Livro da Décima nos traz informações curiosas sobre um certo Dr. Antonio Soares Calheiros, dono do Largo do Bexiga, já que detentor das 12 casas que o compunham, residindo no n. 1, num sobrado de uma loja e um lanço. Possuía um patrimônio de 24 casas, ocupando a 11ª posição na lista dos 20 maiores proprietários de imóveis urbanos. Sobre o Tenente-Coronel Anastácio de Freitas Trancozo, segundo dados fornecidos pelo Livro da Décima, possuía um patrimônio de oito casas, sete delas alugadas, situadas nas ruas do Comércio (um), São Gonçalo (duas), Travessa de Santa Tereza (três) e Rua do Príncipe (duas), totalizando um montante de 102$880rs. Residia na Rua de S. Teresa, num sobrado de dois lanços. Pai de Francisco Pinto do Rego, combinava atividades comerciais à produção agrícola. Seu sítio na Freguesia do Ó, às margens do Rio Tietê, deu origem, mais tarde, ao bairro de Vila Anastácio. Tinha três filhas solteiras que residiam reclusas na Rua do Carmo (MARINS, 1999, p. 221), mantidas pelos cabedais do pai. Em relação às mulheres, as mais ricas em imóveis de renda parecem ser, em geral, as solteiras. Além das filhas do Tenente-Coronel Anastácio, nos referimos às irmãs Toledo Rendon e D. Marianna Angélica Fortes Bustamante Sá Leme. Esta última ganhou luz nos trabalhos de Paulo Garcez Marins (1999, p. 236; 2002), não só como a rica trineta de Fernão Dias Paes, irmã de D. Anna Leoniza de Abilho Fortes, mas sobretudo pelo grande infortúnio sofrido pelo fato de ter tido um filho natural bastardo do Governador da Capitania Bernardo José de Lorena, chegando a receber um “patético atestado da Câmara” ressaltando que, junto da irmã, residia “recolhida” em sua casa “com muita distincção e lei de nobreza”. Rica porém celibatária, D. Marianna Fortes possuía além das 11 casas de aluguel, uma chácara junto da Rua de Santa Ifigênia. Entre os comerciantes, embora não figurando na lista dos 20 mais, destacava-se também o lisboeta Joaquim Jozé de Oliveira, residente na Rua Direita, n. 3, num sobrado de uma loja e um andar, no valor de 38$400rs. Possuía um patrimônio de seis imóveis urbanos (totalizando 72$960rs), seqüestrados em 1810 pela Junta da Cruzada da Corte do Rio de Janeiro (MARINS, 1999, p. Annals of Museu Paulista. v. 13. n.1. Jan.- Jun. 2005. 89 218), dado que suas dívidas excediam seu patrimônio no momento do seu inventário. Quanto ao Coronel Luiz Antonio de Souza, futuro Brigadeiro Luiz Antonio, destacava-se como ilustre fazendeiro no oeste paulista, com múltiplas atividades urbanas. Chefe do prestigioso clã dos Souza Queiroz, seus filhos notabilizaram-se como os futuros Barões de Limeira e de Souza Queiroz. Era casado com D. Genebra de Barros Leite, irmã do 1º Barão de Piracicaba, Antonio Pais de Barros. Detinha cinco imóveis só na Rua do Ouvidor e, embora o de n. 34 correspondesse ao local do seu futuro sobrado (na esquina com a Rua de S. Bento), em 1809, não passava de uma loja e um lanço, destinada ao comércio – "Casa Souza". Como já dissemos, nessa época provavelmente a família residisse na chácara. Também os famosos sobrados do Barão de Iguape e do Brigadeiro Jordão, situados nos Quatro Cantos, famosa esquina das Ruas Direita e S. Bento, ainda não haviam sido construídos em 1809. Nem mesmo Antônio da Silva Prado figurava entre os proprietários de imóveis urbanos em São Paulo nesse período, talvez residindo em Jundiaí, envolvido com o comércio do açúcar, escravos e algodão, estabelecendo-se em São Paulo apenas a partir de 1816 (BRITO, 2000, p. 77-78). Entre os membros dessa prestigiosa família, destacase, em 1809, apenas um dos filhos do Capitão - Mor Martinho da Silva Prado, o Capitão Eleutério da Silva Prado, negociante, residente na Rua de S. Bento n. 22, num sobrado de dois lanços e um andar, e dispondo de dois imóveis que totalizavam 22$080rs. Dos membros da futura elite paulistana do século XIX, figuram no Livro da Décima apenas os nomes de Jaime da Silva Telles, proprietário de quatro casas, mas residente num sobrado alugado, pertencente à Irmandade de São Pedro, no Largo da Sé n. 2. Além dele, destaca-se o nome do Coronel Manuel Rodriguez Jordão – futuro Brigadeiro Jordão e pai do Barão do Rio Claro – ainda não residente no famoso sobrado dos Quatro Cantos da Rua Direita com S. Bento, mas quiçá no seu único imóvel, um sobrado de duas lojas e um andar, sito à Rua de S. Bento, n. 13, avaliado em 24$000rs. Também o Dr. Nicolau Pereira de Campos Vergueiro aparece timidamente como possuidor de um único imóvel, sua provável residência, situada na Rua Direita e avaliada em 30$720rs. O Coronel Francisco Xavier dos Santos aparece como proprietário de seis casas, quatro na Rua de São José (n. 7, 38, 39 e 40) e duas na Rua Direita (n. 26 e 27), totalizando um patrimônio de 92$160rs, além de figurar na planta da cidade de 1810 como o proprietário da chácara do “Chá”. Convém apresentar uma breve comparação com o caso do Rio de Janeiro, no mesmo período. Segundo Fania (FRIDMAN, 1999, p. 47), os beneditinos, entre 1651 e 1750, chegaram a ser proprietários de 37 terrenos e 48 casas de aluguel. Com recursos provenientes dos aluguéis e do gado, a área em torno ao mosteiro foi loteada, sacrificando parte de sua horta. Sua atuação urbana foi ampliada (FRIDMAN, 1999, p. 63-64) e somente no período entre 1743 e 1775 foram erigidas 29 casas na “Rua Nova de São Bento”, como vimos aberta para tanto. De 1751 a 1850, a Ordem acumulou mais de 90 Anais do Museu Paulista. v. 13. n.1. jan.- jun. 2005. 203 casas de aluguel e 29 terrenos foreiros (FRIDMAN, 1999, p. 71). No período colonial, na capital do Brasil, os foros e aluguéis recebidos pelas propriedades no núcleo central da cidade chegavam a superar a renda proveniente dos engenhos, fazendas, denotando o surgimento de uma atividade urbana – a imobiliária – sobretudo nas áreas nobres. Tal como em São Paulo, embora monopolizassem boa parte do tecido urbano da cidade, as ordens religiosas e irmandades não chegaram a ultrapassar, em número, o montante do patrimônio imobiliário laico – sobretudo concentrado nas mãos de homens urbanos, ricos negociantes de “grosso trato” vinculados ao comércio internacional ou negociantes ligados ao comércio regional e local. De longe, no conjunto, os leigos foram os maiores agentes produtores do espaço urbano carioca, ao menos em princípios do século XIX, e outros menos importantes, mas igualmente negociantes e comerciantes, os produtores e detentores de boa parte do tecido urbano de São Paulo. Portanto, embora individualmente as ordens religiosas e irmandades laicas (no caso do Rio de Janeiro) imperassem como os maiores investidores em imóveis urbanos no período colonial, não chegavam a sobrepor os particulares nas somas de conjunto. Implementada por meio do Alvará de 27/6/1808, a Décima Urbana incidiu também nos imóveis da Corte sobre as Freguesias da Sé, Candelária, Santa Rita, parcialmente São José (no trecho que se estendia pelo bairro da Glória e do Catete, indo até a atual Praça José de Alencar, penetrando pelo caminho de Laranjeiras, em direção às Paineiras). Também incluiu pequeno trecho do Engenho Velho, correspondente ao caminho de Mataporcos (atual bairro do Estácio), terminando às margens do Rio Comprido. Esse vasto perímetro urbano foi divido em dois setores – Sé, São José, Engenho Velho; Candelária e Santa Rita –, sendo a área da Freguesia da Candelária a com maior taxa de ocupação e de verticalização. Ao contrário de São Paulo, em que todos os imóveis foram registrados num único livro, a tributação da Décima no caso do Rio de Janeiro resultou em vários volumes. O primeiro aberto em 4/1/1809 (Sé, São José e Engenho Velho) e o terceiro aberto em 12/7/1810, compreendendo as outras freguesias. Entre 1808 e 1810 foram cadastrados 146 logradouros (CAVALCANTI, 2004, p. 259-263), 46,6% catalogados como ruas. As tipologias das edificações variavam entre casas térreas, sobradinhos, sobrados de um, dois ou três andares (com ou sem lojas no térreo). Além dos pavimentos, outros complementos como sótãos e trapeiras também foram registrados. Além das casas, observa-se um maior detalhamento quanto aos outros tipos de imóveis e seus usos: casa de vivenda, loja, sobreloja, armazém, açougue, trapiche, cocheira, senzala, telheiro, casa de banho, pardieiro, corredor, quartos, rótulas, casas com hortas ou quintal, chácaras e terrenos (“chãos”, “terreno devoluto” ou simplesmente “terreno”) (CAVALCANTI, 2004, p. 264), diferentemente do caso paulista, menos detalhado nesse sentido. Tal como em São Paulo, foi estabelecida pela primeira vez uma numeração. Critérios subjetivos orientaram os pontos de partida da numeração, o que hoje dificulta muito o mapeamento dos dados. No Rio de Janeiro, dos 7.548 imóveis arrolados, mais da metade, Annals of Museu Paulista. v. 13. n.1. Jan.- Jun. 2005. 91 4.878, correspondia a edificações de um único pavimento. A capital do Brasil também era uma cidade predominantemente de prédios baixos; 65% horizontal. As edificações térreas predominavam nas extremidades (Engenho Velho, Sé, Santa Rita e São José). A área verticalizada e mais adensada estava concentrada na Freguesia da Candelária, totalizando um conjunto de 858 sobrados, que predominavam sobretudo nas ruas mais importantes, nas proximidades do Largo do Paço, da zona comercial e portuária, no trecho entre a Ponta do Calabouço e o Arsenal da Marinha. A Freguesia da Candelária também abrigava os logradouros e prédios mais importantes – grandes estabelecimentos comerciais dos atacadistas exportadores e importadores. Segundo Nireu (CAVALCANTI, 2004, p. 267), com base no Almanaque de 1794, dos 127 negociantes mais importantes da cidade, 114 tinham seu comércio na Candelária; 67 deles (59%) instalados só na Rua Direita, a principal da cidade. Os imóveis registrados exclusivamente para cocheiras ou de uso misto para tanto – residencial ou comercial – eram pouquíssimos, só 52. Possuir uma cocheira era indicativo de “status” social e somente os comerciantes de “grosso trato” as possuíam em área urbana valorizada; eram os estacionamentos da época. Os prédios não-residenciais eram poucos: oito trapiches, 40 armazéns, duas lojas e 26 telheiros, num total de 76 edificações. Numerosíssimos eram os imóveis de uso misto, 1.456 com lojas e 16 com armazéns. O Almanaque de 1799 atesta a existência de 1.311 lojas de varejo ou oficinas, quantia muito próxima das 1.456 edificações citadas pela Décima. Segundo Nireu, o número reduzido de prédios de uso exclusivamente comercial seria uma conseqüência da não especialização dos espaços urbanos, algo também observável em São Paulo (CAVALCANTI, 2004, p. 271). Foram registrados 26 terrenos – sete deles “devolutos” –, ao passo que em São Paulo, apenas um. A vinda da Corte, induziu o parcelamento das chácaras e uma crescente especulação imobiliária nos arrabaldes da cidade; em São Paulo tal fenômeno é apenas observado a partir de meados do século, intensificando-se com a chegada dos imigrantes. Quanto ao perfil dos proprietários, Nireu contabilizou 2.668 nomes para 7.548 imóveis. Em São Paulo eram 748 nomes para 1.281 imóveis. Do total registrado na Corte, 86,6% pertenciam a pessoas físicas (aí incluídos os padres seculares), 0,4% à Fazenda Real e 12,7% a instituições predominantemente religiosas (ao passo que em São Paulo 81,64% estavam concentrados nas mãos de leigos – 10 imóveis só nas mãos da Câmara – e 18,27% nas mãos das Ordens Religiosas, Irmandades e Padres Seculares). Para o autor, nas somas de conjunto, pouco expressivo era o patrimônio imobiliário concentrado nas mãos das 55 Ordens Religiosas, Irmandades ou Confrarias (Santa Casa de Misericórdia aí inclusa). Pela Décima (1809-1812) verifica-se que esses 55 proprietários institucionais de cunho religioso possuíam 956 imóveis, a saber: Ordem Terceira de São Francisco da Penitência (186); Mosteiro de São Bento (163); Convento do Carmo (125); Santa Casa de Misericórdia (104) (CAVALCANTI, 2004, p. 272-273). No caso de São Paulo a ordem era: Mosteiro de São Bento (61), Convento do Carmo (22), Recolhimento de Santa Tereza (15) e Irmandade do Rosário dos Pretos (9). Como já dissemos, individualmente, 92 Anais do Museu Paulista. v. 13. n.1. jan.- jun. 2005. destacavam-se as ordens religiosas e irmandades; no conjunto destacavam-se os leigos. Dos 7.549 imóveis arrolados, 6.535 estavam nas mãos de 2.585 proprietários particulares. Nota-se uma forte concentração, estando 61,1% de posse de um único imóvel; 17,8% de dois imóveis e 7,6% (ou seja 197 indivíduos) com cinco ou mais imóveis. Maioria esmagadora pertencia ao sexo masculino; 145 eram religiosos (senhores de 273 imóveis) e quatro religiosas (com 12 imóveis). Um conjunto de 89 proprietários (3,4% dos individuais) concentrava 20,7% do total dos imóveis urbanos e deles auferia a vultosa receita anual de 131.863$873rs, equivalente a 42,2% da receita da Alfândega da Capitania do Rio de Janeiro, que em 1806 arrecadara 312632$430rs. Destes, 34 eram negociantes atacadistas. Dos 36 negociantes listados como os mais expressivos nos Almanaques de 1799 e 1811, dez faziam parte da lista dos 89 maiores proprietários de imóveis urbanos do Rio de Janeiro (CAVALCANTI, 2004, p. 274). A mesma concentração nas mãos de poucos, muitos deles negociantes, foi observada em São Paulo. Isso atesta que investimentos no setor imobiliário eram considerados promissores e seguros. A receita sob forma de aluguéis foi a opção de investimento adotada por ricos comerciantes cariocas, tais como os herdeiros de José Caetano de Araújo (41prédios = 8.267$035rs); Cleto Marcelino Ferreira (22 prédios = 5.962$476rs); os herdeiros de Antonio Leite (54 prédios = 4.891$006rs); Manoel Alvares da Fonseca Costa (62 prédios = 4.060$800rs); José Francisco do Amaral (48 prédios = 3.728$160rs); José da Costa Araújo Barros (22 prédios = 3.016$993rs); Domingos Francisco de Araújo Rozo (31 prédios = 2.772$920rs); Antonio José Ribeiro Guimarães (nove prédios = 2.638$240rs); Brás Carneiro Leão (oito prédios = 2.590$400rs); a viúva Francisca Maria da Conceição (52 prédios = 2.501$760rs); Bernardo Francisco de Brito (20 prédios = 2.365$052rs); a viúva Maria Luiza de Souza Dias (nove prédios = 2.318$840rs); Francisco Xavier Pires (20 prédios = 2.302$200rs); Manoel Caetano Pinto (32 prédios = 2.116$542rs) e por fim José Rodrigues de Carvalho (19 prédios = 2.062$200rs). Os montantes dos maiores proprietários paulistas são ínfimos quando comparados aos da Corte; os imóveis e os aluguéis valiam bem mais no Rio de Janeiro; certamente pelo fato de se tratar da Capital do Brasil, com solo urbano mais valorizado, além de dispor de edificações em pedra e cal, mais altas e maiores em área útil. Observam-se 221 edificações em construção, ao passo que em São Paulo se registraram 41 em obras; algo considerável, se compararmos a dinâmica e a demanda (chegada da Família Real e comitiva) entre cidades tão diversas. Dos fogos da Candelária, 86% eram ocupados por inquilinos, significando que o imóvel para aluguel se tornava mais atraente e lucrativo à medida que se aproximava da zona central, do comércio, dos grandes negócios, etc. Nireu calculou que a rentabilidade média era de 6,8%; portanto, mais elevada que os 5% permitidos por lei se a mesma quantia em dinheiro fosse emprestada a juros. No Livro n. 144 do 2º Ofício de Notas – pertencente ao Arquivo Nacional –, no período entre 1807 e 1809, 76 escrituras referentes a transações imobiliárias Annals of Museu Paulista. v. 13. n.1. Jan.- Jun. 2005. 93 de imóveis urbanos foram localizadas por Nireu Cavalcanti, correspondendo a negócios envolvendo 43 prédios térreos, 19 sobrados, cinco chácaras e nove terrenos, movimentando um montante de 95608$000rs – cifra cinco vezes maior que o valor obtido com as exportações da Capitania do Rio de Janeiro no mesmo período (CAVALCANTI, 2004, p. 278-280). Pesquisa semelhante precisa ser realizada para o caso de São Paulo, com intuito de avaliar comparativamente a dinâmica do seu mercado imobiliário. De qualquer forma é possível afirmar que em fins do período colonial era bom negócio empatar capital em casa de aluguel – 1,8% mais rentável que emprestar dinheiro a juro, sobretudo resultando em bem menos riscos. Aspecto pouco contemplado pela historiografia nos seus estudos sobre a cidade colonial brasileira, procuramos enfatizar que boa parte do tecido urbano das cidades brasileiras, desde as suas origens, foi um produto socialmente construído, fruto da iniciativa privada, sendo o solo urbano e o casario passíveis de mercantilização desde tempos muito remotos. Finalizando, apresentamos a prancha com a reconstituição do tecido urbano e cenário do Largo da Sé, de acordo com os dados cadastrados no Livro da Décima, confrontados com as fichas do Fundo Aguirra e a iconografia disponível, incluindo o nome dos proprietários dos imóveis. Trata-se de uma amostragem de um trabalho que envolveu também a reconstituição das ruas Direita, S. Bento, Ouvidor, Comércio, Rosário, Santa Tereza, Carmo e dos largos do Palácio e de S. Gonçalo. É interessante observar que o cenário ganhou vida ao descortinarmos, por detrás das fachadas, os atores sociais envolvidos na produção daqueles espaços (Figura 9). 94 Anais do Museu Paulista. v. 13. n.1. jan.- jun. 2005. Figura 9 – Largo da Sé. Autora da Prancha: Beatriz P. Siqueira Bueno. As imagens dos viajantes foram reproduzidas de LAGO, 2003, p. 37, 69, 92-93. Os desenhos originais pertencem: “Cathedral Kirche zu St. Paul”, Thomas Ender, 1817, Kupferstichkabinett der Akademie der bildenden Künste, Viena; “Cathedral Square, St. Pauls”, 1823, Edmund Pink, Acervo de Artes da BOVESPA, São Paulo; “S. Pedro”, 1827, J. B. Debret, Coleção Beatriz e Mário Pimenta Camargo, São Paulo. As fotografias de Militão Augusto de Azevedo pertencem ao Museu Paulista da USP. Annals of Museu Paulista. v. 13. n.1. Jan.- Jun. 2005. 95 REFERÊNCIAS ARAÚJO, M. L.V. Os caminhos da riqueza dos paulistanos na primeira metade do Oitocentos. Tese (Doutorado) – Departamento de História da Faculdade de Filosofia,Letras e Ciências Humanas da Universidade de São Paulo, 2003. BARBUY, H. M. S. A cidade-exposição: comércio e cosmopolitismo em São Paulo, 1860-1914 (Estudo de História e Cultura Material).Tese (Doutorado) – Faculdade de Arquitetura e Urbanismo, Universidade de São Paulo, 2001. BRITO, M. S. A participação da iniciativa privada na produção do espaço urbano: São Paulo, 1890-1911. Dissertação (Mestrado) – Departamento de Geografia – FFLCH/USP, 2000. CAMPOS, E. Arquitetura paulistana sob o Império. Aspectos da formação da cultura burguesa em São Paulo.Tese (Doutorado) – FAU/USP, 1997. CAMPOS NETO, C. M. Os rumos da cidade. Urbanismo e modernização em São Paulo. São Paulo: Editora Senac, 2002. CAVALCANTI, N. O Rio de Janeiro Setecentista.A vida e a construção da cidade da invasão francesa até a chegada da Corte. Rio de Janeiro: Zahar, 2004. COLLECÇÃO DAS LEIS DO IMPÉRIO DO BRASIL,T. 11, Parte 1ª , Secção 44ª, 1850. DIAS, M. O. L. da S. Quotidiano e poder em São Paulo no século XIX. São Paulo: Basiliense, 1984. FRAGOSO, J. L. Homens de grossa aventura.Acumulação e hierarquia na praça mercantil do Rio de Janeiro (1790-1830). 2ª ed. Rio de Janeiro: Civilização Brasileira, 1998. FRIDMAN, F. Donos do Rio em nome do Rei. Uma história fundiária da cidade do Rio de Janeiro. 2ª ed. Rio de Janeiro: Zahar/Garamond, 1999. GLEZER, R. “Chão de Terra”: um estudo sobre São Paulo colonial. Tese (Livre Docência) – Departamento de História, FFLCH/USP, 1992. HOMEM, M. C. N. O palacete paulistano e outras formas urbanas de morar da elite cafeeeira (1867-1918). São Paulo: Martins Fontes, 1996. LAGO, P. C. do L. Militão Augusto de Azevedo: São Paulo nos anos 1860. Rio de Janeiro: Editora Capivara, 2001. ________. Iconografia paulistana do século XIX. Rio de Janeiro: Editora Capivara, 2003. LEMOS, C. O álbum de Afonso. São Paulo: a reforma de São Paulo. São Paulo: Edições Pinacoteca do Estado, 2001. LÉRIAS, R. A. O Encilhamento e a cidade de São Paulo, 1890-1891. Dissertação (Mestrado) Departamento de História, FFLCH/USP, 1988. 96 Anais do Museu Paulista. v. 13. n.1. jan.- jun. 2005. MARCÍLIO, M. L. A estrutura da população In:A cidade de São Paulo: povoamento e população 1750-1850. São Paulo: Pioneira, 1974, MARINS,P.C.G.Sociabilidades paulistanas.In:Através da rótula.Sociedade e arquitetura urbana no Brasil Sécs. XVII-XX.Tese (Doutorado) – Departamento de História, FFLCH/USP, 1999. p. 150260. ________. Mulheres de elite, filhos naturais – São Paulo, séculos XVIII e XIX. In: FUKUI, L. (org.). Segredos de família. São Paulo: Annablume/ Nemge/ USP/ Fapesp, 2002. MARX, M. Cidade no Brasil, terra de quem? São Paulo: Nobel/Edusp, 1991. MELLO, Z. M. C. de. Metamorfoses da riqueza. São Paulo, 1845-1895. São Paulo: Hucitec, 1985. MEMÓRIA URBANA. A Grande São Paulo até 1940.São Paulo: Arquivo do Estado/Emplasa/Imprensa Oficial, 2001. OLIVEIRA, M. L. F. de. Relações sociais e experiência da urbanização. São Paulo, 1870-1900. Tese (Doutorado) – Departamento de História, FFLCH/USP, 2003. PIRES,W. Configuração territorial, urbanização e patrimônio: Colônia da Glória (1876-1904). Dissertação (Mestrado) – FAU/USP, 2003. REIS FILHO, N. G. São Paulo: vila, cidade e metrópole. São Paulo:Via das Artes, 2004. ROLNIK, R. Mercados: legislação urbana e valorização imobiliária. In: A cidade e a lei. Legislação, política urbana e territórios na cidade de São Paulo. 3ª ed. São Paulo: Fapesp/ Nobel, 2003, p. 101-143. SAINT-HILAIRE,A. Viagem à Província de São Paulo. Belo Horizonte: Itatiaia; São Paulo: Edusp, 1976. SIMÕES JR, J. G. Anhangabaú. História e urbanismo.Tese (Doutorado) – FAU/USP, 1985. TOLEDO, B. L. de. O Triângulo e a configuração de suas ruas e largos. In: Cadernos de fotografia brasileira. São Paulo: Instituto Moreira Salles, 2004. p. 360-380. Artigo apresentado em 03/2005. Aprovado em 06/2005. Annals of Museu Paulista. v. 13. n.1. Jan.- Jun. 2005. 97
https://openalex.org/W1973638930	https://eprint.ncl.ac.uk/fulltext.aspx?url=205884/DD76733E-9A87-4E86-880A-5CA60C4F0BAF.pdf&pub_id=205884	English	null	Socioeconomic position and subjective oral health: findings for the adult population in England, Wales and Northern Ireland	BMC public health	2,014	cc-by	8,945	Newcastle University ePrints Guarnizo-Herreño GC, Watt RG, Fuller E, Steele JG, Shen J, Morris S, Wildman J, Tsakos G. Socioeconomic position and subjective oral health: findings for the adult population in England, Wales and Northern Ireland. BMC Public Health 2014, 14, 827. Newcastle University ePrints Guarnizo-Herreño GC, Watt RG, Fuller E, Steele JG, Shen J, Morris S, Wildman J, Tsakos G. Socioeconomic position and subjective oral health: findings for the adult population in England, Wales and Northern Ireland. BMC Public Health 2014, 14, 827. Copyright: ©2014 Guarnizo-Herreño et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. DOI link to article: http://dx.doi.org/10.1186/1471-2458-14-827 Date deposited: 10-09-2014 This work is licensed under a Creative Commons Attribution 4.0 Unported License ePrints – Newcastle University ePrints http://eprint.ncl.ac.uk Newcastle University ePrints Guarnizo-Herreño GC, Watt RG, Fuller E, Steele JG, Shen J, Morris S, Wildman J, Tsakos G. Socioeconomic position and subjective oral health: findings for the adult population in England, Wales and Northern Ireland. BMC Public Health 2014, 14, 827. Copyright: ©2014 Guarnizo-Herreño et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. DOI link to article: http://dx.doi.org/10.1186/1471-2458-14-827 Date deposited: 10-09-2014 This work is licensed under a Creative Commons Attribution 4.0 Unported License ePrints – Newcastle University ePrints http://eprint.ncl.ac.uk RESEARCH ARTICLE Open Access © 2014 Guarnizo-Herreño et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Abstract Background: The objective of this study was to assess socioeconomic inequalities in subjective measures of oral health in a national sample of adults in England, Wales and Northern Ireland. Methods: We analysed data from the 2009 Adult Dental Health Survey for 8,765 adults aged 21 years and over. We examined inequalities in three oral health measures: self-rated oral health, Oral Health Impact Profile (OHIP-14), and Oral Impacts on Daily Performance (OIDP). Educational attainment, occupational social class and household income were included as socioeconomic position (SEP) indicators. Multivariable logistic regression models were fitted and from the regression coefficients, predictive margins and conditional marginal effects were estimated to compare predicted probabilities of the outcome across different SEP levels. We also assessed the effect of missing data on our results by re-estimating the regression models after imputing missing data. Results: There were significant differences in predicted probabilities of the outcomes by SEP level among dentate, but not among edentate, participants. For example, persons with no qualifications showed a higher predicted probability of reporting bad oral health (9.1 percentage points higher, 95% CI: 6.54, 11.68) compared to those with a degree or equivalent. Similarly, predicted probabilities of bad oral health and oral impacts were significantly higher for participants in lower income quintiles compared to those in the highest income level (p < 0.001). Marginal effects for all outcomes were weaker for occupational social class compared to education or income. Educational and income-related inequalities were larger among young people and non-significant among 65+ year-olds. Using imputed data confirmed the aforementioned results. Conclusions: There were clear socio-economic inequalities in subjective oral health among adults in England, Wales and Northern Ireland with stronger gradients for those at younger ages. ealth, Health inequalities, Adults, Socio-economic factors, Quality of life, Oral health-related quality of life Keywords: Oral health, Health inequalities, Adults, Socio-economic factors, Quality of life, Oral health- * Correspondence: c.guarnizo-herreno.11@ucl.ac.uk 1Department of Epidemiology and Public Health, University College London, 1-19 Torrington Place, WC1E 7HB London, UK Full list of author information is available at the end of the article Socioeconomic position and subjective oral health: findings for the adult population in England, Wales and Northern Ireland Carol C Guarnizo-Herreño1, Richard G Watt1, Elizabeth Fuller2, Jimmy G Steele3, Jing Shen4, Stephen Morris5, John Wildman6 and Georgios Tsakos1 Copyright: Guarnizo-Herreño et al. BMC Public Health 2014, 14:827 http://www.biomedcentral.com/1471-2458/14/827 Background socioeconomic position levels [11-17]. Similarly to in- equalities in general health, the underlying causes of oral health inequalities are related to systematic social disad- vantages and differential access to key resources for health [18-21]. Specifically, studies have indicated that access to material resources, knowledge-related resources and the relative position in the society play a role in the distribu- tion of oral health [22-26]. The association between oral health and socioeconomic position (SEP) has been well established [1-6]. Research has shown consistent inequalities with individuals in lower SEP being more likely to have poorer oral health, as mea- sured by both clinical and subjective indicators [1-13]. Moreover, these socioeconomic inequalities frequently fol- low a gradient with worse oral health at successively lower In the UK, despite a general improvement in adult oral health during the past decades, socioeconomic inequalities in oral health persist [27,28]. Evidence from national oral health surveys has shown consistent inequalities with Guarnizo-Herreño et al. BMC Public Health 2014, 14:827 http://www.biomedcentral.com/1471-2458/14/827 Page 2 of 9 social gradients in different oral health outcomes includ- ing edentulousness, decay experience, periodontal disease, and trauma [1]. Furthermore, income-related inequalities in the number of natural teeth and oral health related quality of life were found in a study using data from the 1998 UK Adult Dental Health Survey [28]. Similarly, an analysis using the English Longitudinal Survey of Aging found significant socioeconomic inequalities in edentu- lousness, self-perceived oral health and oral impacts on daily life for all SEP measures considered (education, in- come, occupational class, wealth, subjective social status, and childhood SEP) [13]. As a response to this clear and consistent evidence, tackling oral health inequalities has become a major goal of the health policy in the UK. In this context, updated information using different SEP indica- tors and measures of oral health is needed to support rele- vant public policy recommendations. probabilistic sample design which provides representa- tive data at national level (England, Wales and Northern Ireland). At every eligible sampled household, all adults aged 16 years and over were invited to participate in an interview and those with at least one natural tooth were also invited to a clinical examination. The survey col- lected interview data from a sample of 11,380 adults, of which 6,469 completed the clinical examination. The overall household response rate was 60% and the indi- vidual response rate within households was 84%. Background We limited our analyses to individuals aged 21 years and older who completed the ADHS interview. Individ- uals aged less than 21 years (n = 591) were not consid- ered in our study due to the very low proportion classified in the highest educational and occupational levels compared to other participants. This presumably reflects the fact that many of the participants aged less than 21 years were still studying and as a result, compar- isons based on current educational attainment and occu- pational social class would not be accurate. The purpose of this study was to assess socioeconomic inequalities in subjective measures of oral health and quality of life in a national sample of adults in England, Wales and Northern Ireland. The ADHS 2009 provides a unique opportunity to study these inequalities since it includes various SEP indicators and a wide range of measures of oral health and quality of life. To our know- ledge, no study has performed an analysis of inequalities in subjective oral health based on the most recent data on adults’ oral health in England, Wales and Northern Ireland. Our study aimed to provide updated informa- tion using different SEP indicators and measures of oral health which is relevant to support public policy recom- mendations. In addition, our analyses used a comprehen- sive analytical approach combining regression modelling and estimation of predictive margins and conditional mar- ginal effects. Our analysis is focused on subjective mea- sures of oral health as we are interested in evaluating how being in different SEP level is related to 1) the general per- ception of oral health status and 2) the functional, social and psychological impacts of oral conditions on the qual- ity of life of people. These subjective measures capture current rather than historic oral health and are associated with general wellbeing, unmet treatment needs and clin- ical outcomes. As such, they are highly relevant for plan- ning and evaluating health services and health promotion interventions as they can influence contemporary deci- sions about oral health and dental care use [29-34]. In addition, we performed complete-case analyses, so only participants with complete information for subject- ive measures of oral health, SEP indicators, and all other covariates were included in the analytical sample. Oral health outcomes Three measures were used as outcome variables: self- rated oral health, Oral Health Impact Profile (OHIP-14), and Oral Impacts on Daily Performance (OIDP). Self- rated oral health is a summary measure that captures multiple dimensions on how people perceive their oral health status [29-32]. For this analysis, based on a 5- item scale, a binary indicator was created combining the answer options of very good/good/fair vs. bad/very bad. This categorization aimed to capture those adults with a clear negative perception about their oral health. The other two outcomes are measures of oral health-related quality of life (OHRQoL) which assess functional, social and psychological effects of oral health [35,36]. Specifically, Background There- fore, from the eligible sample of 10,789 adults aged 21 years and older, we excluded 23 adults with no re- sponses to subjective oral health outcomes, 1,999 with missing information on the SEP indicators, and 2 with missing data on any other covariate. The final sample used for our analyses was 8,765 people. Since most of the participants were excluded from the analyses because they did not have information on income (n = 1,884), we assessed the effect of these missing data on our results. For that purpose, the regression models (described below) were also estimated with data imputed using two ap- proaches, Bayesian multiple imputation techniques, and simple regression techniques and the results were similar to those presented in this paper. Data source and study sample Table 1 Descriptive statistics for study variables, Adult Dental Health Survey 2009 (Based on a study sample of 8,765 individuals) Variables n (weighted %)a Age (years) 21 - 34 1735 (26.65) 35 - 49 2701 (31.30) 50 - 64 2352 (23.66) ≥65 1977 (18.39) Sex Male 3931 (48.44) Female 4834 (51.56) Marital status Single 1901 (27.05) Married/cohabiting 5096 (54.00) Divorced/separated 1087 (11.70) Widowed 681 (7.26) Geographical location (region) North England 2322 (26.17) Midlands England 2366 (27.94) South England (includes London) 2764 (37.46) Wales 751 (5.29) Northern Ireland 562 (3.14) Self-rated general health Very good 3170 (37.00) Good 3764 (43.50) Fair 1363 (14.70) Bad/very bad 468 (4.80) Long standing illness (yes) 2913 (30.56) Educational attainment Degree or equivalent 2137 (26.38) Some educational qualifications 4909 (56.17) No qualifications 1719 (17.44) Occupational social class Managerial and professional 3095 (35.97) Intermediate 1824 (19.98) Routine and manual 3437 (39.13) Other (never worked and long term unemployed) 409 (4.92) Self-rated oral health Very good 2071 (23.48) Good 4112 (46.18) Fair 1895 (22.12) Bad/very bad 687 (8.22) OHIP-14 (Fairly often or very often in at least one item) 1383 (16 04) OHIP-14 measures the frequency of impacts of oral condi- tions on daily life, while OIDP also evaluates the severity of those impacts. Since in this analysis we aimed to identify cases of chronic oral impacts, we derived a binary measure for having one or more responses of ‘very often’ or ‘fairly often’ to any of the OHIP-14 items [37,38]. For the OIDP, we were interested in assessing the presence of severe negative oral impacts and therefore, we created a binary in- dicator for any impact with severity rating of 3 or higher (from a scale 0 to 5). Socioeconomic position measures and other covariates We used three measures of socioeconomic position (SEP): education, occupational social class and equiva- lised household income. Education was measured as the highest qualification attained and was categorized into: de- gree or equivalent, some qualifications, or no qualifica- tions. Occupational social class (current or most recent occupation) was assessed using the UK three-category Na- tional Statistics Socio-Economic Classification scheme (NS-SEC): managerial and professional; intermediate; and routine-manual. These three occupational categories rep- resent broad and well differentiated employment relations and conditions of occupations in modern societies [39]. An additional category of those who never worked and long term unemployed was also considered in our ana- lyses. Data source and study sample We used data from the ADHS 2009, a survey commis- sioned by the NHS Information Centre and conducted by the Office for National Statistics in consortium with the National Centre for Social Research, the Northern Ireland Statistics Research Agency, and experts from UK universities. The survey had a two-stage cluster and Guarnizo-Herreño et al. BMC Public Health 2014, 14:827 http://www.biomedcentral.com/1471-2458/14/827 Guarnizo-Herreño et al. BMC Public Health 2014, 14:827 http://www.biomedcentral.com/1471-2458/14/827 Page 3 of 9 Page 3 of 9 Table 1 Descriptive statistics for study variables, Adult Dental Health Survey 2009 (Based on a study sample of 8,765 individuals) Variables n (weighted %)a Age (years) 21 - 34 1735 (26.65) 35 - 49 2701 (31.30) 50 - 64 2352 (23.66) ≥65 1977 (18.39) Sex Male 3931 (48.44) Female 4834 (51.56) Marital status Single 1901 (27.05) Married/cohabiting 5096 (54.00) Divorced/separated 1087 (11.70) Widowed 681 (7.26) Geographical location (region) North England 2322 (26.17) Midlands England 2366 (27.94) South England (includes London) 2764 (37.46) Wales 751 (5.29) Northern Ireland 562 (3.14) Self-rated general health Very good 3170 (37.00) Good 3764 (43.50) Fair 1363 (14.70) Bad/very bad 468 (4.80) Long standing illness (yes) 2913 (30.56) Educational attainment Degree or equivalent 2137 (26.38) Some educational qualifications 4909 (56.17) No qualifications 1719 (17.44) Occupational social class Managerial and professional 3095 (35.97) Intermediate 1824 (19.98) Routine and manual 3437 (39.13) Other (never worked and long term unemployed) 409 (4.92) Self-rated oral health Very good 2071 (23.48) Good 4112 (46.18) Fair 1895 (22.12) Bad/very bad 687 (8.22) OHIP-14 (Fairly often or very often in at least one item) 1383 (16.04) OIDP (Score of 3 or higher in any item) 1350 (15.62) aFrequencies are weighted but counts are not. Data source and study sample When talking about social gradients, this additional category was not considered as it conceptually does not follow a hierarchical relationship with the other three categories of occupational social class. Total weekly household income was equivalised according to the McClements equivalisation scale and divided into quin- tiles. Age, gender, marital status, geographical location, self-rated general health and long standing illness were included as covariates given the relationship of these characteristics with oral health and SEP. In the models, age was included as a continuous variable, while the other covariates were included using the categories pre- sented in Table 1. Statistical analysis We first estimated the age-standardized prevalence of each subjective measure of oral health by the three SEP indicators. The age-standardization was performed by the direct method, using the UK population (based on the census 2011) as the standard population. Multivari- able regression models were fitted to assess the relation- ships between subjective measures of oral health and SEP indicators. We used binary logistic regression since all oral health outcomes were dichotomous. Models were run for each combination of oral health measure (as dependent variable) and SEP indicator (as categorical explanatory variable) adjusting for age, gender, marital status, geographical location, self-rated general health, Guarnizo-Herreño et al. BMC Public Health 2014, 14:827 http://www.biomedcentral.com/1471-2458/14/827 Page 4 of 9 2) income and OIDP, where the predicted probabilities for the second highest and intermediate quintiles of income were not significantly different from the highest income quintile (Table 2). In addition, persons who never worked or were long term unemployed did not show significantly different probabilities of oral impacts (measured by both OHIP and OIDP) compared to those in managerial and professional occupations (Table 2). Among edentate par- ticipants, analyses indicated non-significant differences in the predicted probabilities of oral health outcomes by all SEP indicators (results not shown). and long-standing illness. From the regression coefficients we then estimated predictive margins and conditional marginal effects to compare predicted probabilities of the outcome across different SEP levels while accounting for all other variables in the model [40,41]. We fitted the models for the total sample and stratified by the presence of natural teeth. Among dentate participants, analyses were further stratified by age groups to evaluate if associa- tions varied by this demographic characteristic. All models took into account the survey design and used the survey sampling probability weights to obtain population-based estimates. All analyses were conducted in Stata 12. Additional analyses stratified by age groups revealed that for self-rated oral health and OHIP the effect of be- ing in the poorest income quintile or in the lowest edu- cational category was larger among younger people (Tables 3 and 4). For the OIDP, results by age groups did not follow exactly the same pattern observed for the other two outcomes, and indicated significant and gener- ally stronger differences by income only for participants aged 35–49 years (Table 5). Discussion We examined socioeconomic inequalities in different subjective oral health measures in a representative sam- ple of adults in England, Wales and Northern Ireland. We found inequalities in the form of social gradients among dentate, but not among edentate participants. This finding is not surprising given the previously de- scribed little variation in oral health among edentate adults and the adaptation of perceptions and expecta- tions usually linked to becoming edentulous [13]. Fur- ther, edentate adults in our study were older, reported poorer general health and were generally more concen- trated in lower SEP categories compared to the whole ana- lytical sample (See Additional file 1: Table S1). Among dentate, important differences in the predicted probability of reporting bad oral health or oral impacts on daily life were observed by educational, occupational and income levels. Between the SEP measures used in this analysis, lar- ger marginal effects were observed for being in lower levels of education and income compared to occupational social class. The effect of being in the poorest income quintile or in the lowest educational category was larger among young people and tended to decrease with age. After adjusting for demographic characteristics, geo- graphical location and general health, analyses showed significant differences in the predicted probabilities of oral health outcomes by SEP level among dentate partic- ipants. For example, the predicted probability of report- ing bad oral health was 9.1 percentage points higher (95% CI: 6.54, 11.68) for participants with no qualifica- tions compared to those with a degree or equivalent and 4.4 percentage points higher (95% CI: 3.30, 5.47) for those with some educational qualifications compared to those with a degree or equivalent (Table 2). Similarly, persons in the lower income quintiles were significantly more likely than those in the highest income level to have reported bad oral health and oral impacts after ad- justment for demographic characteristics, geographical location and general health (p < 0.001). Although lower in magnitude, significant differences were also observed by occupational social class with a higher predicted probability of reporting bad oral health (5 percentage points) for subjects in manual occupations compared to those in the professional/managerial occupational level, and the same was the case for oral impacts. Results We analysed data for 8,765 participants of which 8,171 were dentate and 594 edentate. The sample consisted of 51.6% women, 54% married people, and 17.4% with no educational qualifications. The distributions of study var- iables in the analytical sample are presented in Table 1. In terms of oral health outcomes, more than two thirds of the adults rated their oral health as good or very good, and 16% reported oral impacts on daily life (Table 1). Results of age-standardized prevalence of oral health outcomes by SEP showed consistent and signifi- cant social gradients in subjective oral health by all SEP measures; namely, a higher prevalence of the outcomes at successively lower SEP levels (results not shown). Statistical analysis For all three outcomes, the marginal effect for being in lower income levels tended to be smaller in older age groups and non-significant esti- mates were found among participants aged ≥65 years. Re- sults by occupational social class did not show a clear pattern by age groups for any of the oral health outcomes. Discussion The predicted probabilities of oral health outcomes were significantly dif- ferent from the reference category and gradually increased at lower SEP levels with only two exceptions: 1) income and self-rated oral health (not exactly graded between intermediate and second poorest quintiles of income), and Our results suggest that subjective oral health might be related to diverse social and economic factors that in- fluence people’s position within a society. Our measures of income, education and occupational social class rep- resent different dimensions of SEP and therefore, the Page 5 of 9 Page 5 of 9 Guarnizo-Herreño et al. BMC Public Health 2014, 14:827 http://www.biomedcentral.com/1471-2458/14/827 Guarnizo-Herreño et al. BMC Public Health 2014, 14:827 http://www.biomedcentral.com/1471-2458/14/827 Table 2 Marginal effects (differences in predicted probabilities) of subjective outcomes by SEP level, ADHS 2009, Dentate participants (n = 8,171)a SEP indicator Bad self-rated oral health OHIP-14b OIDPc Percentage points (95% CI) Education Degree or equivalent Reference Reference Reference Some educational qualifications 4.36 (2.95, 5.78)* 5.85 (3.82, 7.87) 5.93 (3.88, 7.97) No qualifications 9.11 (6.54, 11.68) 7.50 (4.30, 10.69) 6.65 (3.54, 9.76) Occupational social class Managerial and professional Reference Reference Reference Intermediate 3.91 (1.94, 5.89) 3.47 (0.76, 6.18)* 2.86 (0.28, 5.44)* Routine and manual 5.06 (3.22, 6.90) 4.64 (2.49, 6.78) 5.14 (2.79, 7.50) Never worked and long term unemployed 5.94 (2.00, 9.87) 2.21 (−3.03, 7.44) 1.38 (−2.81, 5.56) Equivalised Household Income Wealthiest quintile Reference Reference Reference Second wealthiest quintile 2.58 (0.81, 4.35)** 3.26 (0.48, 6.04)* 2.41 (−0.27, 5.09) Intermediate quintile 5.23 (3.32, 7.15) 4.19 (1.38, 7.00) 2.29 (−0.48, 5.06) Second poorest quintile 4.81 (2.82, 6.79) 5.15 (2.35, 7.95) 4.58 (1.45, 7.71) Poorest quintile 7.43 (4.93, 9.92) 8.43 (5.16, 11.71) 7.07 (3.71, 10.42) Asterisks indicate level of significance of the marginal effects compared to the reference category (p < 0.05, p < 0.01). aAll models adjusted for demographic characteristics, geographical location and general health. bFairly often or very often in at least one item. cScore of 3 or higher in any item. g g p g y ( p aAll models adjusted for demographic characteristics, geographical location and general health. Discussion BMC Public Health 2014, 14:827 http://www.biomedcentral.com/1471-2458/14/827 Table 4 Marginal effects (differences in predicted probabilities) of reporting “fairly often or very often” in at least one item in OHIP-14 by age groups and SEP level, ADHS 2009, Dentate participants (n = 8,171)a SEP indicator 21 - 34 years 35 - 49 years 50 - 64 years ≥65 years Percentage points (95% CI) Education Degree or equivalent Reference Reference Reference Reference Some educational qualifications 7.94 (4.47, 11.41)* 6.28 (3.25, 9.30) 3.05 (−1.20, 7.31) 3.14 (−2.52, 8.80) No qualifications 11.32 (4.38, 18.25) 6.92 (0.63, 13.21)* 6.32 (0.76, 11.88)* 5.29 (−0.87, 11.44) Occupational social class Managerial and professional Reference Reference Reference Reference Intermediate 6.18 (0.61, 11.76)* 4.36 (−0.73, 9.45) −1.74 (−5.86, 2.39) 6.87 (1.19, 12.56)* Routine and manual 6.24 (1.80, 10.67) 5.60 (2.01, 9.20) 3.52 (−0.86, 7.91) 3.84 (−0.88, 8.55) Never worked and long term unemployed 4.63 (−3.98, 13.24) 5.18 (−4.25, 14.61) −1.74 (−10.60, 7.12) 2.35 (−5.60, 10.30) Equivalised Household Income Wealthiest quintile Reference Reference Reference Reference Second wealthiest quintile 5.13 (0.32, 9.94)* 2.99 (−1.29, 7.26) 1.72 (−3.78, 7.22) 2.81 (−6.41, 12.02) Intermediate quintile 8.63 (3.60, 13.66)** 6.45 (1.08, 11.82)* −0.29 (−5.55, 4.98) 0.81 (−6.60, 8.23) Second poorest quintile 7.82 (1.97, 13.68)** 6.67 (1.00, 12.34)* 1.14 (−4.77, 7.05) 5.05 (−2.59, 12.69) Poorest quintile 11.69 (5.46, 17.92) 9.85 (4.59, 15.12) 4.82 (−0.91, 10.55) 4.43 (−4.17, 13.03) Asterisks indicate level of significance of the marginal effects compared to the reference category (p < 0.05, p < 0.01). aAll models adjusted for demographic characteristics, geographical location and general health. Table 4 Marginal effects (differences in predicted probabilities) of reporting “fairly often or very o item in OHIP-14 by age groups and SEP level, ADHS 2009, Dentate participants (n = 8,171)a Table 4 Marginal effects (differences in predicted probabilities) of reporting “fairly often or very often” in at least one item in OHIP-14 by age groups and SEP level, ADHS 2009, Dentate participants (n = 8,171)a d cts (differences in predicted probabilities) of reporting “fairly often or very often” in at least one e groups and SEP level, ADHS 2009, Dentate participants (n = 8,171)a Asterisks indicate level of significance of the marginal effects compared to the reference category (p < 0.05, p < 0.01). aAll models adjusted for demographic characteristics, geographical location and general health. adults revealed significant social gradients (by income, education and occupation) in social impacts from oral conditions and poor self-rated oral health [3]. Discussion Table 2 Marginal effects (differences in predicted probabilities) of subjective outcomes by SEP level, ADHS 2009, Dentate participants (n = 8,171)a b Table 2 Marginal effects (differences in predicted probabilities) of subjective outcomes by SEP level, ADHS 2009, al effects (differences in predicted probabilities) of subjective outcomes by SEP level, ADHS 2009, pants (n 8 171)a nces in predicted probabilities) of subjective outcomes by SEP level, ADHS 2009, Table 3 Marginal effects (differences in predicted probabilities) of bad self-rated oral health by age groups and SEP level, ADHS 2009, Dentate participants (n = 8,171)a SEP indicator 21 - 34 years 35 - 49 years 50 - 64 years ≥65 years Percentage points (95% CI) Education Degree or equivalent Reference Reference Reference Reference Some educational qualifications 5.66 (2.77, 8.56) 3.38 (0.75, 6.01)* 5.27 (2.36, 8.18) 0.90 (−2.66, 4.46) No qualifications 12.54 (5.01, 20.06) 8.46 (3.58, 13.34) 9.50 (5.55, 13.44) 5.76 (1.73, 9.79) Occupational social class Managerial and professional Reference Reference Reference Reference Intermediate 1.05 (−3.24, 5.33) 5.78 (2.49, 9.07) 3.31 (−0.47, 7.09) 5.97 (2.37, 9.57) Routine and manual 1.78 (−1.97, 5.52) 7.25 (4.34, 10.15) 5.14 (2.14, 8.14) 7.80 (4.58, 11.01) Never worked and long term unemployed 7.67 (−0.42, 15.76) 6.00 (−0.65, 12.65) 8.44 (−1.69, 18.58) 4.46 (−0.59, 9.50) Equivalised Household Income Wealthiest quintile Reference Reference Reference Reference Second wealthiest quintile 4.05 (0.93, 7.16)* 3.64 (0.73, 6.55)* −0.29 (−4.09, 3.50) 0.32 (−5.51, 6.14) Intermediate quintile 4.92 (1.15, 8.69)* 6.99 (3.45, 10.52)** 4.52 (0.61, 8.42)* 3.06 (−2.35, 8.47) Second poorest quintile 6.53 (1.90, 11.17) 6.33 (2.72, 9.94) 2.83 (−1.71, 7.36) 3.04 (−2.46, 8.54) Poorest quintile 10.39 (5.04, 15.74) 8.13 (4.45, 11.80) 5.27 (0.64, 9.89)* 3.77 (−2.51, 10.05) Asterisks indicate level of significance of the marginal effects compared to the reference category (p < 0.05, p < 0.01). aAll models adjusted for demographic characteristics, geographical location and general health. Table 3 Marginal effects (differences in predicted probabilities) of bad self-rated oral health by age groups and SEP level, ADHS 2009, Dentate participants (n = 8,171)a differences in predicted probabilities) of bad self-rated oral health by age groups and SEP ti i t ( 8 171)a Table 3 Marginal effects (differences in predicted probabilities) of bad self-rated oral health by ag level, ADHS 2009, Dentate participants (n = 8,171)a differences in predicted probabilities) of bad self-rated oral health by age groups and SEP e participants (n = 8,171)a Page 6 of 9 Guarnizo-Herreño et al. Discussion Other stud- ies also highlight the importance of relative SEP and the effects of social hierarchy on subjective oral health as the existence of social gradients has been broadly observed. In the US, based on a nationally representative sample of adults, poorer self-perceived oral health was consistently found at each lower level of education and income [9]. Even among low-income mothers in Washington State (US), self-assessed oral health was significantly worse at each lower educational and income level [47]. Poorer sub- jective oral health outcomes have also been found at each lower level of education, income or occupational social class in studies conducted in Sweden [14], Brazil [17] and the UK [13,28]. In addition, some previous evidence is also in line with our finding of non-significant inequalities among edentate adults [13]. mechanisms through which they influence subjective oral health could be distinctive. First, income has been used as proxy of material-related resources and can affect oral health through access to these resources and also through its influence on self-esteem and social standing [42]. Second, education captures the knowledge-related resources of a person [43] and is a measure of life time SEP since it is generally achieved earlier in life. Education could influence subjective oral health through its effects on social and psychological resources, perceptions, and oral health related behaviours [42,44,45]. Finally, occupa- tional social class refers to people’s relationship to work and to others through a society’s economic structure [46], and it may affect subjective oral health through its influ- ence on support networks, stress in the workplace, control and autonomy. We must acknowledge, however, that as the SEP indicators used in this analysis are likely to be highly correlated with each other, these distinctive mecha- nisms are hypothetical and were not formally tested in this study. mechanisms through which they influence subjective oral health could be distinctive. First, income has been used as proxy of material-related resources and can affect oral health through access to these resources and also through its influence on self-esteem and social standing [42]. Second, education captures the knowledge-related resources of a person [43] and is a measure of life time SEP since it is generally achieved earlier in life. Education could influence subjective oral health through its effects on social and psychological resources, perceptions, and oral health related behaviours [42,44,45]. Discussion Finally, occupa- tional social class refers to people’s relationship to work and to others through a society’s economic structure [46], and it may affect subjective oral health through its influ- ence on support networks, stress in the workplace, control and autonomy. We must acknowledge, however, that as the SEP indicators used in this analysis are likely to be highly correlated with each other, these distinctive mecha- nisms are hypothetical and were not formally tested in this study. In this analysis, inequalities in subjective oral health varied according to age. We generally found larger in- equalities by income and education among younger adults; for self-rated oral health and OHIP these were significant for both younger age groups while for the OIDP they were significant among 35–49 year-olds. However, for all three measures inequalities tended to fade away with age and were not significant among older adults (65+ years). Because we analysed cross-sectional data, it is difficult to disentangle if these results are ex- plained by age effects, mortality selection, or cohort ef- fects. If it is an age effect, our findings could indicate Our findings on inequalities in subjective oral health agree with results of studies from different countries. In the US, analyses have shown that adults in lower income level were significantly more likely to rate their oral health as fair or poor compared to their counterparts in high income [4,10]. Likewise, a multilevel analysis found that Australian adults with lower income were more likely to report their oral health as fair or poor even after adjust- ing for age, gender, education and neighbourhood socio- economic disadvantage [6]. Another study on Australian Page 7 of 9 Page 7 of 9 Guarnizo-Herreño et al. Discussion that material- and knowledge-related resources have a greater impact on subjective oral health in early and mid-adulthood than later in life. This stronger relation- ship between SEP and subjective oral health at younger ages could be explained by the “age-as-leveler” hypoth- esis that states that health inequalities widen in early to mid-adulthood and tend to decrease later in life [48]. One reason to explain this change is that people in higher SEP levels can only delay health decline for a spe- cific period in their lives, and therefore, in later life they “catch up” those with low SEP. Thus, the health of per- sons in high SEP declines relatively slowly in early and mid-adulthood, but at older ages the rate of decline ac- celerates resulting in lower health inequalities [48]. Since subjective measures of oral health are significantly asso- ciated with oral health status and unmet treatment needs [29-31], the age-as-leveler hypothesis seems a sensible explanation for our results. It may also be sug- gested that older people in high SEP are aware of the tendency towards worse oral health with ageing and therefore may have more modest expectations about their oral health compared to earlier in life. If that is the case, lower inequalities in subjective oral health at older ages would partly be the result of adults in high SEP having 1) a faster decline in oral health status and 2) lower expectations about their oral health. Trying to ex- plore this further, we subsequently estimated the gradi- ents by age group for different OHIP and OIDP items. We found steeper gradients by education among the older age group in difficulty eating (measured by OIDP), Although these seem plausible interpretations of our results, we could not rule out the mortality selection and cohort effects as alternative explanations. The former re- fers to the fact that persons in lower SEP tend to die at younger ages, and those who survive longer have rela- tively better health. Then, health inequalities decrease at older ages because people with worse health at low SEP have died, leaving behind a group of relatively healthier people in low SEP levels. Discussion BMC Public Health 2014, 14:827 http://www.biomedcentral.com/1471-2458/14/827 Table 5 Marginal effects (differences in predicted probabilities) of reporting a score of 3 or higher in any OIDP item by age groups and SEP level, ADHS 2009, Dentate participants (n=8,171)a SEP indicator 21 - 34 years 35 - 49 years 50 - 64 years ≥65 years Percentage points (95% CI) Education Degree or equivalent Reference Reference Reference Reference Some educational qualifications 5.85 (2.09, 9.61)* 6.96 (3.88, 10.14)** 5.25 (0.91, 9.57)* 3.04 (−1.83, 7.91) No qualifications 9.32 (1.90, 16.74)* 4.42 (−1.16, 10.00) 5.19 (−0.24, 10.61) 8.17 (2.18, 14.17)** Occupational social class Managerial and professional Reference Reference Reference Reference Intermediate 1.40 (−3.79, 6.58) 2.30 (−1.82, 6.41) 3.47 (−1.29, 8.23) 5.85 (0.34, 11.36)* Routine and manual 5.45 (0.64, 10.27)* 4.58 (1.26, 7.90) 6.64 (2.12, 11.16) 3.89 (−0.86, 8.64) Never worked and long term unemployed 6.90 (−0.88, 14.67) 3.05 (−4.96, 11.07) −4.42 (−11.83, 2.98) −2.32 (−9.22, 4.59) Equivalised Household Income Wealthiest quintile Reference Reference Reference Reference Second wealthiest quintile 0.45 (−4.63, 5.53) 5.33 (1.49, 9.16)** 1.31 (−3.76, 6.39) −1.89 (−10.47, 6.70) Intermediate quintile 2.00 (−3.77, 7.76) 5.08 (0.70, 9.45)* −0.21 (−5.04, 4.61) −0.12 (−8.07, 7.83) Second poorest quintile 4.24 (−1.67, 10.15) 6.53 (1.07, 11.99)* 3.82 (−2.15, 9.80) 2.49 (−5.20, 10.19) Poorest quintile 6.19 (−0.50, 12.87) 10.58 (5.69, 15.47)** 4.65 (−1.55, 10.84) 2.34 (−6.59, 11.26) Asterisks indicate level of significance of the marginal effects compared to the reference category (p < 0.05, p < 0.01). aAll models adjusted for demographic characteristics, geographical location and general health. Table 5 Marginal effects (differences in predicted probabilities) of reporting a score of 3 or highe age groups and SEP level, ADHS 2009, Dentate participants (n=8,171)a Table 5 Marginal effects (differences in predicted probabilities) of reporting a score of 3 or higher in any OIDP item by age groups and SEP level, ADHS 2009, Dentate participants (n=8,171)a SEP indicator 21 - 34 years 35 - 49 years 50 - 64 years ≥65 years cts (differences in predicted probabilities) of reporting a score of 3 or higher in any OIDP item by evel, ADHS 2009, Dentate participants (n=8,171)a Asterisks indicate level of significance of the marginal effects compared to the reference category (p < 0.05, **p < 0.01). aAll models adjusted for demographic characteristics, geographical location and general health. potentially reflecting the worse oral health status among lower SEP groups at this age (results not shown). Discussion Mortality risk has been associ- ated to oral health measures such as periodontal disease and number of teeth [49-52], which in turn are strongly related to subjective oral health, indicating that the mor- tality selection argument might also be a reasonable ex- planation for our results of lower inequalities at older ages. Finally, the cohort effect refers to the fact that people in different age groups have lived under different histor- ical contexts. These contexts could have contributed to weaker or stronger relationships between SEP and health, thereby explaining variations in health inequal- ities across cohorts. In this sense, people aged 65+ years represent a generation that has lived through the Second World War and the decades of economic stability and reconstruction and therefore may be a more cohesive cohort compared to younger adults that have lived pri- marily through a period of excessive economic growth which is linked also to wider inequalities [53]. If findings of our analyses are result of cohort effects, we could Guarnizo-Herreño et al. BMC Public Health 2014, 14:827 http://www.biomedcentral.com/1471-2458/14/827 Guarnizo-Herreño et al. BMC Public Health 2014, 14:827 http://www.biomedcentral.com/1471-2458/14/827 Page 8 of 9 Authors’ contributions JW had the original idea that was developed with RGW, GT, CCG-H, EF, JGS, SM, and JS. CCG-H and GT performed the statistical analysis. CCG-H wrote the first draft and RGW, GT, JW, EF, JS, SM, and JGS contributed to subsequent redrafts. CCG-H, RGW, GT, JW, EF, JGS, SM, and JS read and approved the final draft. References 1 W tt R Turrell G, Sanders AE, Slade GD, Spencer AJ, Marcenes W: The independent contribution of neighborhood disadvantage and individual-level socioeconomic position to self-reported oral health: a multilevel analysis. Community Dent Oral Epidemiol. 2007, 35:195–206. 7. Wamala S, Merlo J, Bostrom G: Inequity in access to dental care services explains current socioeconomic disparities in oral health: the Swedish National Surveys of Public Health 2004–2005. J Epidemiol Community Health. 2006, 60:1027–1033. 7. Wamala S, Merlo J, Bostrom G: Inequity in access to dental care services explains current socioeconomic disparities in oral health: the Swedish National Surveys of Public Health 2004–2005. J Epidemiol Community Health. 2006, 60:1027–1033. 8. Shelley D, Russell S, Parikh NS, Fahs M: Ethnic disparities in self-reported oral health status and access to care among older adults in NYC. J Urban Health 2011, 88:651–662. 9. Sabbah W, Tsakos G, Chandola T, Sheiham A, Watt RG: Social gradients in oral and general health. J Dent Res. 2007, 86:992–996. Received: 7 May 2014 Accepted: 6 August 2014 Published: 9 August 2014 Received: 7 May 2014 Accepted: 6 August 2014 Published: 9 August 2014 Received: 7 May 2014 Accepted: 6 August 2014 Published: 9 August 2014 Acknowledgements Thi k This work was supported by the UK Economic and Social Research Council [Grant Number ES/K004689/1] as part of the Secondary Data Analysis Initiative. Additional file expect to see larger inequalities in the long term as the current cohort of young adults will age and the current cohort of older adults will be phased out. Additional file 1: Table S1. Descriptive statistics for study variables for edentate participants, ADHS 2009. (Based on a study sample of 594 individuals). This study provides comprehensive evidence on socio- economic inequalities in subjective oral health from the most recent survey of adults’ oral health in England, Wales and Northern Ireland. It uses SEP indicators that capture different aspects of individuals’ relative position in a society. Moreover, the oral health measures consid- ered give a comprehensive assessment of how people perceive their oral health and the effects of oral health on functional, social and psychological aspects of daily life. This study, however, has also some limitations. First, because all SEP indicators were self-reported, there could be some measurement bias. This is particularly relevant for income, as people may be reluctant to reveal accurate information [42]. The high proportion of miss- ing income data could be the result of this unwillingness to provide such information. Second, the SEP measures used in our analyses might not be very good markers of SEP among older adults. Wealth may be a better alterna- tive for that age group that consists primarily of pen- sioners, but information on wealth was not collected in this survey. Finally, the prevalence of edentulousness was very low in some SEP categories, and the lack of sig- nificant differences in predicted probabilities among edentate participants could be partially explained by these small sample sizes. References 1 W tt R 1. Watt R, Sheiham A: Inequalities in oral health: a review of the evidence and recommendations for action. Br Dent J 1999, 187:6–12. Our results have important implications for public health. The persistent social gradients in subjective oral health further highlight that health inequalities should be considered a public health priority. More importantly, the variation by age groups indicate that relevant public health interventions should primarily target younger adults as the gradients are steeper among those aged below 50 years. This is in line with adopting a life course approach and focusing earlier in life, as suggested for health inequalities in general [54]. And it may be par- ticularly relevant for these cohorts as they are expected to keep more natural teeth and for more years than older cohorts. Therefore, achieving a good oral health status earlier in life without large differences between the different SEP strata may have a long lasting positive effect on the population’s health and well-being. 2. Locker D: Deprivation and oral health: a review. Community Dent Oral Epidemiol 2000, 28:161–169. 2. Locker D: Deprivation and oral health: a review. Community Dent Oral Epidemiol 2000, 28:161–169. 3. Sanders AE, Spencer AJ: Social inequality in perceived oral health among adults in Australia. Aust N Z J Public Health. 2004, 28:159–166. 3. Sanders AE, Spencer AJ: Social inequality in perceived oral health among adults in Australia. Aust N Z J Public Health. 2004, 28:159–166. 4. Borrell LN, Baquero MC: Self-rated general and oral health in New York City adults: assessing the effect of individual and neighborhood social factors. Community Dent Oral Epidemiol. 2011, 39:361–371. 4. Borrell LN, Baquero MC: Self-rated general and oral health in New York City adults: assessing the effect of individual and neighborhood social factors. Community Dent Oral Epidemiol. 2011, 39:361–371. 5. Celeste RK, Nadanovsky P, Ponce de Leon A, Fritzell J: The individual and contextual pathways between oral health and income inequality in Brazilian adolescents and adults. Soc Sci Med 2009, 69:1468–1475. 5. Celeste RK, Nadanovsky P, Ponce de Leon A, Fritzell J: The individual and contextual pathways between oral health and income inequality in Brazilian adolescents and adults. Soc Sci Med 2009, 69:1468–1475. 6. Turrell G, Sanders AE, Slade GD, Spencer AJ, Marcenes W: The independent contribution of neighborhood disadvantage and individual-level socioeconomic position to self-reported oral health: a multilevel analysis. Community Dent Oral Epidemiol. 2007, 35:195–206. 6. 11. Morita I, Nakagaki H, Yoshii S, Tsuboi S, Hayashizaki J, Igo J, Mizuno K, Sheiham A: Gradients in periodontal status in Japanese employed males. J Clin Periodontol. 2007, 34:952–956. 9. Sabbah W, Tsakos G, Chandola T, Sheiham A, Watt RG: Social gradients in oral and general health. J Dent Res. 2007, 86:992–996. Author details 1Department of Epidemiology and Public Health, University College London, 1-19 Torrington Place, WC1E 7HB London, UK. 2National Centre for Social Research NatCen, 35 Northampton Square, EC1V 0AX London, UK. 3School of Dental Sciences, Newcastle University, Framlington Place, Tyne and Wear NE2 4BW Newcastle Upon Tyne, UK. 4Institute of Health and Society, Newcastle University, Baddiley-Clark Building, Tyne and Wear NE2 4AX Newcastle Upon Tyne, UK. 5Department of Applied Health Research, University College London, 1-19 Torrington Place, WC1E 7HB London, UK. 6Business School, Newcastle University, 5 Barrack Rd, Tyne and Wear NE1 4SE Newcastle Upon Tyne, UK. 12. Elani HW, Harper S, Allison PJ, Bedos C, Kaufman JS: Socio-economic Inequalities and Oral Health in Canada and the United States. J Dent Res. 2012, 91:865–870. 8. Shelley D, Russell S, Parikh NS, Fahs M: Ethnic disparities in self-reported oral health status and access to care among older adults in NYC. J Urban Health 2011, 88:651–662. Guarnizo-Herreño et al. BMC Public Health 2014, 14:827 http://www.biomedcentral.com/1471-2458/14/827 Grembowski D, Spiekerman C, Milgrom P: Social gradients in dental health among low-income mothers and their young children. J Health Care Poor Underserved. 2012, 23:570–588. 19. Fritzell J, Lundberg O: Health Inequalities and Welfare Resources: Continuity and Change in Sweden. Bristol: The Policy Press; 2007. 48. Herd P: Do functional health inequalities decrease in old age? Educational status and functional decline among the 1931–1941 birth cohort. Res Aging 2006, 28:375–392. 20. Whitehead M: The concepts and principles of equity and health. Int J Health Serv. 1992, 22:429–445. 49. Abnet CC, Qiao YL, Dawsey SM, Dong ZW, Taylor PR, Mark SD: Tooth loss is associated with increased risk of total death and death from upper gastrointestinal cancer, heart disease, and stroke in a Chinese population-based cohort. Int J Epidemiol. 2005, 34:467–474. 21. Braveman P, Gruskin S: Defining equity in health. J Epidemiol Community Health. 2003, 57:254–258. 22. Sanders AE, Spencer AJ: Why do poor adults rate their oral health poorly? Aust Dent J. 2005, 50:161–167. 50. Sabbah W, Mortensen LH, Sheiham A, Batty D: Oral health as a risk factor for mortality in middle-aged men: the role of socioeconomic position and health behaviours. J Epidemiol Community Health. 2013, 67:392–397. 23. Locker D: Self-esteem and socioeconomic disparities in self-perceived oral health. J Public Health Dent. 2009, 69:1–8. 24. Sheiham A, Nicolau B: Evaluation of social and psychological factors in periodontal disease. Periodontol 2000 2005, 39:118–131. 51. Aida J, Kondo K, Yamamoto T, Hirai H, Nakade M, Osaka K, Sheiham A, Tsakos G, Watt RG: Oral health and cancer, cardiovascular, and respiratory mortality of Japanese. J Dent Res. 2011, 90:1129–1135. 25. Boyapati L, Wang H-L: The role of stress in periodontal disease and wound healing. Periodontol 2000 2007, 44:195–210. 52. Tu Y-K, Galobardes B, Smith GD, McCarron P, Jeffreys M, Gilthorpe MS: Associations between tooth loss and mortality patterns in the Glasgow Alumni Cohort. Heart. 2007, 93:1098–1103. 26. Sabbah W, Watt RG, Sheiham A, Tsakos G: The role of cognitive ability in socio-economic inequalities in oral health. J Dent Res. 2009, 88:351–355. socio-economic inequalities in oral health. J Dent Res. 2009, 88:351 27. Steele J, O’Sullivan I: Executive Summary: Adult Dental Health Survey 2009. London: NHS Information Centre; 2011. 53. Coburn D: Beyond the income inequality hypothesis: class, neo- liberalism, and health inequalities. Soc Sci Med. 2004, 58:41–56. 28. Guarnizo-Herreño et al. BMC Public Health 2014, 14:827 http://www.biomedcentral.com/1471-2458/14/827 41. Williams R: Using Stata’s Margins Command to Estimate and Interpret Adjusted Predictions and Marginal Effects. Notre Dame: University of Notre Dame; 2011. 13. Tsakos G, Demakakos P, Breeze E, Watt RG: Social gradients in oral health in older adults: findings from the English longitudinal survey of aging. Am J Public Health. 2011, 101:1892–1899. 42. Galobardes B, Shaw M, Lawlor DA, Lynch JW, Davey SG: Indicators of socioeconomic position (part 1). J Epidemiol Community Health. 2006, 60:7–12. 14. Stahlnacke K, Soderfeldt B, Unell L, Halling A, Axtelius B: Perceived oral health: changes over 5 years in one Swedish age-cohort. Community Dent Oral Epidemiol. 2003, 31:292–299. 43. Shaw M, Galobardes B, Lawlor DA, Lynch J, Wheeler B, Davey Smith G: The handbook of inequality and socioeconomic position. Bristol. Policy 2007. 43. Shaw M, Galobardes B, Lawlor DA, Lynch J, Wheeler B, Davey Smith G: The h db k f i lit d i i iti B i t l P li 2007 15. Lopez R, Fernandez O, Baelum V: Social gradients in periodontal diseases among adolescents. Community Dent Oral Epidemiol. 2006, 34:184–196. handbook of inequality and socioeconomic position. Bristol. 16. Geyer S, Schneller T, Micheelis W: Social gradients and cumulative effects 44. Ross CE, C-l W: The links between education and health. Am Sociol Rev. 1995, 60:719–745. 16. Geyer S, Schneller T, Micheelis W: Social gradients and cumulative effects of income and education on dental health in the Fourth German Oral Health Study. Community Dent Oral Epidemiol. 2010, 38:120–128. of income and education on dental health in the Fourth German Oral Health Study. Community Dent Oral Epidemiol. 2010, 38:120–128. 45. Daly MC, Duncan GJ, McDonough P, Williams DR: Optimal indicators of socioeconomic status for health research. Am J Public Health. 2002, 92:1151–1157. 17. Mendonca HL, Szwarcwald CL, Damacena GN: Self-rated oral health: results of the World Health Survey-Primary Care in four municipalities in Rio de Janeiro State, Brazil, 2005. Cad Saude Publica. 2012, 28:1927–1938. 46. Krieger N, Williams DR, Moss NE: Measuring social class in US Public Health Research: concepts, methodologies, and guidelines. Annu Rev Public Health. 1997, 18:341–378. 18. Marmot M, Friel S, Bell R, Houweling TAJ, Taylor S, Commission on Social Determinants of H: Closing the gap in a generation: health equity through action on the social determinants of health. Lancet 2008, 372:1661–1669. 47. Conclusions In summary, our results suggest that socioeconomic in- equalities in subjective oral health still exist among the UK adult population. These inequalities tend to be larger among young adults, imposing challenges for scientific knowledge and policy decision making. Further research should explore the potential contribution of specific mechanisms (material, psychosocial and behavioural) in explaining these inequalities. 10. Borrell LN, Taylor GW, Borgnakke WS, Woolfolk MW, Nyquist LV: Perception of general and oral health in White and African American adults: assessing the effect of neighborhood socioeconomic conditions. Community Dent Oral Epidemiol. 2004, 32:363–373. Community Dent Oral Epidemiol. 2004, 32:363–373. 11. Morita I, Nakagaki H, Yoshii S, Tsuboi S, Hayashizaki J, Igo J, Mizuno K, Sheiham A: Gradients in periodontal status in Japanese employed males. J Clin Periodontol. 2007, 34:952–956. 12. Elani HW, Harper S, Allison PJ, Bedos C, Kaufman JS: Socio-economic Inequalities and Oral Health in Canada and the United States. J Dent Res. 2012, 91:865–870. Page 9 of 9 Page 9 of 9 Guarnizo-Herreño et al. BMC Public Health 2014, 14:827 http://www.biomedcentral.com/1471-2458/14/827 Guarnizo-Herreño et al. BMC Public Health 2014, 14:827 http://www.biomedcentral.com/1471-2458/14/827 Shen J, Wildman J, Steele J: Measuring and decomposing oral health inequalities in an UK population. Community Dent Oral Epidemiol. 2013, 41:481–489. 54. Marmot MG, Allen J, Goldblatt P, Boyce T, McNeish D, Grady M, Geddes I: Marmot Review. Fair Society, Healthy Lives: Strategic Review of Health Inequalities in England Post 2010. London: Institute of Health Equity; 2010. 54. Marmot MG, Allen J, Goldblatt P, Boyce T, McNeish D, Grady M, Geddes I: Marmot Review. Fair Society, Healthy Lives: Strategic Review of Health Inequalities in England Post 2010. London: Institute of Health Equity; 2010. 29. Locker D, Mscn EW, Jokovic A: What do older adults' global self-ratings of oral health measure? J Public Health Dent. 2005, 65:146–152. doi:10.1186/1471-2458-14-827 Cite this article as: Guarnizo-Herreño et al.: Socioeconomic position and subjective oral health: findings for the adult population in England, Wales and Northern Ireland. BMC Public Health 2014 14:827. 30. Chen X, Naorungroj S, Douglas CE, Beck JD: Self-reported oral health and oral health behaviors in older adults in the last year of life. J Gerontol A Biol Sci Med Sci. 2013, 68:1310–1315. 31. Pattussi MP, Peres KG, Boing AF, Peres MA, da Costa JSD: Self-rated oral health and associated factors in Brazilian elders. Community Dent Oral Epidemiol. 2010, 38:348–359. 32. Benyamini Y, Leventhal H, Leventhal EAEA: Self-rated oral health as an independent predictor of self-rated general health, self-esteem and life satisfaction. Soc Sci Med. 2004, 59:1109–1116. 33. Yoon HS, Kim HY, Patton LL, Chun JH, Bae KH, Lee MO: Happiness, subjective and objective oral health status, and oral health behaviors among Korean elders. Community Dent Oral Epidemiol 2013, 41:459–465. 34. Pattussi M, Olinto M, Hardy R, Sheiham A: Clinical, social and psychosocial factors associated with self-rated oral health in Brazilian adolescents. Community Dent Oral Epidemiol. 2007, 35:377–386. Guarnizo-Herreño et al. BMC Public Health 2014, 14:827 http://www.biomedcentral.com/1471-2458/14/827 Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit 35. Sheiham A: Oral health, general health and quality of life. Bull World Health Organ. 2005, 83:644. Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: Submit your next manuscript to BioMed Central and take full advantage of: 36. Slade GD: Measuring Oral Health and Quality of Life. Chapel Hill, NC: University of North Carolina; 1997. 37. Slade GD, Nuttall N, Sanders AE, Steele JG, Allen PF, Lahti S: Impacts of oral disorders in the United Kingdom and Australia. Br Dent J. 2005, 198:489–493. • Convenient online submission 38. Locker D, Quinonez C: To what extent do oral disorders compromise the quality of life? Community Dent Oral Epidemiol. 2011, 39:3–11. 39. Chandola T: Social class differences in mortality using the new UK National Statistics Socio-Economic Classification. Soc Sci Med. 2000, 50:641–649. 40. Gordon RA: Applied Statistics for the Social and Health Sciences. New York: Routledge; 2012. 40. Gordon RA: Applied Statistics for the Social and Health Sciences. New York: Routledge; 2012.
https://openalex.org/W2150528103	http://publikationen.ub.uni-frankfurt.de/files/20668/ar3108.pdf	English	null	Visualization and phenotyping of pro-inflammatory antigen-specific T cells during collagen induced arthritis in a mouse with a fixed collagen type II specific transgenic TCR beta chain	Arthritis research & therapy	2,010	cc-by	13,461	* Correspondence: johan.backlund@ki.se 1Section for Medical Inflammation Research, Department of Medical Biochemistry and Biophysics, Karolinska Institute, Scheeles väg 2, 171 77 Stockholm, Sweden Full list of author information is available at the end of the article RESEARCH ARTICLE Open Access Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Visualization and phenotyping of proinflammatory antigen-specific T cells during collagen-induced arthritis in a mouse with a fixed collagen type II-specific transgenic T-cell receptor b-chain Patrick Merky1, Tsvetelina Batsalova1, Robert Bockermann2, Balik Dzhambazov1, Bettina Sehnert3, Harald Burkhardt4, Johan Bäcklund1* Antigens h The rat CII was obtained from pepsin-digested SWARM chondrosarcoma [10], and subsequently processed as previously described [11]. CII peptides, containing the 259 to 273 sequence of rat CII with a nonmodified lysine at position 264 (K264) or with a b-D-galactopyra- nosyl residue on L-hydroxylysine at position 264 (Gal- HyK264), were synthesized as previously described [12]. Abstract Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Page 2 of 14 cannot be assumed that all, or even the majority, of CD4+ T cells are CII specific, as was the case for the Va11.1/Vb8.3-tg and Va11.1/Vb8.2-tg mouse strains. Indeed, the Vb12-tg mouse can also develop immunity against microorganisms as well as immunity to tubercu- lin-purified protein derivative, an immunogenic compo- nent of complete Freund’s adjuvant (CFA) [6,8]. Materials and methods Mice To generate [B10.Q × DBA/1]F1 mice transgenic for Vb12, B10.Q mice were crossed with DBA/1 mice, expressing a transgenic TCR b-chain (Vb12), obtained from a CII-specific T-cell clone, originally derived from a CII-immunized DBA/1 mouse [6]. All animals were between 7 and 12 weeks of age at the start of the experiments. Mice were housed in a conventional ani- mal facility and all experiments were performed accord- ing to the Swedish ethical committee guidelines. The use of T-cell receptor transgenic (TCR-tg) mice has proven a powerful tool for investigating the nature of self-reactive T cells in tolerance and autoimmunity [3]. To further facilitate the understanding for the role of T cells in CIA, three different CII-specific TCR-tg mouse strains have earlier been described and shown to display an accelerated onset of severe arthritis, com- pared with nontransgenic littermates. Transgenic T cells from all three strains are Aq-restricted and recognize the same region on CII that is located between amino acid positions 260 and 270. This region harbors a lysine residue at position 264, which is naturally subjected to post-translational modifications, through hydroxylation and subsequent glycosylation. Strikingly, each of the three previously described TCR-tg mouse strains in fact recognize different forms of the CII(260-270) epitope, where the Va11.1/Vb8.3-tg mouse [4], the Va11.1/ Vb8.2-tg mouse [5] and the Vb12-tg mouse [6] respond to the nonmodified [4], the hydroxylated [7] and the galactosylated [8] CII(260-270) peptide, respectively. Abstract Introduction: The Vb12-transgenic mouse was previously generated to investigate the role of antigen-specific T cells in collagen-induced arthritis (CIA), an animal model for rheumatoid arthritis. This mouse expresses a transgenic collagen type II (CII)-specific T-cell receptor (TCR) b-chain and consequently displays an increased immunity to CII and increased susceptibility to CIA. However, while the transgenic Vb12 chain recombines with endogenous a- chains, the frequency and distribution of CII-specific T cells in the Vb12-transgenic mouse has not been determined. The aim of the present report was to establish a system enabling identification of CII-specific T cells in the Vb12-transgenic mouse in order to determine to what extent the transgenic expression of the CII-specific b- chain would skew the response towards the immunodominant galactosylated T-cell epitope and to use this system to monitor these cells throughout development of CIA. Methods: We have generated and thoroughly characterized a clonotypic antibody, which recognizes a TCR specific for the galactosylated CII(260-270) peptide in the Vb12-transgenic mouse. Hereby, CII-specific T cells could be quantified and followed throughout development of CIA, and their phenotype was determined by combinatorial analysis with the early activation marker CD154 (CD40L) and production of cytokines. Results: The Vb12-transgenic mouse expresses several related but distinct T-cell clones specific for the galactosylated CII peptide. The clonotypic antibody could specifically recognize the majority (80%) of these. Clonotypic T cells occurred at low levels in the naïve mouse, but rapidly expanded to around 4% of the CD4+ T cells, whereupon the frequency declined with developing disease. Analysis of the cytokine profile revealed an early Th1-biased response in the draining lymph nodes that would shift to also include Th17 around the onset of arthritis. Data showed that Th1 and Th17 constitute a minority among the CII-specific population, however, indicating that additional subpopulations of antigen-specific T cells regulate the development of CIA. Conclusions: The established system enables the detection and detailed phenotyping of T cells specific for the galactosylated CII peptide and constitutes a powerful tool for analysis of the importance of these cells and their effector functions throughout the different phases of arthritis. © 2010 Merky et al.; licensee BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Merky et al. Introduction Collagen-induced arthritis (CIA) is the most commonly used animal model for rheumatoid arthritis. Develop- ment of CIA is dependent on both B cells and T cells. The major role of B cells is to produce collagen type II (CII)-specific antibodies, and passive transfer of such antibodies has the capacity to bind cartilage in vivo and induce an acute arthritis. A major role of T cells is to aid B cells in their production of anti-CII antibodies, but they are also believed to play an active part in the disease via activation of other cell types, such as synovial macrophages. The influence of T cells in established CIA, however, is less clear. Adoptive transfer of CII-spe- cific T cells alone does not induce clinical disease but may lead to microscopic synovitis [1]. Adoptive transfer of CII-specific T cells has also been shown to prolong the otherwise acute arthritis induced by passive transfer of CII antibodies [2]. In the present report we have therefore established and thoroughly characterized a clonotypic antibody, which recognizes a TCR specific for the galactosylated CII(260-270) peptide in the Vb12-tg mouse. Using this antibody we were able to describe the emerging T-cell response and its transition in Vb12-tg mice following challenge with CII. Production of the recombinant 22a1-7E T-cell receptor protein The preparation was subsequently dialyzed for two sequential 16-hour intervals in 100 mM Tris, 400 mM arginine, 2 mM b-mercaptoethanol, 2 mM EDTA, pH 8.0 and 100 mM Tris, 200 mM arginine, 2 mM b-mercaptoethanol, 2 mM EDTA, pH 8.0. Subsequently, the buffer was changed to 20 mM Tris, 150 mM NaCl, 1 mM b-mercaptoethanol, 2 mM EDTA, pH 8.0 and the dialysis procedure continued for another 32 hours (2 × 16 hours). in-house collection), TCR cb-bio (clone H57-597; in- house collection) and propidium iodide (Invitrogen, Eugene, Oregon, USA). Subsequently biotin was detected by allophycocyanin-conjugated streptavidin (BD Pharmingen, San Diego, CA, USA). The cells were then acquired with a flow cytometer (FACSort; BD Bio- sciences, San Jose, CA, USA) by gating on propidium- iodide-negative live cells. For determination of relative frequencies of B22a1+ T cells, their activation and cytokine profile, the follow- ing antibodies were used for extracellular staining: Anti- Fcg receptor (2.4G2) and B22a1-bio from our in-house collection; anti-CD11b-FITC (M1/70), anti-CD45R-FITC (RA3-6B2), anti-MHC class II-FITC (7.16.17), anti-CD49b-FITC (DX5), and antigen-presenting cell (APC)-conjugated streptavidin from BD Pharmingen (San Diego, CA, USA); and anti-CD4-PE-Cy5.5 (RM4-5) from eBioscience (San Diego, CA, USA). For intracellu- lar staining, the following antibodies were used: anti- CD40L-PE (MRI), anti-IFNg-PE-Cy7 (GMX1.2) and anti-IL-17-Pacific blue (TC11-18H10.9) purchased from eBioscience (San Diego, CA, USA), and anti-IL2-Alexa Fluor 700 (JES6-5H4) purchased from BioLegend (San Diego, CA, USA). Homogeneity of the respective preparations was demonstrated by the detection of a single protein band with an electrophoretic mobility corresponding to calcu- lated molecular mass in 14% acrylamide gels upon Coo- massie Blue staining. All antibodies were titrated for optimal saturating con- centration. For exclusion of dead cells, the LIVE/DEAD Fixable Green Dead Cell Stain Kit (Invitrogen, Eugene, Oregon, USA) was used. At least 1 × 105 Th cells were acquired on a flow cytometer (FACS Aria/FACS LSRII; BD Biosciences, San Jose, CA, USA) and were analyzed using FlowJo software (Treestar, Inc. Ashland, OR, USA). Production of the recombinant 22a1-7E T-cell receptor protein Total RNA of the murine T-cell hybridoma clone 22a1- 7E was extracted and cDNA synthesis was performed following standard protocols. The V-gene segments encoding the Va16 and Vb12 chains of the 22a1-7E TCR were amplified by PCR using sequence-specific pri- mers. Following an earlier-described method for genera- tion of single-chain Fv antibodies [13], we used an assembly PCR strategy to connect the Va16 segment and the Vb12 segment by a fragment encoding a gly- cine-serine-linker (-(GSSS)4-) and cloned the entire con- struct into the pTriEx-1.1 vector (Merck, Darmstadt, Germany). Although each of the mentioned post-translationally modified peptides has its importance in Aq-restricted CIA, we have earlier shown that glycosylation of CII is of major importance for T-cell tolerance and pathology in CIA [9]. We therefore found it important to establish an animal model that would allow for identification and tracking of T cells specific for the galactosylated CII peptide. In contrast to the Va11.1/Vb8.3-tg and Va11.1/Vb8.2-tg mouse strains, however, which express both the a-chains and b-chains of the TCR as trans- genes, the galactosylated CII-specific Vb12-tg mouse is only transgenic for the CII-specific b-chain, which may combine with any endogenous a-chain. Although the Vb12-tg mouse displays an increased T-cell and B-cell immunity to CII and is more susceptible to CIA, it Upon control by nucleotide sequencing, the respective construct was transformed into Origami™(DE3)pLacI cells (Merck, Darmstadt, Germany) for bacterial Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Page 3 of 14 expression. The 8 × His-tagged recombinant TCR was purified from inclusion bodies of the lyzed bacteria by affinity chromatography on a Co2+-affinity resin (Takara Bio Europe/Clontech, Saint-Germain-en-Laye, France). The purification was performed according to the manu- facturer’s instructions and the protein was eluted with 150 mM imidazole buffer containing also 6 M guanidi- nium hydrochloride, 50 mM sodium phosphate, 500 mM NaCl, 3 mM b-mercaptoethanol, pH 7.8. Purified fractions of the denatured TCR protein were pooled and diluted to 300 μg/ml for the subsequent refolding proce- dure, consisting of repetitive dialysis steps. A mixture of 0.5 mM GSSH and 5 mM GSH was added and stirred for 1 hour at 4°C. Generation of a monoclonal antibody specific for the 22a1-7E T-cell receptor To generate clonotypic antibodies, E3 rats were immu- nized with a recombinant 22a1-7E TCR protein in CFA (Difco, Detroit, MI, USA). The 22a1-7E TCR originates from a T-cell clone, specific for the galactosylated CII (260-270) epitope, isolated from a CII-immunized Vb12- tg mouse. Lymph node cells from E3 rats immunized with the 22a1-7E TCR protein were prepared 10 days after immunization and were fused with myeloma cells (P3X63-Ag8.653) as described previously [14]. Superna- tant from growing cultures was tested for staining of the 22a1-7E T cell hybridoma clone (Va16/Vb12; RB, unpublished data), HCQ.3 (Va16/Vb8 [15]) and HCQ.4 (Va4/Vb12 [15]). Cells producing antibodies that stained 22a1-7E, but not HCQ.3 or HCQ.4 hybridomas, were selected for expansion, subcloning and retesting. After two additional rounds of selections, one clone spe- cific for the 22a1-7E TCR of the IgG1 isotype was obtained and was denoted B22a1. For antigen-specific detection of CD154 (CD40L) and cytokines, single-cell suspensions were processed as described earlier by Frentsch and colleagues [16] with some modifications. In brief, 1.2 × 107 cells/ml were cultured for 6 hours in the presence of the galactosy- lated CII peptide (GalHyK264, 10 μg/ml) and anti-CD28 (1 μg/ml, clone 61109; R&D Systems, Minneapolis, MN, USA). For intracellular accumulation of CD40L and cytokines, 2 μg/ml Brefeldin A (Sigma-Aldrich, St. Louis, MO, USA) was added after 2 hours of culture. The cells were then washed and stained for 15 minutes at room temperature with the surface markers. After- wards, the cells were fixed with Cytofix/Cytoperm solu- tion (BD Pharmingen, San Diego, CA, USA) and permeabilized with Perm/Wash buffer (BD Pharmingen, San Diego, CA, USA) according to the manufacturer’s instructions, and were stained for CD40L and cytokines. Antibodies and flow cytometry analysis Monoclonal antibody treatments and lymphocyte assay The antibody B22a1 was purified from culture superna- tant using gamma-bind plus sepharose gel matrix (GE Healthcare, Uppsala, Sweden), dialyzed against PBS, The TCR binding specificity of B22a1 and the expres- sion level of TCR on the surface of T-cell hybridoma was determined by flow cytometry using the following antibodies and reagents: B22a1-bio (produced from our Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Page 4 of 14 anti-mouse antibodies specific for IgM, IgG1 or IgG2a (Southern Biotech, Birmingham, AL, USA) were used as detecting antibodies. Binding of biotinylated antibodies was revealed by Extravidin Peroxidase (Sigma-Aldrich). Plates were developed using ABTS (Roche Diagnostics, Mannheim, Germany) as substrate, and the absorbance was measured at 405 nm (Synergy-2; BioTek, Winooski, VT, USA). Total anti-CII IgG levels were measured (μg/ ml) using purified polyclonal anti-CII IgG antibodies of a known concentration as a standard. Isotype levels were measured as arbitrary concentrations using our purified monoclonal anti-CII antibodies with known isotype or pooled sera from arthritic mice. sterile filtered, and kept at -80°C. Antibody concentra- tion was determined either spectrophotometrically (at 280 nm) or by freeze-drying. For in vitro depletion, single-cell suspensions from pooled popliteal and inguinal lymph nodes were pre- pared 10 days post immunization. One-half of each sin- gle-cell suspension was depleted of B22a1+ T cells, whereas the remaining half was kept untouched on ice for further processing. Briefly, lymphocytes were incu- bated with biotinylated antibody B22a1 for 20 minutes at 4°C. Subsequently, biotin binder Dynabeads (Invitro- gen Dynal AS) were added according to the manufac- turer’s recommendation and labeled cells were depleted by placing the sample tubes in a magnetic stand. The samples were depleted twice to reach a final purity > 95% before both the untouched and depleted fractions were set up into triplicate cultures. The cells were sti- mulated with antigen (lathyritic CII (pepsin-free), K264, GalHyK264 and ConA) at a concentration of 106 cells/ well in microtiter plates and were incubated for 72 hours. One microcurie of methyl-[3H]thymidine (Amersham, GE Healthcare, Little Chalfont, Buckin- ghamshire, UK) was added to each well for an additional 15 to 18 hours. Cell proliferation was measured by counting the incorporation of methyl-[3H]thymidine. T-cell hybridoma assays All cells were cultured in DMEM + Glutamax-I (Gibco Life Technologies, Grand Island, NY, USA) supplemen- ted with 5% heat-inactivated FCS and penicillin/ streptomycin. For antibody-induced T-cell hybridoma stimulation, flat-bottomed microtiter 96-well plates (NUNC, Thermo Fisher Scientific, Roskilde, Denmark) were precoated with a titrated concentration of purified immunoglobu- lins (B22a1, anti-TCR cb, anti-CD4 (clone H129.19), all produced from in-house hybridomas) starting at 5 μg/ ml. Plates were then washed and 5 × 104/well T-cell hybridoma were added and cultured for 24 hours, before assaying the supernatant for IL-2 content. For in vivo depletion, B22a1 was injected intraperito- neally as a single dose of 240 μg/mouse. To increase the depletion capacity, MAR18.5 was administered as a sec- ondary antibody 1 hour after the primary antibody by the same method in a dose of 340 μg [17]. The mono- clonal antibodies were given at day 0 (4 hours before immunization), day 6 or day 20, depending on the experiment. For the inhibition of antigen-induced T-cell hyridoma stimulation, 5 × 104/well T-cell hybridoma where pre- incubated with a titrated concentration of purified immunoglobulins (B22a1, anti-TCR cB (clone H57-597), anti-CD4 (clone H129.19; our collection)) starting at 10 μg/ml in flat-bottomed microtiter wells in 50 μl med- ium. After 1 hour of incubation at 37°C, the hybridoma were washed twice with PBS and transferred to a pre- viously prepared microtiter plate containing splenocytes as APCs (5 × 105/well) and peptide (0.3 μg/ml). After 24 hours of incubation at 37°C, the supernatants were ana- lyzed for IL-2. Results The B22a1 antibody failed to stain the T-cell clones not expressing a Va16 chain together with the trans- genic DBA/1-Vb12-tg chain (HCQ.3, HCQ.4, HRC.2 and HM1R.2; Figure 1a). From the clones expressing the DBA/1-Vb12-tg chain together with a closely related Va16 chain, the B22a1 antibody stained the 22a1-7E and 18b4-10 D T-cell clones strongly but not the 3D4- 1.2. Also, when normalizing the B22a1 staining against the level of surface TCR expression, significant staining was only observed against the 22a1-7E and 18b4-10 D Statistical analysis hybridoma clones (18b4-10 D and 3D4-1.2) were also tested. The two latter clones were also obtained from the DBA/1-Vb12-tg mouse and share both TCR a- chains and b-chains with 22a1-7E, but differ at one sin- gle amino acid within the complementarity determining region 3 (CDR3) region of the nontransgenic Va chain (RB, unpublished data). Antibody levels and in vitro lymphocyte assays were analyzed with the Mann-Whitney U test, whereas arthri- tis severity was analyzed with Fisher’s exact test and the flow cytometry data with an unpaired t test. Characterization of B22a1 specificity The clonotypic antibody B22a1 was generated through immunization of a rat with a recombinant protein, cor- responding to the TCR of the 22a1-7E T-cell clone, ori- ginating from a CII-primed DBA/1-Vb12-tg mouse. The purified B22a1 antibody was tested for binding to the 22a1-7E T-cell hybridoma clone as well as to other Aq- restricted and CII-specific clones sharing either one or none of the TCR chains with the 22a1-7E clone (Va16/ Vb12). In addition, two more closely related T-cell 18b4-10D 3D4-1.2 HCQ.3 HM1R.2 HCQ.4 HRC.2 22a1-7E FMO B22a1 TCR cβ A IL-2 (FU) B22a1 0.01 0.1 1 10 100 1000 10000 100000 1000000 10000000 TCR cβ 0.01 0.1 1 10 Ab (μg/ml) CD4 0.01 0.1 1 10 22a1-7E 18b4-10D 3D4-1.2 HCQ.3 HCQ.4 HM1R2 HRC.2 B Figure 1 B22a1 binding requires a combination of Vb12 and Va16 T-cell receptor chain. (a) Flow cytometric analysis of T-cell hybridoma expressing different Va-chain and Vb-chain combinations stained with the biotinylated antibody B22a1 and allophycocyanin-conjugated streptavidin. Histograms show the staining intensities of negative control (FMO; stained with allophycocyanin-conjugated streptavidin only), B22a1 (straight line) for specificity and T-cell receptor (TCR) cb (dotted line) for TCR surface expression levels. (b) For stimulation, T-cell hybridomas were cultured for 24 hours in precoated cell culture plates with titrated amounts of B22a1, TCR cb (positive control) and CD4 (negative control), whereupon IL-2 production was determined by ELISA. Ab, antibody; FU, fluorescence units. 18b4-10D 3D4-1.2 HCQ.3 HM1R.2 HCQ.4 HRC.2 22a1-7E FMO B22a1 TCR cβ A A IL-2 (FU) B22a1 0.01 0.1 1 10 100 1000 10000 100000 1000000 10000000 TCR cβ 0.01 0.1 1 10 Ab (μg/ml) CD4 0.01 0.1 1 10 22a1-7E 18b4-10D 3D4-1.2 HCQ.3 HCQ.4 HM1R2 HRC.2 B Figure 1 B22a1 binding requires a combination of Vb12 and Va16 T-cell receptor chain. (a) Flow cytometric analysis of T-cell hybridoma expressing different Va-chain and Vb-chain combinations stained with the biotinylated antibody B22a1 and allophycocyanin-conjugated streptavidin. Histograms show the staining intensities of negative control (FMO; stained with allophycocyanin-conjugated streptavidin only), B22a1 (straight line) for specificity and T-cell receptor (TCR) cb (dotted line) for TCR surface expression levels. (b) For stimulation, T-cell hybridomas were cultured for 24 hours in precoated cell culture plates with titrated amounts of B22a1, TCR cb (positive control) and CD4 (negative control), whereupon IL-2 production was determined by ELISA. Ab, antibody; FU, fluorescence units. Induction and evaluation of CIA and anti-CII antibodies Induction and evaluation of CIA and anti-CII antibodies Mice were injected by the base of the tail with 50 μl emulsion consisting of 100 μg rat CII emulsified 1:1 in CFA (Difco, Detroit, MI, USA). Development of clinical arthritis was followed three times weekly through visual scoring of the paws, starting 2 weeks after immuniza- tion. The arthritis was scored using a scale ranging from 1 to 15 for each paw, with a maximum score of 60 per mouse [18]. Each arthritic toe and knuckle was scored as 1, with a maximum of 10 per paw. A score of 5 was given to an arthritic ankle. To determine the stimulatory capacity of the clono- typic antibody, the relative production of IL-2 between nonstimulated and stimulated T-cell hybridoma clones was determined by ELISA. Hereby, 50 μl supernatant were removed from the plates after 24 hours of cul- ture. For detection, Jes6-1A12 (5 μg/ml; our collection) and Jes6-5H4-biotin (2 μg/ml; Mabtech, Nacka Strand, Sweden) were used as the capture antibody and detec- tion antibody, respectively. Binding of biotinylated antibodies was revealed by Europium-labeled streptavi- din (PerkinElmer Life Sciences, Inc. Boston, MA, USA) and plates were analyzed using a Victor 1420 multi- label counter (PerkinElmer Life Sciences, Inc. Boston, MA, USA). Blood samples were collected on days 17 and 35 for determination of anti-CII antibody responses. The titers of total anti-CII IgG as well as the IgG1, IgG2a, IgG2b and IgM isotypes were determined through quantitative ELISA [19], where serum was titrated (1:10 to 1:106) in parallel to the standard and titer values were interpo- lated within the linear range and related to the standard curve. Biotinylated rat anti-mouse IgG (clone 187.1; our collection) or peroxidase-conjugated goat Page 5 of 14 Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Characterization of B22a1 specificity IL-2 (FU) B22a1 0.01 0.1 1 10 100 1000 10000 100000 1000000 10000000 TCR cβ 0.01 0.1 1 10 Ab (μg/ml) CD4 0.01 0.1 1 10 22a1-7E 18b4-10D 3D4-1.2 HCQ.3 HCQ.4 HM1R2 HRC.2 B IL-2 (FU) B22a1 0.01 0.1 1 10 100 1000 10000 100000 1000000 10000000 TCR cβ 0.01 0.1 1 10 Ab (μg/ml) CD 0.01 0.1 B 10 TCR cβ 0.01 0.1 1 10 Ab (μg/ml) CD4 0.01 0.1 1 10 22a1-7E 18b4-10D 3D4-1.2 HCQ.3 HCQ.4 HM1R2 HRC.2 B Figure 1 B22a1 binding requires a combination of Vb12 and Va16 T-cell receptor chain. (a) Flow cytometric analysis of T-cell hybridoma expressing different Va-chain and Vb-chain combinations stained with the biotinylated antibody B22a1 and allophycocyanin-conjugated streptavidin. Histograms show the staining intensities of negative control (FMO; stained with allophycocyanin-conjugated streptavidin only), B22a1 (straight line) for specificity and T-cell receptor (TCR) cb (dotted line) for TCR surface expression levels. (b) For stimulation, T-cell hybridomas were cultured for 24 hours in precoated cell culture plates with titrated amounts of B22a1, TCR cb (positive control) and CD4 (negative control), whereupon IL-2 production was determined by ELISA. Ab, antibody; FU, fluorescence units. Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Page 6 of 14 tg mice had increased to 4.5% of CD4+ T cells in drain- ing lymph nodes, while remaining at a low frequency of 0.1% in littermate controls. clones with a sevenfold stronger staining against the 22a1-7E T-cell hybridoma clone (Table 1). Comparison of the amino acid sequence within the CDR3 region of the Va16 chain shows Met-106 (22a1-7E) and Tyr-108 (22a1-7E and 18b4-10D) to be involved in B22a1 recog- nition, as substitution of these to either Ile-106 (18b4- D10) or Asp-108 (3D4-1.2) leads to decreased and insig- nificant antibody binding, respectively. To determine the relative frequency of B22a1+CD4+ T cells among the CII-reactive CD4+ T cells we used the expression of CD154 (CD40L) as a generic marker of antigen-specific T cells [16]. Around 5 to 6% of CD4+ T cells expressed CD40L upon restimulation in vitro with the galactosylated CII peptide, and 70 to 80% of these were also positive for the B22a1 antibody (Figure 2b, middle panel). Roughly 20% of the B22a1+ T cells did not express CD40L, however, and hence appeared to remain in a nonactivated state (Figure 2b, right panel). Characterization of B22a1 specificity As expected, mice devoid of the transgenic CII-specific Vb12 chain mounted a much weaker immune response against the galactosylated CII peptide upon restimulation in vitro, and only very few B22a1 +CD40L+CD4+ T cells were identified. As reported ear- lier [8], the recall response of Vb12-tg mice was strongly biased to the galactosylated CII peptide with only a minimal response directed against the nonglycosylated CII peptide (denoted K264; Figure 2b, left panel insert). Nontransgenic littermates, on the other hand, mounted a low but significant response to both peptides. To test the stimulatory capability of the B22a1 anti- body, T-cell hybridomas were incubated with precoated B22a1 antibody and monitored for activation through IL-2 production (Figure 1b). In agreement with the staining data, B22a1 were highly effective in stimulating the 22a1-7E and 18b4-10 D clones. B22a1 could also sti- mulate the 3D4-1.2 clone at high concentration but failed to activate CII-specific clones not expressing the Va16/Vb12 TCR combination. In an inhibition assay, the B22a1 antibody could also inhibit activation of the 22a1-7E and 18b4-10 D clones as well as partially block activation of the 3D4-1.2 clone, but not that of unre- lated clones (data not shown), again showing that the B22a1 antibody is highly specific for the Va16/DBA/1- Vb12-tg TCR combination. Finally, the B22a1 antibody failed to stain Aq-restricted T cells from mice immu- nized with control (non-CII) antigens as well as an Aq- restricted T-cell hybridoma clone specific for pepsin (see Additional file 1). To further analyze the significance of B22a1+ T cells in Vb12-tg mice, lymph node cells from CII-primed Vb12-tg mice were depleted of B22a1+ T cells in vitro and alteration in the CII-specific recall response was subsequently determined (Figure 3). In vitro depletion of B22a1+ T cells resulted in a reduced recall response to CII and the galactosylated CII peptide by approximately 80%, hence confirming the ex vivo data showing B22a1+ T cells to predominate the CII-specific T-cell response in Vb12-tg mice. B22a1-binding T cells dominate CII-specific response in Vb12-tg mice Having shown that the B22a1 antibody recognizes CII- specific T-cell clones obtained from CII-immunized Vb12-tg mice, we next set out to determine the relative frequency of T cells expressing the clonotypic determi- nant among CII-reactive T cells in primed and non- primed Vb12-tg mice. In nonimmunized Vb12-tg mice, the frequency of B22a1+CD4+ T cells in peripheral lymph nodes and spleen was 0.08%, compared with around 0.007% in wildtype littermates (Figure 2a). Ten days after immunization, the frequency in primed Vb12- Table showing the staining index (SI) of B22a1 and the SI normalized to the T-cell receptor (TCR) expression, respectively. GalHyK264, CII(260-270) with a b-D- galactopyranosyl residue on L-hydroxylysine at position 264; GlcGalHyK264, CII(260-270) with an a-D-glucopyranosyl-(1 > 2)-b-D-galactopyranosyl residue on L- hydroxylysine at position 264; K264, CII(260-270) with a nonmodified lysine at position 264; TCR cb, anti-TCR cb antibody (clone H57-597). a SI normalized to TCR is multiplied by 1,000. dex (SI) of B22a1 and the SI normalized to the T-cell receptor (TCR) expression, respectively. GalHyK264, CII(260-270) with a b-D- -hydroxylysine at position 264; GlcGalHyK264, CII(260-270) with an a-D-glucopyranosyl-(1 > 2)-b-D-galactopyranosyl residue on L- y y y p y g py y b g py y K264, CII(260-270) with a nonmodified lysine at position 264; TCR cb, anti-TCR cb antibody (clone H57-597). a SI normalized to TCR CII-specific T cells expand in the priming phase and retract as the first arthritis symptoms appear Having shown that the B22a1+ T-cell population predo- minates the CII-specific T-cell population in Vb12-tg mice, we next analyzed the dynamics of antigen-specific Table 1 Binding specificities of the B22a1 antibody towards different T-cell receptors Clone TCR Va chain TCR Vb chain Antigen specificity SI TCR cb geometric mean SI normalized to TCRa 22a1-7E 16 12 GalHyK264 24.0 110 218.1 18b4-10D 16 12 GalHyK264 12.3 431 28.6 3D4-1.2 16 12 GalHyK264 and GlcGalHyK264 1.1 431 2.5 HCQ.3 16 8.1 GalHyK264 0.1 357 0.2 HM1R.2 4 12 GalHyK264 0.9 505 1.7 HCQ.4 4 12 K264 0.1 316 0.3 HRC.2 2 20 K264 0.3 591 0.5 Table showing the staining index (SI) of B22a1 and the SI normalized to the T-cell receptor (TCR) expression, respectively. GalHyK264, CII(260-270) with a b-D- galactopyranosyl residue on L-hydroxylysine at position 264; GlcGalHyK264, CII(260-270) with an a-D-glucopyranosyl-(1 > 2)-b-D-galactopyranosyl residue on L- hydroxylysine at position 264; K264, CII(260-270) with a nonmodified lysine at position 264; TCR cb, anti-TCR cb antibody (clone H57-597). a SI normalized to TCR is multiplied by 1,000. Table 1 Binding specificities of the B22a1 antibody towards different T-cell receptors Clone TCR Va chain TCR Vb chain Antigen specificity SI TCR cb geomet 1 Binding specificities of the B22a1 antibody towards different T-cell receptors Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Page 7 of 14 72.8±0.5 0.90±0.37 1.43±0.25 3.72±0.53 0.91±0.06 94.0±0.9 0.45±0.13 0.004±0.002 0.10±0.02 99.5±0.2 B22a1 CD40L Vβ12-tg WT 5.2±0.8 0.50±0.13 CD4 CD40L 0.25±0.04 0.40±0.08 GalHyK264 K264 K264 gated on CD40L gated on CD4 0.08±0.01 4.5±0.3 0.007±0.002 0.1±0.05 B22a1 CD4 Vβ12-tg WT Naive Day 10 post CII priming A B Figure 2 B22a1+ T cells dominate the collagen type II-specific T-cell expansion in Vb12 transgenic mice. Lymph nodes of naïve and collagen type II (CII)-primed mice 10 days post immunization were restimulated for 6 hours in the presence of GalHyK264 and Brefeldin A. Flow cytometry was subsequently performed to determine B22a1 frequencies and CD40L upregulation. (a) Dot plots show B22a1 mean frequency among CD4 cells in Vb12-transgenic (Vb12-tg) mice (n = 5) and wildtype (WT) littermates (n = 4 and n = 5) left untreated (naïve) or immunized 10 days earlier. (b) Left panel: percentage of CD4 T cells upregulating CD40L upon restimluation with GalHyK264 in immunized Vb12-transgenic mice and WT littermates. Insert: frequency of CD40L upregulation following stimulation with the nonmodified K264-peptide. CII-specific T cells expand in the priming phase and retract as the first arthritis symptoms appear Middle panel: distribution of B22a1 staining among CD4+CD40L+ T cells. Right panel: quadrant gates of CD4-gated T cells. Numbers indicate frequencies (mean 0.08±0.01 4.5±0.3 0.007±0.002 0.1±0.05 B22a1 CD4 Vβ12-tg WT Naive Day 10 post CII priming A 0.08±0.01 4.5±0.3 0.007±0.002 0.1±0.05 B22a1 CD4 Vβ12-tg WT Naive Day 10 post CII priming A A 72.8±0.5 0.90±0.37 1.43±0.25 3.72±0.53 0.91±0.06 94.0±0.9 0.45±0.13 0.004±0.002 0.10±0.02 99.5±0.2 B22a1 CD40L Vβ12-tg WT 5.2±0.8 0.50±0.13 CD4 CD40L 0.25±0.04 0.40±0.08 GalHyK264 K264 K264 gated on CD40L gated on CD4 B Figure 2 B22a1+ T cells dominate the collagen type II-specific T-cell expansion in Vb12 transgenic mice. Lymph nodes of naïve and collagen type II (CII)-primed mice 10 days post immunization were restimulated for 6 hours in the presence of GalHyK264 and Brefeldin A. Flow cytometry was subsequently performed to determine B22a1 frequencies and CD40L upregulation. (a) Dot plots show B22a1 mean frequency among CD4 cells in Vb12-transgenic (Vb12-tg) mice (n = 5) and wildtype (WT) littermates (n = 4 and n = 5) left untreated (naïve) or immunized 10 days earlier. (b) Left panel: percentage of CD4 T cells upregulating CD40L upon restimluation with GalHyK264 in immunized Vb12-transgenic mice and WT littermates. Insert: frequency of CD40L upregulation following stimulation with the nonmodified K264-peptide. Middle panel: distribution of B22a1 staining among CD4+CD40L+ T cells. Right panel: quadrant gates of CD4-gated T cells. Numbers indicate frequencies (mean ± standard error of the mean) of two individual mice per group, reproduced in at least two independent experiments (b). 72.8±0.5 0.90±0.37 1.43±0.25 3.72±0.53 0.91±0.06 94.0±0.9 0.45±0.13 0.004±0.002 CD40L Vβ12-tg WT 5.2±0.8 0.50±0.13 CD40L 0.25±0.04 0.40±0.08 GalHyK264 K264 K264 gated on CD40L gated on CD4 B 72.8±0.5 0.90±0.37 1.43±0.25 3.72±0.53 0.91±0.06 94.0±0.9 0.45±0.13 0.004±0.002 0.10±0.02 99.5±0.2 B22a1 CD40L Vβ12-tg WT 5.2±0.8 0.50±0.13 CD4 CD40L 0.25±0.04 0.40±0.08 GalHyK264 K264 K264 gated on CD40L gated on CD4 B Figure 2 B22a1+ T cells dominate the collagen type II-specific T-cell expansion in Vb12 transgenic mice. Lymph nodes of naïve and collagen type II (CII)-primed mice 10 days post immunization were restimulated for 6 hours in the presence of GalHyK264 and Brefeldin A. Flow cytometry was subsequently performed to determine B22a1 frequencies and CD40L upregulation. (a) Dot plots show B22a1 mean frequency among CD4 cells in Vb12-transgenic (Vb12-tg) mice (n = 5) and wildtype (WT) littermates (n = 4 and n = 5) left untreated (naïve) or immunized 10 days earlier. CII-specific T cells expand in the priming phase and retract as the first arthritis symptoms appear (b) Left panel: percentage of CD4 T cells upregulating CD40L upon restimluation with GalHyK264 in immunized Vb12-transgenic mice and WT littermates. Insert: frequency of CD40L upregulation following stimulation with the nonmodified K264-peptide. Middle panel: distribution of B22a1 staining among CD4+CD40L+ T cells. Right panel: quadrant gates of CD4-gated T cells. Numbers indicate frequencies (mean ± standard error of the mean) of two individual mice per group, reproduced in at least two independent experiments (b). Vβ12-tg WT 5.2±0.8 0.50±0.13 CD4 CD40L 0.25±0.04 0.40±0.08 GalHyK264 K264 K264 B B 72.8±0.5 0.90±0.37 1.43±0.25 3.72±0.53 0.91±0.06 94.0±0.9 0.45±0.13 0.004±0.002 0.10±0.02 99.5±0.2 B22a1 CD40L gated on CD40L gated on CD4 gated on CD4 cells dominate the collagen type II-specific T-cell expansion in Vb12 transgenic mice. Lymph nodes of naïve and Figure 2 B22a1+ T cells dominate the collagen type II-specific T-cell expansion in Vb12 transgenic mice. Lymph nodes of naïve and collagen type II (CII)-primed mice 10 days post immunization were restimulated for 6 hours in the presence of GalHyK264 and Brefeldin A. Flow cytometry was subsequently performed to determine B22a1 frequencies and CD40L upregulation. (a) Dot plots show B22a1 mean frequency among CD4 cells in Vb12-transgenic (Vb12-tg) mice (n = 5) and wildtype (WT) littermates (n = 4 and n = 5) left untreated (naïve) or immunized 10 days earlier. (b) Left panel: percentage of CD4 T cells upregulating CD40L upon restimluation with GalHyK264 in immunized Vb12-transgenic mice and WT littermates. Insert: frequency of CD40L upregulation following stimulation with the nonmodified K264-peptide. Middle panel: distribution of B22a1 staining among CD4+CD40L+ T cells. Right panel: quadrant gates of CD4-gated T cells. Numbers indicate frequencies (mean ± standard error of the mean) of two individual mice per group, reproduced in at least two independent experiments (b). Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Page 8 of 14 CPM Vβ12 depleted (4) Vβ12 non-depleted (4) CII K264 Gal HyK264 medium ConA 0 10000 20000 30000 50000 70000 90000 * * Figure 3 In vitro depletion of B22a1+ T cells reduces recall response to collagen type II. Vb12-transgenic mice were immunized with collagen type II (CII) and 10 days later lymph node cells were depleted of B22a1+ T cells in vitro, and alterations in the CII-specific [H3]thymidine incorporation were noted after 4 days of culture with lathyritic CII, K264 peptide, GalHyK264 peptide, medium alone and ConA. The anti-CII response is dominated by proinflammatory cytokine producing B22a1+ T cells during arthritis The anti-CII response is dominated by proinflammatory cytokine producing B22a1+ T cells during arthritis The anti-CII response is dominated by proinflammatory cytokine producing B22a1+ T cells during arthritis To extend our analysis on the dynamics of CII-specific T cells in Vb12-tg mice during development of CIA, we next investigated the activation status and production of proinflammatory cytokines of CII-specific T cells at dif- ferent time points after immunization with CII. We also wanted to include B22a1-CD4+ T cells in these analyses Immunization of Vb12-tg mice with heterologous CII emulsified in CFA induced an onset of arthritis around day 17 that rapidly progressed to severe disease within 0 5 10 15 20 25 30 35 40 0 1 2 3 4 5 6 7 0 10 20 30 40 50 WT LN Vβ12 LN WT spleen Vβ12 spleen WT mLN Vβ12 mLN Vβ12 mice (14) days post immunization % of CD4+ T cells Mean arthritis score Figure 4 Collagen type II-specific T cells expand in priming phase and retract when arthritis symptoms appear. The indicated numbers of Vb12-transgenic mice were immunized with collagen type II (CII) and were followed for development of arthritis (stars, referring to right y axis). In parallel, frequencies of B22a1+ T cells in CII-primed Vb12-transgenic mice and wildtype (WT) littermates were assessed by flow cytometry in draining lymph nodes (LNs, open squares and closed circles), spleen (open and closed triangles) and mesenteric lymph node (mLN, open circles and closed diamonds) (referring to left y axis). Data show the mean ± standard error of the mean of four individual mice (WT littermates), five individual mice (Vb12-transgenic) and 14 individual mice (day 35). 0 5 10 15 20 25 30 35 40 0 1 2 3 4 5 6 7 0 10 20 30 40 50 WT LN Vβ12 LN WT spleen Vβ12 spleen WT mLN Vβ12 mLN Vβ12 mice (14) days post immunization % of CD4+ T cells Mean arthritis score days post immunization Figure 4 Collagen type II-specific T cells expand in priming phase and retract when arthritis symptoms appear. The indicated numbers of Vb12-transgenic mice were immunized with collagen type II (CII) and were followed for development of arthritis (stars, referring to right y axis). CII-specific T cells expand in the priming phase and retract as the first arthritis symptoms appear Data show mean ± standard error of the mean of four individual mice before and after depletion. P < 0.05. CPM, counts per minute. the following 2 weeks (Figure 4, right y axis stars). Ex vivo analyses clearly showed that CII priming at the base of the tail triggered a primary immune response in the draining inguinal lymph nodes, leading to an increased frequency from initially 0.08% of B22a1+CD4+ T cells in nonimmunized animals up to 4.5% within the first 10 days (Figure 4, left y axis). Twenty days post immunization, however, the B22a1+CD4+ T-cell fre- quency subsequently dropped again to 2.2% and slowly stabilized around 1.7% at day 35 (Figure 4, left y axis). A similar pattern could be observed in the spleen, albeit with a somewhat lower frequency. Measurement of the mesenteric lymph node, however, showed that only 0.2% of B22a1+CD4+ T cells infiltrated this organ at peak day 10-indicating that the homing B22a1+CD4+ T cells did not occur at random and that the mesenteric lymph node is not involved in the arthritis-specific immune mechanisms, since these cells had retracted to approxi- mate base levels again at day 20. It was also possible to detect significantly increased frequencies of B22a1+ T cells in wildtype littermates 10 days after immunization (Figure 4). Frequencies were profoundly reduced (0.1 and 0.3% in lymph nodes and spleen, respectively), however, compared with those observed in Vb12-tg mice. CPM Vβ12 depleted (4) Vβ12 non-depleted (4) CII K264 Gal HyK264 medium ConA 0 10000 20000 30000 50000 70000 90000 * Figure 3 In vitro depletion of B22a1+ T cells reduces recall response to collagen type II. Vb12-transgenic mice were immunized with collagen type II (CII) and 10 days later lymph node cells were depleted of B22a1+ T cells in vitro, and alterations in the CII-specific [H3]thymidine incorporation were noted after 4 days of culture with lathyritic CII, K264 peptide, GalHyK264 peptide, medium alone and ConA. Data show mean ± standard error of the mean of four individual mice before and after depletion. P < 0.05. CPM, counts per minute. Figure 3 In vitro depletion of B22a1+ T cells reduces recall Figure 3 In vitro depletion of B22a1+ T cells reduces response to collagen type II. Vb12-transgenic mice were response to collagen type II. CII-specific T cells expand in the priming phase and retract as the first arthritis symptoms appear Vb12 transgenic mice were immunized with collagen type II (CII) and 10 days later lymph node cells were depleted of B22a1+ T cells in vitro, and alterations in the CII-specific [H3]thymidine incorporation were noted after 4 days of culture with lathyritic CII, K264 peptide, GalHyK264 peptide, medium alone and ConA. Data show mean ± standard error of the mean of four individual mice before and after depletion. P < 0.05. CPM, counts per minute. T cells during CIA by determining the frequency of the B22a1+ T cells at different time points after induction of CIA. To relate the frequency of antigen-specific T cells to development of arthritis, a cohort of Vb12-tg mice was immunized with CII and followed for clinical dis- ease in parallel to the ex vivo analyses of representative animals. Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 interested in evaluating the quality of the CII-specific T- cell response over time with regard to production of proinflammatory cytokines, as these have previously been found to dominate the response in Vb12-tg upon immunization with CII in CFA [8,20]. We therefore stratified the B22a1+ and B22a1- T-cell responses according to their production of IL-2, IFNg and IL-17 to detect whether the relative abundance of these cells would shift over time among the CII-specific CD40L- expressing T cells (Figure 5b). Indeed, stratification of CII-specific T cells showed that both B22a1+ and B22a1- T cells mounted an IFNg and IL-2 response that was maintained in the draining lymph node from day 10 post immunization and throughout the development of clinical arthritis. Compensating for the reduction in the total frequency of CII-specific T cells from day 10 onwards (Figure 5a), however, showed that the fre- quency of IL-2-producing and IFNg-producing T cells among CD4+ T cells was highest at the priming phase (day 10) and then decreased with the progressing to ensure that our ex vivo data on the frequency of B22a1+CD4+ T cells could be extrapolated to the whole CII-specific T-cell repertoire in Vb12-tg mice. To achieve this, CII-primed Vb12-tg mice were sacrificed at three different time points during CIA. Pooled draining lymph nodes and the spleen were then analyzed for CD40L and cytokine expression in the priming phase (day 10), shortly after arthritis onset (day 25) and, finally, during severe clinical arthritis (day 35). The CD40L expression in CD4+ T cells followed the same pattern as observed for the frequency of B22a1+ CD4+ T cells, with the most frequent expression detected at day 10 and with fading frequencies at later time points. Importantly, the B22a1+CD4+ T cells pre- vailed as the predominant CII-specific T-cell population, constituting 75 to 80% of all CD40L+CD4+ T cells, throughout the disease course (Figure 5a). Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 g g Having shown that the frequency of activated CII-spe- cific T cells in the draining lymph nodes decline from day 10 and onwards (Figures 4 and 5a), we were also B22a1 CD40L CD4 CD40L CD40L B22a1+ B22a1- 0 1 2 3 4 5 day 10 day 25 day 35 % of CD4+ T cells B22a1 - + A B CD40L B22a1 cytokine 0 2 4 6 8 10 12 14 16 18 day 10 day 25 day 35 B22a1 Cytokine - - - + + + IL-2 IL-2 IFN IFN IL-17 IL-17 * * * ** * * *** * % of CD4+CD40L+ T cells * * * Figure 5 B22a1+ T cells dominate during arthritis and are the main producers of proinflammatory cytokines. Vb12-transgenic mice were primed with collagen type II (CII) and lymphocytes were analyzed 10, 25 and 35 days after immunization, following 6 hours of restimulation in the presence of the GalHyK264 peptide and Brefeldin A. (a) Representative dot plot from day 35 analysis and a summarizing bar graph showing the frequencies of CD4+ T cells being B22a1-and B22a1+ and expressing CD40L. (b) Cytokine production of activated CII-specific T cells, determined by gating on total CD40L+CD4+ T cells. Bars represent mean percentage (± standard error of the mean) of cytokine-producing cells among CD4+CD40L+ T cells, stratified according to B22a1 expression. IL-2 bars show frequency of cells producing IL-2 only, whereas IFNg and IL- 17 bars include IFNg and IL17 single positives as well as IFNg/IL-2 and IL-17/IL-2 double positives, respectively. The IFNg and IL-17 bars are not adjusted for IFNg/IL-17 double-positive cells due to their low abundance. All plots include 12 mice, eight mice and five mice for day 10, day 25 and day 35, respectively. P < 0.05, P < 0.01, P < 0.001. B22a1 CD40L CD40L B22a1+ B22a1- A A 0 1 2 3 4 5 day day day % of CD4+ T cells B22a1 - + * * * ** CD4 CD40L B CD40L B B22a1 cytokine 0 2 4 6 8 10 12 14 16 18 day 10 day 25 day 35 B22a1 Cytokine - - - + + + IL-2 IL-2 IFN IFN IL-17 IL-17 * * * ** * * *** * % of CD4+CD40L+ T cells B22a1 Figure 5 B22a1+ T cells dominate during arthritis and are the main producers of proinflammatory cytokines. The anti-CII response is dominated by proinflammatory cytokine producing B22a1+ T cells during arthritis In parallel, frequencies of B22a1+ T cells in CII-primed Vb12-transgenic mice and wildtype (WT) littermates were assessed by flow cytometry in draining lymph nodes (LNs, open squares and closed circles), spleen (open and closed triangles) and mesenteric lymph node (mLN, open circles and closed diamonds) (referring to left y axis). Data show the mean ± standard error of the mean of four individual mice (WT littermates), five individual mice (Vb12-transgenic) and 14 individual mice (day 35). Page 9 of 14 Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Still, 60 to 85% of the antigen- specific response could not be accounted for by We also investigated IgM, IgG1, IgG2a and IgG2b anti- CII antibodies following antibody treatment, to see whether depletion of B22a1+CD4+ T cells would result in a skewing in the quality of the antibody response. A significant reduced IgM response was observed in mice 0 500 1000 1500 2000 2500 3000 3500 4000 4500 Day 17 Day 35 control (5) depleted (6) aCII IgG total (μg/ml) A B C D * 12 15 18 21 24 27 30 0 20 40 60 80 100 control (15) depleted (16) * days post immunization % Incidence 0.0 0.2 0.4 0.6 * * B22a1 - + control depleted % of CD4+ T cells 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Lymph node Spleen control depleted * *** % of CD4+ T cells Figure 6 Efficient depletion of B22a1+ T cells before arthritis onset has limited effect on clinical disease. (a) The indicated numbers of Vb12-transgenic male mice were treated with B22a1 and MAR18.5 antibodies 4 hours prior to immunization with collagen type II (CII) and followed for development of arthritis. (b) Frequencies of B22a1+ T cells by the end of the arthritis experiment were assessed by flow cytometry in the lymph nodes and spleen. (c) Anti-CII IgG antibody levels at days 17 and 35 after immunization are shown as box plots. (d) Scatter plot showing the frequencies of CD4+ T cells being B22a1- and B22a1+ and expressing CD40L after 6 hours of restimulation in the presence of the GalHyK264 peptide and Brefeldin A by the end of the arthritis experiment at day 35. P < 0.05, P < 0.01, P < 0.001. B 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Lymph node Spleen control depleted * *** % of CD4+ T cells A 12 15 18 21 24 27 30 0 20 40 60 80 100 control (15) depleted (16) * days post immunization % Incidence A B 0 500 1000 1500 2000 2500 3000 3500 4000 4500 Day 17 Day 35 control (5) depleted (6) aCII IgG total (μg/ml) C * D 0.0 0.2 0.4 0.6 * *** B22a1 - + control depleted % of CD4+ T cells C D epletion of B22a1+ T cells before arthritis onset has limited effect on clinical disease. Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Page 10 of 14 Page 10 of 14 arthritis in both B22a1+ and B22a1- CD4+ T cells (data not shown). In marked contrast, CII-specific IL-17 pro- ducing T cells occurred with low frequency in the prim- ing phase and instead peaked at the time of disease onset whereupon the frequency dropped again to low levels by the time of severe disease (Figure 5b). This delay in occurrence was further emphasized by compar- ing the frequency of IL-17-producing T cells among CD4+ T cells (data not shown). measuring the production of IL-2, IL-17 and IFNg. This observation suggests that additional effector and regula- tory functions are probably operating through the remaining CII-specific T cells in order to control the progression and regulation of disease. Data thus far suggest that both B22a1+ and B22a1- CD4+ T cells could be involved in the development of CIA in Vb12-tg mice. To confirm this, we depleted B22a1+CD4+ T cells in vivo prior to immunization with CII and subsequently observed the mice for alternation in the development of CIA. Indeed, in vivo treatment with the B22a1 antibody was efficient in depleting B22a1+CD4+ T cells, as only very few B22a1+ T cells could be identified in the spleen and lymph nodes by the end of the arthritis experiment (Figure 6b) and caused a delay in disease onset (P < 0.05; Figure 6a). In agreement with clinical disease, analysis of the anti-CII antibody titers from the experiments also showed an initial delay of antibody production that was later recov- ered as arthritis had developed (Figure 6c). Importantly, cytokine production was only detected in cells that had upregulated CD40L. In addition, on aver- age 0.4%, 1.6% and 0.6% of the CD40L-expressing T cells were found to produce both IL-17 and IFNg dur- ing the priming, disease-onset and later phases, respec- tively (data not shown). Determination of the frequency of IL-2-producing, IL-17-producing and IFNg-producing T cells, however, could only account for 20 to 35% of the total CII-specific T-cell population. These data col- lectively show that, following priming in vivo, the CII- specific T-cell response in the draining lymph nodes is initially dominated by Th1. As the overall T-cell response subsequently fades, however, the T-cell response shifts to also include Th17 by the time of onset of clinical disease. Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Vb12-transgenic mice were primed with collagen type II (CII) and lymphocytes were analyzed 10, 25 and 35 days after immunization, following 6 hours of restimulation in the presence of the GalHyK264 peptide and Brefeldin A. (a) Representative dot plot from day 35 analysis and a summarizing bar graph showing the frequencies of CD4+ T cells being B22a1-and B22a1+ and expressing CD40L. (b) Cytokine production of activated CII-specific T cells, determined by gating on total CD40L+CD4+ T cells. Bars represent mean percentage (± standard error of the mean) of cytokine-producing cells among CD4+CD40L+ T cells, stratified according to B22a1 expression. IL-2 bars show frequency of cells producing IL-2 only, whereas IFNg and IL- 17 bars include IFNg and IL17 single positives as well as IFNg/IL-2 and IL-17/IL-2 double positives, respectively. The IFNg and IL-17 bars are not adjusted for IFNg/IL-17 double-positive cells due to their low abundance. All plots include 12 mice, eight mice and five mice for day 10, day 25 and day 35, respectively. P < 0.05, P < 0.01, *P < 0.001. Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Discussion Following priming of Vb12-tg mice with CII in vivo, we observed a 50-fold increase in the frequency of B22a1+ T cells - and the majority (80%) of, but not all, B22a1+ T cells would acquire an activated phenotype. While our data consistently indicated CII-specific B22a1 + and B22a1- T cells to exhibit almost identical pheno- types and only differ in their frequency of occurrence, it may be assumed that a similar proportion of nonacti- vated T cells can be found within the CII-specific B22a1- fraction. The lack of activation in 20% of the cells is probably explained by the nonphysiological fre- quency of CII-specific T cells, leading to competition for APCs, as has been demonstrated in other studies [32-34]. Owing to the limited number of APCs and the accessibility of MHCs and costimulatory molecules in vitro, a more stringent selection for activation may be imposed despite an excess of the galactosylated CII peptide. The galactosylated CII(260-270) peptide has been shown to constitute the immunodominant T-cell epitope in Aq-expressing mice [12,15], and T cells specific for the galactosylated variant of the CII(260-270) epitope have been shown to be of major importance for development of arthritis [9]. In the present report, we have used the Vb12-tg mouse to establish a system where it is possible to track T cells specific for galactosylated CII peptide. The Vb12-tg mouse is so far the only described TCR-tg mouse with increased cellular immunity to galactosy- lated CII and increased susceptibility to CIA [6,8]. Despite the transgenic Vb12 chain being expressed on virtually all CD3+ T cells, however, it may still recom- bine freely with endogenously derived a-chains, and the frequency of CII-specific T cells in the Vb12-tg mouse was therefore not known. Recombinant MHC class II tetramers/multimers pre- senting CII peptides have previously been used to iden- tify CII-specific T cells ex vivo in Aq-expressing [21,22], DR4-expressing [23], and DR1-expressing mice [24]. Successful identification of CII-specific T cells in Aq- expressing mice, however, required that the CII peptide had been covalently linked to the recombinant MHC class II molecule. Covalent linkage of the peptide is used in order to increase the stability of the CII peptide: MHC class II molecule complex [25]. Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 (a) The indicated numbers o Figure 6 Efficient depletion of B22a1+ T cells before arthritis onset has limited effect on clinical disease. (a) The indicated numbers of Vb12-transgenic male mice were treated with B22a1 and MAR18.5 antibodies 4 hours prior to immunization with collagen type II (CII) and followed for development of arthritis. (b) Frequencies of B22a1+ T cells by the end of the arthritis experiment were assessed by flow cytometry in the lymph nodes and spleen. (c) Anti-CII IgG antibody levels at days 17 and 35 after immunization are shown as box plots. (d) Scatter plot showing the frequencies of CD4+ T cells being B22a1- and B22a1+ and expressing CD40L after 6 hours of restimulation in the presence of the GalHyK264 peptide and Brefeldin A by the end of the arthritis experiment at day 35. P < 0.05, P < 0.01, *P < 0.001. Page 11 of 14 Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Page 11 of 14 Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 depleted of B22a1+CD4+ T cells on day 17 (P = 0.009) but not on day 35 (P = 0.75; data not shown). Even though levels of anti-CII IgG1, IgG2a and IgG2b levels were uniformly reduced in depleted mice at day 17, the differences did not reach statistical significance (P = 0.178, P = 0.082 and P = 0.125 for IgG1, IgG2a and IgG2b, respectively; data not shown). Analysis of the fre- quency of CII-specific T cells by the end of the arthritis experiments showed that the B22a1-CD4+ T cells in depleted mice had expanded compared with nontreated control mice (P = 0.038). Together with the data on antibody isotype levels and the stratified cytokine pro- files, this suggests that B22a1- and B22a1+ T cells exhi- bit similar qualitatively effector functions and that B22a1- T cells indeed could take over the role of the depleted B22a1+ T cells in Vb12-tg mice (Figure 6d). T cells constitute only a minority of the total TCR repertoire. Still, by performing depletion experiments in vitro and through combinatory analysis of the fre- quencies of clonotypic T cells and the activation status of the total CII-specific T-cell population ex vivo, we found that the B22a1 antibody recognizes the vast majority of the CII-specific T cells in Vb12-tg mice. Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 The frequency of B22a1+ T cells in naïve Vb12-tg mice (~0.1%) was found to be considerably lower than that reported for other TCRb-tg mice using MHC class II tetramer/multimers (1 to 3% [28-30], but still around 10 times higher than in wildtype littermates). This is a considerably higher frequency than that estimated for a specific T-cell clone in a naïve wildtype mouse [31], but may partly be explained by the fact the B22a1 antibody could recognize T cells with closely related but distinct TCRs. Discussion This is presently not a realistic option for the galactosylated CII peptide as hydroxylation of lysine residues of recombinant CII is absent or poor in prokaryotic, yeast and insect cell sys- tems [26,27] but is necessary for subsequent galactosyla- tion of hydroxylysine residues. Because of the poor feasibility to detect galactosylated CII-specific T cells using the MHC class II tetramer technology, we instead decided to generate a clonotypic antibody recognizing T cells specific for the galactosylated CII peptide in the Vb12-tg mouse. Hereby we could show that B22a1+ Our results support an important role for CII-specific T cells in the early phase of the arthritogenic immune response (reviewed in [35]), where depletion of the pre- dominant B22a1+ T-cell population caused a delayed onset of clinical arthritis. CII-specific T cells not tar- geted by the clonotypic antibody were found to expand rapidly and to partially replace the depleted B22a1+ T- cell subset, however, which was also evident when ana- lyzed at the end of the arthritis experiment. It is of note that depletion with the B22a1 antibody before priming with CII caused a strong and rapid as well as longlasting reduction in the frequency of B22a1+ T cells. We found no evidence that antibody treatment would cause TCR downregulation, as has been shown for gδ-T cells after treatment with a monoclonal antibody specific for the constant region of the gδ-TCR and which caused down- regulation of the TCR and generation of persistent invi- sible clones with reduced responsiveness [36] (data not shown). Moreover, the monoclonal antibody MAR18.5 Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Page 12 of 14 Page 12 of 14 mediating effector functions and regulating the initiation or progression of CIA is challenging, however, as the lymphoid response varies over time and between lym- phoid organs and the joints [47,48]. These factors, together with different techniques used for analyses of antigen-specific responses, probably help to explain some of the existing discrepancies with regard to which cell populations and cytokine responses predominate in lymphoid tissues and target organs during CIA [45-50]. Further analysis using the currently described animal model may help elucidate some of these matters. Conclusions Because of the increased precursor frequency of CII-spe- cific T cells in the Vb12-tg mouse and the availability of an antibody recognizing the majority of these cells, the currently described animal model offers a unique possi- bility to further elucidate the role of CII-specific T cells as well as their different subsets during the development and regulation of CIA at different time points and in different tissues. The B22a1 antibody allows for highly sensitive analysis of joint-residing T cells specific for galactosylated CII by flow cytometry, due to its specifi- city and relatively strong affinity to its cognate TCR, compared with MHC class II tetramers. Additional material Additional file 1: B22a1 antibody does not bind to T cells with other tested specificities than CII. Wildtype mice were immunized with either pepsin or ovalbumin (OVA) emulsified in CFA. Seven days later, draining lymph node cells were restimulated in vitro with the immunogenic protein (third and fourth upper graphs from the left) as well as with purified protein derivate (tuberculin-purified protein derivative (PPD), mycobacterial antigens included in CFA; second upper panel from the left) or left unstimulated (upper left panel). After 6 hours of restimulation, cells were recovered and stained for expression of CD4, CD154 and B22a1. The Aq-restricted pepsin-specific T-cell hybridoma HP1 was tested for staining with the B22a1 antibody (lower right panel), and staining of the 22a1-7E T-cell hybridoma clone was stained in parallel as positive control (lower left panel). Representative graphs for each restimulation condition is shown and numbers within the quadrants of the upper graphs shows the mean ± standard error of the mean of three individual mice per group immunized and restimulated in vitro with either pepsin or OVA and of all six mice restimulated in vitro with PPD or left unstimulated. The role of IFNg in CIA is complex and has been associated with both proinflammatory and regulatory functions [38-41], and more recent reports have chal- lenged the concept that CIA is a Th1-mediated disease and have instead suggested the model to be more dependent on Th17 [42-45]. While determined ex vivo at the level of antigen-specific T cells, our findings on the relative abundance of IFNg-responding and IL-17- responding T cells over time add support to the recent report by Lamacchia and colleagues in which investiga- tions of the recall responses of bulk lymph node cells to CII in CIA indicated a shift from an initially dominant Th1 response to a mixed Th1/Th17 response [46]. Determining the role of CII-specific Th1 and Th17 in Discussion We have successfully used the B22a1 antibody for immuno- histochemical identification of B22a1+ T cells in lymph nodes and spleens of CII-primed Vb12-tg mice (data not shown), and we are currently investigating the possibility to stain T cells residing in joints of arthritic animals. was administered as a secondary antibody after B22a1 to increase the depletion capacity as previously shown [17]. Finally, sham-operated and thymectomized mice sub- jected to antibody treatment and subsequent immuniza- tion appeared equally protected from CIA, compared with nondepleted mice, excluding a role for recent B22a1+CD4+ thymic emigrants in the later-occurring arthritis (data not shown). Taking these observations together, therefore, CII-specific B22a1- T cells that are not depleted by the B22a1 antibody seem to be suffi- cient to trigger CIA. Although not conclusively demonstrated, an active role for CII-specific T cells during clinical arthritis is indirectly supported by their presence in the arthritic joints [23,24], and also by experimental therapies specifi- cally targeting Th1 and Th17 during active CIA [35,37]. The inability of the clonotypic antibody to recognize all CII-specific T cells in the Vb12-tg mouse, however, pre- vented us from addressing the role of T cells in the later phases of arthritis development, by means of T-cell depletion studies. Depletion of B22a1+ T cells after immunization with CII caused an even more rapid expansion (within days) of CII-specific B22a1- T cells, and the depletion of B22a1+ T cells was found to be less efficient in draining lymph nodes, compared with the spleen or blood (data not shown). Using combinatorial analysis of expression of the early activation marker CD40L and intracellular cytokines ex vivo, however, we could determine the frequency of CII-specific T cells as well as their cytokine profile at different time points after induction. Our data clearly showed that the fre- quency of CII-specific T cells peaked before onset of clinical disease. In line with this, the frequency of IFNg- producing CII-specific T cells among CD4+ T cells also peaked before onset of clinical arthritis and subse- quently declined with progression of disease. In contrast, the frequency of CII-specific IL-17-producing T cells among CD4+ T cells was less abundant in the early response, and instead increased in frequency to peak at the time of disease onset. References 21. Huang JC, Vestberg M, Minguela A, Holmdahl R, Ward ES: Analysis of autoreactive T cells associated with murine collagen-induced arthritis using peptide-MHC multimers. Int Immunol 2004, 16:283-293. 1. Holmdahl R, Klareskog L, Rubin K, Larsson E, Wigzell H: T lymphocytes in collagen II-induced arthritis in mice. Characterization of arthritogenic collagen II-specific T-cell lines and clones. Scand J Immunol 1985, 22:295-306. 1. Holmdahl R, Klareskog L, Rubin K, Larsson E, Wigzell H: T lymphocytes in collagen II-induced arthritis in mice. Characterization of arthritogenic collagen II-specific T-cell lines and clones. Scand J Immunol 1985, 22:295-306. 22. Backlund J, Treschow A, Firan M, Malmstrom V, Issazadeh-Navikas S, Ward ES, Holmdahl R: Reversal of tolerance induced by transplantation of skin expressing the immunodominant T cell epitope of rat type II collagen entitles development of collagen-induced arthritis but not graft rejection. Eur J Immunol 2002, 32:1773-1783. 2. Nandakumar KS, Backlund J, Vestberg M, Holmdahl R: Collagen type II (CII)- specific antibodies induce arthritis in the absence of T or B cells but the arthritis progression is enhanced by CII-reactive T cells. Arthritis Res Ther 2004, 6:R544-R550. 2. Nandakumar KS, Backlund J, Vestberg M, Holmdahl R: Collagen type II (CII)- specific antibodies induce arthritis in the absence of T or B cells but the arthritis progression is enhanced by CII-reactive T cells. Arthritis Res Ther 2004, 6:R544-R550. j 23. Svendsen P, Andersen CB, Willcox N, Coyle AJ, Holmdahl R, Kamradt T, Fugger L: Tracking of proinflammatory collagen-specific T cells in early and late collagen-induced arthritis in humanized mice. J Immunol 2004, 173:7037-7045. 3. Lafaille JJ: T-cell receptor transgenic mice in the study of autoimmune diseases. J Autoimmun 2004, 22:95-106. 3. Lafaille JJ: T-cell receptor transgenic mice in the study of autoimmune diseases. J Autoimmun 2004, 22:95-106. 4. Brand DD, Myers LK, Whittington KB, Latham KA, Stuart JM, Kang AH, Rosloniec EF: Detection of early changes in autoimmune T cell phenotype and function following intravenous administration of type II collagen in a TCR-transgenic model. J Immunol 2002, 168:490-498. 24. Latham KA, Whittington KB, Zhou R, Qian Z, Rosloniec EF: Ex vivo characterization of the autoimmune T cell response in the HLA-DR1 mouse model of collagen-induced arthritis reveals long-term activation of type II collagen-specific cells and their presence in arthritic joints. J Immunol 2005, 174:3978-3985. 5. Abbreviations APC: antigen-presenting cell; CDR3: complementarity determining region; CFA: complete Freund’s adjuvant; CIA: collagen-induced arthritis; CII: collagen type II; DMEM: Dulbecco’s modified Eagle’s medium; ELISA: enzyme-linked immunosorbent assay; FCS: fetal calf serum; IFN: interferon; GALHYK264: CII (260-270) with a b-D-galactopyranosyl residue on L-hydroxylysine at position Page 13 of 14 Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Author details 1 13. Krebber A, Bornhauser S, Burmester J, Honegger A, Willuda J, Bosshard HR, Pluckthun A: Reliable cloning of functional antibody variable domains from hybridomas and spleen cell repertoires employing a reengineered phage display system. J Immunol Methods 1997, 201:35-55. Author details 1Section for Medical Inflammation Research, Department of Medical Biochemistry and Biophysics, Karolinska Institute, Scheeles väg 2, 171 77 Stockholm, Sweden. 2Section for Medical Inflammation Research, Department of Experimental Medical Sciences, Lund University, Sölvegatan 19, 22184 Lund, Sweden. 3Nikolaus-Fiebiger-Center for Molecular Medicine, Department of Experimental Medicine I, University Erlangen-Nürnberg, Glückstrasse 6, 91054 Erlangen, Germany. 4Division of Rheumatology, Johann Wolfgang Goethe University Frankfurt, Theodor-Stern-Kai 7, 60590 Frankfurt am Main, Germany. 1Section for Medical Inflammation Research, Department of Medical Biochemistry and Biophysics, Karolinska Institute, Scheeles väg 2, 171 77 Stockholm, Sweden. 2Section for Medical Inflammation Research, Department of Experimental Medical Sciences, Lund University, Sölvegatan 19, 22184 Lund, Sweden. 3Nikolaus-Fiebiger-Center for Molecular Medicine, Department of Experimental Medicine I, University Erlangen-Nürnberg, Glückstrasse 6, 91054 Erlangen, Germany. 4Division of Rheumatology, Johann Wolfgang Goethe University Frankfurt, Theodor-Stern-Kai 7, 60590 Frankfurt am Main, Germany. 1Section for Medical Inflammation Research, Department of Medical Biochemistry and Biophysics, Karolinska Institute, Scheeles väg 2, 171 77 Stockholm, Sweden. 2Section for Medical Inflammation Research, Department of Experimental Medical Sciences, Lund University, Sölvegatan 19, 22184 Lund, Sweden. 3Nikolaus-Fiebiger-Center for Molecular Medicine, Department of Experimental Medicine I, University Erlangen-Nürnberg, 4 g y y 14. Arenaz P, Sirover MA: Isolation and characterization of monoclonal antibodies directed against the DNA repair enzyme uracil DNA glycosylase from human placenta. Proc Natl Acad Sci USA 1983, 80:5822-5826. y Glückstrasse 6, 91054 Erlangen, Germany. 4Division of Rheumatology, Johann Wolfgang Goethe University Frankfurt, Theodor-Stern-Kai 7, 60590 Frankfurt am Main, Germany. 15. Corthay A, Backlund J, Broddefalk J, Michaelsson E, Goldschmidt TJ, Kihlberg J, Holmdahl R: Epitope glycosylation plays a critical role for T cell recognition of type II collagen in collagen-induced arthritis. Eur J Immunol 1998, 28:2580-2590. Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 6. Mori L, Loetscher H, Kakimoto K, Bluethmann H, Steinmetz M: Expression of a transgenic T cell receptor beta chain enhances collagen-induced arthritis. J Exp Med 1992, 176:381-388. 264; GLCGALHYK264: CII(260-270) with an a-D-glucopyranosyl-(1 > 2)-b-D- galactopyranosyl residue on L-hydroxylysine at position 264; IL: interleukin; K264: CII(260-270) with a nonmodified lysine at position 264; PBS: phosphate-buffered saline; PCR: polymerase chain reaction; TCR: T-cell receptor; TG: transgenic; TH: T-helper cell. 7. Yamada H, Dzhambazov B, Bockermann R, Blom T, Holmdahl R: A transient post-translationally modified form of cartilage type II collagen is ignored by self-reactive T cells. J Immunol 2004, 173:4729-4735. Acknowledgements 8. Backlund J, Nandakumar KS, Bockermann R, Mori L, Holmdahl R: Genetic control of tolerance to type II collagen and development of arthritis in an autologous collagen-induced arthritis model. J Immunol 2003, 171:3493-3499. The present work was supported by the following grants to JB: the King Gustaf V’s 80-Year Foundation (also granted to TB, RB and BD), Anna Greta Crafoord Foundation for Rheumatology Research, Åke Wiberg Foundation, Greta and Johan Kocks Foundation, Magnus Bergvalls Foundation, Nanna Svartz Foundation, Österlund Foundation, Royal Physiographic Society in Lund, Tore Nilsons Foundation, The Swedish Association Against Rheumatism, The Swedish Research Council, the EU FP6 project AutoCure LSHB-CT-2006-018661, and the EU FP6 project MUGEN LSHG-CT-2005- 005203. The work was also supported by the EU FP6 project EURO-RA MRTN-CT-2004-0005693 (Marie Curie Fellowship to PM), the EU FP7 project Masterswitch HEALTH-F2-2008-223404 (grants to JB and BD), and the Doktor Robert Pfleger-Foundation (grant to HB, support for consumables). The funding bodies had no influence on any other aspect related to the present paper. The authors thank Rikard Holmdahl for support and discussions enabling the study and for critically reading the manuscript. They also thank Carlos Palestro, Kristina Palestro and Tomasz Klaczkowski for taking care of the animals, and Emma Mondoc and Malin Neptin for CII preparation. 9. Backlund J, Treschow A, Bockermann R, Holm B, Holm L, Issazadeh- Navikas S, Kihlberg J, Holmdahl R: Glycosylation of type II collagen is of major importance for T cell tolerance and pathology in collagen- induced arthritis. Eur J Immunol 2002, 32:3776-3784. 10. Smith BD, Martin GR, Miller EJ, Dorfman A, Swarm R: Nature of the collagen synthesized by a transplanted chondrosarcoma. Arch Biochem Biophys 1975, 166:181-186. 11. Andersson M, Holmdahl R: Analysis of type II collagen-reactive T cells in the mouse. I. Different regulation of autoreactive vs. non-autoreactive anti-type II collagen T cells in the DBA/1 mouse. Eur J Immunol 1990, 20:1061-1066. 12. Michaelsson E, Andersson M, Engstrom A, Holmdahl R: Identification of an immunodominant type-II collagen peptide recognized by T cells in H-2q mice: self tolerance at the level of determinant selection. Eur J Immunol 1992, 22:1819-1825. Received: 25 May 2010 Revised: 16 July 2010 Accepted: 3 August 2010 Published: 3 August 2010 Received: 25 May 2010 Revised: 16 July 2010 Accepted: 3 August 2010 Published: 3 August 2010 20. Bockermann R, Holmdahl R: Type II collagen without adjuvant induces eosinophilic arthritis. Eur J Immunol 2007, 37:540-548. Authors’ contributions 16. Frentsch M, Arbach O, Kirchhoff D, Moewes B, Worm M, Rothe M, Scheffold A, Thiel A: Direct access to CD4+ T cells specific for defined antigens according to CD154 expression. Nat Med 2005, 11:1118-1124. BS and HB contributed to the production of recombinant 22a1-7E TCR. RB contributed to the generation of T-cell hybridoma and sequencing of their CDR3 regions. PM, TB and BD contributed to CIA induction and evaluation. JB and PM contributed to generation and production of the monoclonal antibody B22a1. PM contributed to characterization of B22a1 specificity, flow cytometry analysis and T-cell stimulation analysis. PM and JB designed the study and drafted the manuscript, and all authors revised the manuscript. All authors read and approved the final manuscript. BS and HB contributed to the production of recombinant 22a1-7E TCR. RB contributed to the generation of T-cell hybridoma and sequencing of their CDR3 regions. PM, TB and BD contributed to CIA induction and evaluation. JB and PM contributed to generation and production of the monoclonal antibody B22a1. PM contributed to characterization of B22a1 specificity, flow cytometry analysis and T-cell stimulation analysis. PM and JB designed the study and drafted the manuscript, and all authors revised the manuscript. All authors read and approved the final manuscript. 17. Goldschmidt TJ, Holmdahl R, Klareskog L: Depletion of murine T cells by in vivo monoclonal antibody treatment is enhanced by adding an autologous anti-rat kappa chain antibody. J Immunol Methods 1988, 111:219-226. 18. Holmdahl R, Carlsen S, Mikulowska A, Vestberg M, Brunsberg U, Hansson AS, Sundvall M, Jansson L, Pettersson U: Genetic analysis of mouse models for rheumatoid arthritis. Human Genome Methods New York: CRC PressAdolph KW 1998, 215-238. The authors declare that they have no competing interests. 19. Holmdahl R, Klareskog L, Andersson M, Hansen C: High antibody response to autologous type II collagen is restricted to H-2q. Immunogenetics 1986, 24:84-89. Competing interests Th h d l h The authors declare that they have no competing interests. Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Zehn D, Bevan MJ: T cells with low avidity for a tissue-restricted antigen routinely evade central and peripheral tolerance and cause autoimmunity. Immunity 2006, 25:261-270. 49. Roark CL, French JD, Taylor MA, Bendele AM, Born WK, O’Brien RL: Exacerbation of collagen-induced arthritis by oligoclonal, IL-17- producing gamma delta T cells. J Immunol 2007, 179:5576-5583. 31. Jenkins MK, Chu HH, McLachlan JB, Moon JJ: On the composition of the preimmune repertoire of T cells specific for peptide-major histocompatibility complex ligands. Annu Rev Immunol 2010, 28:275-294. 50. Notley CA, Inglis JJ, Alzabin S, McCann FE, McNamee KE, Williams RO: Blockade of tumor necrosis factor in collagen-induced arthritis reveals a novel immunoregulatory pathway for Th1 and Th17 cells. J Exp Med 2008, 205:2491-2497. 32. Garcia Z, Pradelli E, Celli S, Beuneu H, Simon A, Bousso P: Competition for antigen determines the stability of T cell-dendritic cell interactions during clonal expansion. Proc Natl Acad Sci USA 2007, 104:4553-4558. doi:10.1186/ar3108 Cite this article as: Merky et al.: Visualization and phenotyping of proinflammatory antigen-specific T cells during collagen-induced arthritis in a mouse with a fixed collagen type II-specific transgenic T-cell receptor b-chain. Arthritis Research & Therapy 2010 12:R155. doi:10.1186/ar3108 Cite this article as: Merky et al.: Visualization and phenotyping of proinflammatory antigen-specific T cells during collagen-induced arthritis in a mouse with a fixed collagen type II-specific transgenic T-cell receptor b-chain. Arthritis Research & Therapy 2010 12:R155. 33. Hataye J, Moon JJ, Khoruts A, Reilly C, Jenkins MK: Naive and memory CD4 + T cell survival controlled by clonal abundance. Science 2006, 312:114-116. 34. Kedl RM, Rees WA, Hildeman DA, Schaefer B, Mitchell T, Kappler J, Marrack P: T cells compete for access to antigen-bearing antigen- presenting cells. J Exp Med 2000, 192:1105-1113. 35. Corthay A, Backlund J, Holmdahl R: Role of glycopeptide-specific T cells in collagen-induced arthritis: an example how post-translational modification of proteins may be involved in autoimmune disease. Ann Med 2001, 33:456-465. 36. Koenecke C, Chennupati V, Schmitz S, Malissen B, Forster R, Prinz I: In vivo application of mAb directed against the gammadelta TCR does not deplete but generates ‘invisible’ gammadelta T cells. Eur J Immunol 2009, 39:372-379. 37. Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 25. Vollers SS, Stern LJ: Class II major histocompatibility complex tetramer staining: progress, problems, and prospects. Immunology 2008, 123:305-313. arthritis reduces joint inflammation, cartilage destruction, and bone erosion. Arthritis Rheum 2004, 50:650-659. arthritis reduces joint inflammation, cartilage destruction, and bone erosion. Arthritis Rheum 2004, 50:650-659. 45. Fujimoto M, Serada S, Mihara M, Uchiyama Y, Yoshida H, Koike N, Ohsugi Y, Nishikawa T, Ripley B, Kimura A, Kishimoto T, Naka T: Interleukin-6 blockade suppresses autoimmune arthritis in mice by the inhibition of inflammatory Th17 responses. Arthritis Rheum 2008, 58:3710-3719. 26. Myers LK, Brand DD, Ye XJ, Cremer MA, Rosloniec EF, Bodo M, Myllyharju J, Helaakoski T, Nokelainen M, Pihlajaniemi T, Kivirikko K, Yang CL, Ala-Kokko L, Prockop DJ, Notbohm H, Fietzek P, Stuart JM, Kang AH: Characterization of recombinant type II collagen: arthritogenicity and tolerogenicity in DBA/ 1 mice. Immunology 1998, 95:631-639. 46. Lamacchia C, Palmer G, Seemayer CA, Talabot-Ayer D, Gabay C: Enhanced Th1 and Th17 responses and arthritis severity in mice with a deficiency of myeloid cell-specific interleukin-1 receptor antagonist. Arthritis Rheum 2010, 62:452-462. 27. Myers LK, Myllyharju J, Nokelainen M, Brand DD, Cremer MA, Stuart JM, Bodo M, Kivirikko KI, Kang AH: Relevance of posttranslational modifications for the arthritogenicity of type II collagen. J Immunol 2004, 172:2970-2975. 47. Ito Y, Usui T, Kobayashi S, Iguchi-Hashimoto M, Ito H, Yoshitomi H, Nakamura T, Shimizu M, Kawabata D, Yukawa N, Hashimoto M, Sakaguchi N, Sakaguchi S, Yoshifuji H, Nojima T, Ohmura K, Fujii T, Mimori T: Gamma/delta T cells are the predominant source of interleukin-17 in affected joints in collagen-induced arthritis, but not in rheumatoid arthritis. Arthritis Rheum 2009, 60:2294-2303. 28. Malherbe L, Filippi C, Julia V, Foucras G, Moro M, Appel H, Wucherpfennig K, Guery JC, Glaichenhaus N: Selective activation and expansion of high-affinity CD4+ T cells in resistant mice upon infection with Leishmania major. Immunity 2000, 13:771-782. y 29. Malherbe L, Hausl C, Teyton L, McHeyzer-Williams MG: Clonal selection of helper T cells is determined by an affinity threshold with no further skewing of TCR binding properties. Immunity 2004, 21:669-679. 48. Sarkar S, Cooney LA, White P, Dunlop DB, Endres J, Jorns JM, Wasco MJ, Fox DA: Regulation of pathogenic IL-17 responses in collagen-induced arthritis: roles of endogenous interferon-g and IL-4. Arthritis Res Ther 2009, 11:R158. 30. References Osman GE, Cheunsuk S, Allen SE, Chi E, Liggitt HD, Hood LE, Ladiges WC: Expression of a type II collagen-specific TCR transgene accelerates the onset of arthritis in mice. Int Immunol 1998, 10:1613-1622. Page 14 of 14 Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Chiang EY, Kolumam GA, Yu X, Francesco M, Ivelja S, Peng I, Gribling P, Shu J, Lee WP, Refino CJ, Balazs M, Paler-Martinez A, Nguyen A, Young J, Barck KH, Carano RA, Ferrando R, Diehl L, Chatterjea D, Grogan JL: Targeted depletion of lymphotoxin-alpha-expressing TH1 and TH17 cells inhibits autoimmune disease. Nat Med 2009, 15:766-773. 38. Cooper SM, Sriram S, Ranges GE: Suppression of murine collagen-induced arthritis with monoclonal anti-Ia antibodies and augmentation with IFN- g. J Immunol 1988, 141:1958-1962. 38. Cooper SM, Sriram S, Ranges GE: Suppression of murine collagen-induced arthritis with monoclonal anti-Ia antibodies and augmentation with IFN- g. J Immunol 1988, 141:1958-1962. 39. Mauritz NJ, Holmdahl R, Jonsson R, Van der Meide PH, Scheynius A, Klareskog L: Treatment with g-interferon triggers the onset of collagen arthritis in mice. Arthritis Rheum 1988, 31:1297-1304. 39. Mauritz NJ, Holmdahl R, Jonsson R, Van der Meide PH, Scheynius A, Klareskog L: Treatment with g-interferon triggers the onset of collagen arthritis in mice. Arthritis Rheum 1988, 31:1297-1304. 40. Boissier MC, Chiocchia G, Bessis N, Hajnal J, Garotta G, Nicoletti F, Fournier C: Biphasic effect of interferon-g in murine collagen-induced arthritis. Eur J Immunol 1995, 25:1184-1190. 40. Boissier MC, Chiocchia G, Bessis N, Hajnal J, Garotta G, Nicoletti F, Fournier C: Biphasic effect of interferon-g in murine collagen-induced arthritis. Eur J Immunol 1995, 25:1184-1190. 41. Manoury-Schwartz B, Chiocchia G, Bessis N, Abehsira-Amar O, Batteux F, Muller S, Huang S, Boissier MC, Fournier C: High susceptibility to collagen- induced arthritis in mice lacking IFN-g receptors. J Immunol 1997, 158:5501-5506. 41. Manoury-Schwartz B, Chiocchia G, Bessis N, Abehsira-Amar O, Batteux F, Muller S, Huang S, Boissier MC, Fournier C: High susceptibility to collagen- induced arthritis in mice lacking IFN-g receptors. J Immunol 1997, 158:5501-5506. Merky et al. Arthritis Research & Therapy 2010, 12:R155 http://arthritis-research.com/content/12/4/R155 Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: Submit your next manuscript to BioMed Central and take full advantage of: 42. Lubberts E, Joosten LA, Oppers B, van den Bersselaar L, Coenen-de Roo CJ, Kolls JK, Schwarzenberger P, van de Loo FA, van den Berg WB: IL-1- independent role of IL-17 in synovial inflammation and joint destruction during collagen-induced arthritis. J Immunol 2001, 167:1004-1013. 43. Nakae S, Nambu A, Sudo K, Iwakura Y: Suppression of immune induction of collagen-induced arthritis in IL-17-deficient mice. J Immunol 2003, 171:6173-6177. 44. Lubberts E, Koenders MI, Oppers-Walgreen B, van den Bersselaar L, Coenen- de Roo CJ, Joosten LA, van den Berg WB: Treatment with a neutralizing anti-murine interleukin-17 antibody after the onset of collagen-induced 44. Lubberts E, Koenders MI, Oppers-Walgreen B, van den Bersselaar L, Coenen- de Roo CJ, Joosten LA, van den Berg WB: Treatment with a neutralizing anti-murine interleukin-17 antibody after the onset of collagen-induced
https://openalex.org/W3106731144	https://www.research-collection.ethz.ch/bitstream/20.500.11850/456967/2/3428279.pdf	English	null	Finding bugs in database systems via query partitioning	Proceedings of the ACM on programming languages	2,020	cc-by	22,799	ETH Library ACM Reference Format: Manuel Rigger and Zhendong Su. 2020. Finding Bugs in Database Systems via Query Partitioning. Proc. ACM Program. Lang. 4, OOPSLA, Article 211 (November 2020), 30 pages. https://doi.org/10.1145/3428279 MANUEL RIGGER, ETH Zurich, Switzerland ZHENDONG SU, ETH Zurich, Switzerland MANUEL RIGGER, ETH Zurich, Switzerland ZHENDONG SU, ETH Zurich, Switzerland Logic bugs in Database Management Systems (DBMSs) are bugs that cause an incorrect result for a given query, for example, by omitting a row that should be fetched. These bugs are critical, since they are likely to go unnoticed by users. We propose Query Partitioning, a general and effective approach for finding logic bugs in DBMSs. The core idea of Query Partitioning is to, starting from a given original query, derive multiple, more complex queries (called partitioning queries), each of which computes a partition of the result. The individual partitions are then composed to compute a result set that must be equivalent to the original query’s result set. A bug in the DBMS is detected when these result sets differ. Our intuition is that due to the increased complexity, the partitioning queries are more likely to stress the DBMS and trigger a logic bug than the original query. As a concrete instance of a partitioning strategy, we propose Ternary Logic Partitioning (TLP), which is based on the observation that a boolean predicate p can either evaluate to TRUE, FALSE, or NULL. Accordingly, a query can be decomposed into three partitioning queries, each of which computes its result on rows or intermediate results for which p, NOT p, and p IS NULL hold. This technique is versatile, and can be used to test WHERE, GROUP BY, as well as HAVING clauses, aggregate functions, and DISTINCT queries. As part of an extensive testing campaign, we found 175 bugs in widely-used DBMSs such as MySQL, TiDB, SQLite, and CockroachDB, 125 of which have been fixed. Notably, 77 of these were logic bugs, while the remaining were error and crash bugs. We expect that the effectiveness and wide applicability of Query Partitioning will lead to its broad adoption in practice, and the formulation of additional partitioning strategies. 211 CCS Concepts: • Information systems →Database query processing; • Software and its engineering →Software testing and debugging. Additional Key Words and Phrases: database testing, DBMS testing, test oracle, three-valued logic Conference Paper Rights / license: Originally published in: Proceedings of the ACM on Programming Languages 4(OOPSLA), https://doi.org/10.1145/3428279 Originally published in: Proceedings of the ACM on Programming Languages 4(OOPSLA), https://doi.org/10.1145/3428279 This page was generated automatically upon download from the ETH Zurich Research Collection. For more information, please consult the Terms of use. 1 INTRODUCTION SQL is based on a ternary boolean logic, which means that a predicate φ can either evaluate to TRUE, FALSE or NULL. The predicate can be interpreted as a piecewise- defined total function p, with the current row r as an argument: The n partial results can then be composed using a predefined, n-ary composition operator ⋄to obtain a result set RS(Q′ 0)⋄RS(Q′ 1)⋄. . . ⋄RS(Q′ n−1). For simplicity, we denote the composed partial results as RS(Q′). The original query’s result set and the composed partitions must be equal, that is, RS(Q′) = RS(Q). Bugs in the DBMS can then be detected by checking whether the equality indeed holds. While a number of partitioning strategies are imaginable, it is crucial to select one that stresses the DBMS and its query optimizer in different ways, either between partitioning queries of Q′ or between Q′ and Q, so that an inconsistent result set might be observed.f As part of this work, we propose Ternary Logic Partitioning (TLP), which can effectively test the correct implementation and optimization of WHERE clauses, GROUP BY clauses, HAVING clauses, aggre- gate functions, and DISTINCT clauses. SQL is based on a ternary boolean logic, which means that a predicate φ can either evaluate to TRUE, FALSE or NULL. The predicate can be interpreted as a piecewise- defined total function p, with the current row r as an argument: p(r) =   TRUE if φ FALSE if ¬φ NULL if φ IS NULL p(r) =   TRUE if φ FALSE if ¬φ NULL if φ IS NULL Consider a random row r from RS(Q). Irrespective of which predicate we might choose (or ran- domly generate), we know that exactly one of the conditions of the piecewise functionp must hold. Based on this insight, we can partition any Q by deriving three queries that filter records based on whether p holds, ¬p holds, or whether p is NULL, while guaranteeing that the combined result comprises all rows of the original query. If used to test a WHERE clause, the individual subqueries can be aggregated using a union operator. Consider Listing 1, which demonstrates an unknown bug that we reported for MySQL version 8.0.19 and which was fixed for version 8.0.21. Query 1 computes an incorrect result set, and demonstrates the underlying bug. 1 INTRODUCTION One record consisting of the rows in t0 and t1 should be fetched, since 0 and -0 represent the same number, so the comparison should evaluate to TRUE. We found this bug based on the original query O and the partitioning queries P . O lacks a WHERE clause and thus fetches the cross product of all values in t0 and t1; since both tables contain only a single record, only a single record is fetched. P consists of three partitioning queries that are connected by the UNION ALL keyword, which combines the queries’ result sets. We derived these queries by generating a random predicate t0.c0 = t1.c0 for the WHERE clause, and then creating the two other variants with the negated predicate and IS NULL predicate. Thus, P ’s result set is expected to be the same as the one for query O . However, since the query with the predicate t0.c0 = t1.c0 was processed incorrectly, and resulted in the omission of the row, we detected this bug. Based on O and P , we manually created test case 1 to report the bug. Listing 1. A logic bug in MySQL caused a predicate 0=-0 to incorrectly evaluate to FALSE. The check symbol denotes the expected, correct result, while the bug symbol denotes the actual, incorrect result. 1 INTRODUCTION Database Management Systems (DBMSs) are used ubiquitously. Most DBMSs allow inserting, deleting, modifying, and querying data from a database using the Structured Query Language (SQL). DBMSs can be affected by various kinds of bugs. In this work, we consider logic bugs, which are bugs that cause the DBMS to fetch an incorrect result set for a query. For example, for a given query, a DBMS might mistakenly omit a record from the result set, fetch a record that should not be in the result set, or compute an incorrect result for a function or operator. Such bugs are difficult to detect by users and might go unnoticed, especially considering the scale of many databases.f To tackle logic bugs in DBMSs, we propose a general and effective technique to which we refer to as Query Partitioning. The core idea of Query Partitioning is, based on a given query Q with a result set RS(Q), to derive n queries Q′ 0 … Q′ n−1, each of which computes a partial result RS(Q′ i). Authors’ addresses: Manuel Rigger, manuel.rigger@inf.ethz.ch, ETH Zurich, Department of Computer Science, Zurich, Switzerland; Zhendong Su, zhendong.su@inf.ethz.ch, ETH Zurich, Department of Computer Science, Zurich, Switzerland. provided tha the full citati p g py g p y p contact the owner/author(s). © 2020 Copyright held by the owner/author(s). 2475-1421/2020/11-ART211 https://doi.org/10.1145/3428279 This work is licensed under a Creative Commons Attribution 4.0 International License. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:2 Manuel Rigger and Zhendong Su The n partial results can then be composed using a predefined, n-ary composition operator ⋄to obtain a result set RS(Q′ 0)⋄RS(Q′ 1)⋄. . . ⋄RS(Q′ n−1). For simplicity, we denote the composed partial results as RS(Q′). The original query’s result set and the composed partitions must be equal, that is, RS(Q′) = RS(Q). Bugs in the DBMS can then be detected by checking whether the equality indeed holds. While a number of partitioning strategies are imaginable, it is crucial to select one that stresses the DBMS and its query optimizer in different ways, either between partitioning queries of Q′ or between Q′ and Q, so that an inconsistent result set might be observed. As part of this work, we propose Ternary Logic Partitioning (TLP), which can effectively test the correct implementation and optimization of WHERE clauses, GROUP BY clauses, HAVING clauses, aggre- gate functions, and DISTINCT clauses. 1 INTRODUCTION CREATE TABLE t0(c0 INT); CREATE TABLE t1(c0 DOUBLE); INSERT INTO t0 VALUES (0); INSERT INTO t1 VALUES('-0'); 1 SELECT * FROM t0, t1 WHERE t0.c0 = t1.c0; -- {} O SELECT * FROM t0, t1; -- {0, -0} P SELECT * FROM t0, t1 WHERE t0.c0 = t1.c0 UNION ALL SELECT * FROM t0, t1 WHERE NOT(t0.c0 = t1.c0) UNION ALL SELECT * FROM t0, t1 WHERE (t0.c0 = t1.c0) IS NULL; -- {} CREATE TABLE t0(c0 INT); CREATE TABLE t1(c0 DOUBLE); INSERT INTO t0 VALUES (0); INSERT INTO t1 VALUES('-0'); 1 SELECT * FROM t0, t1 WHERE t0.c0 = t1.c0; -- {} O SELECT * FROM t0, t1; -- {0, -0} P SELECT * FROM t0, t1 WHERE t0.c0 = t1.c0 UNION ALL SELECT * FROM t0, t1 WHERE NOT(t0.c0 = t1.c0) UNION ALL SELECT * FROM t0, t1 WHERE NOT(t0.c0 = t1.c0) UNION ALL SELECT FROM t0, t1 WHERE NOT(t0.c0 t1.c0) UNION ALL SELECT * FROM t0, t1 WHERE (t0.c0 = t1.c0) IS NULL; -- {} UNION ALL SELECT * FROM t0, t1 WHERE (t0.c0 = t1.c0) IS NULL; Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:3 Finding Bugs in Database Systems via Query Partitioning Query Partitioning addresses fundamental limitations in existing approaches. Pivoted Query Syn- thesis (PQS) detects logic bugs by checking whether a randomly-selected pivot row is fetched cor- rectly [Rigger and Su 2020c]. To construct a query that fetches the row, PQS relies on an implemen- tation of the DBMS SQL dialect’s supported operators and functions. The technique has proven to be effective. However, unlike Query Partitioning, its implementation effort is high and requires detailed knowledge of the DBMS’ operator and function semantics. Non-optimizing Reference En- gine Construction (NoREC) detects bugs in queries that use a WHERE predicate by rewriting the query to disable the DBMS’ optimizations and addresses PQS’ high implementation effort [Rigger and Su 2020a]. 1 INTRODUCTION A major limitation of both NoREC and PQS is that they are applicable primarily to test WHERE predicates; while they can partially be used to test other features—for example, PQS can test DISTINCT, but cannot detect duplicate rows—it would be unclear, for example, how these approaches could be extended to test aggregate queries.fu We evaluated the effectiveness of Query Partitioning in a large-scale study on six widely-used DBMSs. We found 175 true, previously unknown bugs in five of these systems, which demon- strates the effectiveness and generality of the proposed approach. Many of these were considered important by the developers of the DBMSs, and 125 of these bugs have already been fixed. 77 bugs were logic bugs, while the remaining were error and crash bugs. Furthermore, we compared our proposed approach with NoREC. TLP could detect 17 bugs in features that are out-of-scope for NoREC. We found 12 bugs related to WHERE clauses, which could have also been found by NoREC. Ultimately, Query Partitioning is complementary to PQS, and shares the same advantages and dis- advantages as NoREC, both being metamorphic test oracles [Chen et al. 1998]. Due to the high ef- fectiveness and low implementation effort, we believe that our approach might be widely adopted in practice. For reproducibility, and to facilitate the adoption of TLP, we provide an artifact with the implementation and the bugs we found [Rigger and Su 2020b]. The latest version of SQLancer and the TLP implementation is available at https://github.com/sqlancer. O ll hi ib h f ll i Overall, this paper contributes the following: Overall, this paper contributes the following: • Query Partitioning, a general technique designed for finding logic bugs in DBMSs that use SQL as a query language;u • Ternary Logic Partitioning (TLP), an instantiation of Query Partitioning based on the insight that a boolean predicate can be partitioned to evaluate to TRUE, FALSE, or NULL; • Concrete TLP oracles to test queries using WHERE, HAVING, and GROUP BY clauses as well as aggregate functions and DISTINCT queries; andu • An extensive evaluation of Query Partitioning on six widely-used DBMSs, in which the tech- nique found 175 bugs, and a comparison with the state-of-the-art approach NoREC. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 2 BACKGROUND SQL is a feature-rich language and provides a number of additional features (e.g., window functions and transactions). While our core idea could be generalized to some additional features, we consider them less important and out of scope. Aggregate functions. Various kinds of aggregate functions exist, such as MIN() and MAX() to compute minimum and maximum values for an input expression, SUM() to sum up input values, COUNT() to count the number of rows, and AVG() to compute the average. We base our testing ideas for aggre- gate functions on the optimization of aggregate functions by distributing their computations [Co- hen 2006; Jesus et al. 2015; Yu et al. 2009]. Important properties for aggregate functions were defined [Jesus et al. 2015]. An aggregate function f is self-decomposable when a merge operator ⊕ exists so that, given two non-empty multi-sets X and Y, the following holds: f (X ⊎Y) = f (X)⊕f (Y). Many functions, including MIN(), MAX(), SUM(), and COUNT() are self-composable. For example, con- sider SUM(): SUM({x}) = x and SUM(X ⊎Y) = SUM(X)+SUM(Y). An aggregate function f is com- posable if for some function f and a self-decomposable aggregate function h, it can be expressed as f = д◦h. Every self-composable function is composable, by assigning д as the identity function. The AVG() function is composable when defining д as д((s,c)) = s/c and h as follows: h({x}) = (x, 1) and h(X ⊎Y) = h(X) + h(Y). That is, the AVG() function is computed by dividing the sum of values by the number of rows: AVG(X ⊎Y) = (SUM(X) + SUM(Y))/(COU NT(X) + COU NT(Y)). Automatic testing. We propose a novel automatic testing approach for DBMSs. Two components are essential. First, an effective test case should stress significant portions of the system under test. To this end, a number of database generators have been proposed [Binnig et al. 2007b; Bruno and Chaudhuri 2005; Gray et al. 1994; Houkjær et al. 2006; Khalek et al. 2008; Neufeld et al. 1993], as well as a number of query generators [Bati et al. 2007; Bruno et al. 2006; Jung et al. 2019; Mishra et al. 2008; Poess and Stephens 2004; Seltenreich 2019; Vartak et al. 2010; Zhong et al. 2020]. While these are important components of a testing approach, they are well understood, and not the main focus of this paper. 2 BACKGROUND Relational DBMSs. DBMSs are based on a data model, which abstractly describes how data is orga- nized. We primarily aim to test DBMSs based on the relational data model proposed by Codd [1970], on which most widely-used databases, such as Oracle, Microsoft SQL, PostgreSQL, MySQL, and SQLite are based. Relational DBMSs use a domain-specific language, Structured Query Language (SQL), for interaction. SQL’s data model is based on bags (i.e., multisets), where the same row can occur multiple times [Guagliardo and Libkin 2017]. This contrasts the original relational model, which is based on the concept of sets. Since a query’s result is typically nevertheless referred to as a result set, we also use this term in this paper. In order to merge two bags, without removing duplicates, the multiset addition, denoted as ⊎, is used. To exclude duplicate elements, the union operator, denoted as ∪, is used. In SQL, the UNION ALL operator corresponds to ⊎, and UNION—without ALL—to ∪. Both operators are used in the composition operator of different TLP test oracles. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:4 Manuel Rigger and Zhendong Su SQL. We assume basic familiarity with SQL, and thus provide only a minimal overview of it. In our work, we concentrate on the SELECT statement, which allows querying data from a database. SQL provides various ways of filtering, grouping, and aggregating data. A WHERE clause can be used to specify which rows should be fetched. It contains a boolean predicate, which can evaluate to TRUE, FALSE, or NULL. A number of DBMSs (e.g., SQLite and MySQL) allow the usage of a predicate of any type in a WHERE clause, as they apply implicit conversions to convert values of other types to a boolean value. A GROUP BY clause can be used to aggregate rows. It specifies a number of expressions, based on which the DBMS groups rows for which the expressions evaluate to the same value. They can be used in combination with HAVING clauses, which allow filtering rows after they are grouped. Similar to GROUP BY, a query can contain a DISTINCT clause to compute a result set rather than a bag (i.e., the DISTINCT removes all duplicate rows). Aggregate functions compute values over multiple rows. They can either be used to aggregate the final result set, or in a HAVING clause as part of a predicate. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 2 BACKGROUND We believe that any database generator and query generator that provides control over the format of the queries generated can be used to find bugs using Query Partitioning. In our implementation, we use SQLancer’s database and expression generation mechanism, which is detailed below. Second, an effective test oracle needs to determine whether the generated test case’s result is correct. A specific class of test oracles are metamorphic ones, which can derive a test case and its expected result based on an input and output of a system [Chen et al. 1998]. While the implementation effort for such oracles is often low, they cannot provide a ground truth (i.e., since the output based on which the new test case is generated might be incorrect). The main focus of this paper is metamorphic test oracles, which are based on the general idea of Query Partitioning. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:5 Finding Bugs in Database Systems via Query Partitioning Pivoted Query Synthesis. Pivoted Query Synthesis (PQS) was recently proposed by Rigger and Su [2020c] to find logic bugs in DBMSs. It randomly selects a row, called a pivot row, based on which a query is constructed that must fetch the pivot row. To guarantee that the row is fetched, the test- ing approach executes a randomly-generated predicate and then modifies it so it evaluates to TRUE. While this technique was shown to be effective, a significant limitation is that its implementation effort is high, since the tool needs to implement all operators and functions that are tested. Fur- thermore, it can only effectively test WHERE clauses, since it validates results based on a single row. Although it can generate DISTINCT clauses and GROUP BYs, it cannot detect mistakenly fetched duplicate rows, and omitted duplicate rows. PQS can test aggregate functions only when the table contains a single row, which does not meaningfully test their aggregation functionality. The ap- proach proposed in this paper seeks to complement PQS. Query Partitioning can detect bugs in a wider range of features and requires little implementation effort. PQS can provide a ground truth for an important selection of core operators, and thus fill a gap left open by TLP. NoREC. 2 BACKGROUND Non-optimizing Reference Engine Construction (NoREC) was recently proposed by Rigger and Su [2020a] to find optimization bugs in DBMSs, which are logic bugs that cause the DBMS to incorrectly apply an optimization. The core insight of this approach is that an optimized query can be translated to one that the DBMS cannot effectively optimize. Thus, NoREC is also a metamorphic testing approach. NoREC could also have detected the bug in the motivating example (see Listing 1). Specifically, it would rewrite query 1 to another query SELECT (t0.c0 = t1.c0) IS TRUE FROM t0, t1. The translated query evaluates the predicate that is taken from the WHERE clause of the original query on every row in the table; since only one row is contained, the query would return a single row with a single column whose value is TRUE. In practice, the number of TRUE values would be summed up using the SUM() aggregate function. A predicate must always evaluate to the same value. Thus, it would be expected that the predicate evaluates to TRUE in the WHERE clause, meaning that the result set of the original query should comprise the row. For this query, the result set comprised zero rows, and would allow finding the bug. In fact, the original query was optimized by the DBMS to efficiently fetch the data, while the translated query evaluates the predicate on every row, which made the incorrect optimization inapplicable. As with PQS, NoREC has been successful in detecting a wide range of bugs. However, similar to PQS, a significant limitation is that the approach is applicable only to WHERE clauses (and partially GROUP BY clauses). TLP advances NoREC in two important ways. First, NoREC tackles the test oracle problem by inhibiting DBMS optimizations, while TLP tackles the problem by partitioning a given query, thus, at the conceptual level, they are orthogonal and complement each other. Second, it is unclear how NoREC could be extended to support other features. We address these limitations through TLP. For example, TLP can detect bugs in aggregate functions by partitioning their computations. Our evaluation results demonstrate these distinct benefits of TLP. NoREC. Non-optimizing Reference Engine Construction (NoREC) was recently proposed by Rigger and Su [2020a] to find optimization bugs in DBMSs, which are logic bugs that cause the DBMS to incorrectly apply an optimization. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 2 BACKGROUND The core insight of this approach is that an optimized query can be translated to one that the DBMS cannot effectively optimize. Thus, NoREC is also a metamorphic testing approach. NoREC could also have detected the bug in the motivating example (see Listing 1). Specifically, it would rewrite query 1 to another query SELECT (t0.c0 = t1.c0) IS TRUE FROM t0,h SQLancer. The implementation of our proposed approach is based on SQLancer, in which the PQS and NoREC oracles were also implemented [Rigger and Su 2020a,c]. The syntax and semantics of statements and expressions differ widely among the DBMSs. Thus, SQLancer consists of com- ponents that are specific to each DBMS, such as the database and expression generators. They are implemented manually and naively, and neither databases nor expressions are enumerated systematically. While we believe that both components could be enhanced, for example, by sys- tematically generating databases and expressions, or generating them in a way to maximize the chances of triggering a bug, the focus of this paper is on test oracles, and our findings suggest that the naive generation approach is sufficient to detect many bugs. SQLancer also consists of DBMS-independent components, such as those for logging test cases, randomly generating data, and handling options. SQLancer operates in two phases, by first generating a database and then Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:6 Manuel Rigger and Zhendong Su executing one or multiple test oracles. Since generating a database typically takes significantly more time than executing a query, SQLancer by default attempts to execute 100,000 iterations of the selected test oracle. SQLancer’s Database Generation. In the first phase, SQLancer creates a number of tables using the CREATE TABLE command. Then, statements are generated to change the state of the database and DBMS, for example, by inserting, deleting, and modifying data, setting options, as well as creating indexes and views. The number of statements that are generated of a specific type (e.g., an INSERT) is restricted by an upper limit, which is set to a meaningful, empirically-determined default value, but can be overridden by setting an option. For example, by default, SQLancer generates only up to 30 INSERT statements to restrict the size of the query’ result sets, and thus enable queries to exe- cute quickly. 2 BACKGROUND All statements are syntactically valid, since the database generators are implemented based on the respective SQL dialect’s grammar; however, they might be semantically invalid. For example, an INSERT statement might expectedly fail with an error “UNIQUE constraint failed” when it attempts to insert a duplicate value into a column that is declared as UNIQUE. Each statement is annotated with a list of such “expected” errors. Unexpected errors indicate a bug in the DBMS under test. The outcome of the first phase is the database on which the test oracles are executed. SQLancer’s Expression Generation. In the second phase, the test oracles are used to test the DBMS based on the randomly-generated database. For generating queries, the test oracle implementa- tions request random expressions from the expression generators (which are also used by the statement generators). The oracles proposed in this paper request, for example, predicates used in WHERE and HAVING clauses. The expression generator randomly picks any of the applicable opera- tors, functions, and leaf nodes (i.e., column references and constants). When a specified maximum depth is reached (by default 3), only leaf nodes are considered. Column references are always valid, since SQLancer retrieves the column names of all tables and views from the DBMS, and consid- ers only those column references whose tables and views are referenced in the generated query. Constants are generated by using a random data generator, which uses two strategies to attempt generating meaningful constants. First, boundary values, such as minimum and maximum inte- gers, are generated with an increased probability. Second, constants are cached, and potentially selected in place of a newly generated constant. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 3 APPROACHu Query Partitioning. We envisage Query Partitioning as a versatile technique. The core idea of our approach is to start from a given query and decompose it into multiple equivalent queries, whose results can be composed to obtain the same result as the original query. We refer to the given query as the original query. In the remainder of this paper, we assume that the original query is randomly generated according to the specified format, but it could likewise be given by a user or specified in a test suite. We refer to the multiple queries that are equivalent to the original query as the partitioning queries, each of which computes a partition. We refer to the operator that combines the partitions as the composition operator (denoted by ⋄). Ternary Logic Query Partitioning. In this paper, we consider only a single instance of the general partitioning strategy idea, namely Ternary Logic Partitioning (TLP). The core idea of the technique is that a predicate on a row or intermediate result must either evaluate to TRUE, FALSE, or NULL. Thus, an original query can be decomposed into three partitioning queries. One partitioning query fetches rows where a predicate p holds, one where it does not hold, and one for which it evaluates to NULL. That is, we construct one predicate p, one predicate NOT p, and one predicate p IS NULL. Each predicate is then used in WHERE and HAVING clauses. Accordingly, we refer to these predicates as Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. Finding Bugs in Database Systems via Query Partitioning 211:7 ternary predicate variants. Similar to the original query, we assume this predicate to be randomly generated. In the further, we demonstrate how this idea enables testing WHERE clauses, GROUP BY clauses, HAVING clauses, aggregate functions, and DISTINCT queries. Process. Figure 1 illustrates the process of TLP. Based on an existing database—which is randomly generated in our implementation—a random query Q is generated. We denote the result set of this query as RS(Q) and illustrate a result set using a circle. Based on TLP, we then derive three par- titioning queries Q′ p, Q′ ¬p, and Q′ p IS NU LL from Q. Each partitioning query computes a partition of the result, which we denote as RS(Q′ p), RS(Q′ ¬p), and RS(Q′ p IS NU LL). 3 APPROACHu Based on the composi- tion operator ⋄, the individual partitions are composed to obtain a result set RS(Q′). The equality RS(Q) = RS(Q′) must hold. If we find that the result sets differ, a bug in the DBMS is detected. RS(Q) Ternary Predicate Partitioning RS(Q'p) RS(Q'¬p) RS(Q'p IS NULL) RS(Q') RS(Q'IS NULL) RS(Q'¬p ) RS(Q'p) Q'p Q'¬p IS NULL Composition Q Random Query Generator =  Fig. 1. The idea of Ternary Logic Partitioning (TLP) is to partition a query Q into several partitioning queries Q′p, Q′¬p, and Q′ p IS NU LL, whose partitions are composed to form a result set RS(Q) = RS(Q′). The dashed lines indicate a comparison between the result sets. The check denotes that the result sets match as expected, while the bug represents that the DBMS is affected by a bug, causing a mismatch in the result sets. RS(Q) Ternary Predicate Partitioning RS(Q'p) RS(Q'¬p) RS(Q'p IS NULL) RS(Q') RS(Q'IS NULL) RS(Q'¬p ) RS(Q'p) Q'p Q'¬p IS NULL Composition Q Random Query Generator =  Composition Fig. 1. The idea of Ternary Logic Partitioning (TLP) is to partition a query Q into several partitioning queries Q′p, Q′¬p, and Q′ p IS NU LL, whose partitions are composed to form a result set RS(Q) = RS(Q′). The dashed lines indicate a comparison between the result sets. The check denotes that the result sets match as expected, while the bug represents that the DBMS is affected by a bug, causing a mismatch in the result sets. Intuition on the partitions. Intuitively, the partitioning queries can be considered as computing a subset or a partial bag of Q’s result set (i.e., the partitions are subsets or bags of RS(Q)). For both the WHERE and HAVING test oracles, ⋄corresponds to the multiset addition ⊎. For the DISTINCT and GROUP BY oracles, the partitions can contain duplicate values; for these oracles, the ⋄corresponds to the set union ∪. For the aggregate test oracle, the partitions are not a subset of the original query’s result; rather, they correspond to intermediate values. For example, when testing the MIN() aggregate function, which computes the minimum, the partitions denote the minimum value of their individual partitions. Intuition on the partitions. Intuitively, the partitioning queries can be considered as computing a subset or a partial bag of Q’s result set (i.e., the partitions are subsets or bags of RS(Q)). Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 3 APPROACHu The <e> placeholder refers to an arbitrary expression. An element enclosed in square brackets ([]) denotes that the element is optional. ORDER BYs. A random ORDER BY can be generated for each of the partitioning queries. Since our oracles do not validate the ordering of the result, such clauses must not affect the query’s result. However, they introduce additional complexity (e.g. by causing a DBMS to use an index for sort- ing [Graefe 2011]), which can help to expose additional bugs. In fact, we found bugs that were triggered only when using an ORDER BY clause (see Listing 7). Some DBMSs do not allow individual ORDER BYs in queries joined using UNION or UNION ALL; for them, only a single ORDER BY might be used when the partitions are composed using UNION or UNION ALL operators (see below). ORDER BYs. A random ORDER BY can be generated for each of the partitioning queries. Since our oracles do not validate the ordering of the result, such clauses must not affect the query’s result. However, they introduce additional complexity (e.g. by causing a DBMS to use an index for sort- ing [Graefe 2011]), which can help to expose additional bugs. In fact, we found bugs that were triggered only when using an ORDER BY clause (see Listing 7). Some DBMSs do not allow individual ORDER BYs in queries joined using UNION or UNION ALL; for them, only a single ORDER BY might be used when the partitions are composed using UNION or UNION ALL operators (see below). Composition operator implementation. Every composition operator either contains a ⊎or ∪oper- ator to compose result sets with, and without removing duplicate rows. The testing tool can im- plement them by iterating over each partitioning query’s result set and collecting the rows using an appropriate data structure—for example, a list for ⊎and a set for ∪. Implementing the opera- tor in the testing tool is not applicable when the partition is used for further computations, like in the aggregate oracles. For example, the aggregate MIN oracle computes the minimum value of each partition’s minimum value; the minimum value cannot be easily determined by the testing tool, since, for example, the order of strings can depend on COLLATE clauses that can be part of the query. 3 APPROACHu For these, a more convenient alternative is to use the UNION ALL and UNION operators, which implement the operators’ semantics in SQL (see Section 2). Using them also tests these operators, and, in fact, we found bugs in their implementation. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 3 APPROACHu For both the WHERE and HAVING test oracles, ⋄corresponds to the multiset addition ⊎. For the DISTINCT and GROUP BY oracles, the partitions can contain duplicate values; for these oracles, the ⋄corresponds to the set union ∪. For the aggregate test oracle, the partitions are not a subset of the original query’s result; rather, they correspond to intermediate values. For example, when testing the MIN() aggregate function, which computes the minimum, the partitions denote the minimum value of their individual partitions. Overview. Table 1 shows all the information necessary to fully realize the oracles, which are ex- plained in detail in the subsequent sections. The first column denotes the oracle’s name. The second column describes the format of the randomly-generated query Q. The third column describes the format of a partitioning query Q′ ptern. This query is instantiated with the three ternary predicate variants. The fourth column describes the implementation of the composition operator. Reconsider the motivating example (see Listing 1), which we found using the WHERE oracle, described in the first row of the table. Query O corresponds to the format of Q, while each partitioning query in query P corresponds to the Q′ ptern format. The partitions in query O are composed using the UNION ALL operator, which corresponds to the ⊎operator. Query elements. The <columns> placeholder refers to a set of columns, or expressions that are evalu- ated on each of the rows; this placeholder could also be an asterisk (), specifying that all columns should be fetched. The <tables> placeholder refers to the tables, from which values are fetched. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:8 Manuel Rigger and Zhendong Su The <joins> placeholder can refer to any of the joins (such as inner joins, outer joins, left joins, right joins, and natural joins); although we do not propose an exclusive oracle to test joins, we found that the existing oracles also detect bugs in them. The <e> placeholder refers to an arbitrary expression. An element enclosed in square brackets ([]) denotes that the element is optional. The <joins> placeholder can refer to any of the joins (such as inner joins, outer joins, left joins, right joins, and natural joins); although we do not propose an exclusive oracle to test joins, we found that the existing oracles also detect bugs in them. 3.1 Testing WHERE Clausesh The WHERE oracle tests the correct implementation and optimization of WHERE clauses. It is the most basic test oracle. Nevertheless, our evaluation shows that it is the most effective. Queries. The WHERE oracle assumes an original query that lacks a WHERE clause, and constructs par- titioning queries with a WHERE clause, each of which uses one of the ternary logic predicates. Our intuition is that the original query is unlikely to compute an incorrect result, since it simply fetches all records of a set of tables. In contrast, the partitioning query’s WHERE clauses might result in the incorrect omission or addition of records. Intuition on the test oracles. We believe that the WHERE oracle is sufficient to find the majority of bugs that the TLP oracles can detect. While it specifically generates queries to test WHERE clauses, it also stresses the implementation of a variety of DBMS components and optimizations [Chaudhuri 1998]. Specifically, we found that this test oracle can find bugs in physical access methods (in particular index scans) [Astrahan et al. 1976], common physical operators [Chaudhuri 1998], join algorithms [Graefe 1993], rewriting of queries [Haas et al. 1989], and general optimizations that are applied to predicates (e.g., algebraic simplifications). We quantify this observation in Section 5.3. Existing predicates. It might be desirable to create additional test queries based on queries that already have a WHERE predicate, for example, when the original query is not randomly generated, but when existing queries from a test suite are used. We propose the WHERE Extended oracle for this scenario. Based on an existing WHERE clause and a predicate pexist, partitioning queries are derived that use the AND operator to add an additional ternary variant to the predicate. Comparison to NoREC. We believe that the WHERE oracle has similar bug-finding capabilities as NoREC (see Section 5.2). Both test oracles focus on testing WHERE clauses. NoREC mainly focuses Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:9 Finding Bugs in Database Systems via Query Partitioning 211:9 Table 1. Each of the Ternary Logic Partitioning (TLP) oracles is designed to test a specific SQL feature. 3.1 Testing WHERE Clausesh The WHERE oracle achieves this by introducing three variants of the query, which are optimized to different degrees. For example, an index might only be applicable for one or two of the partitioning queries, but not all of them, enabling it to also find such optimization bugs. on testing for optimization bugs, by evaluating the predicate on every row, which disables most optimizations. The WHERE oracle achieves this by introducing three variants of the query, which are optimized to different degrees. For example, an index might only be applicable for one or two of the partitioning queries, but not all of them, enabling it to also find such optimization bugs. 3.1 Testing WHERE Clausesh Oracle Q Q′ptern ⋄(Q′p, Q′¬p, Q′ p N U LL) WHERE SELECT <columns > FROM <tables > [<joins >] SELECT <columns > FROM <tables > [<joins >] WHERE ptern Q′p ⊎Q′¬p ⊎Q′ p N U LL WHERE Extended SELECT <columns > FROM <tables > [<joins >] WHERE pexist SELECT <columns > FROM <tables > [<joins >] WHERE pexist AND ptern Q′p ⊎Q′¬p ⊎Q′ p N U LL GROUP BY SELECT <columns > FROM <tables > <joins > GROUP BY <columns > SELECT <columns > FROM <tables > <joins > WHERE ptern GROUP BY <columns > Q′p ∪Q′¬p ∪Q′ p N U LL HAVING SELECT <columns > FROM <tables > <joins > [WHERE ...] [GROUP BY ...] SELECT <columns > FROM <tables > <joins > [WHERE ...] [GROUP BY ...] HAVING ptern Q′p ⊎Q′¬p ⊎Q′ p N U LL DISTINCT SELECT DISTINCT <columns > FROM <tables > <joins > SELECT [DISTINCT] <columns > FROM <tables > <joins > WHERE ptern Q′p ∪Q′¬p ∪Q′ p N U LL Aggregate (MIN) SELECT MIN(<e>) FROM <tables > [<joins >] SELECT MIN(<e>) FROM <tables > [<joins >] WHERE ptern MI N (Q′p ⊎Q′¬p ⊎Q′ p N U LL) Aggregate (MAX) SELECT MAX(<e>) FROM <tables > [<joins >] SELECT MAX(<e>) FROM <tables > [<joins >] WHERE ptern MAX (Q′p ⊎Q′¬p ⊎Q′ p N U LL) Aggregate (SUM) SELECT SUM(<e>) FROM <tables > [<joins >] SELECT SUM(<e>) FROM <tables > [<joins >] WHERE ptern SU M(Q′p ⊎Q′¬p ⊎Q′ p N U LL) Aggregate (COUNT) SELECT COUNT(<e>) FROM <tables > [<joins >] SELECT COUNT(<e>) FROM <tables > [<joins >] WHERE ptern SU M(Q′p ⊎Q′¬p ⊎Q′ p N U LL) Aggregate (AVG) SELECT AVG(<e>) FROM <tables > [<joins >] SELECT SUM(<e>) as s, COUNT(<e>) as c FROM <tables > [<joins >] WHERE ptern SU M(s(Q′p ⊎Q′¬p ⊎Q′ p N U LL)) SU M(c(Q′p ⊎Q′¬p ⊎Q′ p N U LL)) Finding Bugs in Database Systems via Query Partitioning 211 9 Table 1. Each of the Ternary Logic Partitioning (TLP) oracles is designed to test a specific SQL feature. Table 1. Each of the Ternary Logic Partitioning (TLP) oracles is designed to test a specific SQL feature. SU M(Q′p ⊎Q′¬p ⊎Q′ p N U LL) SU M(s(Q′p ⊎Q′¬p ⊎Q′ p N U LL)) SU M(c(Q′p ⊎Q′¬p ⊎Q′ p N U LL)) on testing for optimization bugs, by evaluating the predicate on every row, which disables most optimizations. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 3.2 Testing Grouping CREATE TABLE t0(c0 INT); CREATE VIEW v0(c0) AS SELECT CAST(t0.c0 AS INTEGER) FROM t0; INSERT INTO t0(c0) VALUES (0); O SELECT DISTINCT FROM t0 LEFT OUTER JOIN v0 ON v0.c0 >= '0'; -- {0\|0} P SELECT * FROM t0 LEFT OUTER JOIN v0 ON v0.c0 >= '0' WHERE TRUE UNION SELECT * FROM t0 LEFT OUTER JOIN v0 ON v0.c0 >= '0' WHERE NOT TRUE UNION SELECT * FROM t0 LEFT OUTER JOIN v0 ON v0.c0 >= '0' WHERE TRUE IS NULL; - {0\| NULL} CREATE TABLE t0(c0 INT); CREATE VIEW v0(c0) AS SELECT CAST(t0.c0 AS INTEGER) FROM t0; INSERT INTO t0(c0) VALUES (0); Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 3.2 Testing Grouping The DISTINCT, GROUP BY, and HAVING test oracles are closely related, as they all test the grouping and filtering of rows. We refer to them collectively as grouping oracles. Queries. The DISTINCT oracle is based on the composition operator, which excludes duplicate rows using the ∪operator. The partitioning queries themselves can thus optionally omit the DISTINCT keyword. The GROUP BY test oracle, similarly to the DISTINCT test oracle, relies on the ∪operator Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:10 Manuel Rigger and Zhendong Su to exclude duplicate rows. The columns in the GROUP BY must correspond to those columns that are fetched. If the GROUP BY clause would contain additional columns that are not represented in <columns>, then the additional groups would be invisible for the composition operator. Similarly, if columns that are fetched are not represented in the GROUP BY clause, duplicate values would nev- ertheless be removed by ∪. Note that this might prevent some bugs from being found. The HAVING oracle validates that HAVING clauses, which are logically applied after the GROUP BY is performed, are performed correctly. Thus, unlike the DISTINCT and GROUP BY oracles, the ternary predicates are used in the HAVING clause, rather than in the WHERE clause. Example. Listing 2 gives a representative example for the grouping oracles, specifically for the DISTINCT oracle, to illustrate the format of the queries and to give an example of a bug they can detect. The original query O contains a DISTINCT clause and computes the correct value {0\|0}. The partitioning queries P compute an incorrect result {0\|NULL}, since the affinity of the view column c0 is mistakenly discarded—the affinity of a column determines what implicit conversions are performed and is a concept unique to SQLite. Note that when removing the DISTINCT clause, the query computes the same incorrect result as the partitioning queries, which is why the WHERE test oracle cannot detect this bug. As discussed, the subqueries can optionally discard the DISTINCT clause; in this example, either option would have detected the bug. Listing 2. This simplified DISTINCT test case found a bug in SQLite and exemplifies the structure of the queries of the grouping oracles. 3.3 Aggregate Functionsh The aggregate query partitioning test oracles are used to test aggregate functions. We consider the most-commonly used aggregate functions MIN(), MAX(), COUNT(), SUM(), and AVG(). Aggregate func- tions can be optimized by decomposing the computation and distributing it [Jesus et al. 2015]. We use the core idea of distributing the computation as a basis for testing aggregate functions. Self-composable aggregate functions. The simplest test oracles for aggregate functions are for self- composable aggregate functions (i.e., MIN(), MAX(), SUM(), and COUNT()). Unlike for the oracles intro- duced above, the partition for aggregate functions is an intermediate result, rather than a subset of the original query’s result set. For example, a partition of the MIN oracle computes the mini- mum value of the respective partitioning query. To compute the partitions, an additional step is necessary; for example, for MIN(), the overall minimum value must be computed. To this end, an- other aggregate function can be applied; for example, to compute the overall minimum value, MIN() can be applied once more. The aggregate function for the composition operator is not necessarily the same as for the partitioning query. Consider, for example, the COUNT oracle. The partitioning queries compute the number of rows in their partition using COUNT(). They are then summed up using the SUM() aggregate function. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:11 Finding Bugs in Database Systems via Query Partitioning Self-composable aggregate functions example. Listing 3 shows an example for an original query O , and the partitioning queries P , which is an actual test case generated by the MAX oracle, which de- tected a bug in CockroachDB. In this example, the original query fetched NULL, rather than 0, which was the result set returned by the composed partitioning queries. This bug affected interleaved ta- bles, which are used to implement parent-child relationships between tables, when experimental vectorization features were turned on. The developers explained that an incorrect predicate was used to skip interleaved child rows when performing a reverse scan. Note that the partitioning queries’ results must be assigned an alias (as aggr), so that the partitions can be composed. Listing 3. This simplified MAX() test case detected a bug in CockroachDB, and exemplifies the structure of the queries for self-composable aggregate functions. 3.3 Aggregate Functionsh SET vectorize=experimental_on; CREATE TABLE t0(c0 INT); CREATE TABLE t1(c0 BOOL) INTERLEAVE IN PARENT t0(rowid); INSERT INTO t0(c0) VALUES (0); INSERT INTO t1(rowid , c0) VALUES(0, TRUE); O SELECT MAX(t1.rowid) FROM t1; -- {NULL} P SELECT MAX(aggr) FROM ( SELECT MAX(t1.rowid) as aggr FROM t1 WHERE '+' >= t1.c0 UNION ALL SELECT MAX(t1.rowid) as aggr FROM t1 WHERE NOT('+' >= t1.c0) UNION ALL SELECT MAX(t1.rowid) as aggr FROM t1 WHERE ('+' >= t1.c0) IS NULL O SELECT MAX(t1.rowid) FROM t1; -- {NULL} P SELECT MAX(aggr) FROM ( SELECT MAX(t1.rowid) as aggr FROM t1 WHERE '+' >= t1.c0 UNION ALL SELECT MAX(t1.rowid) as aggr FROM t1 WHERE NOT('+' >= t1.c0) UNION ALL SELECT MAX(t1.rowid) as aggr FROM t1 WHERE ('+' >= t1.c0) IS NULL ); -- {0} Other composable aggregate functions. For aggregate functions that are not self-composable, but composable, such as AVG(), we can compute the results using a result tuple, rather than a single value. For example, to compute AVG(), we utilize that AVG(Q) corresponds to SUM(Q)/COU NT(Q). Accordingly, each partitioning query computes a tuple (SUM(Qp),COU NT(Qp)), which is then composed by dividing the sum of the first tuple values by the sum of the second. Listing 4. This simplified AVG test case demonstrates a bug in DuckDB, and shows the structure of the queries for composable, but not self-composable aggregate functions. Listing 4. This simplified AVG test case demonstrates a bug in DuckDB, and shows the structure of the queries for composable, but not self-composable aggregate functions. CREATE TABLE t0(c0 BIGINT); INSERT INTO t0(c0) VALUES (2); INSERT INTO t0(c0) VALUES (9223372036854775807); O SELECT AVG(t0.c0) FROM t0; -- {4.611686018427388e+18} P SELECT SUM(s)/SUM(c) FROM ( SELECT SUM(t0.c0) AS s, COUNT(t0.c0) AS c FROM t0 WHERE c0 UNION ALL SELECT SUM(t0.c0) AS s, COUNT(t0.c0) AS c FROM t0 WHERE NOT c0 UNION ALL SELECT SUM(t0.c0) AS s, COUNT(t0.c0) AS c FROM t0 WHERE c0 IS NULL ); -- { -4611686018427387903} INSERT INTO t0(c0) VALUES (9223372036854775807); O SELECT AVG(t0.c0) FROM t0; {4.611686018427388e 18} P SELECT SUM(s)/SUM(c) FROM ( SELECT SUM(t0.c0) AS s, COUNT(t0.c0) AS c FROM t0 WHERE c0 UNION ALL SELECT SUM(t0.c0) AS s, COUNT(t0.c0) AS c FROM t0 WHERE NOT c0 UNION ALL SELECT SUM(t0.c0) AS s, COUNT(t0.c0) AS c FROM t0 WHERE c0 IS NULL ); -- { -4611686018427387903} Composable aggregate function example. 3.3 Aggregate Functionsh Listing 4 gives a concrete example on an AVG oracle test case that found a bug in DuckDB. Query O shows an original query that computes the AVG() of the values contained in column c0. Each partitioning query P computes two values, one being the sum (aliased as s) and one being the count of values in c0 (aliased as c). The expression SUM(s)/SUM(c) is associated with the composition operator; it divides the accumulated sums with the accumulated counts. For this test case, DuckDB computed the correct result for the original query. For the partitioning queries, only the first aggregate query fetches a row, which is expected. However, the addition of 9223372036854775807 and 2 in SUM() overflowed, which was an unexpected result Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:12 Manuel Rigger and Zhendong Su and caused a silent wraparound. The bug was confirmed as a real bug. However, the developers were already aware of it, and had not addressed it, since addressing this bug without significant performance impact was non-trivial. After we reported it, they nevertheless decided to fix it. Commutativity. All the aggregate functions that we considered are commutative. For our purpose, we also assume SUM() and AVG() to be commutative, although the processing order matters for floating-point numbers. To account for rounding errors caused by this, we compare floating-point numbers in the result sets using an epsilon. Other non-commutative aggregate functions, such as GROUP_CONCAT(), which concatenates strings, exist. In order to support these, an operator-specific comparator could be implemented. For example, a comparator for GROUP_CONCAT() could split the concatenated string by its delimiter(s), sort the tokens, and use the sorted tokens for comparison. Such an implementation would be more tedious compared to the other test oracles. Furthermore, non-commutative functions provide less optimization potential for the DBMSs. Thus, we did not consider non-commutative functions further in our work. 4 SELECTED BUGSh This section gives an overview of interesting bugs that we found using TLP. This selection is nec- essarily subjective, and we sought to demonstrate the range of different bugs that the individual oracles detected. For brevity, we show only reduced test cases that demonstrate the underlying core problem, rather than the original and partitioning queries that found the bugs. 4.1 WHERE Clauses This section presents bugs detected by the WHERE oracle. Unless otherwise noted, these bugs can also be detected by NoREC and PQS. Note that in Section 5.2, we systematically investigate the relationship between the WHERE oracle and NoREC. MySQL comparison bug. Listing 5 shows a bug where a comparison of numbers yielded an incorrect result. The comparison 0.9 > t0.c0 should evaluate to TRUE for c0=0 and fetch the row in t0. However, MySQL failed to fetch the row. This is one of multiple basic bugs that we found in MySQL. We still consider it interesting, since it shows that even mature DBMSs are prone to such bugs. Listing 5. MySQL incorrectly evaluated the comparison and failed to fetch the row. Listing 5. MySQL incorrectly evaluated the comparison and failed to fetch the row. CREATE TABLE t0(c0 INT); INSERT INTO t0(c0) VALUES (0); SELECT * FROM t0 WHERE 0.9 > t0.c0; -- {0} {} CREATE TABLE t0(c0 INT); INSERT INTO t0(c0) VALUES (0); SELECT * FROM t0 WHERE 0.9 > t0.c0; -- {0} {} SELECT * FROM t0 WHERE 0.9 > t0.c0; -- {0} {} TiDB comparison bug. We found a bug in TiDB where fetching from a view unexpectedly omitted a row (see Listing 6). The WHERE clause should evaluate to TRUE and fetch a row, since it refers to the constant value 1 in the view. However, TiDB unexpectedly did not fetch a row. The bug was classified as a P1 bug, which is the second-highest severity category. We believe that this bug is interesting, since it demonstrates that our approach can detect bugs in views, without specifically aiming to test them. Listing 6. TiDB failed to fetch a row from a view. CREATE TABLE t0(c0 INT); CREATE VIEW v0(c0, c1) AS SELECT t0.c0, 1 FROM t0; INSERT INTO t0 VALUES (0); SELECT v0.c0 FROM v0, t0 WHERE v0.c1; -- {0} {} Listing 6. TiDB failed to fetch a row from a view. CREATE TABLE t0(c0 INT); CREATE VIEW v0(c0, c1) AS SELECT t0.c0, 1 FROM t0; INSERT INTO t0 VALUES (0); SELECT v0.c0 FROM v0, t0 WHERE v0.c1; -- {0} {} Listing 6. TiDB failed to fetch a row from a view. CREATE TABLE t0(c0 INT); CREATE VIEW v0(c0, c1) AS SELECT t0.c0, 1 FROM t0; INSERT INTO t0 VALUES (0); SELECT v0.c0 FROM v0, t0 WHERE v0.c1; -- {0} {} Proc. ACM Program. Lang., Vol. CREATE TABLE t0 (c0 DECIMAL PRIMARY KEY , c1 INT UNIQUE); 487610} {1.81948761E+9} Missing error for invalid regular expression. We found a bug in CockroachDB where an invalid regular expression caused a SELECT to retrieve an empty result set, rather than printing an error message (see Listing 8). We found this bug because none of the partitioning queries fetched any rows. Both PQS and NoREC could not detect such bugs, since for these approaches, the original query would result in the expected error above. Rigger and Su [2020a] specifically explain that NoREC cannot detect errors due to nondeterminism in the evaluation of queries. Rather than exiting with an error, CockroachDB returned an empty result set for this query. Listing 8. Rather than exiting with an error, CockroachDB returned an empty result set for t CREATE VIEW v0(c0) AS SELECT COUNT_ROWS () FROM t0; CREATE VIEW v0(c0) AS SELECT COUNT_ROWS () FROM t0; SELECT * FROM v0 WHERE '' !~ '+'; -- error parsing regexp: missing argument to repetition operator: + {} Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 4.1 WHERE Clauses 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:13 Finding Bugs in Database Systems via Query Partitioning ORDER BY affects a query’s result. We found a bug in CockroachDB, where a value was unexpect- edly represented using the E notation (see Listing 7). Specifically, while the default row engine encodes the fetched decimal value as 1819487610, the vector-based engine, which was used for the partitioning queries, represented the value as 1.81948761E+9. While this was confirmed as a bug, it was not deemed to be very important, considering that both represent the same value. How- ever, we believe that this bug is interesting, since it demonstrates that an ORDER BY can incorrectly influence a query’s result. Listing 7. The ORDER BY clause affected the representation of the decimal value 1819487610 when vector-based execution engine in CockroachDB. SET SESSION VECTORIZE=on; SET SESSION VECTORIZE=on; SET SESSION VECTORIZE=on; CREATE TABLE t0 (c0 DECIMAL PRIMARY KEY , c1 INT UNIQUE); INSERT INTO t0(c0) VALUES (1819487610); SELECT t0.c0 FROM t0 ORDER by t0.c1; -- {1819487610} {1.81948761E+9} CREATE TABLE t0 (c0 DECIMAL PRIMARY KEY , c1 INT UNIQUE); SELECT t0.c0 FROM t0 GROUP BY t0.c0 HAVING NOT (VARIANCE (0) IS NULL); --{} {0} SELECT t0.c0 FROM t0 GROUP BY t0.c0 HAVING NOT (VARIANCE (0) IS NULL); --{} {0} Non-deterministic MAX(). We found a bug in DuckDB, where a complex query using GROUP BY and HAVING clauses, as well as UNION resulted in a nondeterministic result (see Listing 11). As explained by the developers, this bug was caused since non-inlined strings were not being properly copied into the hash table when stored as MAX() values. Since this lead to a user-after-free error, this bug might have also been detected by undefined-behavior checkers [Regehr 2010; Stepanov and Serebryany 2015]. We believe that this bug is interesting nevertheless, since it demonstrates the range of bugs that TLP can detect. Listing 11. DuckDB nondeterministically fetched two and three rows for this query. g y q y CREATE TABLE t0(c0 INT); CREATE TABLE t1(c0 VARCHAR); INSERT INTO t1 VALUES (0.9201898334673894) , (0); INSERT INTO t0 VALUES (0); SELECT * FROM t0, t1 GROUP BY t0.c0, t1.c0 HAVING t1.c0!=MAX(t1.c0) UNION ALL SELECT * FROM t0, t1 GROUP BY t0.c0, t1.c0 HAVING NOT t1.c0>MAX(t1.c0); -- nondeterministic result INSERT INTO t1 VALUES (0.9201898334673894) , (0); Non-deterministic GROUP BY. We found a bug in TiDB, where a SELECT nondeterministically fetched a duplicate row (see Listing 12). We could reproduce the bug only with a large number of rows; note that we removed the INSERTs from the listing for brevity. We believe that is likely a bug that is caused by a race condition. TiDB is written in Go, for which race detectors seem to exist, indicating that such a bug might have been found by them. However, race condition checkers are known to be slow [Serebryany et al. 2011], and TLP might be a viable and cheaper alternative to identify test cases that trigger race conditions. Listing 12. TiDB computed a non-deterministic result for this query. CREATE TABLE t0(c0 INT , c1 INT); CREATE TABLE t1(c0 INT , c1 INT); CREATE TABLE t2(c0 INT , c1 INT); -- 27 INSERTS ANALYZE TABLE t1, t2; SELECT t1.c0 LIKE t1.c0 FROM t1, t2, t0 GROUP BY t1.c0 LIKE t1.c0; -- nondeterministic result Listing 12. TiDB computed a non-deterministic result for this query. 4.2 Grouping Bugs is section presents bugs that were detected by the GROUP BY, HAVING, and DISTINCT orac GROUP BY disregards COLLATE. We found a bug in DuckDB, where the GROUP BY operator disre- garded a COLLATE NOCASE (see Listing 9). Note that a COLLATE clause controls the behavior of compar- isons for strings; in this example, it specifies that string comparisons should be performed without considering the case of the strings. While the SELECT was expected to return a result set containing either 'a' or 'A', it fetched both. The GROUP BY oracle detected this bug, since, unlike the GROUP BY operator, the UNION operator respected the COLLATE. This bug is interesting, since it demonstrates a basic bug in the operator itself, rather than an optimization bug, to which NoREC is limited. How- ever, we found this bug shortly after COLLATEs were merged to master, and before this feature was released, suggesting that this feature was not yet thoroughly tested. Listing 9. The GROUP BY operator disregarded that c0 has a COLLATE NOCASE in DuckDB CREATE TABLE t0(c0 VARCHAR COLLATE NOCASE); INSERT INTO t0(c0) VALUES ('a'), ('A'); SELECT t0.c0 FROM t0 GROUP BY t0.c0; -- {'a'} or {'A'} {'a', 'A'} } Incorrect VARIANCE(0) optimization. We found a bug in CockroachDB where VARIANCE(0) IS NULL was unexpectedly optimized to FALSE (see Listing 10). Interestingly, VARIANCE(0) evaluates to NULL if the table contains zero or one rows; if the table contains at least two rows, it evaluates to 0. The Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:14 Manuel Rigger and Zhendong Su optimization was thus incorrect for this case, where the table contained only one row. We believe that this case is interesting, since aggregate functions cannot be used in WHERE clauses, so although this is an optimization bug, it could not have been found by NoREC. Listing 10. CockroachDB unexpectedly optimized VARIANCE(0) to FALSE. CREATE TABLE t0(c0 INT); INSERT INTO t0(c0) VALUES (0); SELECT t0.c0 FROM t0 GROUP BY t Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 4.3 Aggregate Bugs DuckDB assumed that no MIN() value was set, since the minimum value corresponds CREATE TABLE t0(c0 INT); INSERT INTO t0 VALUES (-1); SELECT MIN(CAST(c0 AS BIGINT) << 63) FROM t0; -- { -9223372036854775808} {N SELECT MIN(CAST(c0 AS BIGINT) << 63) FROM t0; - SELECT MIN(CAST(c0 AS BIGINT) << 63) FROM t0; -- { -9223372036854775808} {NULL} Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 4.3 Aggregate Bugs Similar to grouping bugs, aggregate functions are interesting since they cannot be detected by either NoREC or PQS. Similar to grouping bugs, aggregate functions are interesting since they cannot be detected by either NoREC or PQS. MAX() and UTF-16 bug. We found a bug in SQLite, where MAX() computed an incorrect result for the ordering of UTF-16 strings and non-ASCII characters (see Listing 13). The SQLite developers Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:15 Finding Bugs in Database Systems via Query Partitioning explained that SQLite was incorrectly using the UTF-8 collating sequence with some, but not all expressions in the database having a UTF16LE encoding. Although this bug might seem obscure, we believe that it is interesting, because it would go likely undetected by users, but result in unex- pected results when an application relies on it. Listing 13. SQLite computed an unexpected ordering for special non-ASCII characters and UTF-16LE encod- ing. PRAGMA encoding = 'UTF -16'; CREATE TABLE t0(c0 TEXT); INSERT INTO t0(c0) VALUES ('฀'), (1); SELECT MAX(CASE 1 WHEN 1 THEN t0.c0 END) FROM t0; -- {0} {'฀'} SUM() optimization. Listing 14 demonstrates an optimization bug in DuckDB. As explained by the developers, to sum up the constants, an optimization sum += input * count was applied, where input refers to the constant -1. Since count was declared as an unsigned integer, the result was cast to an unsigned number, resulting in an underflow [Dietz et al. 2012]. This finding demonstrates that also aggregate functions are affected by optimization bugs, which NoREC is unable to find. Listing 14. DuckDB computed an incorrect result due to an optimization that summed up constants by using an unsigned, rather than a signed integer. CREATE TABLE t0 (c0 INT); INSERT INTO t0 VALUES (0); SELECT SUM(-1) FROM t0; -- CREATE TABLE t0 (c0 INT); INSERT INTO t0 VALUES (0); } ELECT SUM(-1) FROM t0; -- {-1} MIN() initialization bug. Listing 15 demonstrates a bug in MIN() in DuckDB. The culprit was that the minimum value of the domain, −263 for integers, was used to indicate whether a minimum value has been set. Since the expression CAST(c0 as BIGINT)<< 32 sets the minimum value for c0=-1, the implementation mistakenly assumed that no minimum value was set, and returned NULL. Listing 15. 1For PostgreSQL, MySQL, and SQLite, only (inofficial) GitHub mirrors are available. 2These numbers are not accurate, but represent a best-effort estimate. We omitted counting tests, where this was possible (using cloc). For TiDB, we counted the repositories of PD, TiKV, and TiDB, which are all necessary to run TiDB. 3PingCAP implemented PQS for TiDB; for the other DBMSs, the approaches were implemented as part of the evaluation of the respective papers [Rigger and Su 2020a,c]. 5 EVALUATION Online Transactional Processing (OLTP) workloads are those that con- sist of frequent inserts, updates, and deletes. In contrast, Online Analytical Processing (OLAP) workloads typically involve complex queries with aggregates. Traditional systems, NewSQL sys- tems, and SQLite are mostly optimized towards OLTP workloads. DuckDB is a representative of an OLAP systems, and stores its data column-wise. CockroachDB and TIDB are mainly developed commercially (by Cockroach Labs and PingCAP); they provide an open version of their DBMSs on GitHub, which we tested. DuckDB has been developed by a research group, but “is intended to be a stable and mature database system.” SQLite is developed by a small development team lead by D. Richard Hipp. MySQL has open-source contributors, and is also developed by Oracle. PostgreSQL is backed by open-source contributors. an application’s process. Traditional systems like MySQL [2020] and PostgreSQL [2020] are stan- dalone, meaning that they run in a dedicated process. NewSQL systems like CockroachDB [Taft et al. 2020] and TiDB [Huang et al. 2020] are distributed relational DBMSs, which aim to provide a high degree of scalability by splitting up the database [Pavlo and Aslett 2016]; however, we tested only their SQL component. Online Transactional Processing (OLTP) workloads are those that con- sist of frequent inserts, updates, and deletes. In contrast, Online Analytical Processing (OLAP) workloads typically involve complex queries with aggregates. Traditional systems, NewSQL sys- tems, and SQLite are mostly optimized towards OLTP workloads. DuckDB is a representative of an OLAP systems, and stores its data column-wise. CockroachDB and TIDB are mainly developed commercially (by Cockroach Labs and PingCAP); they provide an open version of their DBMSs on GitHub, which we tested. DuckDB has been developed by a research group, but “is intended to be a stable and mature database system.” SQLite is developed by a small development team lead by D. Richard Hipp. MySQL has open-source contributors, and is also developed by Oracle. PostgreSQL is backed by open-source contributors. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 5 EVALUATION We evaluated both the effectiveness and generality of TLP in finding bugs, compared TLP to NoREC, investigated the overlap between the individual test oracles, and systematically studied how additions to the database and expression generator affect TLP’s bug-finding capabilities. Implementation. We implemented our approach in SQLancer, in which also PQS and NoREC were implemented. Since SQLancer did not support the generation of databases and queries for TiDB and DuckDB, we added support for these systems. Furthermore, we implemented the generation of many small, previously unsupported features in SQLancer (e.g., generating arrays and array operations in CockroachDB). The test oracles are implemented in about 500 LOC for each DBMS under test. Note that our implementation is available at https://github.com/sqlancer. Tested DBMSs. In our evaluation, we considered six popular and widely-used DBMSs with a wide range of characteristics to demonstrate the generality of TLP (see Table 2). SQLite [2020] and DuckDB [Raasveldt and Mühleisen 2020] are both embedded DBMSs, meaning that they run within Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:16 211:16 211:16 Manuel Rigger and Zhendong Su Table 2. We tested a diverse set of popular and emerging DBMSs; all numbers are the latest as of May 2020. Table 2. We tested a diverse set of popular and emerging DBMSs; all numbers are the latest as of May 2020. Popularity Rank DBMS DB- Engines Stack Over- flow GitHub Stars 1 LOC 2 First Release Kind Tested By SQLite 9 4 1.5k 0.3M 2000 Embedded, OLTP PQS, NoREC MySQL 2 1 5.0k 3.8M 1995 Traditional PQS PostgreSQL 4 2 6.3k 1.4M 1996 Traditional PQS, NoREC CockroachDB 77 - 17.7k 1.1M 2015 NewSQL NoREC TiDB 118 - 23.1K 0.8M 2017 NewSQL PQS3 DuckDB - - 0.5k 59k 2018 Embedded, OLAP - Table 2. We tested a diverse set of popular and emerging DBMSs; all numbers are the latest as of May 2020. ed a diverse set of popular and emerging DBMSs; all numbers are the latest as of May 2020. an application’s process. Traditional systems like MySQL [2020] and PostgreSQL [2020] are stan- dalone, meaning that they run in a dedicated process. NewSQL systems like CockroachDB [Taft et al. 2020] and TiDB [Huang et al. 2020] are distributed relational DBMSs, which aim to provide a high degree of scalability by splitting up the database [Pavlo and Aslett 2016]; however, we tested only their SQL component. 5.1 Effectiveness For both TiDB and MySQL, we omitted to report a number of suspected bugs found by the WHERE oracle, due to the large num- ber of open bugs. For TiDB, 35 bugs were verified, but have not yet been addressed. For MySQL, 6 bugs were confirmed, but not fixed. Many of the MySQL bugs were rather basic (e.g. see Listing 5), which prevented us from testing MySQL more comprehensively. Furthermore, we found a large number of open bugs in its bug tracker; in fact, even a large number of the bugs reported by the authors of PQS have not yet been addressed. In addition, MySQL has a closed development pro- cess, with only the release versions being publicly available. These appear every 2–3 months. We found one bug in MySQL version 8.0.19, which was fixed quickly, but will appear only in MySQL 8.0.21 (i.e., potentially half a year later). We implemented the WHERE Extended oracle for only one DBMS—CockroachDB—after the simple WHERE oracle could not find additional bugs. It did not de- tect any additional bugs, which is expected; as explained in Section 3.1, this oracle is useful mainly to utilize an existing test suite that contains queries that have WHERE clauses. Found bugs. Table 3 shows the number of bugs we found and the status of the corresponding bug reports. We opened 181 bug reports, 175 which were either fixed or confirmed by the developers. 125 bugs have been fixed, which demonstrates that the DBMS developers considered the majority of the bugs to be important. Almost all bugs were addressed by code changes; only 1 bug was addressed by a documentation change. The behavior in 3 bug reports was surprisingly considered to be intended by the bug verifiers of MySQL; we discuss one of them in detail below. We opened only 3 duplicate bug reports, as we carefully checked the bug tracker for similar bugs. We could comprehensively test SQLite, CockroachDB, PostgreSQL, and DuckDB. That is, we were not re- stricted by any open bug reports that could have prevented us from testing by making it difficult to filter out duplicate test cases for bugs. For TiDB and MySQL, we stopped testing due to the large number of open bug reports. We found more bugs in DuckDB and TiDB than in the other DBMSs, since the other DBMSs were comprehensively tested by NoREC and PQS (see Table 2). 5.1 Effectiveness Study methodology and challenges. We started testing the DBMSs while implementing our ap- proach, and tested them over a period of roughly three months. A significant factor limiting our bug-finding efforts were duplicate test cases for bugs. For a single bug, SQLancer typically gener- ated many test cases that would trigger it, making it infeasible to filter out such test cases man- ually, which was also observed by Rigger and Su [2020a,c]. While automatic bug prioritization approaches were proposed for compilers [Chen et al. 2013], applying them for DBMSs would be more challenging and slow due to needing to install, set up, and stop many versions of a single DBMS. To address this, we typically avoided the generation of features that induce already known bugs; however, this was not always possible—for example, when we could not discover the neces- sary conditions to reproduce the bug—or restricted the bug-finding capabilities significantly (e.g., by avoiding the generation of comparisons). When we found a bug, we first automatically reduced it [Regehr et al. 2012], to then manually produce a minimal test case that demonstrated the under- lying bug. Before reporting a bug, we checked the public bug trackers for similar bugs, to avoid creating duplicate bug reports. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:17 Finding Bugs in Database Systems via Query Partitioning Finding Bugs in Database Systems via Query Partitioning Table 3. We found 175 previously unknown bugs, 125 of which have been fixed. Closed DBMS Fixed Verified Intended Duplicate SQLite 4 0 0 0 MySQL 1 6 3 0 CockroachDB 23 8 0 0 TiDB 26 35 0 1 DuckDB 72 0 0 2 Oracle implementation. Since the WHERE oracle is the simplest oracle, we implemented it first for every DBMS. We found that the other test oracles detected bugs that also the WHERE oracle could find. Since the other oracle’s test cases were typically more complex than the ones generated by the WHERE oracle (e.g. the GROUP BY oracle could detect many of the same bugs, but by generating unnecessary GROUP BYs), it was not desirable to use them before the WHERE oracle’s bug-finding satu- rated. Consequently, we implemented the other oracles only for DBMSs for which the WHERE oracle saturated, namely SQLite, PostgreSQL, DuckDB, and CockroachDB. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 5.1 Effectiveness The other oracles detected 17 bugs in total; although many of these bugs were serious, the number of found bugs is low when compared to the bugs found by the WHERE oracle. Our analysis suggests that the fea- tures tested by these oracles rely mostly on functionality in the DBMSs that is tested by the WHERE oracle; in Section 5.3, we investigate this hypothesis based on coverage information. Besides logic bugs, we found 25 crash bugs and 62 error bugs. Crash bugs refer to process crashes (e.g., a mem- ory error resulting in a SEGFAULT). Error bugs were due to unexpected errors in the DBMSs (e.g., internal errors printing a stack trace). The higher number of crash bugs in DuckDB is explained by us using the debug build for testing, which resulted in assertion violations, which accounted for 11 of the crash bugs. While the developer also appreciated those bug reports, finding them was not a goal for us, since they could have been found by existing approaches, such as fuzzers. We report these numbers nevertheless, to put the numbers of found logic bugs into relation. As with PQS and NoREC, we found a larger number of error and crash bugs than logic bugs. False positives. In theory, our approach should not be affected by false positives (i.e., when SQLancer reports a bug, it is always a real bug). However, the MySQL bug verifiers considered 3 of our bug reports as false positives. For example, consider the test case in Listing 16. The WHERE oracle found an inconsistency in the result, since neither of the partitioning queries fetched a row, while the original query fetched the row with c0=0. While reducing the bug, we found strong evi- dence that led us to believe that this was indeed a bug. First, rewriting the query to evaluate the predicate indicated that the predicate should evaluate TRUE, so also NoREC would have considered it to be a bug. Second, the query’s behavior changes when omitting the UNIQUE constraint and yields the result we would expect; this is unexpected, because an index should never affect a query’s re- sult. Third, an earlier version of MySQL computed the result we expected. Fourth, TiDB, which strives to be compatible with MySQL, computed the result we would expect. Despite these, the re- port was closed, based on the argument that Oracle !8c computes the same result. 5.1 Effectiveness We did not find any bugs in PostgreSQL, which is why we omitted this DBMS from the table. This is not surprising. PQS detected only one logic bug in it, and NoREC did not detect any logic bugs. Test oracles. Table 4 shows how many bugs each individual test oracle detected. In total, we found 77 logic bugs. The WHERE oracle detected 60 bugs, suggesting that—even though it is conceptu- ally the simplest oracle—it is the most effective one. For DBMSs that were intensively tested, like Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. Manuel Rigger and Zhendong Su 211:18 Table 4. We found 60 bugs with the WHERE oracle, 10 with the aggregate oracle, 3 with the HAVING oracle, and the others by internal DBMS errors and crashes. Table 4. We found 60 bugs with the WHERE oracle, 10 with the aggregate oracle, 3 with the HAVING oracle, and the others by internal DBMS errors and crashes. Query Partitioning Oracle DBMS WHERE Aggregate GROUP BY HAVING DISTINCT Error Crash SQLite 0 3 0 0 1 0 0 CockroachDB 3 3 0 1 0 22 2 TiDB 29 0 1 0 0 27 4 MySQL 7 0 0 0 0 0 0 DuckDB 21 4 1 2 1 13 19 Listing 16. The report associated with this test was considered a false positive by the MySQL bug verifiers. CREATE TABLE t0(c0 DECIMAL UNIQUE); Listing 16. The report associated with this test was considered a false positive by the MySQL bug verifiers. CREATE TABLE t0(c0 DECIMAL UNIQUE); INSERT INTO t0(c0) VALUES (0); Listing 16. The report associated with this test was considered a false positive by the MySQL bug verifiers. CREATE TABLE t0(c0 DECIMAL UNIQUE); INSERT INTO t0(c0) VALUES (0); SELECT * FROM t0 WHERE '' BETWEEN t0.c0 AND t0.c0; -- {0} {} CREATE TABLE t0(c0 DECIMAL UNIQUE); INSERT INTO t0(c0) VALUES (0); SELECT * FROM t0 WHERE '' BETWEEN t0.c0 AND t0.c0; -- {0} {} SELECT * FROM t0 WHERE '' BETWEEN t0.c0 AND t0.c0; -- {0} {} SQLite and CockroachDB, this oracle was less effective. In Section 5.2, we will thus closely inves- tigate the relationship between NoREC and the query partitioning WHERE oracle. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 5.2 Comparison with NoREC We studied how NoREC relates to TLP. Both NoREC and TLP are metamorphic oracles and have similar advantages (e.g., the small implementation effort required to realize them) as well as the same disadvantage (i.e., they cannot establish a ground truth). We did not compare to PQS, which is complementary to TLP and NoREC, and whose advantages and disadvantages were already studied [Rigger and Su 2020a]. Both PQS and NoREC are mostly limited to finding bugs in WHERE clauses and are not applicable to the other features TLP can test (see Section 2). Consequently, we compare only the TLP WHERE oracle with NoREC. Methodology. Fairly comparing the two techniques is challenging. Optimally, we could apply each technique to the same DBMS and compare the number of distinct bugs that they find. However, determining whether a test case triggers a specific bug would be difficult and labor-intensive to determine [Marcozzi et al. 2019]. Given that NoREC had been used to test two DBMSs before we tested them using TLP—SQLite and CockroachDB—analyzing any additional bugs that the WHERE oracle found gives an insight into what additional bugs it can find. Similarly, for DBMSs in which the WHERE oracle did not find any additional bugs, NoREC could be applied to validate whether it can find any additional bugs. Given that DuckDB is the only DBMS that had not been tested by NoREC, and on which our testing efforts saturated, we implemented and tested NoREC only on this DBMS. In addition, we sought to give an estimate on the oracles’ overlap based on a manual analysis of the found bugs. Specifically, we tried to translate a NoREC test case to a WHERE oracle test case and vice versa, by following a similar methodology as for the comparison of NoREC and PQS [Rigger and Su 2020a]. For the majority of cases, this is straightforward. To translate a NoREC test case to a WHERE oracle test case, we can take the original query with a WHERE clause, and create the two other partitioning queries by assuming the WHERE clause predicate to be the randomly- generated predicate based on which the ternary variants are derived. To obtain the original query, the WHERE clause must be removed. Similarly, to translate a WHERE oracle test case to a NoREC test case, one of the partitioning queries can be assumed as the original query for NoREC. 5.1 Effectiveness After we further inquired, a second bug verifier subsequently elaborated that an empty STRING cannot represent a valid DECIMAL value, referring to the SQL standard. While indeed a warning that the empty string is not a valid DECIMAL is printed, such warnings are printed for many other queries too, which do not violate the TLP assumptions. Thus, we still believe that TLP does not result in false positives. 211:19 Finding Bugs in Database Systems via Query Partitioning Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 5.2 Comparison with NoREC The more complex translated query does not hinder the effectiveness of NoREC’s bug-finding capabilities, since, as with the original approach, the expression has to be evaluated on every record of the target tables, disabling many optimizations. Overall, we did not find any bugs using NoREC on DuckDB. Note that we verified that NoREC could have detected bugs that were found by the WHERE oracle. Manual analysis of the NoREC bugs. In total, NoREC found 50 bugs. We could mechanically trans- late 42 NoREC test cases so that the bugs could have been found using the WHERE oracle. For 8 test cases, a mechanic translation as described above was not possible. We identified two root causes for this. The first one was that NoREC detected the bug in an aggregate function that was used to efficiently sum up for how many records a predicate evaluates to TRUE in the translated, unopti- mized query, and affected 4 cases. We speculate that these bugs might have been found by one of the TLP aggregate oracles. The second root cause was that the bug was unexpectedly triggered in the translated, unoptimized query, which evaluates the predicate on every row, which affected 4 cases. We believe that the WHERE oracle might overlook these bugs, since it does not compare to which value a predicate is evaluated when used in a different context. Manual analysis of the TLP WHERE bugs. We analyzed all 60 bugs found by the WHERE oracle (count- ing also the 3 bugs described above). For 48 bugs, we could mechanically derive NoREC test cases. In 5 of these cases, comparing the record count was insufficient to detect the bug; also the contents had to be compared, contrary to prior suggestions [Rigger and Su 2020a]. For the other 12 bugs, it is doubtful that NoREC could have detected them. 3 test cases triggered bugs related to joins and did not require a WHERE clause. Although the WHERE clauses were redundantly generated by the WHERE oracle, it detected these bugs, because the overall number of fetched rows mismatched. 3 test cases triggered bugs in operators, both in NoREC’s unoptimized and optimized case. Further- more, we found 1 bug that was triggered in the UNION operator, which is out-of-scope for NoREC. 5.2 Comparison with NoREC In fact, this was not necessary for many queries, as we typically used a NoREC test case to demonstrate the underlying bug, which is more compact than a TLP test case. The limitation of this manual analysis is that for bugs for which we cannot derive an equivalent test case, we cannot necessarily conclude that no such test case exists, because a different test case might trigger the same underlying bug. Additional WHERE bugs in DBMSs tested by NoREC. SQLite and CockroachDB were extensively tested by NoREC, and we found 3 additional bugs in them using the WHERE oracle (all in Cock- roachDB). In a first step, we closely analyzed these bugs to determine whether NoREC could have found them, using the methodology to translate test cases described above. One bug could have been found directly by NoREC; we speculate that it was not found because the test case triggering the bug relied on the INTERVAL data type, which we added to SQLancer, and which was not present when NoREC was evaluated. The bug in Listing 7 could have been found by NoREC, but only if the content of the records is fetched in the translated query, which was described as unnecessary by Rigger and Su [2020a]. The bug in Listing 8 could not have been found by NoREC, since the translated query results in the expected error, rather than yielding an unexpected result. Additional NoREC bugs in DBMSs tested by TLP WHERE. DuckDB is the only DBMS for which our bug-finding efforts saturated, and which has not yet been tested by NoREC. Thus, we implemented NoREC for this DBMS to determine whether NoREC could find any bugs in this system. Note that DuckDB does not provide the IS TRUE and IS FALSE operators, which are used in the translated query that the DBMS is unlikely to optimize. However, this is not problematic, since the translation can be implemented using other operators. Specifically, an original, potentially optimized query with a Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:20 Manuel Rigger and Zhendong Su predicate p can be translated to a query SELECT SUM(count) FROM (SELECT (p IS NOT NULL AND p)::INT as count FROM <tables>), so that the size of the original query’s result set must be equal to the count obtained by the second query. 5.2 Comparison with NoREC 1 bug was due to a hint to the query optimizer, which also took effect when used in the translated, unoptimized query, but not in all of the partitioning queries. As mentioned above, one test case resulted in an incorrect result, rather than an error. 3 test cases induced undefined behavior, but did not result in an unexpected result when using NoREC. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 5.3 Test Oracle Coverage During our experiments, we found that different oracles can detect the same underlying bugs in a number of cases, which is an expected behavior. For example, the WHERE oracle specifically aims at testing WHERE clauses, but also the subsequent oracles generate WHERE clauses, and thus might detect bugs in their handling. However, subsequent oracles are not guaranteed to find all bugs; for example, the GROUP BY oracle might overlook bugs in the handling of WHERE since an optimization might no longer be applicable when using a GROUP BY. Furthermore, it would be preferable to use the WHERE oracle even for bugs that also the GROUP BY oracle can find, since developers typically strive to understand a bug based on a minimal example, where redundant GROUP BY clauses would slow down triaging and the reduction of the bug, presenting an impediment. To investigate the overlap quantitatively, we measured the coverage of individual and combined oracles on DuckDB. DuckDB is a good choice for this, since we tested this DBMS comprehensively, and since every oracle found bugs that were not found by the other oracles. Figure 2 displays the line coverage, when running each of the 15 configurations for 10 hours. The barplots show the coverage of the individual oracles. The dotted red line, which rises starting from the left, illustrates Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. Finding Bugs in Database Systems via Query Partitioning 211:21 55.3 55.9 56 56 56.1 56.1 56 56.1 56 55.6 54 55 56 57 WHERE GROUP BY HAVING DISTINCT Aggregate Test oracle line coverage in % Fig. 2. The bar plots present the coverages of the individual oracles. The red, dotted line and the numbers above it denote the accumulated coverage of the oracles to the left side. The blue, dashed line and the numbers below it denote the same for the oracles to the right side. We executed SQLancer using 10 threads on an AMD Ryzen Threadripper 2990WX processor with 32 cores and 32GB RAM running an Ubuntu 18.04 64-bit OS. 55.3 55.9 56 56 56.1 56.1 56 56.1 56 55.6 54 55 56 57 WHERE GROUP BY HAVING DISTINCT Aggregate Test oracle line coverage in % Fig. 2. The bar plots present the coverages of the individual oracles. 5.3 Test Oracle Coverage Nevertheless, a coverage increase can be observed when adding oracles, and indeed each test oracle found unique bugs. Despite this evidence that there is a large overlap, it should be noted that coverage information id l li it d i i ht f DBMS J t l [2019] f d th t th t f 5.3 Test Oracle Coverage The maximum coverage that is achieved by utilizing all test oracles is 56.1%. The coverage is rather low, because we did not test components such as subqueries, window functions, transactions, and sequences, as well as due to code that is never executed (e.g., due to external dependencies). In comparison, PQS achieved only a coverage ranging from 23.7% to 43.0%. By generating databases alone, a test coverage of 48.3% is already achieved. Each test oracle achieves a similar coverage; the range of test coverages is 0.6% (i.e., ranging from 55.3% to 55.9%). When using test oracles in combination, a small coverage increase can be observed, independent from in which order oracles are combined. However, the HAVING oracle seems to decrease the coverage, presumably since it lowers the throughput of the other oracles. Overall, we believe that these findings confirm our assumption that the oracles stress a large common part of the DBMSs. Nevertheless, a coverage increase can be observed when adding oracles, and indeed each test oracle found unique bugs. Despite this evidence that there is a large overlap, it should be noted that coverage information provides only limited insights for DBMSs. Jung et al. [2019] found that the core components of DBMSs achieve a coverage of >95% after running only tens of queries. Rigger and Su [2020a] argued, for NoREC, that coverage information is not insightful, and that they found many bugs in SQLite despite its impressive test suite, which provides 100% MC/DC coverage. The maximum coverage that is achieved by utilizing all test oracles is 56.1%. The coverage is rather low, because we did not test components such as subqueries, window functions, transactions, and sequences, as well as due to code that is never executed (e.g., due to external dependencies). In comparison, PQS achieved only a coverage ranging from 23.7% to 43.0%. By generating databases alone, a test coverage of 48.3% is already achieved. Each test oracle achieves a similar coverage; the range of test coverages is 0.6% (i.e., ranging from 55.3% to 55.9%). When using test oracles in combination, a small coverage increase can be observed, independent from in which order oracles are combined. However, the HAVING oracle seems to decrease the coverage, presumably since it lowers the throughput of the other oracles. Overall, we believe that these findings confirm our assumption that the oracles stress a large common part of the DBMSs. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 5.3 Test Oracle Coverage The red, dotted line and the numbers above it denote the accumulated coverage of the oracles to the left side. The blue, dashed line and the numbers below it denote the same for the oracles to the right side. We executed SQLancer using 10 threads on an AMD Ryzen Threadripper 2990WX processor with 32 cores and 32GB RAM running an Ubuntu 18.04 64-bit OS. the aggregated coverage by summing up all the coverage of the oracles to the left. The dashed blue line, which rises starting from the right, illustrates the same for all oracles to the right.hh the aggregated coverage by summing up all the coverage of the oracles to the left. The dashed blue line, which rises starting from the right, illustrates the same for all oracles to the right. The maximum coverage that is achieved by utilizing all test oracles is 56.1%. The coverage is rather low, because we did not test components such as subqueries, window functions, transactions, and sequences, as well as due to code that is never executed (e.g., due to external dependencies). In comparison, PQS achieved only a coverage ranging from 23.7% to 43.0%. By generating databases alone, a test coverage of 48.3% is already achieved. Each test oracle achieves a similar coverage; the range of test coverages is 0.6% (i.e., ranging from 55.3% to 55.9%). When using test oracles in combination, a small coverage increase can be observed, independent from in which order oracles are combined. However, the HAVING oracle seems to decrease the coverage, presumably since it lowers the throughput of the other oracles. Overall, we believe that these findings confirm our assumption that the oracles stress a large common part of the DBMSs. Nevertheless, a coverage increase can be observed when adding oracles, and indeed each test oracle found unique bugs. Despite this evidence that there is a large overlap, it should be noted that coverage information provides only limited insights for DBMSs. Jung et al. [2019] found that the core components of DBMSs achieve a coverage of >95% after running only tens of queries. Rigger and Su [2020a] argued, for NoREC, that coverage information is not insightful, and that they found many bugs in SQLite despite its impressive test suite, which provides 100% MC/DC coverage. gg g g y g p g line, which rises starting from the right, illustrates the same for all oracles to the right. 5.4 Case Study features in six phases (see Table 5).4 In the initial phases, we implemented the most basic SQL fea- tures of H2, and deliberately avoided using SQLancer’s optimizations for generating data, that is, generating boundary values with an increased probability, and caching constants, as described in Section 2. In the last two phases, we implemented H2-specific options as well as functions and en- abled SQLancer’s optimized data generation. For each phase, we studied which bugs the additional functionality allowed us to find. We considered only the WHERE oracle, since this oracle found most of the bugs in the other DBMSs, and since comprehensively testing H2 was a non-goal. features in six phases (see Table 5).4 In the initial phases, we implemented the most basic SQL fea- tures of H2, and deliberately avoided using SQLancer’s optimizations for generating data, that is, generating boundary values with an increased probability, and caching constants, as described in Section 2. In the last two phases, we implemented H2-specific options as well as functions and en- abled SQLancer’s optimized data generation. For each phase, we studied which bugs the additional functionality allowed us to find. We considered only the WHERE oracle, since this oracle found most of the bugs in the other DBMSs, and since comprehensively testing H2 was a non-goal. Phase 1. Initially, we set up the database generator to create a single table using CREATE TABLE, with 1–3 columns declared as either INT or BOOL without any constraints. We generated 0–30 INSERT statements. As constants, we generated uniformly-distributed 32-bit integers, booleans, and NULL values. For the SELECT statements generated by the WHERE oracle, we considered only the most basic syntax (i.e., SELECT * FROM t0 WHERE ptern). As expressions used in the WHERE clause, we consid- ered constants, column references, the most basic comparison operators, logical operators, and unary prefix and postfix operators (see Table 5 for the full list of operators), and a maximum op- erator depth of 3. We could complete the phase 1 implementation within two hours. SQLancer found the first two bugs within seconds. While one of the bugs was an error bug—causing a NullPointerException in H2—the other bug was a logic bug (see Listing 17). 4The commits were added as part of the pull request at https://github.com/sqlancer/sqlancer/pull/217. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 5.4 Case Study We conducted a case study on one additional DBMS, to better understand how comprehensive an implementation must be to find bugs, by starting with a minimal-working implementation, and then systematically adding features. The case study is aimed at addressing two potential concerns. The first potential concern is that the approach could be effective only when fully realized, for example, by setting DBMS-specific options (e.g., see Listing 7). The second potential concern is that the database and query generation of SQLancer could be a significant factor for finding bugs, in particular its optimized data generation. While we considered adding TLP to existing tools for database and query generation, we found that most tools support only DBMSs that we comprehen- sively tested with TLP, which would make it less likely to find interesting new bugs. Furthermore, we found that most tools described in research papers are not publicly available. Thus, we resorted to systematically studying an additional implementation in SQLancer.ti We tested H2, which is an embedded DBMS written in Java. It is mature and was first released in 2005. It is also highly popular on GitHub, and has been starred more than 2.4k times. To sys- tematically study the effect of supporting additional statements and keywords, we implemented 211:22 Manuel Rigger and Zhendong Su Table 5. We implemented and extended the testing support for H2 in six phases. Additional Functionality Bugs # Commit Statements Types Expressions Misc LOC WHERE Error 1 f1817d4 INSERT BOOL, INT =, >, >=, <, <=, !=, AND, OR, NOT, -, +, IS NULL 1 table, no constraints 501 1 1 2 61e6f1c CREATE INDEX, ANALYZE VARCHAR, DOUBLE IN, BETWEEN, CASE, LIKE, IS DISTINCT FROM, REGEXP 1–3 tables, UNIQUE/NOT NULL/PRIMARY KEY constraints 639 1(+1) 1 3 f53a88e CREATE VIEW BINARY \|\|, +, -, , // Generated columns, JOINs, ORDER BY 985 0 4 4 e81df01 UPDATE, DELETE type variants CAST, IS TRUE, IS FALSE, IS UNKNOWN Foreign keys, CHECK constraints, DEFAULT values 1,213 0 2 5 3d8d28a SET, MERGE INTO 93 functions SELECTIVITY 1,404 0 8 6 5b16b77 SQLancer Data Generation 1,414 0 0 Table 5. We implemented and extended the testing support for H2 in six phases. 5.4 Case Study CREATE TABLE T0(c0 VARCHAR UNIQUE); INSERT INTO T0(C0) VALUES (-1), (-2); SELECT FROM T0 WHERE c0 >= -1; -- expected: {-1, -2}, actual: {-2} additional keywords); the latter collects statistics on the tables’ contents to be used when creating a query plan. Since DBMSs typically create secondary indexes for UNIQUE and PRIMARY KEY constraints, we added support for generating those (and NOT NULL) constraints as well. We implemented the VARCHAR and DOUBLE data types, both without size specifications. For string constants, we considered only a single lowercase alphabetic character, rather than relying on SQLancer’s string generation. For DOUBLE constants, we generated uniformly-distributed floating-point numbers. As expressions, we added the IN operator, which checks whether a value is contained in a list of values, the ternary comparison operator BETWEEN, the switch-like CASE operator, and additional comparison operators, such as the regular expression operators LIKE and REGEXP. We found two logic bugs using the WHERE oracle due to the addition of the VARCHAR type and UNIQUE constraints. While one bug report was considered a duplicate to the first logic bug that we reported, the bug required a separate fix, by producing the expected result rather than rejecting the query as invalid. The second bug was due to comparisons of strings with integers (see Listing 18). While the bug was considered a release blocker, it was not fixed within 30 days, which is why we subsequently disabled generating VARCHAR columns. Based on our experience with testing other DBMSs, we were surprised that we did not find more logic bugs after introducing indexes. We found one error bug, caused by a bug in the implementation of the CASE operator. Phase 3. Next, we additionally created generated columns, which are table columns that are com- puted based on other columns. We generated 0–1 CREATE VIEW statements; to this end, we added a generator that randomly generates SELECT statements, which potentially included WHERE, JOIN, HAVING, GROUP BY, ORDER BY, LIMIT, and OFFSET clauses. We added support for all join types that H2 supports, namely LEFT, RIGHT, INNER, CROSS, and NATURAL joins. We could reuse these parts to enhance the WHERE or- acle test generation, by additionally generating JOIN and ORDER BY clauses as well as by considering views in JOIN and FROM clauses. 5.4 Case Study This bug was con- sidered minor by one of the H2 maintainers, who argued that the predicate is invalid according to the SQL standard, since boolean and numeric types cannot be converted to each other; however, the maintainer also stated that the test case exposed a minor bug since H2 optimizes this expres- sion in an unexpected way, causing both the non-negated and negated ternary predicate variants to evaluate to TRUE. The maintainer addressed this by disallowing such queries. We believe that finding the first logic bug even with the phase 1 implementation is strong evidence that TLP is effective even with preliminary database and expression generation components. Phase 2. Next, we allowed the creation of 1–3 tables, one or more of which could be referenced by the WHERE oracle. Indexes often exposed bugs in other DBMSs that we tested. Thus, we added support for generating 0–5 CREATE INDEX as well as the ANALYZE statements (without specifying any Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:23 Finding Bugs in Database Systems via Query Partitioning Listing 17. We found this H2 optimization bug in the first phase. CREATE TABLE T0(c0 BOOL); INSERT INTO T0(c0) VALUES (true); SELECT * FROM t0 WHERE NOT (c0 != 2 AND c0) -- expected: {} or query is rejected , actual: {TRUE} Listing 17. We found this H2 optimization bug in the first phase. CREATE TABLE T0(c0 BOOL); INSERT INTO T0(c0) VALUES (true); SELECT * FROM t0 WHERE NOT (c0 != 2 AND c0) -- expected: {} or query is rejected , actual: {TRUE} Listing 17. We found this H2 optimization bug in the first phase. Listing 18. We found this H2 bug in the second phase after adding the VARCHAR type and UNIQUE constraints CREATE TABLE T0(c0 VARCHAR UNIQUE); INSERT INTO T0(C0) VALUES (-1), (-2); SELECT * FROM T0 WHERE c0 >= -1; -- expected: {-1, -2}, actual: {-2} Listing 18. We found this H2 bug in the second phase after adding the VARCHAR type and UNIQUE constraints. CREATE TABLE T0(c0 VARCHAR UNIQUE); INSERT INTO T0(C0) VALUES (-1), (-2); SELECT * FROM T0 WHERE c0 >= -1; -- expected: {-1, -2}, actual: {-2} Listing 18. We found this H2 bug in the second phase after adding the VARCHAR type and UNIQUE constraints. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 5.4 Case Study In addition, we implemented the BINARY data type, which represents a byte array; as constants, we derived byte arrays from uniformly-distributed integers, which im- plicitly restricts the length of the generated arrays. Furthermore, we implemented support for string concatenation (i.e., the \|\| operator) and binary arithmetic operators. Surprisingly, while we failed to find any new logic bugs—generating joins and views revealed logic bugs in most other DBMSs—we identified 4 new error bugs. The first two bugs were in the implementations of CASE and BETWEEN operators and caused unexpected syntax errors when querying views. The third bug resulted in a ClassCastException when using the newly-added remainder operator % on a DOUBLE table column. The fourth bug caused an unexpected NullPointerException for cyclic references in the generated columns of a table. Phase 4. Next, we additionally generated 0–10 UPDATE and DELETE statements, which exposed bugs in other DBMSs when, for example, used on tables with indexes. As additional data types, we Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:24 Manuel Rigger and Zhendong Su considered variants of the already-supported data types. That is, we considered 1-, 2-, 4-, and 8- byte integers, and all their supported aliases (e.g., 8-byte integers can be specified using BIGINT and INT8). We considered 4- and 8-byte floating-point types, as well as optional precision arguments. For VARCHAR and BINARY, we additionally considered size specifications. To better utilize the additional types, we added the CAST operator (as well as missing unary comparison operators). For tables, we considered foreign keys, CHECK constraints, and DEFAULT values. SQLancer could detect two new error bugs. The first bug caused a NegativeArraySizeException when using a large size specification on the BINARY data type. The second bug was that the YEAR alias for a 2-byte integer no longer worked. Phase 5. In this phase, we primarily added DBMS-specific functionality. First, we added the SET statement, which allows setting DBMS-specific options; we selected 19 options that we believed could influence a query’s result. Furthermore, we implemented the MERGE INTO statement, which either inserts or updates values; the other DBMSs we tested supported similar functionality using other keywords and statements. For tables, we considered the SELECTIVITY keyword to specify how selective a column is expected to be in comparisons, which is utilized by the query planner. 5.4 Case Study Finally, we added support for 38 numeric functions, 38 string functions, and 17 general-purpose functions. While we again could not find any logic bugs, we could detect 8 error bugs. Of those bugs, 4 caused unexpected exceptions in functions, 2 resulted in exceptions when using the MERGE INTO statement, and 2 resulted in unexpected syntax errors when using specific functions in views. Phase 6. In the last phase, we started using SQLancer’s support for improved data generation. By doing so, we implicitly lifted the restriction of generating single-character alphabetic strings, by generating potentially multiple-character strings that also included special characters. We could not detect any further bugs. This was surprising to us, because we could detect a number of bugs related to special strings in other DBMSs. However, we found that in phase 5, one bug with a call to STRINGDECODE(X'5c38') triggered a StringIndexOutOfBoundsException. The binary constant 5c38 was randomly generated as described for phase 3, and was interpreted as the string "/8", specifying an invalid octal number, which triggered the bug. Discussion. Overall, we found only 2(+1) logic bugs in H2; this low number was surprising to us, since H2 had not been tested by PQS or NoREC. That we found both of these bugs after imple- menting a minimal prototype that generated only the most basic constructs (i.e., the phase 1 and 2 constructs) suggests that TLP’s effectiveness does not rely on the support of sophisticated fea- tures or data generation. This corresponds to our experience that most logic bugs are found after supporting the DBMSs’ core functionality, which is often optimized, and thus might more likely result in programming errors. In contrast, “exotic” features seem to more commonly result in error bugs (such as for the function errors). SQLancer’s optimized data generation did not enable TLP to find any additional bugs, suggesting that TLP does not require an efficient database or query generator to find bugs. While the H2 implementation is not fully realized—for example, types such as DATE, and statements such as ALTER TABLE are missing—the testing implementation is not signifi- cantly smaller than the other implementations; for example, the DuckDB implementation, which supports all TLP oracles, consists of 2,200 LOC. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 6 DISCUSSION Bug importance. It is difficult to measure the importance of the bugs we found. The developers of the DBMSs we tested explicitly told us that they appreciated our bug-finding efforts, and consid- ered many of the bugs to be important. For example, an engineering manager from Cockroach Labs wrote on a social media platform that we are “doing the database industry a great service. Thank you!”. Similarly, the most-contributing committer to DuckDB told us: This work is tremendously Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:25 Finding Bugs in Database Systems via Query Partitioning helpful for us, and I imagine anyone working on a DBMS. Usually these bugs would be slowly found by users over the years, not only negatively affecting the experience of those users but also requiring much more effort to debug and reproduce […]. For us especially it is extremely helpful because we have not yet gone through decades of users using the system, so this testing allows us to take a mas- sive shortcut and squeeze out many bugs that would otherwise be found by users. PingCAP started a bug bounty program for a release candidate of their DBMS TiDB, while we were testing it. As part of this, PingCAP also assigned severities to our bug reports, reaching from P0 (for the most serious issues) to P3 (documentation bugs). We reported 28 bugs as part of this program. While, based on the bug-bounty guidelines, incorrect query results should result in a P0 classification, PingCAP updated the guidelines after we reported the first batch of bugs, to reserve them the right to downgrade bugs, to which we agreed. Consequently, 22 bugs were classified as P1, and 6 as P2, that is, the second-highest and third-highest severities, demonstrating that the bugs we found were deemed important. In fact, we could redeem the points we received to obtain more than 100 T-shirts. Found bugs. As shown in Table 4, most bugs that we found were crash and error bugs. We do not consider these a contribution of this paper, since they could have also been found by fuzzers and other testing approaches. We list them merely for completeness, and since the give insight on the distribution of errors. That we found more errors and crash bugs might indicate that these are more common, or easier to find than logic bugs. 6 DISCUSSION Although the DBMS developers also greatly valued these bugs, we consider logic bugs to be more dangerous. For error and crash bugs, users of the DBMS obtain direct feedback that the query failed (e.g., since the process exits with an error). For logic bugs, however, errors might go unnoticed. NoREC and PQS. Compared to NoREC and PQS, TLP can detect bugs in GROUP BY clauses, DISTINCT queries, HAVING clauses, and aggregate functions. PQS and NoREC are not applicable for testing most of these features, except partially in corner cases (e.g., when a table contains only a single row, aggregate functions can partially be tested by PQS). TLP is a metamorphic testing approach, and similar to NoREC, it cannot establish a ground truth (i.e., an operator or function might consistently behave incorrectly, so that no bugs can be detected). In fact, due to this, Rigger and Su [2020a] found that NoREC can detect only about half of the bugs that PQS can find. Thus, TLP is complementary to PQS, and not a replacement for it. The WHERE oracle overlaps with NoREC as demonstrated in Section 5.2. Our manual analysis suggest that the WHERE oracle can find 12 bugs that NoREC can find, and that NoREC can find 8 bugs that the WHERE oracle cannot find. A threat to this is that the manual analysis was only a best-effort comparison. Limitations. Our testing does not apply to transactions, window functions, sequences, and non- deterministic functions. Queries can have ambiguous results, which limits the technique; this af- fects subqueries in particular [Rigger and Su 2020a], which we did not test. We found that especially SQLite has some peculiarities, such as treating the integer 0 and floating-point number 0.0 as the same number. CockroachDB and TiDB are distributed DBMSs, and we tested only their SQL com- ponents. We considered only the most commonly used aggregate functions, many of which were straightforward to decompose. An overview of various decompositioning strategies, also for other classes of aggregate functions, is given by Jesus et al. [2015]. Implementation order. Developers might wonder in which order to implement test oracles. The WHERE oracle is the simplest, but most effective oracle to implement. Only when this oracle does not find any more bugs is it useful to implement the subsequent oracles, which generate additional clauses in addition to the WHERE clause. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 7 RELATED WORK The closest related work is Pivoted Query Synthesis (PQS) and Nonoptimizing Reference Engine Construction (NoREC), which both aim to find logic bugs and were both extensively discussed. A number of approaches to test various aspects of DBMSs and related software have been proposed. Differential testing of DBMSs. Differential testing [McKeeman 1998] refers to a testing technique where a single input is passed to multiple systems that are expected to produce the same output; if the systems disagree on the output, a bug in at least one of the systems has been detected. It has proven to be effective in many domains [Brummayer and Biere 2009; Kapus and Cadar 2017; McKeeman 1998; Yang et al. 2011]. Slutz [1998] applied this technique for testing DBMSs in a system called RAGS by generating SQL queries that are sent to multiple DBMSs and then observing differences in the output sets. While the approach was effective, the author stated that the small common core and the differences between different DBMSs were a challenge, which was also noted by Rigger and Su [2020a,c]. Differential testing was found to be useful to compare query plans within a DBMS, or the performance of multiple versions of a DBMS. Specifically, Gu et al. [2012] used options and hints to force the generation of different query plans, to then rank the accuracy of the optimizer based on the estimated cost for each plan. Jung et al. [2019] used differential testing in a system called APOLLO to find performance regression bugs in DBMSs, by executing a SQL query on an old and newer version of a DBMS. Solver-based testing of DBMSs. ADUSA is a query-aware database generator that generates inputs as well as the expected result for a query [Khalek et al. 2008]. It translates the schema and query to an Alloy specification, which is subsequently solved. The approach could reproduce various known and injected bugs in MySQL, HSQLDB, and also find a new bug in Oracle Database. Similar approaches have also been proposed in related domains; for example, QED uses a solver to tackle the data generation and test oracle problems in the context of the SPARQL query language [Thost and Dolby 2019]. We believe that the high overhead that solver-based approaches incur might inhibit such approaches from finding more DBMS bugs. Random and targeted queries. Many query generators have been proposed for purposes such as bug-finding and benchmarking. 6 DISCUSSION Generating the simplest test case possible is preferable, since it speeds up the triaging, reduction, and understanding of bugs. Similarly, the HAVING oracle should Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:26 Manuel Rigger and Zhendong Su be implemented only after the GROUP BY oracle cannot find any additional bugs, since the HAVING oracle also generates GROUP BY clauses. The aggregate test oracles are more complex and specialized to an individual aggregate function; thus, we believe that these could be implemented last. Other partitioning strategies. Besides TLP, a number of additional partitioning strategies are imag- inable. As one example, the partitioning could be specific to operators or functions. For example, MySQL provides an operator <=>, which is similar to the equality operator, but evaluates also to a boolean value when comparing to a NULL value. A query using the operator in a predicate could be partitioned by replacing it with a series of IS NULL checks and an equality comparison. 7 RELATED WORK SQLsmith is a widely-used, open-source random query generator, which has found over 100 bugs in widely-used DBMSs [Seltenreich 2019]. Bati et al. proposed an ap- proach based on genetic algorithms to incorporate execution feedback for generating queries [Bati et al. 2007]. SQLFUZZ [Jung et al. 2019] also utilizes execution feedback and randomly generates queries using only features that are supported by all the DBMSs they considered. Abdul Khalek and Khurshid [2010] proposed generating both syntactically and semantically valid queries based on a solver-backed approach. Zhong et al. [2020] proposed a mutation-based, coverage-guided fuzzing approach that focuses on generating queries that are valid both syntactically and semantically, which they realized as a tool called SQUIRREL. All these random-query generators can be used to Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:27 Finding Bugs in Database Systems via Query Partitioning find bugs such as crashes and hangs in DBMSs. When paired with the test oracles proposed in this paper, they could also be used to find logic bugs. find bugs such as crashes and hangs in DBMSs. When paired with the test oracles proposed in this paper, they could also be used to find logic bugs. Random and targeted databases. Many approaches have been proposed to automatically generate databases. Given a query and a set of constraints, QAGen [Binnig et al. 2007b; Lo et al. 2010] gener- ates a database that matches the desired query results by combining traditional query processing and symbolic execution. Reverse Query Processing takes a query and a desired result set as an input to then generate a database that could have produced the result set [Binnig et al. 2007a]. As discussed above, ADUSA is a query-aware database generator [Khalek et al. 2008]. Gray et al. [1994] discussed a set of techniques utilizing parallel algorithms to quickly generate billions-record databases. DGL is a domain-specific language that generates input data following various distribu- tions and inter-table correlations based on iterators that can be composed [Bruno and Chaudhuri 2005]. An improved database generation might enable TLP to find additional bugs. Metamorphic testing. Metamorphic testing [Chen et al. 1998] addresses the test oracle problem by, based on an input and output of a system, generating a new input for which the result is known. Central in this approach is the metamorphic relation, which can be used to infer the expected result. 8 CONCLUSION This paper has presented the general idea of Query Partitioning, and a concrete instantiation of this idea, termed Ternary Logic Partitioning (TLP). The core idea of Query Partitioning is to partition a query into multiple so-called partitioning queries, each of which computes a partition of the result. By using a composition operator, the partitions can be combined to yield the same result as the original query; if the result differs, a bug in the DBMS has been detected. TLP partitions queries based on a boolean predicate, which can either evaluate to TRUE, FALSE, or NULL. TLP can detect bugs in various features, including WHERE clauses, GROUP BY clauses, HAVING clauses, DISTINCT queries, and aggregate functions. Our evaluation on six widely-used DBMSs has demonstrated that TLP is highly effective and general, as it could detect 77 logic bugs, at least 17 of which cannot be detected by existing techniques. Despite TLP’s effectiveness, we believe that a number of additional query partitioning strategies can be devised, which might allow finding additional bugs in DBMSs. 7 RELATED WORK This technique has been applied successfully in various domains [Chen et al. 2018; Donaldson et al. 2017; He et al. 2020; Le et al. 2014; Segura and Zhou 2018; Winterer et al. 2020]. The test oracle proposed as part of this paper is a metamorphic one, since based on the original query and its result set, we generate partitioning queries, whose composed result sets must be equal to the original query’s result set. Optimizing aggregate functions. We adopted ideas from optimizing aggregate functions to testing them. Cohen [2006] studied user-defined aggregate functions in the context of query optimization, query rewriting, and view maintenance. Yu et al. [2009] studied the interfaces and implementation of user-defined aggregate functions in the context of distributed aggregation. Since we decompose a query to partitioning queries, which can be computed independently, we study a similar problem. However, in contrast to their work, our goal is not to decompose the query for optimization, but to test the DBMSs. Jesus et al. [2015] surveyed the techniques in distributed data aggregation, provided a formal framework, and characterized the types of aggregate functions. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. REFERENCES Shadi Abdul Khalek and Sarfraz Khurshid. 2010. Automated SQL Query Generation for Systematic Testing of Database Engines. In Proceedings of the IEEE/ACM International Conference on Automated Software Engineering (Antwerp, Belgium) (ASE ’10). ACM, New York, NY, USA, 329–332. https://doi.org/10.1145/1858996.1859063fi ( ) p g M. M. Astrahan, M. W. Blasgen, D. D. Chamberlin, K. P. Eswaran, J. N. Gray, P. P. Griffiths, W. F. King, R. A. Lorie, P. R. hl G R P l T d d S R R l l A h M. M. Astrahan, M. W. Blasgen, D. D. Chamberlin, K. P. Eswaran, J. N. Gray, P. P. Griffiths, W. F. King, R. A. Lorie, P. R. McJones, J. W. Mehl, G. R. Putzolu, I. L. Traiger, B. W. Wade, and V. Watson. 1976. System R: Relational Approach to M. M. Astrahan, M. W. Blasgen, D. D. Chamberlin, K. P. Eswaran, J. N. Gray, P. P. Griffiths, W. F. King, R. A. Lorie, P. R. McJones, J. W. Mehl, G. R. Putzolu, I. L. Traiger, B. W. Wade, and V. Watson. 1976. System R: Relational Approach to Database Management. ACM Trans. Database Syst. 1, 2 (June 1976), 97–137. https://doi.org/10.1145/320455.320457 McJones, J. W. Mehl, G. R. Putzolu, I. L. Traiger, B. W. Wade, and V. Watson. 1976. System R: Relational Approach to Database Management. ACM Trans. Database Syst. 1, 2 (June 1976), 97–137. https://doi.org/10.1145/320455.320457 Database Management. ACM Trans. Database Syst. 1, 2 (June 1976), 97–137. https://doi.org/10.1145/320455 rdik Bati, Leo Giakoumakis, Steve Herbert, and Aleksandras Surna. 2007. A Genetic Approach for Random Tes Hardik Bati, Leo Giakoumakis, Steve Herbert, and Aleksandras Surna. 2007. A Genetic Approach for Random Testing of Database Systems. In Proceedings of the 33rd International Conference on Very Large Data Bases (Vienna, Austria) (VLDB ’07). VLDB Endowment, 1243–1251. Database Systems. In Proceedings of the 33rd International Conference on Very Large Data Bases (Vienna, Aust ’07). VLDB Endowment, 1243–1251.u Carsten Binnig, Donald Kossmann, and Eric Lo. 2007a. Reverse Query Processing. Proceedings - International Conference on Data Engineering, 506–515. https://doi.org/10.1109/ICDE.2007.367896u Carsten Binnig, Donald Kossmann, Eric Lo, and M. Tamer Özsu. 2007b. QAGen: Generating Query-Aware Test Databases. In Proceedings of the 2007 ACM SIGMOD International Conference on Management of Data (Beijing, China) (SIGMOD ’07). Association for Computing Machinery, New York, NY, USA, 341–352. https://doi.org/10.1145/1247480.1247520 Robert Brummayer and Armin Biere. 2009. Fuzzing and Delta-Debugging SMT Solvers. ACKNOWLEDGMENTS We want to thank the DBMSs’ developers for verifying and addressing our bug reports as well as for their feedback on our work. Furthermore, we are grateful for the feedback received by the anonymous reviewers, as well as by the members of the AST Lab at ETH Zurich. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. Manuel Rigger and Zhendong Su 211:28 REFERENCES In Proceedings of the 7th Interna- tional Workshop on Satisfiability Modulo Theories (Montreal, Canada) (SMT ’09). Association for Computing Machinery, New York, NY, USA, 1–5. https://doi.org/10.1145/1670412.1670413 Nicolas Bruno and Surajit Chaudhuri. 2005. Flexible Database Generators. In Proceedings of the 31st International Conference on Very Large Data Bases (Trondheim, Norway) (VLDB ’05). VLDB Endowment, 1097–1107. Nicolas Bruno, Surajit Chaudhuri, and Dilys Thomas. 2006. Generating Queries with Cardinality Constraints for DBMS Testing. IEEE Trans. on Knowl. and Data Eng. 18, 12 (Dec. 2006), 1721–1725. https://doi.org/10.1109/TKDE.2006.190 olas Bruno, Surajit Chaudhuri, and Dilys Thomas. 2006. Generating Queries with Cardinality Constraints for T i IEEE T K l d D E 18 12 (D 2006) 1721 1725 h //d i /10 1109/TKDE 2006 Testing. IEEE Trans. on Knowl. and Data Eng. 18, 12 (Dec. 2006), 1721–1725. https://doi.org/10.1109/TKDE.2006 Surajit Chaudhuri. 1998. An Overview of Query Optimization in Relational Systems. In Proceedings of the Seventeenth ACM SIGACT-SIGMOD-SIGART Symposium on Principles of Database Systems (Seattle, Washington, USA) (PODS ’98). Association for Computing Machinery, New York, NY, USA, 34–43. https://doi.org/10.1145/275487.275492 Surajit Chaudhuri. 1998. An Overview of Query Optimization in Relational Systems. In Proceedings of the Seventeenth ACM SIGACT-SIGMOD-SIGART Symposium on Principles of Database Systems (Seattle, Washington, USA) (PODS ’98). Surajit Chaudhuri. 1998. An Overview of Query Optimization in Relational Systems. In Proceedings of the Seventeenth ACM SIGACT-SIGMOD-SIGART Symposium on Principles of Database Systems (Seattle, Washington, USA) (PODS ’98). Association for Computing Machinery, New York, NY, USA, 34–43. https://doi.org/10.1145/275487.275492 Tsong Y Chen, Shing C Cheung, and Shiu Ming Yiu. 1998. Metamorphic testing: a new approach for generating next test cases. Technical Report. Technical Report HKUST-CS98-01, Department of Computer Science, Hong Kong.u ng Y Chen, Shing C Cheung, and Shiu Ming Yiu. 1998. Metamorphic testing: a new approach for generating g , g g, g p g pp f g cases. Technical Report. Technical Report HKUST-CS98-01, Department of Computer Science, Hong Kong.u Tsong Yueh Chen, Fei-Ching Kuo, Huai Liu, Pak-Lok Poon, Dave Towey, T. H. Tse, and Zhi Quan Zhou. 2018. Metamorphic Testing: A Review of Challenges and Opportunities. ACM Comput. Surv. 51, 1, Article 4 (Jan. 2018), 27 pages. https: //doi.org/10.1145/3143561 Yang Chen, Alex Groce, Chaoqiang Zhang, Weng-Keen Wong, Xiaoli Fern, Eric Eide, and John Regehr. 2013. Taming Compiler Fuzzers. In Proceedings of the 34th ACM SIGPLAN Conference on Programming Language Design and Imple- mentation (Seattle, Washington, USA) (PLDI ’13). Association for Computing Machinery, New York, NY, USA, 197–208. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. REFERENCES Compiler Fuzzing: How Much Doe ? P ACM P L OOPSLA A i l 55 (O ) h //d i / 5/ 6 5 Michaël Marcozzi, Qiyi Tang, Alastair F. Donaldson, and Cristian Cadar. 2019. Compiler Fuzzing: How Mu Matter? Proc. ACM Program. Lang. 3, OOPSLA, Article 155 (Oct. 2019), 29 pages. https://doi.org/10.1145/336 William M McKeeman. 1998. Differential testing for software. Digital Technical Journal 10, 1 (1998), 100–107.u Chaitanya Mishra, Nick Koudas, and Calisto Zuzarte. 2008. Generating Targeted Queries for Database Testing. In Pro- ceedings of the 2008 ACM SIGMOD International Conference on Management of Data (Vancouver, Canada) (SIGMOD ’08). ACM, New York, NY, USA, 499–510. https://doi.org/10.1145/1376616.1376668 MySQL. 2020. MySQL Homepage. https://www.mysql.com/t Andrea Neufeld, Guido Moerkotte, and Peter C. Lockemann. 1993. Generating Consistent Test Data: Restricting the Search Space by a Generator Formula. The VLDB Journal 2, 2 (April 1993), 173–214.tt Andrew Pavlo and Matthew Aslett. 2016. What’s Really New with NewSQL? SIGMOD Rec. 45, 2 (Sept. 2016), 45–55. https://doi.org/10.1145/3003665.3003674hu Meikel Poess and John M. Stephens. 2004. Generating Thousand Benchmark Queries in Seconds. In Proceedings of the Thirtieth International Conference on Very Large Data Bases - Volume 30 (Toronto, Canada) (VLDB ’04). VLDB Endowment, 1045–1053. g Q g Q p g p p g q g Mark Raasveldt and Hannes Mühleisen. 2020. Data Management for Data Science - Towards Embedded A g Q g Q p g p p g q g Mark Raasveldt and Hannes Mühleisen. 2020. Data Management for Data Science - Towards Embedded Analytics. In CIDR. J h R h 2010 A G id t U d fi d B h i i C d C htt //bl h / hi /213 Mark Raasveldt and Hannes Mühleisen. 2020. Data Management for Data Scie Mark Raasveldt and Hannes Mühleisen. 2020. Data Management for Data Science - Towards Embedded Analytics. In CIDR. John Regehr. 2010. A Guide to Undefined Behavior in C and C++. https://blog.regehr.org/archives/213 Raasveldt and Hannes Mühleisen. 2020. Data Management for Data Science - Towards Embedded Analytics. In CID egehr. 2010. A Guide to Undefined Behavior in C and C++. https://blog.regehr.org/archives/213 John Regehr. 2010. A Guide to Undefined Behavior in C and C++. https://blog.regehr.org/archives/213 John Regehr, Yang Chen, Pascal Cuoq, Eric Eide, Chucky Ellison, and Xuejun Yang. 2012. Test-Case Reduction for C Compiler Bugs. In Proceedings of the 33rd ACM SIGPLAN Conference on Programming Language Design and Imple- mentation (Beijing, China) (PLDI ’12). REFERENCES https://doi.org/10.1145/2491956.2462173 E. F. Codd. 1970. A Relational Model of Data for Large Shared Data Banks. Commun. ACM 13, 6 (June 1970), 377–387. https://doi.org/10.1145/362384.362685ih Sara Cohen. 2006. User-Defined Aggregate Functions: Bridging Theory and Practice. In Proceedings of the 2006 ACM SIG- MOD International Conference on Management of Data (Chicago, IL, USA) (SIGMOD ’06). Association for Computing Machinery, New York, NY, USA, 49–60. https://doi.org/10.1145/1142473.1142480l Will Dietz, Peng Li, John Regehr, and Vikram Adve. 2012. Understanding Integer Overflow in C/C++. In Proceedings of the 34th International Conference on Software Engineering (Zurich, Switzerland) (ICSE ’12). IEEE Press, 760–770. Alastair F. Donaldson, Hugues Evrard, Andrei Lascu, and Paul Thomson. 2017. Automated Testing of Graphics Shader Compilers. Proc. ACM Program. Lang. 1, OOPSLA, Article 93 (Oct. 2017), 29 pages. https://doi.org/10.1145/3133917 Goetz Graefe. 1993. Query Evaluation Techniques for Large Databases. ACM Comput. Surv. 25, 2 (June 1993), 73–169. https://doi.org/10.1145/152610.152611 Goetz Graefe. 2011. Modern B-Tree Techniques. Found. Trends Databases 3, 4 (April 2011), 203–402. https://doi.org/10. 1561/1900000028u Jim Gray, Prakash Sundaresan, Susanne Englert, Ken Baclawski, and Peter J. Weinberger. 1994. Quickly Generating Billion- R d S h i D b SIGMOD R (M ) h //d i / / J y g J g Q y Record Synthetic Databases. SIGMOD Rec. 23, 2 (May 1994), 243–252. https://doi.org/10.1145/191843.19188u Zhongxian Gu, Mohamed A. Soliman, and Florian M. Waas. 2012. Testing the Accuracy of Query Optimizers. In Proceedings of the Fifth International Workshop on Testing Database Systems (Scottsdale, Arizona) (DBTest ’12). ACM, New York, NY, USA, Article 11, 6 pages. https://doi.org/10.1145/2304510.2304525 p g p g Paolo Guagliardo and Leonid Libkin. 2017. A Formal Semantics of SQL Queries, Its Validation, and Applications. Proc. VLDB Endow. 11, 1 (Sept. 2017), 27–39. https://doi.org/10.14778/3151113.3151116u L. M. Haas, J. C. Freytag, G. M. Lohman, and H. Pirahesh. 1989. Extensible Query Processing in Starburst. In Proceed- ings of the 1989 ACM SIGMOD International Conference on Management of Data (Portland, Oregon, USA) (SIGMOD ’89). Association for Computing Machinery, New York, NY, USA, 377–388. https://doi.org/10.1145/67544.66962 Finding Bugs in Database Systems via Query Partitioning 211:29 Pinjia He, Clara Meister, and Zhendong Su. 2020. Structure-Invariant Testing for Machine Translation. In Proceedings of the ACM/IEEE 42nd International Conference on Software Engineering (Seoul, South Korea) (ICSE ’20). Association for Computing Machinery, New York, NY, USA, 961–973. https://doi.org/10.1145/3377811.3380339 Kenneth Houkjær, Kristian Torp, and Rico Wind. 2006. Simple and Realistic Data Generation. REFERENCES In Proceedings of the 32nd International Conference on Very Large Data Bases (Seoul, Korea) (VLDB ’06). VLDB Endowment, 1243–1246. Dongxu Huang, Qi Liu, Qiu Cui, Zhuhe Fang, Xiaoyu Ma, Fei Xu, Li Shen, Liu Tang, Yuxing Zhou, Menglong Huang, Wan Wei, Cong Liu, Jian Zhang, Jianjun Li, Xuelian Wu, Lingyu Song, Ruoxi Sun, Shuaipeng Yu, Lei Zhao, Nicholas Cameron, Liquan Pei and Xin Tang 2020 TiDB: A Raft Based HTAP Database Proc VLDB Endow 13 12 (Aug 2020) 3072 3084 Dongxu Huang, Qi Liu, Qiu Cui, Zhuhe Fang, Xiaoyu Ma, Fei Xu, Li Shen, Liu Tang, Yuxing Zhou, Menglong Huang, Wan Wei, Cong Liu, Jian Zhang, Jianjun Li, Xuelian Wu, Lingyu Song, Ruoxi Sun, Shuaipeng Yu, Lei Zhao, Nicholas Cameron, Liquan Pei, and Xin Tang. 2020. TiDB: A Raft-Based HTAP Database. Proc. VLDB Endow. 13, 12 (Aug. 2020), 3072–3084. https://doi.org/10.14778/3415478.3415535 Wei, Cong Liu, Jian Zhang, Jianjun Li, Xuelian Wu, Lingyu Song, Ruoxi Sun, Shuaipeng Yu, Lei Zhao, Nichola Liquan Pei, and Xin Tang. 2020. TiDB: A Raft-Based HTAP Database. Proc. VLDB Endow. 13, 12 (Aug. 2020), https://doi.org/10.14778/3415478.3415535 P. Jesus, C. Baquero, and P. S. Almeida. 2015. A Survey of Distributed Data Aggregation Algorithms. IEEE Communications Surveys Tutorials 17, 1 (2015), 381–404. Jinho Jung, Hong Hu, Joy Arulraj, Taesoo Kim, and Woonhak Kang. 2019. APOLLO: Automatic Detection and Diagnosis of Performance Regressions in Database Systems. Proc. VLDB Endow. 13, 1 (Sept. 2019), 57–70. https://doi.org/10.14778/ 3357377.3357382 Timotej Kapus and Cristian Cadar. 2017. Automatic Testing of Symbolic Execution Engines via Program Generation and Differential Testing. In Proceedings of the 32Nd IEEE/ACM International Conference on Automated Software Engineering (Urbana-Champaign, IL, USA) (ASE 2017). IEEE Press, Piscataway, NJ, USA, 590–600.u S. A. Khalek, B. Elkarablieh, Y. O. Laleye, and S. Khurshid. 2008. Query-Aware Test Generation Using a Relational Constraint Solver. In Proceedings of the 2008 23rd IEEE/ACM International Conference on Automated Software Engineering (ASE ’08). IEEE Computer Society, Washington, DC, USA, 238–247. https://doi.org/10.1109/ASE.2008.34 Vu Le, Mehrdad Afshari, and Zhendong Su. 2014. Compiler Validation via Equivalence Modulo Inputs. In Proceedings of the 35th ACM SIGPLAN Conference on Programming Language Design and Implementation (Edinburgh, United Kingdom) (PLDI ’14). ACM, New York, NY, USA, 216–226. https://doi.org/10.1145/2594291.2594334 Eric Lo, Carsten Binnig, Donald Kossmann, M. Tamer Özsu, and Wing-Kai Hon. 2010. A framework for testing DBMS features. The VLDB Journal 19, 2 (01 Apr 2010), 203–230. https://doi.org/10.1007/s00778-009-0157-y ël Marcozzi, Qiyi Tang, Alastair F. Donaldson, and Cristian Cadar. 2019. REFERENCES Association for Computing Machinery, New York, NY, USA, 335–346. https: //doi.org/10.1145/2254064.2254104 Manuel Rigger and Zhendong Su. 2020a. Detecting Optimization Bugs in Database Engines via Non-Optimizing Reference Engine Construction. In Proceedings of the 2020 28th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering (Sacramento, California, United States) (ESEC/FSE 2020).u Manuel Rigger and Zhendong Su. 2020b. OOPSLA 20 Artifact for “Finding Bugs in Database Systems via Query Partitioning”. https://doi.org/10.5281/zenodo.4032401u Manuel Rigger and Zhendong Su. 2020c. Testing Database Engines via Pivoted Query Synthesis. In 14th USENIX Symposium on Operating Systems Design and Implementation (OSDI 20). USENIX Association, Banff, Alberta. https://www.usenix. org/conference/osdi20/presentation/riggeru Sergio Segura and Zhi Quan Zhou. 2018. Metamorphic Testing 20 Years Later: A Hands-on Introduction. In Proceedings of the 40th International Conference on Software Engineering: Companion Proceeedings (Gothenburg, Sweden) (ICSE ’18). Association for Computing Machinery, New York, NY, USA, 538–539. https://doi.org/10.1145/3183440.3183468 Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. 211:30 Manuel Rigger and Zhendong Su Manuel Rigger and Zhendong Su Andreas Seltenreich. 2019. SQLSmith. https://github.com/anse1/sqlsmith Andreas Seltenreich. 2019. SQLSmith. https://github.com/anse1/sqlsmith Konstantin Serebryany, Alexander Potapenko, Timur Iskhodzhanov, and Dmitriy Vyukov. 2011. Dynamic Race Detection with LLVM Compiler. In Proceedings of the Second International Conference on Runtime Verification (San Francisco, CA) (RV’11). Springer-Verlag, Berlin, Heidelberg, 110–114. https://doi.org/10.1007/978-3-642-29860-8_9 ld l h f l Konstantin Serebryany, Alexander Potapenko, Timur Iskhodzhanov, and Dmitriy Vyukov. 2011. Dynamic Race Detection with LLVM Compiler. In Proceedings of the Second International Conference on Runtime Verification (San Francisco, CA) (RV’11). Springer-Verlag, Berlin, Heidelberg, 110–114. https://doi.org/10.1007/978-3-642-29860-8_9 ( ) p g g g p g Donald R Slutz. 1998. Massive stochastic testing of SQL. In VLDB, Vol. 98. 618–622. Massive stochastic testing of SQL. In VLDB, Vol. 98. 618– ald R Slutz. 1998. Massive stochastic testing of SQL. In V SQLite. 2020. SQLite Homepage. https://www.sqlite.org/ Evgeniy Stepanov and Konstantin Serebryany. 2015. MemorySanitizer: Fast Detector of Uninitialized Memory Use in C++. In Proceedings of the 13th Annual IEEE/ACM International Symposium on Code Generation and Optimization (San Francisco, California) (CGO ’15). IEEE Computer Society, USA, 46–55.t Rebecca Taft, Irfan Sharif, Andrei Matei, Nathan VanBenschoten, Jordan Lewis, Tobias Grieger, Kai Niemi, Andy Woods, Anne Birzin, Raphael Poss, Paul Bardea, Amruta Ranade, Ben Darnell, Bram Gruneir, Justin Jaffray, Lucy Zhang, and Peter Mattis. 2020. CockroachDB: The Resilient Geo-Distributed SQL Database. Proc. ACM Program. Lang., Vol. 4, No. OOPSLA, Article 211. Publication date: November 2020. REFERENCES In Proceedings of the 2020 ACM SIG- MOD International Conference on Management of Data (Portland, OR, USA) (SIGMOD ’20). International Foundation for Autonomous Agents and Multiagent Systems, Richland, SC, 1493–1509. https://doi.org/10.1145/3318464.3386134huh Veronika Thost and Julian Dolby. 2019. QED: Out-of-the-Box Datasets for SPARQL Query Evaluation. In The Semantic Web, Pascal Hitzler, Miriam Fernández, Krzysztof Janowicz, Amrapali Zaveri, Alasdair J.G. Gray, Vanessa Lopez, Armin Haller, and Karl Hammar (Eds.). Springer International Publishing, Cham, 491–506.uh Manasi Vartak, Venkatesh Raghavan, and Elke A. Rundensteiner. 2010. QRelX: Generating Meaningful Queries That Pro- vide Cardinality Assurance. In Proceedings of the 2010 ACM SIGMOD International Conference on Management of Data (Indianapolis, Indiana, USA) (SIGMOD ’10). Association for Computing Machinery, New York, NY, USA, 1215–1218. https://doi.org/10.1145/1807167.1807323 https://doi.org/10.1145/1807167.1807323 Dominik Winterer, Chengyu Zhang, and Zhendong Su. 2020. Validating SMT Solvers via Semantic Fusion. In Proceedings of the 41st ACM SIGPLAN Conference on Programming Language Design and Implementation (London, UK) (PLDI 2020). Association for Computing Machinery, New York, NY, USA, 718–730. https://doi.org/10.1145/3385412.3385985 Xuejun Yang, Yang Chen, Eric Eide, and John Regehr. 2011. Finding and Understanding Bugs in C Compilers. In Proceedings of the 32Nd ACM SIGPLAN Conference on Programming Language Design and Implementation (San Jose, California, USA) (PLDI ’11). ACM, New York, NY, USA, 283–294. https://doi.org/10.1145/1993498.1993532 Yuan Yu, Pradeep Kumar Gunda, and Michael Isard. 2009. Distributed Aggregation for Data-Parallel Computing: Interfaces and Implementations. In Proceedings of the ACM SIGOPS 22nd Symposium on Operating Systems Principles (Big Sky, Montana, USA) (SOSP ’09). Association for Computing Machinery, New York, NY, USA, 247–260. https://doi.org/10. 1145/1629575.1629600 Rui Zhong, Yongheng Chen, Hong Hu, Hangfan Zhang, Wenke Lee, and Dinghao Wu. 2020. SQUIRREL: Testing Database Management Systems with Language Validity and Coverage Feedback. In Proceedings of the 27th ACM Conference on Computer and Communications Security (CCS).
https://openalex.org/W3172213311	https://www.frontiersin.org/articles/10.3389/fnsyn.2021.656377/pdf	English	null	Nitric Oxide Regulates GluA2-Lacking AMPAR Contribution to Synaptic Transmission of CA1 Apical but Not Basal Dendrites	Frontiers in synaptic neuroscience	2,021	cc-by	9,735	ORIGINAL RESEARCH published: 03 June 2021 doi: 10.3389/fnsyn.2021.656377 Edited by: Wayne S. Sossin, McGill University, Canada Reviewed by: Anne McKinney, McGill University, Canada Eric Hanse, University of Gothenburg, Sweden Vidar Jensen, University of Oslo, Norway Correspondence: Violetta O. Ivanova v.o.ivanova@ihna.ru Correspondence: Violetta O. Ivanova v.o.ivanova@ihna.ru Nitric Oxide Regulates GluA2-Lacking AMPAR Contribution to Synaptic Transmission of CA1 Apical but Not Basal Dendrites Violetta O. Ivanova, Pavel M. Balaban and Natalia V. Bal Cellular Neurobiology of Learning Lab, Institute of Higher Nervous Activity and Neurophysiology of the Russian Academy of Science, Moscow, Russia The mechanisms of synaptic plasticity differ in distinct local circuits. In the CA1 region of the hippocampus, the mechanisms of long-term potentiation (LTP) at apical dendrites in stratum radiatum and basal dendrites in stratum oriens involve different molecular cascades. For instance, participation of nitric oxide in LTP induction was shown to be necessary only for apical dendrites. This phenomenon may play a key role in information processing in CA1, and one of the reasons for this difference may be differing synaptic characteristics in these regions. Here, we compared the synaptic responses to stimulation of apical and basal dendrites of CA1 pyramidal neurons and found a difference in the current–voltage characteristics of these inputs, which is presumably due to a distinct contribution of GluA2-lacking AMPA receptors to synaptic transmission. In addition, we obtained data that indicate the presence of these receptors in pyramidal dendrites in both stratum radiatum and stratum oriens. We also demonstrated that inhibition of NO synthase reduced the contribution of GluA2-lacking AMPA receptors at apical but not basal dendrites, and inhibition of soluble guanylate cyclase did not affect this phenomenon. INTRODUCTION Received: 20 January 2021 Accepted: 19 April 2021 Published: 03 June 2021 The CA1 region of the hippocampus is a prevailing part of the brain for studying the phenomena of synaptic plasticity. This region with its widespread projections (Cenquizca and Swanson, 2007) is a key structure for the propagation of signals from the hippocampus to other parts of the brain. The apical and basal dendrites of the CA1 pyramidal cells extend in two main directions: stratum radiatum/stratum lacunosum moleculare and stratum oriens, respectively. Most of the excitatory inputs to CA1 pyramidal neurons originate from CA3 pyramidal cells through their ipsilateral Schaffer collaterals and contralateral commissural fibers to str. radiatum (Van Strien et al., 2009), as well as from the entorhinal cortex through the perforant pathway into the stratum lacunosum moleculare (Masurkar et al., 2017). Keywords: AMPA, nitric oxide, hippocampus, synaptic transmission, pyramidal neurons Citation: Ivanova VO, Balaban PM and Bal NV (2021) Nitric Oxide Regulates GluA2-Lacking AMPAR Contribution to Synaptic Transmission of CA1 Apical but Not Basal Dendrites. Front. Synaptic Neurosci. 13:656377. doi: 10.3389/fnsyn.2021.656377 June 2021 \| Volume 13 \| Article 656377 Frontiers in Synaptic Neuroscience \| www.frontiersin.org AMPARs Contribution to Synaptic Transmission Ivanova et al. Synaptic plasticity at basal dendrites may be an important part of information processing in CA1, since synaptic plasticity of CA1 apical dendrite synapses can be homeostatically regulated by the cell-wide history of synaptic activity (metaplasticity) including the activity of basal dendrites (Hulme et al., 2012). While synaptic plasticity at the apical dendrites of CA1 pyramidal neurons has been extensively studied, relatively little is known whether the mechanisms of synaptic plasticity are the same at basal dendrites. was shown to affect the incorporation of different AMPAR subunits to the cell membrane via several different pathways (for review, see Ivanova et al., 2020). Thus, the interaction of nitric oxide with AMPAR subunits could be crucial in the trafficking of GluA2-lacking AMPARs [calcium-permeable (CP- AMPARs)]. In our study, we demonstrate the differences in AMPAR-NO interactions between apical and basal dendrites. We show that the contribution of CP-AMPARs to synaptic transmission in apical dendrites is higher than in basal dendrites, and the NO synthase blockade flattens this difference. This effect does not involve sGC-dependent cascades. Our results confirm previous studies demonstrating different NO-dependent mechanisms in the apical and basal dendrites of CA1 pyramidal neurons. Synaptic plasticity at the basal and apical dendrites of the hippocampal CA1 region has some similarities (Bradshaw et al., 2003), but the clear spatial separation of these synapses, combined with differences in the innervation of str. radiatum and str. oriens, suggests that the molecular mechanisms of plasticity may vary significantly. For instance, knockout of both the endothelial and neuronal forms of nitric oxide synthase (NOS) caused a drop in amplitude responses after induction of long-term potentiation (LTP) only at apical dendrites (Son et al., 1996), but not at basal dendrites, which is consistent with other data regarding the involvement of NOS in synaptic plasticity in str. radiatum and str. oriens (Haley et al., 1996). In addition, differences were found in the molecular cascades dependent on nitric oxide (NO): inhibition of cyclic guanosine monophosphate (cGMP), cGMP-dependent protein kinase (PKG), and soluble guanylyl cyclase (sGC) during LTP induction were demonstrated to be more effective in blocking LTP in str. Slice Preparation p Horizontal brain slices (300 µm thick) containing the ventral hippocampus and entorhinal cortex were prepared from the brains of 25–35-day-old C57Bl/6 female and male mice killed by decapitation. The slicing chamber contained an oxygenated ice-cold solution (modified from Dugue et al., 2005) composed of the following (in mM) K-gluconate, 140; N-(2-hydroxyethyl) piperazine-N′-ethanesulfonic acid (HEPES), 10; Na-gluconate, 15; ethylene glycol-bis(2-aminoethyl)-N,N,N′,N′-tetraacetic acid (EGTA), 0.2; and NaCl, 4 (pH 7.2 with KOH). Brain slices were cut using a Vibratome (Leica VT1000S, Germany). Slices were incubated for at least 40 min at 35◦C before being stored at room temperature in artificial CSF (ACSF) containing the following (in mM): NaCl, 125; NaHCO3, 25; KCl, 2.5; NaH2PO4, 1.25; MgCl2, 3.9; CaCl2, 1; and glucose, 25; bubbled with 95% O2, and 5% CO2. One of the key participants in LTP is the α-amino-3-hydroxy- 5-methyl-4-isoxazolepropionic acid receptor (AMPAR) which is the main provider of excitatory transmission in the mammalian CNS (Malinow and Malenka, 2002; Diering and Huganir, 2018). Most AMPARs are heterotetramers combined from the GluA1, GluA2, GluA3, and GluA4 subunits. In the adult brain, almost all GluA2 subunit mRNA undergoes posttranscriptional editing, which leads to a replacement of the neutral amino acid glutamine with positively charged arginine in the polypeptide chain of the subunit. This replacement alters the electrophysiological properties of GluA2-containing AMPARs and makes them impermeable to calcium (Higuchi et al., 1993). Nitric oxide Citation: radiatum than in str. oriens (Son et al., 1998). Authors suggest that this difference is due to the fact that endothelial NO synthase (eNOS) is not present in str. oriens, which is the main source of nitric oxide in hippocampal LTP according to O’Dell et al. (1994). However, these data contradict other studies. For instance, it was shown that there is more eNOS in str. radiatum than in str. oriens due to the higher density of blood vessels (Blackshaw et al., 2003) and that LTP maintenance requires involvement of at least one form of NOS (Son et al., 1996). It is important to note that LTP sensitivity to NO in stratum radiatum also depends on the stimulation protocol for LTP induction (Lu et al., 1999; Bal et al., 2017; Maltsev et al., 2019). Taken together, these data indicate that the mechanism of LTP at basal dendrites of CA1 pyramidal cells is likely to be NO-independent. However, it should be noted that the neural NOS is still present in str. oriens (O’Dell et al., 1994; Blackshaw et al., 2003), and application of NO donors caused cGMP production in str. oriens (Bartus et al., 2013), which suggests that nitric oxide has some other functions in this region. Animals and Ethical Approval Animals and Ethical Approval All experiments followed the European Convention for the Protection of Vertebrate Animals used for Experimental and other Scientific Purposes 1986 86/609/EEC and were approved by the Ethical Committee of the Institute of Higher Nervous Activity and Neurophysiology, Russian Academy of Sciences (IHNA RAS). The mice were purchased from the Nursery for laboratory animals of the Branch of the Institute of Bioorganic Chemistry of the Russian Academy of Sciences in Pushchino. The mice were maintained in a temperature-controlled vivarium (22 ± 2◦C) under a 12-h light/dark cycle (lights on at 08.00 h) with food and water ad libitum. All efforts were made to minimize animal suffering and to reduce the number of animals used. Frontiers in Synaptic Neuroscience \| www.frontiersin.org Rectiﬁcation Index (RI) 30; HEPES, 10; NaCl, 8; EGTA, 0.2; MgATP, 4; Na3GTP, 0.3; and phosphocreatine, 10 (pH 7.3 with CsOH) osmolarity ∼290 mOsm. The polyamine-containing solution was identical except for the addition of 10 µM spermine. 30; HEPES, 10; NaCl, 8; EGTA, 0.2; MgATP, 4; Na3GTP, 0.3; and phosphocreatine, 10 (pH 7.3 with CsOH) osmolarity ∼290 mOsm. The polyamine-containing solution was identical except for the addition of 10 µM spermine. 30; HEPES, 10; NaCl, 8; EGTA, 0.2; MgATP, 4; Na3GTP, 0.3; and phosphocreatine, 10 (pH 7.3 with CsOH) osmolarity ∼290 mOsm. The polyamine-containing solution was identical except for the addition of 10 µM spermine. Experiments started after baseline stabilization (∼100 sweeps). For the −70-mV and +35-mV holding points, at least 30 and 50 sweeps were collected, respectively. The RI was calculated as the ratio of EPSCs measured at −70 mV and +35 mV (EPSC-70/EPSC+35). For each +35-mV point, only the last 30 sweeps were analyzed due to potential space clamp problems. CA1 pyramidal cells were identified visually using an Olympus microscope fitted with infrared differential interference contrast optics (Olympus BX51WI). Whole-cell recordings from these neurons were made in a voltage-clamp mode using the ELC-03XS amplifier (NPI Electronic, Tamm, Germany) and Clampex software (Axoclamp, Molecular Devices). Cells were held at −70 mV. Cells with unhealthy morphology and resting membrane potential above −50 mV (before correction for the liquid junction potential) were excluded from the experiments. To evoke synaptic current, glass electrodes filled with ACSF were placed in the dendritic region of stratum radiatum and stratum oriens, ∼50–100 µm from the cell body, to stimulate the inputs at interstimulus intervals of 6 s. Inhibitory synaptic transmission was blocked during recordings by adding 50 µM picrotoxin to the perfusion ACSF. In all the experiments except AMPA/NMDA ratio measurements, the NMDA-mediated component was blocked by adding 50 µM APV to the ACSF. For the experiments with NOS inhibition, slices were incubated for 40–120 min in L-NAME, 15 min in 3-bromo-7-nitroindazole or carboxy-PTIO before being placed in the perfusion chamber. Whole-cell recordings typically started 5–10 min after break-in, when the balance between intracellular milieu and patch solution was established and a steady-state current was reached, except the experiments with GluR2- lacking AMPA receptor inhibition. The stimulation intensity was adjusted to produce an EPSC with an amplitude of ∼50 pA at the beginning of each recording. The experiments were not started if there was an unstable baseline. Electrophysiology Electrophysiological recording was performed in an acrylic glass perfusion chamber (Luigs and Neumann, Germany) with the bath temperature kept at 30 ± 2◦C and perfused at a constant rate of 3 ml/min. Patch electrodes (resistance 4–5 MΩ) were pulled from borosilicate capillary glass (Narishige PC-100 Puller, Japan) and were filled with either a polyamine-free or a polyamine-containing solution. The polyamine-free solution consisted of the following (in mM): Cs-gluconate, 110; CsCl, June 2021 \| Volume 13 \| Article 656377 Frontiers in Synaptic Neuroscience \| www.frontiersin.org 2 AMPARs Contribution to Synaptic Transmission Ivanova et al. Drugs ODQ from Sigma-Aldrich was prepared as a 25-mM stock solution in DMSO and diluted down to achieve a final bath concentration of 30 µM. 3-Bromo-7-nitroindazole (3-Br-7- ni; Enzo Life Sciences) was dissolved as a 100-mM stock in DMSO and diluted down to achieve a final bath concentration of 50 µM. Nω-Nitro-L-arginine methyl ester hydrochloride (L-NAME; Sigma-Aldrich) was prepared as a 200-mM stock in milli-Q water diluted down to achieve a final bath concentration of 200 µM. Carboxy-PTIO (2-(4-Carboxyphenyl)-4,4,5,5- tetramethylimidazoline-1-oxyl-3-oxide potassium salt) from Enzo Life Sciences was prepared as a 100-mM stock in DMSO and diluted down to achieve a final bath concentration of 50 µM. Naspm (1-naphthylacetyl spermine) was prepared as a 100-mM Rectiﬁcation Index (RI) Series resistance was monitored, and data from cells in which series resistance varied by >15% during recording were discarded from the analysis. In all the experiments, the command voltage was corrected for the liquid junction potential (−10 mV). To test synapses for polyamine-dependent facilitation (PdF), we applied four stimuli with an interstimulus interval of 33 ms ×40 for each of the inputs, 5 and 20 min after whole-cell patch formation. To evaluate changes in PdF, we normalized each of the EPSCs to the first in the train, averaged the obtained values, and separately compared the data for apical and basal dendrites in different conditions. For PdF analysis, an additional selection criterion was applied: traces with undetectable peaks of the first EPSC in the train were retracted from the analysis. The number of the retracted traces was <1% of the total number of traces. Paired-pulse ratio was monitored by applying two stimuli with a 50-ms interstimulus interval. The experiments started after baseline stabilization (∼100 sweeps), and at least 20 sweeps were collected. GluR2-Lacking AMPA Receptor Inhibition GluR2-Lacking AMPA Receptor Inhibition Experiments were performed using QX-314-containing spermine-free intracellular solution. Recording started after the holding current stabilizes (1–2 min after the beginning of whole-cell recording). The amplitude of test responses stabilizes after 15–20 min of recording. The GluR2-lacking AMPA receptor antagonist Naspm (200 µM) was applied 30 min after the start of the recording. For analysis, 10 successive responses were averaged and normalized to the mean EPSC amplitude obtained between 20 and 30 min of the recording session. The degree of the blockade was evaluated as the ratio of the average steady-state current amplitude without and after Naspm application. Cells that did not correspond to the standard criteria of electrophysiological properties, such as input resistance, series resistance, and baseline holding current, were excluded from the analysis. AMPA/NMDA Current Ratio The AMPA/NMDA ratio was measured in Mg2+-free ACSF. AMPA and NMDA receptor-mediated EPSCs were pharmacologically isolated by sequential bath application of APV and CNQX, respectively. First, the compound AMPAR and NMDA-mediated current was recorded in Mg2+-free ASCF. After collecting at least 100 sweeps, the AMPA-mediated component was blocked by application of 50 µM CNQX. An additional 100 sweeps of the NMDA-mediated currents were collected and the NMDA nature of these currents was confirmed by subsequent application of APV. The AMPA-mediated component was obtained by subtraction of the averaged NMDA-mediated currents from the averaged compound response. For subsequent analysis, the mean amplitude of the AMPA currents was normalized to the amplitude of the NMDAR EPSCs. Statistical Analysis y Results are presented as mean ± standard error (S.E.M.) of n cells. All statistical tests were performed using SigmaPlot 11.0 (Systat Software Inc., USA). For pairwise comparisons (Figures 1B,C), one-way ANOVA was used. For multiple comparisons (Figures 1F, 3A,B, 4C,D, 5C,D, 6, Supplementary Figure 1), we used two-way ANOVA. A significant main effect or interaction was followed by post-hoc comparison using Multiple Comparisons vs. Control Group (Holm–Sidak test); for between-subject analysis, the untreated cells were scored as ‘‘Control Group.’’ For Figures 2, 5A,B, and Supplementary Figure 2, we used two-way repeated-measures ANOVA. A significant main effect or interaction was followed by post-hoc comparisons using Multiple Comparisons vs. Control Group (Holm–Sidak Test). For between-subject analysis, the control cells were scored as ‘‘Control Group,’’ and for within-subject analysis all measurements were compared to the first EPSC in the train. A probability level of 0.05 or less was considered statistically significant (∗p ≤0.05, ∗∗p ≤0.005). p In addition, Rozov and colleagues have shown previously that application of several high-frequency stimuli to the input with CP-AMPARs causes relief of the polyamine block which, in turn, results in polyamine-dependent facilitation (PdF; Rozov and Burnashev, 1999; Rozov et al., 2018). We applied a similar protocol to the apical and basal inputs of CA1 pyramidal cells and found that this type of short-term plasticity was characteristic for synapses of these cells. We applied four stimuli with an interstimulus interval of 33 ms to each of the inputs at the beginning of the recording and 20 min later. We used spermine-free and spermine-containing intracellular solutions in order to test whether polyamine washout would affect the EPSCs. For experiments with polyamine washout, we adjusted the stimuli strength after 20 min to avoid spikes due to potential space clamp problems. Figures 2A,B demonstrate the results of these experiments. Current–Voltage Relationship The experiments started after baseline stabilization (∼100 sweeps). For each holding potential point, at least 20 sweeps were collected. Responses were averaged and normalized to the mean EPSC amplitude obtained at −70 mV. To evaluate differences, the normalized values at +50 mV were compared. June 2021 \| Volume 13 \| Article 656377 Frontiers in Synaptic Neuroscience \| www.frontiersin.org 3 AMPARs Contribution to Synaptic Transmission Ivanova et al. Therefore, we tested whether the responses to stimulation of apical and basal inputs are sensitive to extracellular Naspm (200 µM) application (Figures 1D–G). To evaluate the degree of amplitude reduction, we averaged the values from each cell in the 5-min segment at the end of the recordings (Figure 1F). We observed a decrease in AMPAR currents in both of the inputs (73% ± 4 of the baseline in str. radiatum and 83% ± 6.5 in str. oriens, n = 10), which may indicate the presence of CP-AMPARs in the basal and apical dendrites of CA1 pyramidal neurons. However, comparison to the untreated cells (n = 6) showed significant difference only in apical (73% ± 4 vs. 109% ± 7, p = 0.001, two-way ANOVA) but not in basal inputs (83% ± 6.5 vs. 94% ± 4, p = 0.354, two-way ANOVA). Even at physiological resting membrane potentials (∼−70 mV), a substantial portion of GluA2-lacking channels could still be blocked by polyamines during single unitary EPSCs. Thus, we used a spermine-free intracellular solution in these experiments. As a result of the endogenous polyamine washout from the dendrites, the amplitude of responses increased up to 20 min. To avoid subsequent spike generation, we added 30 µM QX-314 to the intracellular solution. Due to the different morphology of apical and basal dendrites (Benavides-Piccione et al., 2020), the response amplitude at the basal inputs tends to stabilize earlier than at the apical inputs. stock in milli-Q water and diluted down to achieve a final bath concentration of 200 µM (Tocris Bioscience). Picrotoxin from Sigma-Aldrich was prepared as a 100-mM stock in DMSO and diluted down to achieve a final bath concentration of 50 µM. DL-2-Amino-5-phosphonopentanoic acid sodium salt (APV) from Tocris was prepared as a 100-mM stock in milli-Q water and diluted down to achieve a final bath concentration of 50 µM. Current–Voltage Relationship CNQX disodium salt from Tocris Bioscience was prepared as a 100-mM stock in milli-Q water and diluted down to achieve a final bath concentration of 50 µM. Statistical Analysis When we used a spermine-free intracellular solution, the 3rd and 4th EPSCs in the train were significantly lower 20 min after polyamine washout than at the beginning of the recording at apical dendrites (p = 0.037 for the 3rd EPSCs and p = 0.014 for the 4th EPSCs, two-way RM ANOVA, n = 8), and we observed the same tendency at basal dendrites (p = 0.117 for the 3d EPSCs and p = 0.066 for the 4th EPSCs, two-way RM ANOVA, n = 8), whereas with the presence of polyamines in the patch pipette we did not observe such differences (apical: p = 0.46 for the 3rd EPSCs and p = 0.101 for the 4th EPSCs; basal: p = 0.47 for the 3rd EPSCs and p = 0.26 for the 4th EPSCs, two-way RM ANOVA, n = 6). Thus, we showed polyamine- dependent facilitation at the synapses of CA1 pyramidal cells, which indicates the presence of CP-AMPARs at both apical and basal dendrites of these cells. However, experiments measuring the current–voltage characteristics of these inputs revealed that the CP-AMPA receptors contribute more to synaptic RESULTS Calcium-Permeable AMPA Receptors Contribute More to Synaptic Transmission at Apical Dendrites of CA1 Pyramidal Neurons Than at Basal Dendrites Calcium-Permeable AMPA Receptors Contribute More to Synaptic Transmission at Apical Dendrites of CA1 Pyramidal Neurons Than at Basal Dendrites First, to evaluate the contribution of GluA2-lacking AMPARs to currents in apical and basal compartments of CA1 pyramidal neurons, we measured the RI by stimulating apical and basal dendrites with glass electrodes located as shown in Figure 1A. The rectification index reflects inward rectification of GluA2-lacking AMPARs and is scored as the ratio of the current amplitude measured at −70 mV and +35 mV (EPSC–70 mV/EPSC+35 mV). We found that the rectification index in responses to stimulation of apical dendrites (RI = 3.4 ± 0.2) is significantly higher than in responses to stimulation of basal dendrites (RI = 2.5 ± 0.2; p = 0.005, one-way ANOVA, n = 14, Figure 1C). The current–voltage (IV) characteristics for each of the inputs supported the difference revealed by RI measurement; the apical IV demonstrated significantly higher inward rectification (p = 0.035, one-way ANOVA, n = 7, Figure 1B). The most common approach to detecting the contribution of GluA2-lacking AMPARs to synaptic transmission is to test sensitivity to CP-AMPAR antagonists, for example, 1-naphthylacetyl spermine trihydrochloride (Naspm). June 2021 \| Volume 13 \| Article 656377 Frontiers in Synaptic Neuroscience \| www.frontiersin.org 4 AMPARs Contribution to Synaptic Transmission Ivanova et al. FIGURE 1 \| Contribution of calcium-permeable AMPA receptors to synaptic transmission. (A) A schematic representation of the location of the stimulating electrode. (B) Current–voltage characteristics of apical (purple) and basal (orange) inputs. n = 7 cells from four animals, p ≤0.05. (C) Comparison of rectiﬁcation indices in stratum radiatum (purple) and stratum oriens (orange). Open gray circles represent individual data points, n = 14 cells from nine animals, p ≤0.005. (D) The time course of averaged EPSC amplitude changes at apical inputs during washout of polyamine (black, n = 6 cells from three animals) and subsequent Naspm (200 µM) application (purple, n = 10 cells from eight animals), when CA1 pyramidal neurons were dialyzed with a polyamine-free intracellular solution. For analysis, 10 successive responses were averaged and normalized to the mean EPSC amplitude obtained between the 20th and 30th minutes of the recording session (baseline marked with gray dashed lines). (E) The same time course as (D) but for basal inputs. Blue, untreated cells, n = 6 cells from ﬁve animals. Orange, cells treated with Naspm (200 µM), (n = 10 cells from eight animals). (F) A histogram demonstrating normalized EPSCs for the last 5 min of the curves D and E. Calcium-Permeable AMPA Receptors Contribute More to Synaptic Transmission at Apical Dendrites of CA1 Pyramidal Neurons Than at Basal Dendrites A decrease in the intracellular concentration of nitric oxide significantly reduced RIs at the apical dendrites compared to control cells (DMSO: 3.1 ± 0.1, p = 0.321, n = 8; L-NAME: 2.6 ± 0.2, p = 0.025, n = 8; 3-br-7-ni: 2.4 ± 0.3, p = 0.007, n = 10; PTIO: 2.4 ± 0.3, p = 0.002, n = 7, two-way ANOVA), but not basal dendrites (DMSO: 2.1 ± 0.2, p = 0.28, n = 8; L-NAME: 2.6 ± 0.4, p = 0.78, n = 8; 3-br-7-ni: 2.1 ± 0.2, p = 0.24, n = 10; PTIO: 1.8 ± 0.1, p = 0.07, n = 7, two-way ANOVA), in various ways, thus leveling the significant difference between these two inputs. The current–voltage characteristics of the studied inputs under NOS blockade by L-NAME demonstrated a loss of inward rectification at synaptic inputs in apical but not basal dendrites (p = 0.031 for apical, p = 0.4 for basal, two-way ANOVA, n = 7, Figure 3A). We also compared the AMPA/NMDA ratio at apical and basal dendrites with Calcium-Permeable AMPA Receptors Contribute More to Synaptic Transmission at Apical Dendrites of CA1 Pyramidal Neurons Than at Basal Dendrites Open gray circles represent individual data points, p ≤0.005. (G) Representative averaged traces at the indicated times of the curves D and E are shown in black and purple for apical dendrites, and in blue and orange for basal dendrites. FIGURE 1 \| Contribution of calcium-permeable AMPA receptors to synaptic transmission. (A) A schematic representation of the location of the stimulating electrode. (B) Current–voltage characteristics of apical (purple) and basal (orange) inputs. n = 7 cells from four animals, p ≤0.05. (C) Comparison of rectiﬁcation indices in stratum radiatum (purple) and stratum oriens (orange). Open gray circles represent individual data points, n = 14 cells from nine animals, p ≤0.005. (D) The time course of averaged EPSC amplitude changes at apical inputs during washout of polyamine (black, n = 6 cells from three animals) and subsequent Naspm (200 µM) application (purple, n = 10 cells from eight animals), when CA1 pyramidal neurons were dialyzed with a polyamine-free intracellular solution. For analysis, 10 successive responses were averaged and normalized to the mean EPSC amplitude obtained between the 20th and 30th minutes of the recording session (baseline marked with gray dashed lines). (E) The same time course as (D) but for basal inputs. Blue, untreated cells, n = 6 cells from ﬁve animals. Orange, cells treated with Naspm (200 µM), (n = 10 cells from eight animals). (F) A histogram demonstrating normalized EPSCs for the last 5 min of the curves D and E. Open gray circles represent individual data points, p ≤0.005. (G) Representative averaged traces at the indicated times of the curves D and E are shown in black and purple for apical dendrites, and in blue and orange for basal dendrites. transmission at apical dendrites of CA1 pyramidal neurons than basal dendrites. RI of DMSO alone (25 µl). Nitric Oxide Synthase Blockade Alters CP-AMPAR Contribution to Currents in Apical but Not Basal Dendrites Since nitric oxide can affect the incorporation of CP-AMPARs into the postsynaptic membrane of cells, we tested whether the blockade of its synthesis by various NO synthase inhibitors affects the characteristics of the apical and basal synapses of CA1 pyramidal cells. Figure 3B shows the rectification indices of the apical and basal inputs under the NO-synthase inhibitor L-NAME (200 µM), the selective inhibitor of neuronal NO-synthase 3-bromo-7-nitroindazole (50 µM), and the NO scavenger carboxy-PTIO (50 µM). The latter two inhibitors were dissolved in DMSO, and we also tested the effect on June 2021 \| Volume 13 \| Article 656377 Frontiers in Synaptic Neuroscience \| www.frontiersin.org 5 AMPARs Contribution to Synaptic Transmission Ivanova et al. FIGURE 2 \| Polyamine-dependent facilitation at the synapses of CA1 pyramidal cells. (A) Polyamine-dependent facilitation at apical (upper) and basal (lower) dendrites recorded with spermine-containing intracellular solution 5 and 20 min after whole-cell patch establishment, n = 6 cells from four animals. (B) Polyamine-dependent facilitation at apical (upper) and basal (lower) dendrites recorded with spermine-free intracellular solution 5 and 20 min after whole-cell patch establishment, n = 8 cells from ﬁve animals, p ≤0.05; ns, nonsigniﬁcant. Averaged traces of responses to stimuli with a 33-ms interstimulus interval at the indicated times are displayed above each graph. EPSC numbering corresponds to the x-axis of each graph. FIGURE 2 \| Polyamine-dependent facilitation at the synapses of CA1 pyramidal cells. (A) Polyamine-dependent facilitation at apical (upper) and basal (lower) dendrites recorded with spermine-containing intracellular solution 5 and 20 min after whole-cell patch establishment, n = 6 cells from four animals. (B) Polyamine-dependent facilitation at apical (upper) and basal (lower) dendrites recorded with spermine-free intracellular solution 5 and 20 min after whole-cell patch establishment, n = 8 cells from ﬁve animals, p ≤0.05; ns, nonsigniﬁcant. Averaged traces of responses to stimuli with a 33-ms interstimulus interval at the indicated times are displayed above each graph. EPSC numbering corresponds to the x-axis of each graph. or without NOS inhibition (Figure 5D). In control cells, this ratio was significantly higher at apical dendrites than at basal dendrites (6.7 ± 0.6 vs. 4.9 ± 0.6, n = 5, p = 0.049, two-way ANOVA), and this difference disappeared after NOS inhibition (4.3 ± 0.8 vs. 3.9 ± 0.8, n = 5, p = 0.788, two-way ANOVA), supporting the results of the experiments with RI measurement. 5 min of recording (89.4% ± 3.2 vs. Frontiers in Synaptic Neuroscience \| www.frontiersin.org Nitric Oxide Synthase Blockade Alters CP-AMPAR Contribution to Currents in Apical but Not Basal Dendrites 83% ± 6.5, n = 10, p = 0.264, two-way ANOVA, Figure 4D). Interestingly, the increase in the current amplitude at the beginning of the recording, which is associated with polyamine washout in both str. radiatum and str. Oriens, persisted after incubation in L-NAME (Figures 4A,B). This might be due to either specific mechanisms associated with polyamine washout or the possibility of NO-dependent regulation of CP-AMPAR sensitivity to polyamines. In the latter case, polyamines could still block the receptor pore but the blockade is more easily relieved which causes smaller EPSC increase and faster baseline stabilization. Additionally, NOS inhibition prevented the drop in the current amplitude during CP-AMPAR blockade by Naspm in str. radiatum (109% ± 7.3 vs. 3% ± 4, n = 10, p = 0.001, two-way ANOVA, Figure 4C); however, in str. oriens we did not find a significant difference in the last June 2021 \| Volume 13 \| Article 656377 Frontiers in Synaptic Neuroscience \| www.frontiersin.org 6 AMPARs Contribution to Synaptic Transmission Ivanova et al. FIGURE 3 \| Rectiﬁcation properties of the apical and basal inputs under the NO-synthase inhibitors. (A) Current–voltage characteristics of apical (left) and basal (right) inputs in control cells (n = 7) and under nitric oxide synthase (NOS) blockade (n = 7), four animals, p ≤0.05. (B) Comparison of rectiﬁcation indices in stratum radiatum (purple) and stratum oriens (orange) in control cells (n = 14 cells from nine animals) and under treatment with DMSO (n = 8 cells from four animals), L-NAME (n = 8 cells from six animals), 3-bromo-7-nitroindazole (n = 10 cells from seven animals), and PTIO (n = 7 cells from four animals). Top, example traces at +35 mV and −70 mV. p ≤0.05, p ≤0.005. Open gray circles represent individual data points. FIGURE 3 \| Rectiﬁcation properties of the apical and basal inputs under the NO-synthase inhibitors. (A) Current–voltage characteristics of apical (left) and basal (right) inputs in control cells (n = 7) and under nitric oxide synthase (NOS) blockade (n = 7), four animals, p ≤0.05. (B) Comparison of rectiﬁcation indices in stratum radiatum (purple) and stratum oriens (orange) in control cells (n = 14 cells from nine animals) and under treatment with DMSO (n = 8 cells from four animals), L-NAME (n = 8 cells from six animals), 3-bromo-7-nitroindazole (n = 10 cells from seven animals), and PTIO (n = 7 cells from four animals). Nitric Oxide Synthase Blockade Alters CP-AMPAR Contribution to Currents in Apical but Not Basal Dendrites Top, example traces at +35 mV and −70 mV. p ≤0.05, p ≤0.005. Open gray circles represent individual data points. Next, we tested how the decrease of nitric oxide in the cell would affect polyamine-dependent facilitation at the studied inputs in the presence of spermine in the recording pipette (Figures 5A,B). NOS inhibition significantly affects the 4th EPSCs at apical dendrites (p = 0.006, two-way RM ANOVA, n = 7), but not at basal dendrites (p = 0.499, two-way RM ANOVA, n = 7). PdF recording was performed at the fifth minute of recording after a stable baseline was reached. Considering the wide range of nitric oxide action in the presynapse of cells (Hardingham et al., 2013), we tested whether the discovered effect was due to the presynaptic effect of nitric oxide. One of the possible approaches for evaluation of the presynaptic contribution to synaptic transmission is paired-pulse ratio measurement (Schulz et al., 1994; Christie and Jahr, 2006). Recording of paired-pulse ratio at the synapses of CA1 pyramidal cells during NOS inhibition did not reveal significant differences in these cells compared to control cells (control: 1.6 ± 0.1 in str. radiatum, and 1.4 ± 0.06 in str. oriens, n = 8; L-NAME: 1.5 ± 0.1 in str. radiatum, and 1.4 ± 0.07 in str. oriens, n = 7; 3-bromo-7-ni: 1.4 ± 0.06 in str. radiatum, and 1.4 ± 0.1 in str. oriens, n = 7). Thus, the results suggest that the disappearance of PdF in str. radiatum after inhibition of nitric oxide synthesis was associated with the interaction of nitric oxide with CP- AMPARs. We examined whether the sGC-dependent pathway is involved in this interaction by treating slices with the sGC inhibitor ODQ (30 µM). We measured RIs under this condition and found that sGC inhibition did not affect the RIs of neither apical (RI = 3.5 ± 0.2, n = 5) nor basal (RI = 2.5 ± 0.3, n = 5) inputs (Figure 6, p = 0.027, two-way ANOVA). Next, we blocked NOS and sGC simultaneously and found that treating with L-NAME also leveled the RIs of apical (RI = 3.3 ± 0.3, n = 5) June 2021 \| Volume 13 \| Article 656377 Frontiers in Synaptic Neuroscience \| www.frontiersin.org 7 AMPARs Contribution to Synaptic Transmission Ivanova et al. Nitric Oxide Synthase Blockade Alters CP-AMPAR Contribution to Currents in Apical but Not Basal Dendrites FIGURE 4 \| NOS inhibition prevented CP-AMPAR current blockade by Naspm in the apical dendrites, but not in the basal dendrites. (A) The time course of EPSC amplitude changes in apical inputs during washout of polyamine (yellow, n = 6 cells from ﬁve animals), Naspm (200 µM) application (purple, n = 10, eight animals), and under NOS inhibition (black, n = 6 cells from four animals) and under NOS inhibition with Naspm application (blue, n = 10 cells from eight animals). For analysis, 10 successive responses were averaged and normalized to the mean EPSC amplitude obtained between 20 and 30 min of the recording session (baseline marked with gray dashed lines). Top, example traces. (B) The same time course as (A) but for basal inputs. Pink, untreated cells, n = 6 cells from ﬁve animals. Yellow, cells treated with Naspm, n = 10 cells from eight animals. Blue, cells treated with L-NAME and Naspm, n = 10 cells from eight animals. Black, cells treated with L-NAME, n = 6 cells from four animals. Top, example traces. (C,D) Histograms demonstrating normalized EPSCs for the last 5 min of the curves A and B, respectively. Open gray circles represent individual data points, p ≤0.005. FIGURE 4 \| NOS inhibition prevented CP-AMPAR current blockade by Naspm in the apical dendrites, but not in the basal dendrites. (A) The time course of EPSC amplitude changes in apical inputs during washout of polyamine (yellow, n = 6 cells from ﬁve animals), Naspm (200 µM) application (purple, n = 10, eight animals), and under NOS inhibition (black, n = 6 cells from four animals) and under NOS inhibition with Naspm application (blue, n = 10 cells from eight animals). For analysis, 10 successive responses were averaged and normalized to the mean EPSC amplitude obtained between 20 and 30 min of the recording session (baseline marked with gray dashed lines). Top, example traces. (B) The same time course as (A) but for basal inputs. Pink, untreated cells, n = 6 cells from ﬁve animals. Yellow, cells treated with Naspm, n = 10 cells from eight animals. Blue, cells treated with L-NAME and Naspm, n = 10 cells from eight animals. Black, cells treated with L-NAME, n = 6 cells from four animals. Top, example traces. (C,D) Histograms demonstrating normalized EPSCs for the last 5 min of the curves A and B, respectively. Nitric Oxide Synthase Blockade Alters CP-AMPAR Contribution to Currents in Apical but Not Basal Dendrites Open gray circles represent individual data points, p ≤0.005. and basal (RI = 3 ± 0.4) inputs, as in the cells treated with the NOS inhibitors alone (p = 0.680, two-way ANOVA). observed only a tendency (Figure 2). However, taking into account that basal dendrites are located closer to the soma than apical dendrites, polyamines are washed out via the patch pipette faster. According to this, PdF at the basal inputs should slightly decrease by the fifth minute after whole-cell patch formation. Indeed, if the recording starts at ∼1 min, the p-value for the 4th EPSC decreases in basal dendrites (Supplementary Figure 2, p = 0.024). However, we observed a significantly higher contribution of CP-AMPARs to glutamatergic synaptic transmission in stratum radiatum, than in stratum oriens by measuring the rectification index (Figure 1C), which suggests that physiology or number of the receptors differs in these compartments. Frontiers in Synaptic Neuroscience \| www.frontiersin.org DISCUSSION In the present study, we described data supporting the presence of CP-AMPARs not only in the apical dendrites of CA1 pyramidal cells but also in the basal dendrites. Recording of basic transmission during whole-cell patch-clamp with spermine-free intracellular solution showed a gradual increase in the EPSC amplitudes at both inputs (Figures 1D,E), which is associated with the release of GluR2-lacking AMPARs from the polyamine block and leading to an increase in the conductance of these receptors (Rozov et al., 2012). This growth does not occur with 10 µM spermine in the patch pipette (Supplementary Figure 3). GluR2-lacking AMPAR blockade decreased the response amplitudes significantly in the apical dendrites and insignificantly in the basal dendrites. In addition, application of high-frequency stimulation to the inputs revealed a significant polyamine-dependent facilitation (Rozov and Burnashev, 1999; Rozov et al., 2018) in str. radiatum, while in str. oriens we The presence of GluA2-lacking AMPARs in CA1 cells prompts the question of their localization, as studies have shown different data on their presence in the postsynapse after LTP induction (Plant et al., 2006; Adesnik and Nicoll, 2007; Moult et al., 2010). Some studies have demonstrated that GluA2-lacking AMPARs constitute a small subpopulation of the synaptic AMPA receptors in non-potentiated CA1 pyramidal neurons in adult rodents (Rozov et al., 2012; Mattison et al., June 2021 \| Volume 13 \| Article 656377 Frontiers in Synaptic Neuroscience \| www.frontiersin.org Frontiers in Synaptic Neuroscience \| www.frontiersin.org 8 AMPARs Contribution to Synaptic Transmission Ivanova et al. FIGURE 5 \| NOS inhibition affected polyamine-dependent facilitation in the apical but not basal dendrites. (A) Facilitation at apical dendrites recorded with spermine-containing intracellular solution in untreated cells (purple, n = 6 cells from four animals) and under NOS blockade (orange, n = 6 cells from four animals). p ≤0.05. Top, example traces. (B) Basal dendrite responses recorded with spermine-containing intracellular solution in untreated cells (purple, n = 7 cells from four animals) and under NOS blockade (orange, n = 7 cells from four animals). Top, example traces. EPSC numbering corresponds to the x-axis of each graph. ns, nonsigniﬁcant. (C) Comparison of paired-pulse ratios in stratum radiatum (purple) and stratum oriens (orange) in untreated cells and under NOS blockade (control: n = 8 cells from ﬁve animals; L-NAME: n = 7 cells from three animals; 3-bromo-7-ni: n = 7 cells from three animals). Open gray circles represent individual data points. Top, example traces. DISCUSSION (D) Comparison of AMPA-NMDA ratios in stratum radiatum (purple) and stratum oriens (orange) in untreated cells (n = 5 cells from three animals) and under NOS blockade (n = 5 cells from three animals). Top, example traces. p ≤0.05, p ≤0.005. FIGURE 5 \| NOS inhibition affected polyamine-dependent facilitation in the apical but not basal dendrites. (A) Facilitation at apical dendrites recorded with spermine-containing intracellular solution in untreated cells (purple, n = 6 cells from four animals) and under NOS blockade (orange, n = 6 cells from four animals). p ≤0.05. Top, example traces. (B) Basal dendrite responses recorded with spermine-containing intracellular solution in untreated cells (purple, n = 7 cells from four animals) and under NOS blockade (orange, n = 7 cells from four animals). Top, example traces. EPSC numbering corresponds to the x-axis of each graph. ns, nonsigniﬁcant. (C) Comparison of paired-pulse ratios in stratum radiatum (purple) and stratum oriens (orange) in untreated cells and under NOS blockade (control: n = 8 cells from ﬁve animals; L-NAME: n = 7 cells from three animals; 3-bromo-7-ni: n = 7 cells from three animals). Open gray circles represent individual data points. Top, example traces. (D) Comparison of AMPA-NMDA ratios in stratum radiatum (purple) and stratum oriens (orange) in untreated cells (n = 5 cells from three animals) and under NOS blockade (n = 5 cells from three animals). Top, example traces. p ≤0.05, p ≤0.005. 2014), and in other studies, it was shown that GluR2- lacking AMPARs in pyramidal cells in the hippocampus are replaced with GluR2-containing AMPARs in mature animals (Ho et al., 2007; Malkin et al., 2016). This discrepancy can be explained by the presence of polyamines in the patch pipette which affects the ability of GluA2-lacking AMPAR antagonists to block them (Rozov et al., 2012). In addition, the presence of polyamines in the pipette also determines the rectification characteristics of synapses with GluR2-lacking AMPARs in their membrane (Kamboj et al., 1995): the current–voltage characteristics of such synapses in the absence of spermine are linear, as in mutant GluA1-/- mice (Rozov et al., 2012). polyamines in CA1 pyramidal cells is unknown; however, it is known that the concentration varies in different regions of the rat brain (Shaskan et al., 1973), as well as in other mammals (Igarashi and Kashiwagi, 2010). Frontiers in Synaptic Neuroscience \| www.frontiersin.org DISCUSSION However, according to our data, the increase in current amplitude during PdF under NOS inhibition disappears, which indicates an unlikelihood of an increased concentration of intracellular polyamines in apical dendrites. Nitric oxide may also act through the regulation of CP-AMPAR trafficking or through the modification of incorporated receptors. NO inhibition could disrupt one of the possible mechanisms involved in the trafficking of AMPAR subunits: the indirect sGC-dependent pathway (Serulle et al., 2007), direct nitrosylation of GluR1 subunits (Selvakumar et al., 2013; Von Ossowski et al., 2017), or different protein–protein interactions (Chen et al., 2000; Zhang et al., 2015; for review, see Ivanova et al., 2020). One such interaction was shown for GluR2 incorporation via NSF-dependent declustering of the PICK1–GluR2 complex (Hanley et al., 2002; Huang et al., 2005; Sossa et al., 2007); however, in the two latter studies the NO donor application resulted in increased GluR1 surface expression (Hanley et al., 2002, Figure 5B and Sossa et al., 2007, Figure 4C); thus, there is a possibility of NSF involvement in GluR2-lacking AMPAR trafficking. We tested the cGMP-dependent pathway by blocking sGC (Figure 6) but did not find any differences in the rectification characteristics of the studied synapses. Nitric oxide was shown to be involved in LTP maintenance in str. radiatum, but not in str. Oriens; however, the presence of nNOS was shown in both str. radiatum and str. oriens (see the ‘‘Introduction’’ section). We assumed that this contrast was due to the difference in the modulation of synaptic characteristics by nitric oxide in these areas. In particular, nitric oxide could differently affect the CP-AMPAR contribution to synaptic transmission of apical and basal dendrites. We found that NOS inhibition by two different inhibitors and treatment with an NO scavenger reduced inward rectification and caused a drop in the rectification index at apical dendrites (Figure 3), which reflected a decrease in the contribution of GluR2-lacking AMPARs to the currents in these synapses. In addition, NOS inhibition prevented the decrease in response amplitude under Naspm treatment (Figure 4) and reduced the polyamine-dependent facilitation (Figures 5A,B), whereas, at basal inputs the NOS inhibition did not affect any of the synaptic characteristics. Thus, our data indicate that nitric oxide does not affect the contribution of GluR2-lacking AMPARs to synapses of CA1 pyramidal cell basal dendrites, while inhibition of nitric oxide synthesis significantly reduced the contribution of these receptors to apical dendrite synaptic currents. DISCUSSION In our study, we used 10 µM spermine in the intracellular solution to record the rectification properties of inputs. One can argue that the spermine concentration in our experiments was insufficient to successfully block GluR2- lacking AMPARs; however, the experiments with polyamine- dependent facilitation (Figure 2) indicated the opposite: 10 µM spermine in the patch pipette prevented polyamine washout from the dendritic compartments. Moreover, using 10 µM spermine- containing intracellular solution does not cause an increase in EPSC amplitude (Supplementary Figure 3). In addition, Hu et al. (1994) showed that 100 µM spermine inhibits [3H]L- citrulline formation, which reflects NOS activity, by ∼60% in cerebellar cells, whereas 10 µM only slightly inhibits this The polyamine concentration in the patch pipette varies in different studies (Rozov et al., 2012; Mattison et al., 2014; Malkin et al., 2016). The precise concentration of intracellular free June 2021 \| Volume 13 \| Article 656377 Frontiers in Synaptic Neuroscience \| www.frontiersin.org 9 AMPARs Contribution to Synaptic Transmission Ivanova et al. FIGURE 6 \| Soluble guanylyl cyclase inhibition did not affect CP-AMPAR contribution to synaptic transmission of CA1 pyramidal cells. Comparison of rectiﬁcation indices in stratum radiatum (purple) and stratum oriens (orange) in untreated cells (n = 14 cells from nine animals) and under treatment with ODQ (30 µM, n = 5 cells from three animals) and L-NAME (200 µM, n = 5 cells from three animals). Top, example traces. p ≤0.05. FIGURE 6 \| Soluble guanylyl cyclase inhibition did not affect CP-AMPAR contribution to synaptic transmission of CA1 pyramidal cells. Comparison of rectiﬁcation indices in stratum radiatum (purple) and stratum oriens (orange) in untreated cells (n = 14 cells from nine animals) and under treatment with ODQ (30 µM, n = 5 cells from three animals) and L-NAME (200 µM, n = 5 cells from three animals). Top, example traces. *p ≤0.05. reaction, Figure 1). This might indicate that high spermine concentration in the patch pipette can cause NOS inhibition. Indeed, when 100 µM spermine was added to the patch pipette, the difference in RIs between apical and basal dendrites of CA1 pyramidal cells was not statistically significant (apical: 3 ± 0.4; basal: 2.3 ± 0.2, n = 7, p = 0.12, two-way ANOVA; Supplementary Figure 1), as in the case of NOS blockade. et al., 1998; Boucher et al., 1999), which in turn determines the conductivity of GluR2-lacking AMPARs. Frontiers in Synaptic Neuroscience \| www.frontiersin.org REFERENCES Cenquizca, L. A., and Swanson, L. W. (2007). Spatial organization of direct hippocampal field CA1 axonal projections to the rest of the cerebral cortex. Brain Res. Rev. 56, 1–26. doi: 10.1016/j.brainresrev.2007.05.002 Adesnik, H., and Nicoll, R. A. (2007). Conservation of glutamate receptor 2-containing AMPA receptors during long-term potentiation. J. Neurosci. 27, 4598–4602. doi: 10.1523/JNEUROSCI.0325-07.2007 Chen, L., Chetkovich, D. M., Petralia, R. S., Sweeney, N. T., Kawasaki, Y., Wenthold, R. J., et al. (2000). Stargazin regulates synaptic targeting of AMPA receptors by two distinct mechanisms. Nature 408, 936–943. doi: 10.1038/35050030 Bal, N., Roshchin, M., Salozhin, S., and Balaban, P. (2017). Nitric oxide upregulates proteasomal protein degradation in neurons. Cell. Mol. Neurobiol. 37, 763–769. doi: 10.1007/s10571-016-0413-9 Christie, J. M., and Jahr, C. E. (2006). Multivesicular release at Schaffer collateral-CA1 hippocampal synapses. J. Neurosci. 26, 210–216. doi: 10.1523/JNEUROSCI.4307-05.2006 Bartus, K., Pigott, B., and Garthwaite, J. (2013). Cellular targets of nitric oxide in the hippocampus. PLoS One 8:e57292. doi: 10.1371/journal.pone.00 57292 Diering, G. H., and Huganir, R. L. (2018). The AMPA receptor code of synaptic plasticity. Neuron 100, 314–329. doi: 10.1016/j.neuron.2018.10.018 Benavides-Piccione, R., Regalado-Reyes, M., Fernaud-Espinosa, I., Kastanauskaite, A., Tapia-González, S., León-Espinosa, G., et al. (2020). Differential structure of hippocampal CA1 pyramidal neurons in the human and mouse. Cereb. Cortex 30, 730–752. doi: 10.1093/cercor/ bhz122 Dugue, G. P., Dumoulin, A., Triller, A., and Dieudonne, S. (2005). Target- dependent use of co-released inhibitory transmitters at central synapses. J. Neurosci. 25, 6490–6498. doi: 10.1523/JNEUROSCI.1500-05.2005 Haley, J. E., Schaible, E., Pavlidis, P., Murdock, A., and Madison, D. V. (1996). Basal and apical synapses of CA1 pyramidal cells employ different LTP induction mechanisms. Learn. Mem. 3, 289–295. doi: 10.1101/lm.3.4.289 Blackshaw, S., Eliasson, M. J. L., Sawa, A., Watkins, C. C., Krug, D., Gupta, A., et al. (2003). Species, strain and developmental variations in hippocampal neuronal and endothelial nitric oxide synthase clarify discrepancies in nitric oxide- dependent synaptic plasticity. Neuroscience 119, 979–990. doi: 10.1016/s0306- 4522(03)00217-3 Hanley, J. G., Khatri, L., Hanson, P. I., and Ziff, E. B. (2002). NSF ATPase and alpha-/beta-SNAPs disassemble the AMPA receptor-PICK1 complex. Neuron 34, 53–67. doi: 10.1016/s0896-6273(02)00638-4 Boucher, J. L., Moali, C., and Tenu, J. P. (1999). Nitric oxide biosynthesis, nitric oxide synthase inhibitors and arginase competition for L-arginine utilization. Cell. Mol. Life Sci. 55, 1015–1028. doi: 10.1007/s000180050352 Hardingham, N., Dachtler, J., and Fox, K. (2013). The role of nitric oxide in pre-synaptic plasticity and homeostasis. Front. Cell. Neurosci. 7:190. doi: 10.3389/fncel.2013.00190 Bradshaw, K. FUNDING This work was supported by the Russian Science Foundation, grant #20-15-00398. DISCUSSION So far, it is unclear which nitric oxide-dependent mechanism exerts this effect in str. radiatum. For instance, the concentration of free nitric oxide in the cell can affect synthesis of intracellular polyamines (Buga Obtained results suggest that nitric oxide upregulates the CP-AMPAR sensitivity to polyamines: this might explain changes in the current–voltage relationships, decreased PdF, and persisting EPSC growth under NOS inhibition. Moreover, such modulation was demonstrated for GluR2-lacking AMPARs by the auxiliary protein stargazin (Soto et al., 2007). However, the mechanism of this modulation requires further clarification. One June 2021 \| Volume 13 \| Article 656377 Frontiers in Synaptic Neuroscience \| www.frontiersin.org Frontiers in Synaptic Neuroscience \| www.frontiersin.org 10 AMPARs Contribution to Synaptic Transmission Ivanova et al. possible explanation of the NOS inhibition effect is a reduced GluR2-lacking AMPAR surface expression. However, that does not explain the increase in EPSC amplitudes during polyamine washout after incubation in L-NAME. possible explanation of the NOS inhibition effect is a reduced GluR2-lacking AMPAR surface expression. However, that does not explain the increase in EPSC amplitudes during polyamine washout after incubation in L-NAME. AUTHOR CONTRIBUTIONS VI designed and performed the experiments, analyzed the data, and wrote the article. PB managed the project and wrote the article. NB conceived and designed the experiments, managed the project, and wrote the article. All authors contributed to the article and approved the submitted version. ACKNOWLEDGMENTS We thank David Jappy for useful comments on the manuscript. DATA AVAILABILITY STATEMENT The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fnsyn. 2021.656377/full#supplementary-material. The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. ETHICS STATEMENT The animal study was reviewed and approved by the Ethical committee of the Institute of Higher Nervous Activity and Neurophysiology, Russian Academy of Sciences (IHNA RAS). In conclusion, this study demonstrates the effects of NOS inhibition on GluA2-lacking AMPA receptor-mediated currents at apical but not basal dendrites of the CA1 pyramidal neurons. This effect could underlie the differences in synaptic plasticity of the aforementioned synapses, although the mechanisms of this effect require further study. Many studies demonstrated differences in the mechanisms of synaptic plasticity between different neuron’s compartments; the concept of a specialized ‘‘memory synapse’’ is discussed (Sossin, 2018). We believe that our study highlights the importance of such phenomena as synaptic heterogeneity which may underlie the features of information processing in the hippocampus. In addition, the importance of AMPA receptors for such aspects of cell life as synaptic plasticity and homeostasis is undeniable. AMPAR GluA1–4 subunit trafficking, subunit-specific protein interactions, auxiliary subunits, and posttranslational modifications could predict the types and extent of synaptic plasticity; this is the so-called ‘‘AMPA receptor code of synaptic plasticity’’ (Diering and Huganir, 2018), and our data reveal more details of this complex code. REFERENCES D., Emptage, N. J., and Bliss, T. V. P. (2003). A role for dendritic protein synthesis in hippocampal late LTP. Eur. J. Neurosci. 18, 3150–3152. doi: 10.1111/j.1460-9568.2003.03054.x Higuchi, M., Single, F. N., Köhler, M., Sommer, B., Sprengel, R., and Seeburg, P. H. (1993). RNA editing of AMPA receptor subunit GluR-B: A base-paired intron-exon structure determines position and efficiency. Cell 75, 1361–1370. doi: 10.1016/0092-8674(93)90622-w Buga, G. M., Wei, L. H., Bauer, P. M., Fukuto, J. M., and Ignarro, L. J. (1998). N(G)- hydroxy-L-arginine and nitric oxide inhibit Caco-2 tumor cell proliferation by distinct mechanisms. Am. J. Physiol. 275, R1256–R1264. doi: 10.1152/ajpregu. 1998.275.4.R1256 Ho, M. T. W., Pelkey, K. A., Topolnik, L., Petralia, R. S., Takamiya, K., Xia, J., et al. (2007). Developmental expression of Ca2+-permeable AMPA receptors underlies depolarization-induced long-term depression at mossy fiber-CA3 June 2021 \| Volume 13 \| Article 656377 Frontiers in Synaptic Neuroscience \| www.frontiersin.org 11 AMPARs Contribution to Synaptic Transmission Ivanova et al. pyramid synapses. J. Neurosci. 27, 11651–11662. doi: 10.1523/JNEUROSCI. 2671-07.2007 Rozov, A., Sprengel, R., and Seeburg, P. H. (2012). GluA2-lacking AMPA receptors in hippocampal CA1 cell synapses: evidence from gene-targeted mice. Front. Mol. Neurosci. 5:22. doi: 10.3389/fnmol.2012.00022 pyramid synapses. J. Neurosci. 27, 11651–11662. doi: 10.1523/JNEUROSCI. 2671-07.2007 Hu, J., Mahmoud, M. I., and El-Fakahany, E. E. (1994). Polyamines inhibit nitric oxide synthase in rat cerebellum. Neurosci. Lett. 175, 41–45. doi: 10.1016/0304- 3940(94)91073-1 Rozov, A., Zakharova, Y., Vazetdinova, A., and Valiullina-Rakhmatullina, F. (2018). The role of polyamine-dependent facilitation of calcium permeable ampars in short-term synaptic enhancement. Front. Cell. Neurosci. 12:345. doi: 10.3389/fncel.2018.00345 Huang, Y., Man, H. Y., Sekine-Aizawa, Y., Han, Y., Juluri, K., Luo, H., et al. (2005). S-nitrosylation of N-ethylmaleimide sensitive factor mediates surface expression of AMPA receptors. Neuron 46, 533–540. doi: 10.1016/j.neuron. 2005.03.028 Schulz, P. E., Cook, E. P., and Johnston, D. (1994). Changes in paired- pulse facilitation suggest presynaptic involvement in long-term potentiation. J. Neurosci. 14, 5325–5337. doi: 10.1523/JNEUROSCI.14-09-05325.1994 Hulme, S. R., Jones, O. D., Ireland, D. R., and & Abraham, W. C. (2012). Calcium- dependent but action potential-independent BCM-like metaplasticity in the hippocampus. J. Neurosci. 32, 6785–6794. doi: 10.1523/JNEUROSCI.0634-12. 2012 Selvakumar, B., Jenkins, M. A., Hussain, N. K., Huganir, R. L., Traynelis, S. F., and Snyder, S. H. (2013). S-nitrosylation of AMPA receptor GluA1 regulates phosphorylation, single-channel conductance and endocytosis. Proc. Natl. Acad. Sci. U S A 110, 1077–1082. doi: 10.1073/pnas.1221295110 Igarashi, K., and Kashiwagi, K. (2010). Modulation of cellular function by polyamines. REFERENCES Int. J. Biochem. Cell Biol. 42, 39–51. doi: 10.1016/j.biocel.2009. 07.009 Serulle, Y., Zhang, S., Ninan, I., Puzzo, D., McCarthy, M., Khatri, L., et al. (2007). A GluR1-cGKII interaction regulates AMPA receptor trafficking. Neuron 56, 670–688. doi: 10.1016/j.neuron.2007.09.016 Ivanova, V. O., Balaban, P. M., and Bal, N. V. (2020). Modulation of AMPA receptors by nitric oxide in nerve cells. Int. J. Mol. Sci. 21:981. doi: 10.3390/ijms21030981 Shaskan, E. G., Haraszti, J. H., and Snyder, S. H. (1973). Polyamines: developmental alterations in regional disposition and metabolism in rat brain. J. Neurochem. 20, 1443–1452. doi: 10.1111/j.1471-4159.1973.tb00256.x Kamboj, S. K., Swanson, G. T., and Cull-Candy, S. G. (1995). Intracellular spermine confers rectification on rat calcium-permeable AMPA and kainate receptors. J. Physiol. 486, 297–303. doi: 10.1113/jphysiol.1995.sp020812 Son, H., Hawkins, R. D., Martin, K., Kiebler, M., Huang, P. L., Fishman, M. C., et al. (1996). Long-term potentiation is reduced in mice that are doubly mutant in endothelial and neuronal nitric oxide synthase. Cell 87, 1015–1023. doi: 10.1016/s0092-8674(00)81796-1 Lu, Y. F., Kandel, E. R., and Hawkins, R. D. (1999). Nitric oxide signaling contributes to late-phase LTP and CREB phosphorylation in the hippocampus. J. Neurosci. 19, 10250–10261. doi: 10.1523/JNEUROSCI.19-23-10250.1999 Son, H., Lu, Y. F., Zhuo, M., Arancio, O., Kandel, E. R., and Hawkins, R. D. (1998). The specific role of cGMP in hippocampal LTP. Learn. Mem. 5, 231–245. Malinow, R., and Malenka, R. C. (2002). AMPA receptor trafficking and synaptic plasticity. Annu. Rev. Neurosci. 25, 103–126. doi: 10.1146/annurev.neuro.25. 112701.142758 Sossa, K. G., Beattie, J. B., and & Carroll, R. C. (2007). AMPAR exocytosis through NO modulation of PICK1. Neuropharmacology 53, 92–100. doi: 10.1016/j. neuropharm.2007.04.005 Malkin, S. L., Amakhin, D. V., Veniaminova, E. A., Kim, K. K., Zubareva, O. E., Magazanik, L. G., et al. (2016). Changes of ampa receptor properties in the neocortex and hippocampus following pilocarpine-induced status epilepticus in rats. Neuroscience 327, 146–155. doi: 10.1016/j.neuroscience.2016. 04.024 Sossin, W. S. (2018). Memory synapses are defined by distinct molecular complexes: a proposal. Front. Synaptic Neurosci. 10:5. doi: 10.3389/fnsyn.2018. 00005 Soto, D., Coombs, I. D., Kelly, L., Farrant, M., and Cull-Candy, S. G. (2007). Stargazin attenuates intracellular polyamine block of calcium-permeable AMPA receptors. Nat. Neurosci. 10, 1260–1267. doi: 10.1038/nn1966 Maltsev, A. V., Bal, N. V., and Balaban, P. M. (2019). LTP suppression by protein synthesis inhibitors is NO-dependent. Neuropharmacology 146, 276–288. doi: 10.1016/j.neuropharm.2018.12.009 Van Strien, N. M., Cappaert, N. L. M., and Witter, M. P. (2009). REFERENCES The anatomy of memory: an interactive overview of the parahippocampal- hippocampal network. Nat. Rev. Neurosci. 10, 272–282. doi: 10.1038/nrn2614 Masurkar, A. V., Srinivas, K. V., Brann, D. H., Warren, R., Lowes, D. C., and Siegelbaum, S. A. (2017). Medial and lateral entorhinal cortex differentially excite deep versus superficial CA1 pyramidal neurons. Cell Rep. 18, 148–160. doi: 10.1016/j.celrep.2016.12.012 Von Ossowski, L., Li, L. L., Moykkynen, T., Coleman, S. K., Courtney, M. J., and Keinanen, K. (2017). Cysteine 893 is a target of regulatory thiol modifications of GluA1 AMPA receptors. PLoS One 12:e0171489. doi: 10.1371/journal.pone. 0171489 Mattison, H. A., Bagal, A. A., Mohammadi, M., Pulimood, N. S., Reich, C. G., Alger, B. E., et al. (2014). Evidence of calcium-permeable AMPA receptors in dendritic spines of CA1 pyramidal neurons. J. Neurophysiol. 112, 263–275. doi: 10.1152/jn.00578.2013 Zhang, P., Fu, W. Y., Fu, A. K., and Ip, N. Y. (2015). S-nitrosylation-dependent proteasomal degradation restrains Cdk5 activity to regulate hippocampal synaptic strength. Nat. Commun. 6:8665. doi: 10.1038/ncomms9665 Moult, P. R., Cross, A., Santos, S. D., Carvalho, A. L., Lindsay, Y., Connolly, C. N., et al. (2010). Leptin regulates AMPA receptor trafficking via PTEN inhibition. J. Neurosci. 30, 4088–4101. doi: 10.1523/JNEUROSCI.3614-09.2010 Conﬂict of Interest: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. O’Dell, T. J., Huang, P. L., Dawson, T. M., Dinerman, J. L., Snyder, S. H., Kandel, E. R., et al. (1994). Endothelial NOS and the blockade of LTP by NOS inhibitors in mice lacking neuronal NOS. Science 265, 542–546. Plant, K., Pelkey, K. A., Bortolotto, Z. A., Morita, D., Terashima, A., McBain, C. J., et al. (2006). Transient incorporation of native GluR2-lacking AMPA receptors during hippocampal long-term potentiation. Nat. Neurosci. 9, 602–604. doi: 10.1038/nn1678 Copyright © 2021 Ivanova, Balaban and Bal. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Rozov, A., and Burnashev, N. (1999). Polyamine-dependent facilitation of postsynaptic AMPA receptors counteracts paired-pulse depression. Nature 401, 594–598. REFERENCES doi: 10.1038/44151 June 2021 \| Volume 13 \| Article 656377 Frontiers in Synaptic Neuroscience \| www.frontiersin.org 12
https://openalex.org/W4232744056	https://www.qeios.com/read/QXROMU/pdf	English	null	Merkel Cell Carcinoma pM1a TNM Finding v7	Definitions	2,020	cc-by	87	Qeios · Definition, February 7, 2020 Open Peer Review on Qeios Open Peer Review on Qeios v7 National Cancer Institute National Cancer Institute Merkel Cell Carcinoma pM1a TNM Finding v7 Qeios ID: QXROMU · https://doi.org/10.32388/QXROMU Source National Cancer Institute. Merkel Cell Carcinoma pM1a TNM Finding v7. NCI Thesaurus. Code C88534. National Cancer Institute. Merkel Cell Carcinoma pM1a TNM Finding v7. NCI Thesaurus. Code C88534. Merkel cell carcinoma with metastasis to skin, subcutaneous tissues or distant lymph nodes. (from AJCC 7th Ed.) Qeios ID: QXROMU · https://doi.org/10.32388/QXROMU 1/1
https://openalex.org/W2103874701	https://europepmc.org/articles/pmc3211143?pdf=render	English	null	Assessment of Influence of Magnetic Forces on Aggregation of Zero-valent Iron Nanoparticles	Nanoscale research letters	2,010	cc-by	5,410	Introduction Zero-valent iron nanoparticles (nZVI) composed of iron and its oxides are spherical particles with dia- meter approximately 50 nm and with a large specific surface. These particles are used for the decontamina- tion of groundwater and soil and especially for the decontamination of organic pollutants such as haloge- nated hydrocarbons [1]. Nanoparticles migrate through the soil and can reach the contamination in-situ. Prop- erties of the nZVI and remediation possibilities depend on methods of production [2]. At the Technical Uni- versity of Liberec, experiments with iron nanoparticles TODA produced by the company Toda Kogyo Corp. [3] and with the nanoparticles NANOFER, produced by the company NANO IRON s.r.o. [4], are made. During a remedial intervention, transport of the iron nanoparticles is slowed down due to rapid aggregation of them. The rate of aggregation increases with grow- ing concentration of particles in solution and with growing ionic strength of the solution [5]. For the pre- servation of the transport properties, it is advisable to stabilize the particles. A lot of methods of stabilization were published [6-10]. We simulate the transport of the iron nanoparticles and that is why we examine the interactions among them causing the aggregation. Models of aggregation of small particles were pub- lished in many articles (e.g. [11-13]). They are mostly The extended model of the aggregation of iron nano- particles will be included into a solver of particle trans- port in groundwater. It would allow to simulate the transport of iron nanoparticles and to predict the effi- ciency of the remedial intervention. That could be useful for the proposal of optimal remedial intervention, which would enable to decontaminate an affected area effi- ciently and economically. * Correspondence: dana.rosicka@tul.cz Institute of Novel Technologies and Applied Informatics, Technical University of Liberec, Liberec, Czech Republic. © 2010 Rosická and Šembera. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Abstract Aggregation of zero-valent nanoparticles in groundwater is influenced by several physical phenomena. The article shortly introduces preceding works in modeling of aggregation of small particles including influence of sedimentation, velocity profile of water, heat fluctuations, and surface electric charge. A brief description of inclusion of magnetic forces into the model of aggregation follows. Rate of influence of the magnetic forces on the aggregation depends on the magnitude of magnetization of the particles, radius of nanoparticles, size of the aggregates, and their concentration in the solution. Presented results show that the magnetic forces have significant influence on aggregation especially of the smallest iron particles. based on the publications [14,15]. However, this gener- ally used model is insufficient for our case. A surface charge established on the surface of particles causes repulsive electrostatic forces between them. The influ- ence on the aggregation into the known aggregation model was implemented [16]. The iron particles cor- rode in the water, and this process causes change of the surface charge as well as the change of the rate of aggregation [17]. Because the particles are made from iron, they also have magnetic properties, which signifi- cantly affects the rate of aggregation [2,18-21]. That is why we want to derive a mathematical model of mag- netic forces among particles and to add it into the aggregation model. There is shown a procedure of the derivation of the model in this paper and there is made also an evaluation of the rate of aggregation influenced by magnetic forces here. Assessment of Influence of Magnetic Forces on Aggregation of Zero-valent Iron Nanoparticles Dana Rosická*, Jan Šembera Rosická and Šembera Nanoscale Res Lett 2011, 6:10 http://www.nanoscalereslett.com/content/6/1/10 Rosická and Šembera Nanoscale Res Lett 2011, 6:10 http://www.nanoscalereslett.com/content/6/1/10 Open Access Open Access Aggregation of Colloids and Small Particles in Groundwater The particles in groundwater aggregate easily. They cre- ate clumps of particles up to the size of several μm [20] Rosická and Šembera Nanoscale Res Lett 2011, 6:10 http://www.nanoscalereslett.com/content/6/1/10 Rosická and Šembera Nanoscale Res Lett 2011, 6:10 http://www.nanoscalereslett.com/content/6/1/10 Page 2 of 6 Page 2 of 6 that cohere and decrease the possibility of migration of particles through pores of the ground. The aggregation of particles is proven by experiments described in many articles. In [22], particle size, size distribution and sur- face composition were characterized by transmission electron microscopy (TEM), X-ray diffraction, high- resolution X-ray photoelectron spectroscopy, X-ray absorption near edge structure, and acoustic/electroa- coustic spectrometry. There are presented micrographs of a single particle and aggregates of iron particles in the article. In [3], characterization of iron nanoparticles using TEM according to methods of its preparation is done. In [20], a type of aggregation according to begin- ning concentration of iron nanoparticles is studied by dynamic light scattering, optical microscopy, and sedi- mentation measurements. in the middle of the pore the velocity of flowing is lar- gest. Since the particles have different velocities, accord- ing to the place where they are in the pore, the particles can meet each other and create the aggregate. Following [11], the mass transport coefficient for the velocity gra- dients of particles is ij i j G d d 2 3 1 6 = + ( ) , (4) (4) where G is the average velocity gradient in a pore. In the case of the small particles, the heat fluctuation of particles has a significant effect on the particle aggre- gation. Brownian diffusion causes a random movement of the particles and again it facilitates the aggregation. Following [11], the mass transport coefficient for the Brownian diffusion is The aggregation of the particles is caused by many processes that generally occur during the particle migra- tion. The decrease in the mobility can be formulated by a rate of aggregation that is given by the mass transport coefficients b [m3 s-1]. It was published in many papers (e.g. [11,15]). The coefficients give a probability Pij of creation of aggregate from particle i and particle j together with concentrations ni, nj of particles i and par- ticles j (1). Particle i means the aggregate created from i elementary nanoparticles. Aggregation of Colloids and Small Particles in Groundwater   ij B i j i j k T d d d d 1 2 2 3 = + ( ) , (5) (5) where kB stands for Boltzmann constant and T denotes absolute temperature. A statistical assessment of the importance of the parti- cular processes to the creation of the aggregates was done. The Table 1 shows that for the smallest particles the Brownian diffusion is most considerable. The sedi- mentation is most significant when the difference between sizes of the aggregates is largest. It is a conse- quence of the fact that the difference between the velo- cities of the particles is largest. The velocity gradients depend on the pore size. When the size of pores is small, the aggregation is most influenced by the velocity gradient, if the difference between the particles is large. P n n ij ij i j =  , (1)     ij ij ij ij = + + 3 2 1. (2) P n n ij ij i j =  , (1) (2) The ij 3 is the mass transport coefficient of aggrega- tion caused by the gravity, ij 2 is the mass transport coefficient for the velocity gradients, and ij 1 stands for the mass transport coefficient of the heat fluctuations. The notation is adopted from [11]. The mass transport coefficient for the velocity gradi- ents is quantified for the case with a small size of pores The first process that causes the aggregation of parti- cles is sedimentation. Due to the gravitation forces, the particles fall. According to their size, the velocity of the sedimentation is different for the different aggregates. It implies that the particles have a chance to meet the others and aggregate due to the attachment forces. Aggregation of Colloids and Small Particles in Groundwater Fol- lowing [11], the mass transport coefficient for sedimen- tation is i- he he It he l- n- 3) d- ng he d Table 1 The mass transport coefficients for Brownian diffusion, velocity gradients, and sedimentation, for different sizes of aggregates i j ij 1 ij 2 ij 3 1 1 1.0 × 10-17 2.2 × 10-20 0 1 10 1.3 × 10-17 8.8 × 10-20 5.9 × 10-22 1 102 1.9 × 10-17 5.0 × 10-19 1.0 × 10-20 1 103 3.3 × 10-17 3.7 × 10-18 2.0 × 10-19 1 104 6.5 × 10-17 3.2 × 10-17 3.8 × 10-18 1 105 1.3 × 10-16 3.0 × 10-16 7.9 × 10-17 1 106 2.8 × 10-16 3.0 × 10-15 1.7 × 10-15 1 107 6.0 × 10-16 2.8 × 10-14 3.5 × 10-14 10 10 1.1 × 10-17 2.2 × 10-19 0 102 102 1.3 × 10-17 8.8 × 10-19 1.2 × 10-20 103 103 1.1 × 10-17 2.2 × 10-17 5.9 × 10-18 104 104 1.3 × 10-17 8.8 × 10-17 0 Electrostatic Forces Among Iron Nanoparticles Values of mass transport coefficients for aggregates with the sizes between 50 nm and 5 μm are com- puted. The surface charge depends on ζ potential (see e.g. [24]). ζ potential depends on pH of the water. The measured results of this dependence acquired using the Malvern ZetaSizer are shown in the Figure 1. Zero-valent iron provides alkaline reaction in water, so the measurement was done for higher pH values only. g A mathematical model of aggregation for the case of iron particles was compiled. The sedimentation, velo- city gradients, and Brownian diffusion are not suffi- cient for the description of the process aggregation. The surface of iron particles oxidizes. The ions on the surface attract ions from the electrolyte and the elec- tric double layer arises. Therefore, the influence of the electrostatic forces was added in the mass transport coefficients. The detailed derivation of it is published in [16,23]. l The statistical assessment of the importance of the electrostatic forces to the creation of the aggregates was done in [16,23]. The results are that the mass transport coefficient for Brownian diffusion is limited by the elec- trostatic forces mostly for large aggregates. The mass transport coefficient for the velocity gradients is not lim- ited by the surface charge for measured ζ potential. The ζ potential would have to be minimally ten times larger to affect the rate of aggregation. The mass transport coefficient for sedimentation is limited by the surface charge mostly for the small particles. For sedimentation, the mass transport coefficient ij el 3 has the following form:      ij el ij i j i j i j d d d d 3 3 2 2 0 12 1 1 = − − , (6) (6) where si and sj stand for surface charge on particle i and particle j, respectively, ε0 is permittivity of the med- ium. If the term that reduces the mass transport coeffi- cient is greater than the mass transport coefficient without the influence of electrostatic forces ij 3 , the probability of collision of particles i and j should be equal to zero. That is why Figure 1 Dependence of ζ potential of nZVI on pH measured with Malvern ZetaSizer.   ij el ij el 3 3 0 = max( , ). Table 1 The mass transport coefficients for Brownian diffusion, velocity gradients, and sedimentation, for different sizes of aggregates Table 1 The mass transport coefficients for Brownian diffusion, velocity gradients, and sedimentation, for different sizes of aggregates    ij p i j i j g d d d d 3 2 2 2 72 = − + − ( )( ) \| \|,   (3) (3) where g is gravity acceleration, h is viscosity of med- ium, ϱ is density of medium, ϱp is density of aggregating particles, di is diameter of particle i. where g is gravity acceleration, h is viscosity of med- ium, ϱ is density of medium, ϱp is density of aggregating particles, di is diameter of particle i. The process that works similarly is water drifting. The flowing water in a pore in soil has a velocity profile and Rosická and Šembera Nanoscale Res Lett 2011, 6:10 http://www.nanoscalereslett.com/content/6/1/10 Page 3 of 6 Page 3 of 6 particles. The total probability of the aggregation of a particle i with a particle j is then particles. The total probability of the aggregation of a particle i with a particle j is then and a large flux (flux = 3.67 × 10-4 ms-1, porosity = 0.39, velocity gradient G = 50 s-1). In the other cases, the mass transport coefficient would be much smaller than the others. P n n ij ij el ij el ij el i j = + + ( ) .       3 2 1 (12) (12) Magnetic Forces Between Two Spherical Iron The scalar potential of the magnetic field around one homogeneous spherical iron nanoparticle with radius a located at the point [0, 0, 0] was determined: According to [25], the electromagnetic potential in the point r near a permanent magnet is equal to      ( ) ( ( )) ( ) ( ) r = −′ ′ + + −′ ∫∫∫ M x r r x x x r d a 3 2 1 2 2 2 3 2 2 2 0 0 0 2 3 cos sin ′r d d , (16) (16) ( ) , r MR = ∫R dV V 3 (13) (13) where a is the radius of the nanoparticle and [x1, x2, where a is the radius of the nanoparticle and [x1, x2, x3] are the coordinates of the point r. The direction of the vector of polarization M is set to the direction x3, M is the magnitude of the vector M. where a is the radius of the nanoparticle and [x1, x2, x3] are the coordinates of the point r. The direction of the vector of polarization M is set to the direction x3, M is the magnitude of the vector M. where the vector M is the vector of magnetic polariza- tion at the point dV, the vector R is the difference between the source of magnetic field dV and the point r, R is the length of R. After integration, the magnetic potential around a fer- romagnetic sphere is obtained:   ( ) arctan r = − + + − ⎛ ⎝ ⎜ ⎜ ⎞ ⎠ ⎟ ⎟ + + − ⎛ ⎝ ⎜ 4 3 1 2 2 2 3 2 2 1 2 2 2 3 2 2 M x a a x x x a x x x a ⎜ ⎞ ⎠ ⎟ ⎟ + + − x x x a 1 2 2 2 3 2 2 . (17) Figure 2 Hysteresis loop of the zero-valent iron nanoparticles for temperature of 300 K measured with magnetometer MPMS XL by Jiří Tuček at the Palacký University Olomouc. Electrostatic Forces Among Iron Nanoparticles (7) (7) For the velocity gradients, the mass transport coeffi- cient ij el 2 is equal to      ij el ij i j i j i j d d d d 2 2 2 2 0 12 1 1 = − + , (8) (8)   ij el ij el 2 2 0 = max( , ). (9) (9) And finally, the Brownian diffusion gives the mass transport coefficient ij el 1 :      ij el ij i j i j i j d d d d 1 1 2 2 0 3 = − + ( ) , (10) (10)   ij el ij el 1 1 0 = max( , ). (11) (11) Figure 1 Dependence of ζ potential of nZVI on pH measured with Malvern ZetaSizer. Figure 1 Dependence of ζ potential of nZVI on pH measured with Malvern ZetaSizer. The probability of particle collision decreases quadra- tically with the quantum of the surface charge of Rosická and Šembera Nanoscale Res Lett 2011, 6:10 http://www.nanoscalereslett.com/content/6/1/10 Page 4 of 6 Magnetic Field of Iron Particles Intensity of the magnetic field H can be subsequently computed as The iron particles NANOFER that are used for the reme- dial intervention were measured by magnetometer MPMS XL, an equipment based on the SQUID effect (Supercon- ducting quantum interference device), owned by the Palacký University Olomouc, Czech Republic. The iron particles are ferromagnetic, a hysteresis loop of the iron particles measured by SQUID is in Figure 2. That is the reason why it is necessary to include the influence of the magnetic forces among particles to the model of aggrega- tion of the particles. However, it is a very complicated pro- cess which cannot be described analytically. That is why only the description of the magnetic forces between two particles is shown in this paper. Although it is not a suffi- cient model, it can be used for determination of something what we could call “effective range” of the magnetic forces. Also the limits of the sizes of the magnetic forces can be determined. It can be used for the assessment of the aggre- gation of the particles depending on their concentration. H r grad r ( ) ( ( )). = −  (14) (14) Finally, the magnetic force between the source of the intensity of magnetic field H and a permanent magnet with the vector of polarization M0 at the point r is equal to F M grad H ( ) ( ) ( ) . r r dV V = − ⋅ ∫ 0 (15) (15) Magnetic Forces Between Two Spherical Iron Figu for According to (14), the components of the vector of intensity of the magnetic field around a spherical ferro- magnetic particle is H C a a C C i i ( ) ( ) arctan ( ) ( ) r r r r = − − ⎛ ⎝ ⎜ ⎞ ⎠ ⎟ ⎛ ⎝ ⎜⎜ ⎞ ⎠ ⎟⎟ ⎡ ⎣ ⎢ ⎢ ⎤ ⎦ ⎥ ⎥ −    3 4 4 x C ax x C a C x x C a i i i 3 3 3 4 ( ) ( ) arctan ( ) ( ) r r r r r r ⋅ − ⎛ ⎝ ⎜ ⎞ ⎠ ⎟ ⎛ ⎝ ⎜⎜ ⎞ ⎠ ⎟⎟ +  − ⎛ ⎝ ⎜ ⎞ ⎠ ⎟ ⎛ ⎝ ⎜⎜ ⎞ ⎠ ⎟⎟ arctan ( ) ( ) , a C C r r (18) (18) where δi3 stands for Kronecker delta and i = 1, 2, 3. The symbol C(r) replaces Figure 2 Hysteresis loop of the zero-valent iron nanoparticles for temperature of 300 K measured with magnetometer MPMS XL by Jiří Tuček at the Palacký University Olomouc. C x x x a ( ) . r = + + − 1 2 2 2 3 2 2 (19) (19) Rosická and Šembera Nanoscale Res Lett 2011, 6:10 http://www.nanoscalereslett.com/content/6/1/10 Rosická and Šembera Nanoscale Res Lett 2011, 6:10 http://www.nanoscalereslett.com/content/6/1/10 Rosická and Šembera Nanoscale Res Lett 2011, 6:10 http://www.nanoscalereslett.com/content/6/1/10 Page 5 of 6 where i is potential of magnetic field of the nanopar- ticle located in the point ri. where i is potential of magnetic field of the nanopar- ticle located in the point ri. The derived formula of the size of the magnetic forces between two iron nanoparticles is very extensive; hence, it is not presented here. Though, an example of the numerical result is shown. In Figure 3, there is the visualisation of a part of the vector field of the magnetic forces between two nanoparticles. First nanoparticle is in an arbitrary point near second nanoparticle with radius a which is touching the center of the upper right side of the figure. The figure is created by the software Mathematica 5, copyrighted by Wolfram Research, Inc. The influence of magnetic forces in comparison with the gravitational forces is investigated. Magnetic Field Around an Aggregate The aggregate of iron nanoparticles is in fact a clump of many permanent magnets. It is impossible to establish an analytical model of interaction of two such aggregates. To analyze statistically the influence of the magnetic forces on aggregation of two nanoparticle aggregates, a script for the examination of the worst possibility (the largest forces) and the averaged possibility of influencing the aggregation by the magnetic forces was written down. In Figure 4, the dashed line characterizes the effective ranges for interaction of chosen aggregates interacting with a single nanoparticle. The solid line in the graph characterizes the effective range for the interaction of two aggregates of the same size. The absolute value of the magnetic force and consequently also effective range quadratically depends on the magnitude of the magnetic polarization. The graph is plotted for the magnetic polarization 170 emu g-1. More important information than absolute values is the trends of the lines. y gg g To analyze statistically the influence of the magnetic forces on aggregation of two nanoparticle aggregates, a script for the examination of the worst possibility (the largest forces) and the averaged possibility of influencing the aggregation by the magnetic forces was written down. The statistical model of the aggregate is made so that the volume of the aggregate is filled by uniformly distrib- uted nanoparticles (small homogeneous magnets) with randomly uniformly distributed direction of the magnetic polarization. The magnitude of the magnetic polarization of all nanoparticles is the same. The magnetic potential of the aggregate is then the superposition of magnetic potentials of all nanoparticles creating the aggregate: According to the results from the Table 1 in the Sec- tion 2, sedimentation influences the aggregation mostly when the difference between sizes of the two aggregates is large. Consequently, the dashed line in the Figure 4 gives a good information about the influence of mag- netic forces on aggregation of aggregates of different sizes. The solid line comparing the influence of mag- netic and gravitational forces between two similar aggre- gates does not include the real information about the influence of magnetic forces because another force than gravitational governs the aggregation process in such a case. Magnetic Forces Between Two Spherical Iron It could be com- pared also with other affecting forces but the gravitation force was chosen for the reason of small number of variables. If one aggregate is in a fixed position and another one is located somewhere vertically under it, there should be a unique distance (“effective range”) between them so that if they are closer than it, the lower aggregate would attach to higher one by magnetic forces. If they are more distant, they would sediment separately. The distance is measured between the sur- faces of particles. Magnetic Field Around an Aggregate   ( ) ( ), r r r = − ∑ i i i (20) (20) Figure 3 Visualization of the vector field of the magnetic forces between two spherical particles of nZVI, using software Mathematica 5, copyrighted by Wolfram Research, Inc. One nanoparticle is in an arbitrary point near a nanoparticle with radius a which is touching the center of the upper right side of the figure. Figure 4 Effective range of the magnetic forces of chosen aggregates. The dashed line characterizes the effective ranges for the interaction of aggregates interacting with a single nanoparticle. The solid line characterizes the effective ranges for the interaction of two aggregates of the same size. The graph is plotted for the magnetic polarization 170 emu g-1. Figure 4 Effective range of the magnetic forces of chosen aggregates. The dashed line characterizes the effective ranges for the interaction of aggregates interacting with a single nanoparticle. The solid line characterizes the effective ranges for the interaction of two aggregates of the same size. The graph is plotted for the magnetic polarization 170 emu g-1. Figure 4 Effective range of the magnetic forces of chosen aggregates. The dashed line characterizes the effective ranges for the interaction of aggregates interacting with a single nanoparticle. The solid line characterizes the effective ranges for the interaction of two aggregates of the same size. The graph is plotted for the magnetic polarization 170 emu g-1. Figure 4 Effective range of the magnetic forces of chosen Figure 3 Visualization of the vector field of the magnetic forces between two spherical particles of nZVI, using software Mathematica 5, copyrighted by Wolfram Research, Inc. One nanoparticle is in an arbitrary point near a nanoparticle with radius a which is touching the center of the upper right side of the figure. Figure 3 Visualization of the vector field of the magnetic forces between two spherical particles of nZVI, using software Mathematica 5, copyrighted by Wolfram Research, Inc. One nanoparticle is in an arbitrary point near a nanoparticle with radius a which is touching the center of the upper right side of the figure. Rosická and Šembera Nanoscale Res Lett 2011, 6:10 http://www.nanoscalereslett.com/content/6/1/10 Page 6 of 6 Rosická and Šembera Nanoscale Res Lett 2011, 6:10 http://www.nanoscalereslett.com/content/6/1/10 Page 6 of 6 9. Kanel SR, Manning B, Charlet L, Choi H: Removal of Arsenic(III) from groundwater by nano scale zero-alent iron. Acknowledgements Thi lt i li d 21. Zhang LY, Wang J, Wei LM, Liu P, Wei H, Zhang YF: Synthesis of Ni nanowires via a hydrazine reduction route in aqueous ethanol solutions assisted by external magnetic fields. Nano-Micro Lett 2009, 1:49-52. This result is realized under the state subsidy of the Czech Republic within the research and development project “Advanced Remediation Technologies and Processes Center” 1M0554—Program of Research Centers PP2-DP01 supported by Ministry of Education and within the research project FR-TI1/ 456 “Development and implementation of the tools additively modulating soil and water bioremediation”—Program MPO-TIP supported by Ministry of Industry and Trade. Kind thanks to Jiří Tuček from the Palacký University Olomouc for granting of Figure 2. 22. Sun Y-P, Cao J, Zhang W-X, Wang HP: Characterization of zero-valent iron nanoparticles. Adv Colloid Interface Sci 2006, 120:47-56. 23. Pelikánová D: Nanoparticle Aggregation Model [in Czech], Diploma thesis, Technical University of Liberec. 2008. 23. Pelikánová D: Nanoparticle Aggregation Model [in Czech], Diploma thesis, Technical University of Liberec. 2008. 24. Stumm W, Morgan JJ: Aquatic Chemistry. A Wiley-Interscience Publication, New York; 1996. 24. Stumm W, Morgan JJ: Aquatic Chemistry. A Wiley-Interscience Publication, New York; 1996. 25. Votrubík V: Theory of the electromagnetic field [in Czech]. Czechoslovak Academy of Science Publication, Praha; 1958. Received: 21 June 2010 Accepted: 10 August 2010 Published: 24 August 2010 Received: 21 June 2010 Accepted: 10 August 2010 Published: 24 August 2010 doi:10.1007/s11671-010-9753-4 Cite this article as: Rosická and Šembera: Assessment of Influence of Magnetic Forces on Aggregation of Zero-valent Iron Nanoparticles. Nanoscale Res Lett 2011 6:10. Magnetic Field Around an Aggregate Environ Sci Technol 2005, 39(5):1291-1298. On the basis of this result, it is obvious that the mag- netic forces among particles have significant effect on the rate of aggregation of the particles. On the basis of the effective range, concentration of particles uniformly dispersed in solution can be computed. According to distances between particles, the concentration would have to be very small (about 15 mg l-1) to be possible to neglect the influence of the magnetic forces. 10. Song H, Carraway ER: Reduction of chlorinated methanes by nano-sized zero-valent iron. Kinetics, pathways, and effect of reaction conditions. Environ Eng Sci 2006, 23(2). 11. Buffle J, Van Leeuweh H, eds.: Environmental Particles, Lewis publishers. 1993, 2:353-360. 12. Somasundaran P, Runkana V: Modeling flocculation of colloidal mineral suspensions using population balances. Int J Miner Process 2003, 72:33-55. 13. Garrick S, Zachariah M, Lehtinen K: Modeling and simulation of nanoparticle coagulation in a high reynolds number incompressible flows. Proceeding of the National Conference of the Combustion Institute Oakland; 2001, 25-27. Conclusion The influence of the magnetic forces among iron parti- cles on the rate of aggregation in terms of the “effective range” was assessed. The effective range is a distance in which the magnetic forces outweigh the gravitation force that causes the aggregation. To assess the mag- netic field around more interacting aggregates, it should be made a model using the Finite Element Method (FEM) or another numerical method. The next steps in the studying of nanoparticle aggregation due to mag- netic forces can be the evaluation of the effective range in comparison with other forces indicated in this paper and/or building a FEM model of nZVI aggregate. 14. Thomas B, Camp R: Velocity gradients and internal work in fluid motion. J Boston Soc Civil Eng 1943, 30:4. 15. Smoluchowski MV: Test of a mathematical theory of coagulation kinetics of colloid solutions [in German]. Zeitschrift f physik Chemie 1916, XCII:129-168. 16. Rosická D, Šembera J: Mathematical model of aggregation of nanoscale particles with surface charge. 2009, submitted. 16. Rosická D, Šembera J: Mathematical model of aggregation of nanoscale particles with surface charge. 2009, submitted. 17. Reardon EJ, Fagan R, Vogan JL, Przepiora A: Anaerobic corrosion reaction kinetics of nanosized iron. Environ Sci Technol 2008, 42(7). 17. Reardon EJ, Fagan R, Vogan JL, Przepiora A: Anaerobic corrosion reaction kinetics of nanosized iron. Environ Sci Technol 2008, 42(7). 18. Horák D, Petrovský E, Kapicka A, Frederichs T: Synthesis and characterization of magnetic Poly(Glycidyl Methacrylate) microspheres. J Magnet Magn Mater 2007, 500-506. 18. Horák D, Petrovský E, Kapicka A, Frederichs T: Synthesis and characterization of magnetic Poly(Glycidyl Methacrylate) microspheres. J Magnet Magn Mater 2007, 500-506. 19. Masheva V, Grigorova M, Nihtianova D, Schmidt JE, Mikhov M: Magnetization processes of small γ-Fe2O3 particles in non-magnetic matrix. Phys D Appl Phys 1999, 32:1595-1599. y y 20. Phenrat T, Saleh N, Sirk K, Tilton RD, Lowry GV: Aggregation and sedimentation of aqueous nanoscale zerovalent iron dispersions. Environ Sci Technol 2007, 41(1):284-290. References 1. Zhang W-X: Nanoscale iron particles for environmental remediation: an overview. J Nanopart Res 2003, 5:323-332. 1. Zhang W-X: Nanoscale iron particles for environmental remediation: an overview. J Nanopart Res 2003, 5:323-332. 2. Li L, Fan M, Brown CR, Van Leeuwen JH, Wang J, Wang W, Song Y, Zhang P: Synthesis, properties, and environmental applications of nanoscale iron-based materials: a review. Crit Rev Env Sci Technol 2006, 36:405-431. 2. Li L, Fan M, Brown CR, Van Leeuwen JH, Wang J, Wang W, Song Y, Zhang P: Synthesis, properties, and environmental applications of nanoscale iron-based materials: a review. Crit Rev Env Sci Technol 2006, 36:405-431. 3. Nurmi JT, Tratnyek PG, Sarathy V, Baer DR, Amonette JE, Pecher K, Wang C, Linehan JC, Matson DW, Penn RL, Driessen MD: Characterization and properties of metallic iron nanoparticles: spectroscopy, electrochemistry and kinetics. Environ Sci Technol 2005, 39(5):1221-1230. 4. Filip J, Zboril R, Schneeweiss O, Zeman J, Cernik M, Kvapil P, Otyepka M: Environmental applications of chemically pure natural ferrihydrite. Environ Sci Technol 2007, 41:4367-4374. Submit your manuscript to a journal and beneﬁ t from: 7 Convenient online submission 7 Rigorous peer review 7 Immediate publication on acceptance 7 Open access: articles freely available online 7 High visibility within the ﬁ eld 7 Retaining the copyright to your article Submit your next manuscript at 7 springeropen.com Submit your manuscript to a journal and beneﬁ t from: 7 Convenient online submission 7 Rigorous peer review 7 Immediate publication on acceptance 7 Open access: articles freely available online 7 High visibility within the ﬁ eld 7 Retaining the copyright to your article Submit your next manuscript at 7 springeropen.com Submit your manuscript to a journal and beneﬁ t from: Submit your manuscript to a journal and beneﬁ t from: 5. Saleh N, Kim H-J, Phenrat T, Matyjaszewski K, Tilton RD, Lowry GV: Ionic strength and composition affect the mobility of surface-modified FeO nanoparticles in water-saturated sand columns. Environ Sci Technol 2008, 42(9):3349-3355. 6. Johnson RL, Johnson GO, Nurmi JT, Tratnyek PG: Natural organic matter enhanced mobility of nano zerovalent. Environ Sci Technol 2009, 43(14):5455-60. 6. Johnson RL, Johnson GO, Nurmi JT, Tratnyek PG: Natural organic matter enhanced mobility of nano zerovalent. Environ Sci Technol 2009, 43(14):5455-60. 7. Kanel SR, Greneche J-M, Choi H: Arsenic (V) removal from groundwater using nanoscale zero-valent iron as a colloidal reactive barrier material. Environ Sci Technol 2006, 40(6):2045-2050. 7. Kanel SR, Greneche J-M, Choi H: Arsenic (V) removal from groundwater using nanoscale zero-valent iron as a colloidal reactive barrier material. Environ Sci Technol 2006, 40(6):2045-2050. 8. Tiraferri A, Chen KL, Sethi R, Elimelech M: Reduced aggregation and sedimentation of zero-valent iron nanoparticles in the presence of guar gum. J Colloid Interface Sci 2008, 324:71-79.
https://openalex.org/W1580439447	https://jrnl.nau.edu.ua/index.php/ZI/article/download/3358/3391	Russian	null	Analysis of vulnerabilities of protocols of authentification of WEB	Zahist ìnformacìï	2,012	cc-by	2,978	Коломыцев М.В., Носок С.А. Коломыцев М.В., Носок С.А. НАУКОВО-ПРАКТИЧНИЙ ЖУРНАЛ «ЗАХИСТ ІНФОРМАЦІЇ» № 3, 2012 НАУКОВО-ПРАКТИЧНИЙ ЖУРНАЛ «ЗАХИСТ ІНФОРМАЦІЇ» № 3, 2012 УДК 004.056.53 УДК 004.056.53 УДК 004.056.53 АНАЛИЗ УЯЗВИМОСТЕЙ ПРОТОКОЛОВ АУТЕНТИФИКАЦИИ WEB Современная тенденция создания WEB-приложений состоит в персонализации контента и управляемом доступе к предоставляемым сервисам. В связи с этим, повышаются требования к аутентификации клиентов WEB-сайтов. Существующие подходы к построению механизма аутентификации зачастую уязвимы к атакам. В статье рассматриваются различные ограничения и меры, обеспечение которых позволит снизить риск нарушений безопасности. ру Ключевые слова: WEB-приложения, протокол, ресурс. Ключевые слова: WEB-приложения, протокол, ресурс. Современная тенденция создания WEB-приложений состоит в персонализации контента и управляемом доступе к предоставляемым сервисам. В связи с этим, повышаются требования к аутентификации клиентов WEB-сайтов. Существующие подходы к построению механизма аутентификации зачастую уязвимы к атакам. В статье рассматриваются различные ограничения присущие используемым процедурам аутентификации и меры, обеспечение которых позволит снизить риск нарушений безопасности. р р ру Актуальность задачи безопасности систем аутентификации Консорциум Web Application Security Consortium (http://www.webappsec.org) периодически представляет статистику уязвимостей и угроз безопасности Web-приложений. В предлагаемой ими классификации угроз отдельный класс угроз связан с атаками на используемые методы аутентификации [1]. Связаны они с особенностями используемых в WEB схем аутентификации и их потенциальными ограничениями. Ограничивающими факторами при выборе схемы аутентификации являются: - ограничения на сложность реализации. Используемые технологии должны поддерживаться на разных платформах и не создавать существенной нагрузки на сервер. Поэтому технологии, реализующие сложные вычисления на стороне клиента (с использованием например, Javascript, Java, ActiveX и FlashВ) используются не часто. Наиболее распространенной формой обмена аутентифицирующей информацией в последовательности HTTP запросов являются cookie, - восприятие пользователями. Форма диалога должна быть максимально простой, и исключать необходимость установки дополнительных компонент на компьютере клиента, либо организации процедуры аутентификации в виде длинной последовательности диалоговых окон, - восприятие пользователями. Форма диалога должна быть максимально простой, и исключать необходимость установки дополнительных компонент на компьютере клиента, либо организации процедуры аутентификации в виде длинной последовательности диалоговых окон, - производительность. Защищенные протоколы аутентификации, криптографические преобразования вообще, существенно снижают производительность сервера. Это в частности, касается протокола SSL. Уязвимости протоколов аутентификации в WEB Уязвимости протоколов аутентификации в WEB Базовая аутентификация Базовая аутентификация Это простой протокол проверки подлинности, поддерживаемый всеми браузерами [3]. Его еще иногда называют HTTP аутентификацией. Протокол работает следующим образом: - ресурсы, для доступа к которым необходимо пройти аутентификации, помещаются в отдельный каталог, и создается файл конфигурации, описывающий местоположение базы учетных записей и набор ограничений, - ресурсы, для доступа к которым необходимо пройти аутентификации, помещаются в отдельный каталог, и создается файл конфигурации, описывающий местоположение базы учетных записей и набор ограничений, - браузер отправляет HTTP запрос на доступ к защищенному ресурсу, который выглядит так: GET / / HTTP/1 0 - браузер отправляет HTTP запрос на доступ к защищенному ресурсу, который выглядит так: GET /test/secure HTTP/1 0 GET /test/secure HTTP/1.0, - серверный HTTP процесс определяет, что необходим доступ к защищенному ресурсу и считывает из файла конфигурации параметры (имя домена). Выяснив, что пользовательский запрос не содержит аутентифицируещей информации, он его отвергает, сообщая клиенту, что требуется аутентификация с помощью пароля: HTTP/1.1 401 Unauthorized WWW-Authenticate: Basic realm="luxor", 41 НАУКОВО-ПРАКТИЧНИЙ ЖУРНАЛ «ЗАХИСТ ІНФОРМАЦІЇ» № 3, 2012 - браузер, получив отказ, проверяет, нет ли в кеше пароля и логина для указанной области данного сервера. Если не находит, то выводит диалоговое окно для ввода имени и пароля. Введенные пользователем данные сохраняются для дальнейшего использования, - браузер повторяет запрос к серверу, но уже с информацией о имени и пароле пользователя. Пароль пользователя передается в зашифрованном (Base64) виде: - браузер повторяет запрос к серверу, но уже с информацией о имени и пароле пользователя. Пароль пользователя передается в зашифрованном (Base64) виде: Authorization: Basic dGVzdDp0ZXN0, Authorization: Basic dGVzdDp0ZXN0, - серверный HTTP процесс сверяет полученную информацию с той, которая хранится в базе учетных записей. Если учетная запись с таким именем не существует, или переданный пароль не совпадает с тем, который хранится в базе, сервер отвергает запрос. Получив отказ, браузер повторно выводить окно для ввода имени или пароля, - если все проверки оказались успешными, сервер пересылает браузеру нужную HTTP страницу. - если все проверки оказались успешными, сервер пересылает браузеру нужную HTTP страницу. Теоретически, для доступа к каждой защищенной страничке необходимо указать имя и пароль. На практике, браузер пытается получить доступ к защищаемым страничкам, используя имя и пароль, хранимые в кеше браузера. у р р р у р Недостаток протокола– пароль шифруется, но очень слабым алгоритмом (Base64). Имя пользователя и пароль пересылаются в заголовке Authorization и могут быть перехвачены. Зашифрованный пароль несложно расшифровать, например, с помощью следующей программы на Perl: #!/usr/bin/perl # bd64.pl d d f b # decode from base 64 use MIME::Base64; print decode_base64($ARGV[0]); Программу можно выполнить, указав перехваченный пароль как параметр: C:\bd64.pl dGVzdDp0ZXN0 test:test . Другим, важным с точки зрения безопасности, обстоятельством является то, что установленное доверительное соединение может быть разорвано только путем закрытия браузера. Даже если пользователь переключил браузер на сайт, не требующий аутентификации, имя и пароль хранятся в памяти. Таким образом, оставленный без присмотра компьютер может служить средством, для беспрепятственного доступа других пользователей к защищенным ресурсам. Аутентификация на основе хеша Аутентификация на основе хеша Данный метод аутентификации был добавлен в протокол HTTP, для исправления основных недостатков базовой аутентификации. Во многом этот метод работает так же, как и базовая аутентификация [4]. При попытке доступа браузера к защищенному ресурсу, запрос отвергается. Но в сообщении сервера указывается, что требуется аутентификация на основе хеша и содержится значение (nonce), индивидуальное для каждого запроса. Это значение может формироваться, например, по значению текущего времени и ip-адреса клиента. По умолчанию, при вычислении хеша используется алгоритм MD5. Затем: - браузер получает пароль пользователя (из диалогового окна или из памяти); - браузер получает пароль пользователя (из диалогового окна или из пам - объединяет в одну строку имя пользователя, имя области аутентификации и пароль, затем вычисляет хеш полученного значения (хешА); - объединяет в одну строку запрашиваемый URL с названием метода, используемого в запросе (PUT, GET, …) и вычисляет хеш полученной строки (хешВ); - выполняет конкатенацию хешА и хешВ со значением nonce и вычисляет результирующий хеш; р у ру - полученное значение хеша отправляется серверу; - полученное значение хеша отправляется серверу; - когда сервер получает клиентский запрос, он извлекает из хранилища пароль пользователя, на его основе повторяет вычисления контрольных значений хеша. 42 Аутентификация на основе форм Данный протокол аутентификации пожалуй, чаще всего используется в Web- приложениях. Хотя стандартной реализации данного метода нет, его преимуществом является возможность гибкого формирования алгоритма реализации. Разработчики сами определяют детали его функционирования. В этом протоколе для ввода имени и пароля пользователя используются HTML формы. Форма – это основной способ ввода информации на стороне клиента, для отправки ее серверу (точнее, Web-приложению) с помощью тегов FORM и INPUT. Например, для ASP.NET способ аутентификации задается в файле web.config строками р ) , для ASP.NET способ аутентификации задается в файле web.config строками НАУКОВО-ПРАКТИЧНИЙ ЖУРНАЛ «ЗАХИСТ ІНФОРМАЦІЇ» № 3, 2012 Возможна реализация алгоритма, когда хранится не пароль пользователя, а вычисленное значение хешА. Процедура аутентификации считается успешной, если результат вычислений совпадают с тем, что получено от клиента. Протокол аутентификации на основе хеша является стойким к перехвату пароля, однако, его нельзя считать полностью защищенным. Он уязвим к атакам повтора (replay) запроса. Если трафик не шифруется с помощью SSL или HTTPS, то атакующий может просматривать все запросы в текстовом виде. Он может сохранить текст запроса целиком и позднее повторить его. Поскольку оригинальный запрос браузера находится в хеше, то атакующий может получить доступ только к определенным ресурсам. Если запрашиваемым ресурсом является статическая HTML страница, то он получит ее копию (которую, впрочем, он и так мог получить, просто перехватывая трафик между клиентом и сервером). Если же HTML страницы формируются динамически, то ресурсом является сценарий сервера, и последствия такой атаки могут быть гораздо серьезнее. Решением этой проблемы может быть хранение отправленного клиенту значения nonce и запрет его повторного использования. Однако такой вариант является достаточно затратным для ресурсов сервера. Более простой ( и менее надежный) способ – формирование nonce на основе информации, которую трудно подделать: ip-адреса клиента, текущей даты и временной метки. HTTP сервер примет запрос на аутентификацию, только если он поступил из того же ip-адреса и не является слишком старым. Другая уязвимость протокола состоит в том, что сервер ограничен в возможности защищенного хранения паролей. Поскольку для вычисления хеша используется пароль, сервер должен иметь доступ к паролю в открытом виде, либо он должен хранить вычисленное значение хешА. Таким образом, данный протокол нельзя использовать, если пароли защищены с помощью необратимого шифрования. В этом смысле данный протокол более уязвим, чем протокол базовой аутентификации. Если неавторизованный пользователь получает доступ к базе паролей, он получает доступ к хешу, сформированному на основе имени пользователя и пароля. В этом случае, ему нет необходимости знать пароль, он может воспользоваться значением хешА для атаки данного сайта. Таким образом, безопасность базы паролей при использовании данного протокола является важнейшей задачей. Аутентификация на основе форм <authentication mode="Forms"> </authentication> . <authentication mode="Forms"> </authentication> . Этот файл определяет защищаемые ресурсы, а так же содержит имена пользователей и пароли (или их хешы). Имена пользователей и пароли могут храниться и в специально организованном хранилище, например, базе данных. ме того, должна присутствовать форма для ввода логина и пароля (login.aspx). цедура аутентификации происходит следующим образом: Кроме того, должна присутствовать форма для ввода логина и пароля (login.aspx) Процедура аутентификации происходит следующим образом: - пользователь отправляет запрос на доступ к ресурсу (странице default.aspx) GET /default.aspx HTTP/1.0 - сервер перенаправляет запрос к форме для ввода пароля HTTP/1.1 302 Found Location: /login aspx?ReturnUrl=%2fdefault aspx Location: /login.aspx?ReturnUrl=%2fdefault.aspx , 43 НАУКОВО-ПРАКТИЧНИЙ ЖУРНАЛ «ЗАХИСТ ІНФОРМАЦІЇ» № 3, 2012 - в браузере пользователя отображается форма login.aspx . Она содержит поля для ввода имени пользователя и его пароля. После того, как пользователь укажет имя, пароль и нажмет кнопку «Ок», введенная информация будет отправлена серверу: - в браузере пользователя отображается форма login.aspx . Она содержит поля для ввода имени пользователя и его пароля. После того, как пользователь укажет имя, пароль и ввода имени пользователя и его пароля. После того, как пользователь укажет имя, пароль и нажмет кнопку «Ок», введенная информация будет отправлена серверу: р у р нажмет кнопку «Ок», введенная информация будет отправлена серверу: нажмет кнопку «Ок», введенная информация будет отправлена серверу: POST /login.aspx?ReturnUrl=%2fDefault.aspx HTTP/1.0 g p p STATE=gibberish&txtUser=test&txtPassword=test . g p p STATE=gibberish&txtUser=test&txtPassword=test . При использовании метода GET введенные данные пересылаются в URL, в методе POST они пересылаются в теле запроса. Всегда должен использоваться метод POST, чтобы избежать хранения аутентификаторов на стороне клиента (в истории браузера), кеширования их прокси серверами. Да и сервера приложений зачастую сохраняют информацию о URL для сбора статистики. Поскольку информация об аутентификаторах помещается в ответ браузера в открытом виде, она передаются на сервер, обычно по протоколу SSL или TLS. р у р р д , р д р р, р у Проблема безопасности HTML-форм заключается в том, что по своей природе Web- приложения являются приложениями «без предыстории» (stateless). Если для доступа к ресурсу пользователь должен быть аутентифицирован, то возникает необходимость сохранения учетных данных, введенных в поля формы, чтобы передать на сервер как часть следующей транзакции. Чтобы каким-то образом сохранить аутентифицирующую информацию, разработчики могут использовать два подхода – сохранять необходимую информацию на сервере либо на стороне клиента, причем большинство разработчиков предпочитают второй вариант, считая его более легким в реализации. Информация на стороне клиента может сохраняться в файлах cookies браузера, в виде значений, добавляемых к URL и в скрытых полях HTML-форм. Уязвимость последних двух вариантов очевидна [2]. В случае, если информация о текущей сессии пользователя хранится на сервере (в виде временных файлов или временных таблиц в базе данных), ссылки на эту информацию должны быть доступны на стороне клиента. Как правило, они хранятся в виде Id сессии в файлах cookie . - получив логин и пароль сервер сверяет их теми, что хранятся в файле web.config (или другом, выбранном разработчиками хранилище). Если проверка прошла успешно, сервер перенаправляет клиента к той странице, которую клиент запрашивал: HTTP/1.1 302 Found Location: /Default.aspx Set-Cookie: AuthCookie=45F68E1F33159A9158etc.; path=/ htmlheadtitleObject moved/title/headbody . НАУКОВО-ПРАКТИЧНИЙ ЖУРНАЛ «ЗАХИСТ ІНФОРМАЦІЇ» № 3, 2012 Аутентификация на основе форм уязвима к атакам brute force. Кроме того, это протокол уязвим к атакам прослушивания (eavesdropping ) сети и атакам повтора сообщений (replay attacks), если не используется защита сообщений с помощью HTTPS или других криптопротоколов. Если cookie содержат конфиденциальную информацию, для них могут быть установлены два атрибута: secure и HTTPOnly. Если установлен атрибут secure, браузер не станет передавать cookie вне HTTPS соединения. Атрибут HTTPOnly предолжен Microsoft и предназначен для защиты от атак перехвата сессии (session hijacking). Браузеры, поддерживающие этот атрибут, не позволяют программам на JavaScript получать доступ к cookie, даже если политика безопасности такой доступ разрешает. у р р Обзор атак на схемы аутентификации Web-приложений приведены в [2]. Заключение у р р Обзор атак на схемы аутентификации Web-приложений приведены в [2]. З р Заключение Аутентификация является важнейшим механизмом безопасности Web-сервера. Поскольку требования к функционалу и безопасности Web-сервером могут существенно различаться, не существует одного, «наилучшего» способа аутентификации. Однако можно сформулировать основные требования к реализации данного механизма, реализация которых К б существенно снижает риск осуществления атак. К таким требованиям можно отн - строгая политика управления паролями и учетными записями пользователей; - невозможность обхода механизма аутентификации, например, с помощью инъекций; - невозможность обхода механизма аутентификации, например, с помощ инъекций; рет использования персональных данных в качестве аутентификаторов; - запрет использования персональных данных в качестве аутентификаторов - пересылаемые по сети аутентификаторы должны быть защищены с ротокола HTTPS; вся получаемая от пользователей информация должна тщательн роваться; контролироваться; - идентификаторы сессии должны создаваться таким образом, чтобы их нельзя было предсказать. При разработке подсистемы аутентификации необходимо определить требуемый уровень защищенности. Выбор уровня защищенности можно рассматривать как компромисс между удобством пользователя, производительностью и защищенностью. Необходимо учитывать все особенности используемых криптографических методов и протоколов. При реализации криптографических алгоритмов, таких как протокол формирования хеша SHA-1, протокол аутентификации сообщений HMAC, протоколы высокого уровня, например, SSL необходимо понимать круг задач, решаемых с их помощью.Рекомендации по построению схем аутентификации можно найти в прилагаемом списке литературы. НАУКОВО-ПРАКТИЧНИЙ ЖУРНАЛ «ЗАХИСТ ІНФОРМАЦІЇ» № 3, 2012 сервер перенаправляет клиента к той странице, которую HTTP/1.1 302 Found Location: /Default.aspx Set-Cookie: AuthCookie=45F68E1F33159A9158etc.; path=/ htmlheadtitleObject moved/title/headbody . Заголовок Set-Cookie содержит аутентификатор клиента. Метод шифрования cookie (в данном примере 3DES), в случае ASP.NET задается в файле web.config. оловок Set-Cookie содержит аутентификатор клиента. Метод шифрования cookie римере 3DES), в случае ASP.NET задается в файле web.config. - клиент повторяет запрос к ресурсу, но уже с только что установленным аутентификатором (cookie), который автоматически добавляется в заголовок HTTP: GET /Default.aspx HTTP/1.0 p Cookie: AuthCookie=45F68E1F33159A9158etc - сервер проверяет cookie, и в случае успеха, предоставляет требуемую страницу с HTTP сообщением 200 OK. - сервер проверяет cookie, и в случае успеха, предоставляет требуемую страницу с HTTP сообщением 200 OK. Cookie может быть действительна только для текущей сессии браузера, но может быть создана и для длительного хранения. Каждый раз, когда пользователь запрашивает страницу с сервера, браузер автоматически отправляет cookie серверу. Сервер проверяет cookie по своей базе идентификаторов и, при наличии в базе такого идентификатора, разрешает пользователю доступ. Файлы cookie – это текстовые файлы, содержащие пары имя/значение. Кроме того, они могут содержать информацию о сроке действия, пути и доменном имени (RFC2965). Браузер MS Internet Explorer хранит их в виде отдельных файлов в папке Application Data\cookies. Браузеры на основе Netscape и Mozilla хранят их в одном файле, называемом cookies.txt. А поскольку они сохраняются в файл, то могут быть изменены с помощью текстового редактора.\ Сессионные cookie подделать сложнее, однако существуют программные средства, например, Burp Proxy, которые позволяют просматривать и изменять сессионные cookie . 44 1. The WASC Threat Classification v2.0. [Электронный ресурс] - Режим доступа. 2. http://projects.webappsec.org/w/page/13246978/Threat-Classification свободный. — Загл. с экрана. 3. Authentication and Session Management on the Web [Электронный ресурс] - Режим доступа http://www.westpoint.ltd.uk/advisories/Paul_Johnston_GSEC.pdf свободный. — Загл. с экрана. 4. Web Authentication Security [Электронный ресурс] - Режим доступа. 5. http://www.sans.org/reading_room/whitepapers/webservers/web-authentication- security_1250 свободный. — Загл. с экрана. 6. A Guide to Web Authentication Alternatives [Электронный ресурс] - Режим доступа http://unixpapa.com/auth/ свободный. — Загл. с экрана. 7. Хогланд Г. Взлом программного обеспечения: анализ и использование кода [Текст]/ Хогланд Г. Мак-Гроу Г. // Издательский дом "Вильяме". - М.: -2005, 389 с.: ил. - ISBN 5-8459-0785-3, 0-201-78695-8. 8. CodeNet - Все для программиста! [Электронный ресурс]: Поиск уязвимостей в программах помощью анализаторов кода. - Режим доступа: http://www.codenet.ru/progr/other/code-analysers.php свободный. — Загл. с экрана. Надійшла: 24.07.2012 р. ЛИТЕРАТУРА 1. The WASC Threat Classification v2.0. [Электронный ресурс] - Режим доступа. 2. http://projects.webappsec.org/w/page/13246978/Threat-Classification свободный. — Загл. с экрана. 3. Authentication and Session Management on the Web [Электронный ресурс] - Режим доступа. http://www.westpoint.ltd.uk/advisories/Paul_Johnston_GSEC.pdf свободный. — Загл. с экрана. 4. Web Authentication Security [Электронный ресурс] - Режим доступа. 5. http://www.sans.org/reading_room/whitepapers/webservers/web-authentication- security_1250 свободный. — Загл. с экрана. 6. A Guide to Web Authentication Alternatives [Электронный ресурс] - Режим доступа: http://unixpapa.com/auth/ свободный. — Загл. с экрана. 7. Хогланд Г. Взлом программного обеспечения: анализ и использование кода [Текст]/ Хогланд Г., Мак-Гроу Г. // Издательский дом "Вильяме". - М.: -2005, 389 с.: ил. - ISBN 5-8459-0785-3, 0-201-78695-8. 8. CodeNet - Все для программиста! [Электронный ресурс]: Поиск уязвимостей в программах с помощью анализаторов кода. - Режим доступа: http://www.codenet.ru/progr/other/code-analysers.php свободный. — Загл. с экрана. Надійшла: 24 07 2012 р 1. The WASC Threat Classification v2.0. [Электронный ресурс] - Режим доступа. 1. The WASC Threat Classification v2.0. [Электронный ресурс] - Режим доступа. 2. http://projects.webappsec.org/w/page/13246978/Threat-Classification свободный. — Загл. с экрана. 2. http://projects.webappsec.org/w/page/13246978/Threat-Classification свободный. — Загл. с экрана. 3. Authentication and Session Management on the Web [Электронный ресурс] - Режим www.westpoint.ltd.uk/advisories/Paul_Johnston_GSEC.pdf свободный. — Загл. с экрана. 3. Authentication and Session Management on the Web [Электронный ресурс] - Режим доступа. http://www.westpoint.ltd.uk/advisories/Paul_Johnston_GSEC.pdf свободный. — Загл. с экрана. 4 W b A th ti ti S it [Э й ] Р p _ _ p д 4. Web Authentication Security [Электронный ресурс] - Режим доступа. 5. http://www.sans.org/reading_room/whitepapers/webservers/web-authentication- security_1250 свободный. — Загл. с экрана. 5. http://www.sans.org/reading_room/whitepapers/webservers/web-authentication- security_1250 свободный. — Загл. с экрана. 6. A Guide to Web Authentication Alternatives [Электронный ресурс] - Режим доступа: http://unixpapa.com/auth/ свободный. — Загл. с экрана. 6. A Guide to Web Authentication Alternatives [Электронный ресурс] - Режим доступа: http://unixpapa.com/auth/ свободный. — Загл. с экрана. 7. Хогланд Г. Взлом программного обеспечения: анализ и использование кода [Текст]/ Хогланд Г., Мак-Гроу Г. // Издательский дом "Вильяме". - М.: -2005, 389 с.: ил. - ISBN 5-8459-0785-3, 0-201-78695-8. 8. CodeNet - Все для программиста! [Электронный ресурс]: Поиск уязвимостей в программах с помощью анализаторов кода. - Режим доступа: http://www.codenet.ru/progr/other/code-analysers.php свободный. — Загл. с экрана. 8. CodeNet - Все для программиста! [Электронный ресурс]: Поиск уязвимостей в программах с помощью анализаторов кода. - Режим доступа: http://www.codenet.ru/progr/other/code-analysers.php свободный. — Загл. с экрана. Надійшла: 24.07.2012 р. Рецензент: д.т.н., професор Юдін О.К. 45
https://openalex.org/W4388078352	https://churchhistory.elpub.ru/jour/article/download/133/178	Russian	null	Synodic of the Studenitsa Monastery as a historical source	Rossijskij žurnal istorii cerkvi	2,023	cc-by	3,594	Белянкин Ю. С. Белянкин Ю. С. Белянкин Ю. С. Задачей настоящей статьи является раскрытие и анализ содержания рукописного милостинного Синодика, принадлежащего с середины XVII в. одному из важнейших сербских монастырей — Студеница. Наша цель состоит в том, чтобы показать источниковедческую ценность Синодика из Студеницы, в первую очередь, для истории русско-сербских церковных связей. Такие рукописи известны в единичных экземплярах в хранили- щах Балканского полуострова и Греции, все они отражают историю отношений России с южнославянскими и другими православными народами юга Европы преимущественно в период правления, преимущественно, династии Романовых. Ценность рукописи состоит в том, что она маркирует начало установления регулярной практики хождений духовенства Студеницы за милостыней в Россию, продолжавшейся столетие. Монастыр- ский Синодик содержит богатый материал по генеалогии многих русских, украинских, сербских и иных семей и фамилий, сыгравших важную роль в исторических событиях юга и востока Европы во второй половине XVII- XVIII вв. Ключевые слова: монастырь Студеница, рукописи, милостинный Синодик, генеалогия, царь Алексей Михай- лович. Отношения и деятельность: не оказывают влияния на представленный материал. Белянкин Юрий Сергеевич — зав. сектором изучения особо ценных фондов Центра по исследованию про- блем развития библиотек в информационном обществе Российской государственной библиотеки, Москва, Россия. ORCID: 0000-0003-1236-7953. Автор, ответственный за переписку (Corresponding author): bys86@yandex.ru Автор, ответственный за переписку (Corresponding author): bys86@yandex.ru Рукопись получена 28.06.2023 Рецензия получена 16.07.2023 Принята к публикации 22.07.2023 Для цитирования: Белянкин Ю. С. Синодик монастыря Студеница как исторический источник. Российский журнал истории Церкви. 2023;4(3):103-115. doi:10.15829/2686-973X-2023-133. EDN OYUMXM Рукопись получена 28.06.2023 Рецензия получена 16.07.2023 Рецензия получена 16.07.2023 Принята к публикации 22.07.2023 Принята к публикации 22.07.2023 Для цитирования: Белянкин Ю. С. Синодик монастыря Студеница как исторический источник. Российс журнал истории Церкви. 2023;4(3):103-115. doi:10.15829/2686-973X-2023-133. EDN OYUMXM ISSN 2686-973X (Print) ISSN 2687-069X (Online) НОВЫЕ ОТКРЫТИЯ ISSN 2686-973X (Print) ISSN 2687-069X (Online) НОВЫЕ ОТКРЫТИЯ РОССИЙСКИЙ ЖУРНАЛ ИСТОРИИ ЦЕРКВИ 2023; 4(3): 103-115 doi:10.15829/2686-973X-2023-133 Российский журнал истории Церкви Российский журнал истории Церкви of many Russian, Ukrainian, Serbian and other families that played a historical role in the events of Southern an Eastern Europe in the second half of the XVII-XVIII centuries. Keywords: Studenitsa Monastery, manuscripts, Synodic, genealogy, tsar Alexei Mikhailovich. 3 Турилов, А. А., Мошкова, Л. В. (2016). Каталог славянских рукописей Афонских обителей. Белград, 387. Synodic of the Studenitsa Monastery as a historical source Yuriy S. Beliankin Yuriy S. Beliankin The article analyzes the content of the handwritten Synodic, which has belonged to one of the most important Serbian monasteries — Studenica — since the middle of the XVII century. Such manuscripts are known in single copies in the repositories of the Balkan Peninsula and Greece, all of them reflect the history of Russia’s relations with the South Slavic and other Orthodox peoples of southern Europe during the reign, mainly, of the Romanov dynasty. The value of this manuscript is that it marks the beginning of the establishment of the regular practice of the clergy of the Studenitsa for alms in Russia, which lasted a century. Synodic contains a wealth of material on the genealogy 103 2023; 4 (3) 1 Крамер, А. В. (2011). Раскол Русской Церкви в середине XVII в. СПб.: Алетейя, 22-52. 2 Каталог на славянските ръкописи в библиотеката на Зографския манастир в Света гора. (1994). София, (77), 63. 3 Турилов, А. А., Мошкова, Л. В. (2016). Каталог славянских рукописей Афонских обителей. Белград, 387. 2 Каталог на славянските ръкописи в библиотеката на Зографския манастир в Света гора. (1994). Софи (77), 63. 1 Крамер, А. В. (2011). Раскол Русской Церкви в середине XVII в. СПб.: Алетейя, 22-52. 4 Каталог славянских рукописей…, 386. Relationship and Activities: none. Yuriy S. Beliankin — head of the Sector for the study of particularly Valuable Funds of the Center for the Study of Problems of Library Development in the Information Society of the Russian State Library, Moscow, Russia. ORCID: 0000-0003-1236-7953. Corresponding author: bys86@yandex.ru Received: 28.06.2023 Revision Received: 16.07.2023 Accepted: 22.07.2023 For citation: Yuriy S. Beliankin. Synodic of the Studenitsa Monastery as a historical source. Russian Journal of Church History. 2023;4(3):103-115. (In Russ.) doi:10.15829/2686-973X-2023-133. EDN OYUMXM Синодики в XVII-XVIII вв. для индивидуальных или коллективных (родовых и тому подобных) поминовений о здравии и упокоении, запол- нявшиеся при сборе пожертвований от поминаемых донаторов, были явлением достаточно распространенным, а сам факт сбора милостыни на русских землях для бедствующих афонских и балканских обителей — традиционным. Многочисленные свидетельства об этом присутствуют в актовых источниках эпохи1, а современные хранилища рукописей содер- жат подобные памятники. В частности, в болгарcком монастыре Зограф на Афоне в настоящее время хранится Зографский помянник, именуемый "русским". Рукопись, украшенная орнаментами и заставками в харак- терном "старопечатном" стиле и созданная, судя по качеству и богатству исполнения, в московской Оружейной палате, была изготовлена в Москве около 1639 г., в нее записывались вкладчики и донаторы Зографской обители2 (рис. 1; выходная запись Синодика; описан в каталоге как "Руски Зографски Поменик. Писан в Москва за Зографския манастир"). Еще один пример — Синодик-помянник Великой Лавры, составленный около 1642-1645 гг. также на территории России. В Синодике в общей части поминаются русские великие князья и цари от Владимира до Василия Шуйского, великие княгини и царицы, митрополиты и патриархи вплоть до патриарха Иоасафа, а также перечни боярских и княжеских фамилий 3. Аналогичная рукопись имеется в сербском Хиландарском монастыре на Афоне. Синодик милостинный датируется серединой XVIII в. В нем поми- наются русские правители вплоть до Петра II и Елизаветы Петровны, патриархи до Филарета, фамилии семей из т.н. Славяносербии — сербских 104 doi:10.15829/2686-973X-2023-3 Новые открытия doi:10.15829/2686-973X-2023-3 Новые открытия с 1 Выходная запись русского Синодика монастыря Зограф См : Каталог на Рис. 1. Выходная запись русского Синодика монастыря Зограф. См.: Каталог на славянските ръкописи в библиотеката на Зографския манастир в Света гора. (1994). София, Рис. 78. 105 Российский журнал истории Церкви 2023; 4 (3) 2023; 4 (3) 2023; 4 (3) Российский журнал истории Церкви Рис. 2. Синодик монастыря Студеница. Л. 10 об. Выходная запись. Рис. 2. Синодик монастыря Студеница. Л. 10 об. Выходная запись. поселений на территории Слободской Украины 4. Аналогичным образом освещает связи Синая с Россией и Украиной в XVII — середине XVIII вв. поселений на территории Слободской Украины 4. Аналогичным образом освещает связи Синая с Россией и Украиной в XVII — середине XVIII вв. милостинный Синодик монастыря вмч. Екатерины на Синае (Sinait. 5 Altbauer, M. (1992). An East-Slavic Sinodik from the Sinai. Köln; Weimar; W. 6 Богданович, Д. (1982). Инвентар ћирилских рукописа у Jyгославиjи (XI-XVII века). Белград, 82. Relationship and Activities: none. ловича монастырь получал регулярные денежные пожертвования, то есть царское расположение к гостям из Студеницы было явным, вследствие чего для них, очевидно, изготовили за счет казны милостинный Синодик. Выходная дата рукописи (рис. 3) прослеживается в орнаментирован- ном рукописном картуше "старопечатного" стиля на первом листе: "Сий Синодик монастыря Студенице храма Успения Пресвятые Богородицы <...> лета 7163". Писцовая выходная запись отсутствует. Таким образом, рукопись была изготовлена в 1655 г., ее появление связано с прибытием в Москву архимандрита Неофита в январе этого года и подачей им чело- битной царю Алексею Михайловичу с просьбой о вспомоществовании, поскольку это был первый визит духовенства Студеницы в Москву7. Судя по составу поминаемых персоналий рассматриваемого Синодика, его бытование в существенной мере связано с украинскими землями, через которые по обыкновению лежал большой участок пути сербских просителей в Москву. В самом начале Синодика, вслед за общей пред- исловной частью поминаются, прописанные золотом, имена дочерей Алексея Михайловича (в частности, сестра Петра I Наталья Алексеевна), а также дети соправителя Петра I Иоанна Алексеевича (рис. 4). Под 1686 г. поминается патриарх Иоаким (рис. 5), при этом имя патриарха Никона в череде поминаемых отсутствует (вероятно, в связи с осужде- нием Никона восточными иерархами и сведением его с престола). Поми- нается также скончавшийся во младенчестве царевич Дмитрий Алексе- евич и длинным списком — все дети царя Михаила Федоровича, а также род Бориса Годунова. р р у Одним из первых в рукописи зафиксирован род "его пресветлого вели- чества" гетмана Ивана Степановича Мазепы с припиской на поле "писан 1691 [года] месеца иуния 21". В Синодике о здравии записана мать гетмана Марина Мокиевская, бывшая настоятельницей Киево-Печерского Воз- несенского монастыря, а в поминании усопших — Адам-Степан Мазепа, отец гетмана и соратник Богдана Хмельницкого. В Синодике также при- сутствует большая запись на помин рода кошевого гетмана и гетмана Запо- рожской Сечи Ивана Брюховецкого, сподвижника Богдана Хмельницкого, внесшего большой вклад во вхождение украинских земель в состав Рос- сии. Имеется поминание родов петровского сподвижника, героя Полтавы, изюмского полковника Федора Владимировича Шидловского, известного участника Северной войны (судя по большому количеству записей с оди- наковой локацией, в городах Изюм и Харьков сербские посланники про- вели изрядное количество времени), а также представителя известной на Слобожанщине семьи, сына основателя города Сумы, полковника и столь- ника Андрея Герасимовича Кондратьева. Следующим вписан род имере- тинского царя Арчила II, который с конца XVII в. постоянно жил в Москве и стал основателем грузинской диаспоры в "царствующем граде". Среди усопших поминаются скончавшийся около 1693 г. его сын Матвей, сын Александр (ум. ок. 7 Православные монастыри. Путешествие по святым местам. [Монастырь Студеница] № 39/2007. С. 26. Relationship and Activities: none. милостинный Синодик монастыря вмч. Екатерины на Синае (Sinait. 106 doi:10.15829/2686-973X-2023-3 Новые открытия Р 3 С С Л 11 Н doi:10.15829/2686-973X-2023-3 Новые открытия Рис. 3. Синодик монастыря Студеница. Л. 11. Начало основного текста. slav. 9b) 5. Кроме того, в Пантелеймоновом монастыре на Афоне сохра- нился милостинный Синодик 1705 г. (Ризница. № 8). Что касается серб- ских монастырей, то в настоящее время помимо Синодика Студеницы известен такой же "русский" Синодик XVII в., заметно меньшего объема, монастыря Крка в Северной Далмации 6. slav. 9b) 5. Кроме того, в Пантелеймоновом монастыре на Афоне сохра- нился милостинный Синодик 1705 г. (Ризница. № 8). Что касается серб- ских монастырей, то в настоящее время помимо Синодика Студеницы известен такой же "русский" Синодик XVII в., заметно меньшего объема, монастыря Крка в Северной Далмации 6. Текст рассматриваемого в статье Синодика монастыря Студеница (Жичская епархия Сербской Православной церкви) ранее не публико- 107 2023; 4 (3) Российский журнал истории Церкви Российский журнал истории Церкви Рис. 4. Синодик монастыря Студеница. Л. 17. Поминание детей царя Алексея Михайловича и царя Ивана Алексеевича. Рис. 4. Синодик монастыря Студеница. Л. 17. Поминание детей царя Алексея Михайловича и царя Ивана Алексеевича. вался и не исследовался, он хранится в настоящее время на своем искон- ном месте, в монастырской ризнице, и представляет собой рукопись in quarto объемом 156 листов в цельнокожаном переплете XIX в., основной почерк — крупный каллиграфический русский полуустав, с приписками многими скорописными и полууставными почерками до второй половины XVIII в. (рис. 2). Судя по основным группам дат в поминальных статьях рукописи, было три периода заполнения Синодика, когда монахи из Сту- 108 doi:10.15829/2686-973X-2023-3 Новые открытия Рис. 5. Синодик монастыря Студеница. Л. 18. Поминание патриархов Московских и всея Руси. Рис. 5. Синодик монастыря Студеница. Л. 18. Поминание патриархов Московских и всея Руси. деницы, по-видимому, имели возможность отправиться за милостыней, в XVII в. и первой половине XVIII в., а именно: третья четверть XVII в., начало и середина XVIII в. Следует заметить, что монастырь Студеница — один из важнейших и древнейших в Южной Сербии, а с утратой Косово стал главным духовным центром сербского Православия, поскольку Студеница была заложена основателем сербского государства Стефа- ном Неманей еще в XII в. Факт изготовления Синодика для Студеницы в Москве неудивителен, учитывая, что по решению царя Алексея Михай- 109 2023; 4 (3) Российский журнал истории Церкви ловича монастырь получал регулярные денежные пожертвования, то есть царское расположение к гостям из Студеницы было явным, вследствие чего для них, очевидно, изготовили за счет казны милостинный Синодик. Relationship and Activities: none. 1711 г.), Давид и Дарья, также на момент составления записи уже скончавшиеся (1758 г.). В студеницком Синодике встречается 110 doi:10.15829/2686-973X-2023-3 Новые открытия и род гетмана Войска Запорожского Ивана Самойловича (запись "року 1685"). В Синодик внесен род Лопухиных, среди которых поминается боя- рыня Евдокия, которую можно идентифицировать как первую супруга Петра I Евдокию Лопухину, не поминаемую здесь как царица. Петра I Евдокию Лопухину, не поминаемую здесь как царица. Записи в Синодике подчеркивают состав ближайшего круга Петра I, так, в нем имеется запись о здравии одного из самых известных сподвижников Петра I, именитого человека Григория Дмитриевича Строганова, а также одного из крупнейших солепромышленников того времени Григория Федо- ровича Шустова. В числе родственников Строганова вписаны вторая жена Мария Новосильцева, сыновья, бароны Александр и Николай, его усоп- шая мать Анна, первая жена Васса. В Синодик внесен род Ивана Алексее- вича Мусина-Пушкина — начальника Монастырского приказа и Печатного двора и ближайшего сподвижника Петра I. Интересно, что среди поминае- мых в роду князя Михаила Яковлевича Черкасского — "царь" Симеон Бекбу- латович, "во иноцех Стефан", "царствовавший" по велению Ивана Грозного, а также отец князя Михаила Яков, его сын Петр, с которым Михаил Яков- левич по указанию Петра I воеводствовал в Тобольске, его жена Марфа. Кроме того, отдельной статьей в Синодике за 1663 г. записан отец Миха- ила Яков Куденетович Черкасский. Многие синодичные записи датируются в приписках на полях и в тексте "7212" годом, т.е. 1703/04 гг., и "7171" годом (1662 г.), что, по всей видимости, свидетельствует о хождениях за мило- стыней монастырской братии в указанные периоды, в процессе которых и совершались индивидуальные и групповые записи донаторов в изучае- мый кодекс. Действительно, архимандрит Неофит посетил Москву второй раз именно летом 1662 г. и получил разрешение от царя Алексея Михайло- вича приезжать за милостыней раз в пять лет. В Синодике помянут и Борис Петрович Шереметев — правая рука Петра I, генерал-фельдмаршал и герой Северной войны, с упоминанием упокоения его жены Евдокии Чирико- вой. На листах Синодика нам встречается имя Петра Ивановича Прозоров- ского — крупного государственного деятеля, наставника Петра I, "дядьки" царя Иоанна Алексеевича. Надо заметить, что Синодик Студеницы пред- ставляет собой прекрасный источник по просопографии и генеалогии пер- вых лиц петровской эпохи, содержащий достаточно подробные росписи родов и семей, даты поминаний etc., т.е. даже более точную информацию, чем могли сохранить актовые источники. Большая часть записей Синодика, несмотря на писцовую дату, отно- сится к последней четверти XVII — первой половине XVIII в. Relationship and Activities: none. и в большой мере отражает исторический контекст конца правления Алексея Михай- ловича и весь период эпохи Петра I в широком смысле. Так, в Синодике записан дьяк Козьма Никитич Нефимонов — дипломат, готовивший дого- воры России с Австрией и Венецией в 1697 г., полковник Войска Запо- рожского Войца Сербин, активный участник политических событий на Украине в последней четверти XVII в.; представители родов Нарышки- ных (в т.ч. дядя царя Петра I, глава Посольского приказа Лев Кирилло- вич Нарышкин), Хованских, Прозоровских, Хворостининых, Трубецких, Бутурлиных, Одоевских, Елизаровых, Салтыковых, Мусиных-Пушкиных, Урусовых, Долгоруковых, Щербатовых, Шереметевых, Львовых, Плеще- 111 Российский журнал истории Церкви 2023; 4 (3) 2023; 4 (3) евых, Барятинских, Стрешневых, Кольцовых-Мосальских. Здесь же ука- заны: дядька Петра I Борис Алексеевич Голицын; род одного из главных придворных книжников и поэтов последней четверти XVII в. Кариона Истомина (поминание иерея Тита, схимонахини Елены, Гавриила, Анны, иеросхимонаха Феофана, монаха Софрониа, Сильвестра); род гетмана Войска Запорожского Даниила Апостола, полковника Миргородского (запись 1730-х гг. и явно прижизненная, поминаются среди прочих сам гет- ман и его жена Ульяна). Даниил Апостол, как известно, вместе с гетманом Мазепой перешел на сторону шведского короля в Северной войне, однако с момента очевидных неудач Мазепы вернулся на службу Петру I. у р у у у ру Наиболее ранняя запись относится к 1655 г. (т.е. времени появления Синодика — род дипломата боярина Григория Гавриловича Пушкина) и 1662/63 гг. (посвящена роду князя Даниила Степановича Великогагина, известного своей деятельностью на Левобережной Украине). Кроме того, в рукопись внесен целый ряд представителей духовенства из великорус- ских и западнорусских земель, которым удалось "записаться" в Синодик и внести пожертвования, в их числе члены братии и фамилии Ахтыр- ского Благовещенского, московского Симонова монастыря, Кирилло- Белозерского монастыря, московской церкви Николая Чудотворца в Котельниках, Троице-Сергиева монастыря, епископа Мстиславльского и Оршанского, Молченского монастыря в Путивле, митрополита Рязан- ского и Муромского Авраамия, митрополита псковского Маркелла, Дум- ницкого монастыря, Иоанна архиепископа Черниговского, Исаии митро- полита Нижегородского и Алатырского, митрополита Иустина (Базиле- вича) Белгородского и Обоянского, московского Успенского собора. Периодически на полях листов встречаются детали вкладов: "дал вклад книгу Обед душевный [Симеона Полоцкого] и 2 рубли дал", "дал вкладу ризы патрахильные поручи", "даден пояс един сребрян и позлащен с каме- нием украшен". 8 О нем: Крылов, Г. прот. (2009). Книжная справа XVII века. Богослужебные Минеи. М. 9 О нем: Оборнева, З.Е. (2017). Переводчик Посольского приказа Борис Богомольцев. Древняя Русь Вопросы медиевистики, 1, 50-61. Relationship and Activities: none. Из других представителей высшей аристократии в Синодике встреча- ются записи о крестном отце Петра I князе Алексее Никитиче Трубец- ком, участнике Конотопской битвы и военных действий на Украине; сыне Дмитрия Пожарского Иване Дмитриевиче Пожарском; участнике Смуты боярине Федоре Семеновиче Куракине; патриаршем ризничем иеродиа- коне Филарете (с пометой "дал вкладу Чиновник", вероятно, московское издание "Архиерейского чиновника" 1677 г.); окольничем и знамени- том фаворите царя Алексея Михайловича Федоре Михайловиче Ртищеве (запись 1663 г.). В Синодике, что примечательно, среди прочих, поми- нается род известного книжного справщика Московского Печатного двора Иосифа Белого8. В Синодике записаны знаменитые украинские церковные деятели той эпохи: Иннокентий Гизель, архимандрит Киево- Печерской Лавры, осыпанный благодарностями царя Алексея Михайло- вича, автор хорошо известного "Синопсиса"; а также наместник Киево- Печерской Лавры, автор многочисленных церковных сочинений, спод- вижник Гизеля Антоний Радивиловский. 112 doi:10.15829/2686-973X-2023-3 doi:10.15829/2686-973X-2023-3 Новые открытия Новые открытия doi:10.15829/2686-973X-2023-3 Новые открытия Ценность представляет и большая запись о поминовении под 7171 г. известного "греческого" переводчика Посольского приказа Бориса Бого- мольцева (рис. 6) 9. Запись за 1686 г. (в этом году состоялся очередной визит 113 2023; 4 (3) Российский журнал истории Церкви в Россию просителей из Студеницы) перечисляет родственников Васи- лия Кочубея, легендарной фигуры, казненного за якобы ложный донос на гетмана Мазепу, который вскоре действительно изменил Петру I в войне с Карлом XII. Следующие герои петровской эпохи — Алексей Семенович Шеин — участник Крымских и Азовских походов кон. XVII в., генералис- симус, расследователь стрелецких бунтов (запись посмертная, т.е. после 1700 г.); и думный дьяк Емельян Украинцев, выдающийся дипломат, посол в Европе и Турции, глава Посольского приказа (прижизненная запись 1663 г.). Вторую половину Синодика занимают поминания, имеющие мест- ные балканские корни. Так, наиболее тожественная орнаментированная запись сделана о поминовении родственников Иоанна, епископа Каран- себешского Темишварского баната (Румыния), от 1753 г., она заверена его собственноручной подписью. Синодик монастыря Студеница является не только достаточно редким бытовавшим в Сербии рукописным памятником эпохи османского угнете- ния, ставшего ключевой причиной "хождений" греческих и южнославян- ских православных в пределы последнего православного царства — Мос- ковского государства, но и содержательным источником по восточноев- ропейской внешнеполитической повестке конца XVII — начала XVIII вв., русско-украинским отношениям, отраженным в рассмотренном памят- нике в достаточной полноте через поминаемых исторических персона- лий и отдельные их группы, участвовавшие в событиях от присоединения Гетманщины к России в середине-второй половине XVII в. до победы в Северной войне и образования Российской империи в первой-второй четверти XVIII в. Литература Каталог на славянските ръкописи в библиотеката на Зографския манастир в Света гора. (1994). София, CIBAL, 63 с. 20 6 б 38 1. Каталог на славянските ръкописи в библиотеката на Зографския манастир в Света гора. (1994). София, CIBAL, 63 с. 1. Каталог на славянските ръкописи в библиотеката на Зографския манастир в Света гора. (1994). София, CIBAL, 63 с. 2. Турилов, А. А. Мошкова, Л. В. (2016). Каталог славянских рукописей Афонских обителей. Белград: Чигоja штампа, 387 с. 3. Богданович, Д. (1982). Инвентар ћирилских рукописа у Jyгославиjи (XI-XVII века). Белград: САНУ, 82 с. Турилов, А. А. Мошкова, Л. В. (2016). Каталог славянских рукописей Афонских обителей. Белград: Чигоja штампа, 387 с. Богданович, Д. (1982). Инвентар ћирилских рукописа у Jyгославиjи (XI-XVII века). Белград: САНУ, 82 с. 4. Крылов, Г., прот. (2009). Книжная справа XVII века. Богослужебные Минеи. М.: Индрик. Крылов, Г., прот. (2009). Книжная справа XVII века. Богослужебные Минеи. М.: Индрик. 5. Оборнева, З. Е. (2017). Переводчик Посольского приказа Борис Богомольцев. Древняя Русь. Вопросы медиевистики. М.: Индрик, 1, 50-61. 5. Оборнева, З. Е. (2017). Переводчик Посольского приказа Борис Богомольцев. Древняя Русь. Вопросы медиевистики. М.: Индрик, 1, 50-61. doi:10.15829/2686-973X-2023-3 Новые открытия Relationship and Activities: none. Кроме того, историческая география записей в Сино- дике, благодаря повторявшимся в них от раза к разу населенным пунктам, городам, монастырям etc еще раз показывает традиционный маршрут всех путешествий славянских и греческих просителей и паломников в Рос- сию — через Балканский полуостров, Румынию, Украину и западные рус- ские земли — в Москву (далее маршруты могли варьироваться в пределах Центральной России). Студеницкий синодик, изначально возникший судя по всему на основании царского распоряжения, весьма неординарен по "личному" составу записанных в него персоналий, поскольку в нем зафик- сирована в большой мере военная, политическая и церковная "элита" эпохи, охотно желавшая внесения в этот помянник, что отличает его от других подобных синодиков-помянников, рядового характера и с более предсказуемым содержанием , имеющих лишь локальное значение. 114 doi:10.15829/2686-973X-2023-3 References 1. Katalog na slavianskite rukopisi v bibliotekata na Zografskia manastir v Sveta gora (1994). Sofia: CIBAL, 63 р. (In Bulgarian) 2. Turilov, A. A. Moshkova, L. V. (2016). Katalog slavianskih rukopisey Afonskih obiteley. Beograd: Chigoja shtampa, 387 р. (In Russ.) 3. Bogdanovich, D. (1982). Inventar chirilskih rukopisa u Jugoslaviji (XI-XVII cent.). Beograd: Serbian Academy of sciences and arts, 82 р (In Russ ) Katalog na slavianskite rukopisi v bibliotekata na Zografskia manastir v Sveta gora (1994). Sofia: CIBAL, 63 р. (In Bulgarian) T il A A M hk L V (2016) K t l l i kih k i Af kih bit l B d Chi j ht 387 (I R ) 1. Katalog na slavianskite rukopisi v bibliotekata na Zografskia manastir v Sveta gora (1994). Sofia: CIBAL, 63 р. (In Bulgarian) 2. Turilov, A. A. Moshkova, L. V. (2016). Katalog slavianskih rukopisey Afonskih obiteley. Beograd: Chigoja shtampa, 387 р. (In Russ 2. Turilov, A. A. Moshkova, L. V. (2016). Katalog slavianskih rukopisey Afonskih obiteley. Beograd: Chigoja shtampa, 387 р. (In Russ.) 3. Bogdanovich, D. (1982). Inventar chirilskih rukopisa u Jugoslaviji (XI-XVII cent.). Beograd: Serbian Academy of sciences and arts, 82 р. (In Russ.) 3. Bogdanovich, D. (1982). Inventar chirilskih rukopisa u Jugoslaviji (XI-XVII cent.). Beograd: Serbian Academy of sciences and arts, 82 р. (In Russ.) 4. Krylov, G. (2009). Knizhnaya sprava XVII veka. Bogosluzhebnye Minei. M.: Indrik. (In Russ.) 4. Krylov, G. (2009). Knizhnaya sprava XVII veka. Bogosluzhebnye Minei. M.: Indrik. (In Russ.) 5. Oborneva, Z. E. (2017). Perevodchik Posolskogo prikaza Boris Bogomoltsev. Drevnya Rus. Voprosy medievistiki. M.: Indrik, 50-61. (In Russ.) 115
https://openalex.org/W2128681952	https://europepmc.org/articles/pmc4423211?pdf=render	English	null	Mechanisms of splicing-dependent trans-synaptic adhesion by PTPδ–IL1RAPL1/IL-1RAcP for synaptic differentiation	Nature communications	2,015	cc-by	11,089	ARTICLE Received 30 Dec 2014 \| Accepted 16 Mar 2015 \| Published 24 Apr 2015 ARTICLE ARTICLE NATURE COMMUNICATIONS \| DOI: 10.1038/ncomms7926 M Glu-Leu-Arg-Glu, whereas there are three variations in meA: a three-residue peptide Glu-Ser-Ile (meA3), a six-residue peptide Gly-Gly-Thr-Pro-Ile-Arg (meA6) and their tandem combination (Glu-Ser-Ile-Gly-Gly-Thr-Pro-Ile-Arg) (meA9). Binding to IL-1RAcP is signiﬁcantly enhanced by the presence of both meA and meB, independently of the meA variation. In contrast, the length and sequence of meA are critically important for binding to IL1RAPL1. Only PTPd variants containing meA9 and meA6 can bind to IL1RAPL1. The meB insertion increases binding of the meA9-containing variant, whereas the meA6- containing variant absolutely requires the combination with meB. These ﬁndings indicate that information of synaptogenic postsynaptic ligands for the type-IIa RPTPs is encoded in their splice inserts. Similarly, in another presynaptic organizer neurexin, the insert in the splice site 4 regulates the speciﬁcity to postsynaptic organizers Cbln1-GluRd2 and LRRTM1/2 (refs 15,16). The ‘splice-insert signaling code’ concept for speciﬁc synaptic connections was proposed originally for trans- synaptic adhesion between neurexin and neuroligin, but appears likely more general for synaptogenic adhesion3,7–9,15–18. However, structural mechanisms underlying this concept remain elusive. To elucidate the structural basis for decoding the ‘splice-insert signaling code’ in the type-IIa RPTPs, we determined the crystal structures of PTPd-ECD in complex with IL-1RAcP-ECD and IL1RAPL1-ECD. Together with structure- based mutational analyses by surface plasmon resonance (SPR) spectroscopy and synaptogenic assays, we reveal the structural basis of the splicing-dependent trans-synaptic adhesion by PTPd and IL1RAPL1/IL-1RAcP for synaptic differentiation. M ammalian brains are composed of at least a few hundred billions of neurons, which are connected by synapses in a spatiotemporally organized manner to establish higher-order brain functions such as learning, memory and emotion. Synapse formation is triggered through trans- synaptic interactions between selective pairs of pre- and postsynaptic adhesion molecules known as ‘synaptic organizers’ or ‘synaptogenic proteins’, many of which are associated with neurodevelopmental disorders such as intellectual disability and autism1–6. Recent studies have highlighted a major role of type- IIa receptor protein tyrosine phosphatases (RPTPs) as synaptic organizers in pre- and postsynaptic differentiation during neural development5,6. Vertebrate type-IIa RPTPs comprise LAR, PTPs and PTPd, which share common domain architecture with a single transmembrane (TM) helix ﬂanked by a large extracellular domain (ECD) and cytoplasmic tandem PTP domains: the membrane-proximal PTP domain (D1) is catalytically active, whereas the membrane-distal PTP domain (D2) is inactive. The ECD consists of three immunogloblin-like (Ig) domains and four to eight ﬁbronectin type-III (Fn) domains. Results Structure of the complex between PTPd and IL1RAPL1. To elucidate the structural basis of the PTPd–IL1RAPL1 interaction for trans-synaptic adhesion, we determined the crystal structures of IL1RAPL1-ECD in complex with PTPd Ig1-Ig2 and the full- length PTPd-ECD at 2.7 and 4.4 Å resolutions, respectively (Fig. 1a,b, Table 1). Description of PTPd Ig1-Ig2 will be based on the 2.7-Å-complex structure, whereas that of the other regions of PTPd-ECD will be based on the 4.4-Å-complex structure. For crystallography, we used a PTPd isoform containing four Fn domains and both meA9 and meB, which is the most abundant in the developing brain and has the highest afﬁnity for IL1RAPL1 (Kd: 0.15 mM) among all the PTPd variants (Table 2). Hereafter, ‘PTPd’ indicates this isoform, unless otherwise noted. The Ig1-Ig2 domains of PTPd retain the binding ability for IL1RAPL1-ECD, as shown in our serial deletion analysis of PTPd (Supplementary Fig. 1a). PTPd-ECD binds to IL1RAPL1-ECD at a ratio of 1:1 and the asymmetric unit contains one PTPdIL1RAPL1 complex (Fig. 1b and Supplementary Fig. 1b). IL1RAPL1-ECD consists of three Ig domains, which are arranged in an L-shape (Fig. 1a,b). PTPd-ECD exhibits an elongated shape with a length of 180 Å along the longest axis. The Ig1 and Ig2 domains of PTPd tightly interact with each other to form a compact V-shaped unit, which is almost identical to the isolated Ig1-Ig2 structures of PTPd19 (PDB:2YD6, 2YD7) (rmsd values of 0.8–0.9 Å for 195 Ca atoms). No substantial change in the backbone structure of PTPd Ig1-Ig2 occurred upon binding to IL1RAPL1 (Supplementary Fig. 1c). The Ig3 domain is spatially separated by the meB-containing linker and positioned apart from the Ig1-Ig2 unit. The following Fn1 and Fn2 domains are aligned in a straight line, similarly to the isolated PTPd Fn1-Fn2 structure, which we determined at 2.0 Å resolution (Supplementary Fig. 1d). The Fn3 domain of PTPd is oriented perpendicular to the Fn1-Fn2 domains, as observed in the PTPs Ig1-Fn3 structure (Supplementary Fig. 1e). Electron density of the Fn4 domain and subsequent C-terminal Varieties of the combinations between pre- and postsynaptic organizers rely upon structural diversities of their ECDs, which can be drastically increased by multiple isoforms and splice variants. In fact, several isoforms and splice variants of type-IIa RPTPs are expressed in mouse brain, and bind selectively to the corresponding postsynaptic organizers5,6. Type-IIa RPTPs commonly have three Ig domains (Ig1–Ig3) in their amino (N)-terminal regions. ARTICLE In mammals, presynaptic type-IIa RPTPs form trans-synaptic adhesion by speciﬁcally binding to their cognate postsynaptic organizers through their ECDs for inducing synaptic differentiation of neurons. To date, ﬁve different types of postsynaptic organizers for type-IIa RPTPs have been reported: interleukin-1 receptor (IL-1R) accessory protein (IL-1RAcP)7, IL-1RAcP-like-1 (IL1RAPL1)8, neurotrophin receptor tyrosine kinase C (TrkC)9, netrin-G ligand-3 (NGL-3)10,11 and Slit- and Trk-like (Slitrk) family proteins12,13. For example, co-culture with either IL-1RAcP- or IL1RAPL1-expressing non-neuronal HEK293T cells can stimulate presynaptic differentiation in contacting axons of cultured cortical neurons, however, the IL1RAPL1- and IL-1RAcP-induced presynaptic differentiations are completely abolished and decreased by B70% in the neurons lacking PTPd, respectively7,8. Similarly, PTPd can induce postsynaptic differentiation of cultured cortical neurons in a manner speciﬁc to IL1RAPL1 and IL-1RAcP (more precisely, a central nervous system-restricted IL-1RAcP isoform, which differs only in the carboxy (C)-terminal from a widely distributed isoform). On the other hand, Slitrk family proteins induce presynaptic differentiation through the interaction with type-IIa RPTPs12,13. Among Slitrk family proteins, Slitrk3 selectively induces inhibitory synapse formation, suggesting potential roles of synapse organizers in balancing excitatory and inhibitory synapses12. Various combinations of trans-synaptic adhesion between pre- and postsynaptic organizers might play an important role in generating extremely diverged but highly organized synaptic connections of neurons. f h b b d NATURE COMMUNICATIONS \| 6:6926 \| DOI: 10.1038/ncomms7926 \| www.nature.com/naturecommunications & 2015 Macmillan Publishers Limited. All rights reserved. Mechanisms of splicing-dependent trans-synaptic adhesion by PTPd–IL1RAPL1/IL-1RAcP for synaptic differentiation Atsushi Yamagata1,2,3, Tomoyuki Yoshida4,5,6, Yusuke Sato1,2,3, Sakurako Goto-Ito1, Takeshi Uemura3,5,7,8, Asami Maeda1,3, Tomoko Shiroshima1,3, Shiho Iwasawa-Okamoto4,6, Hisashi Mori4, Masayoshi Mishina5,9 & Shuya Fukai1,2,3 Synapse formation is triggered through trans-synaptic interaction between pairs of pre- and postsynaptic adhesion molecules, the speciﬁcity of which depends on splice inserts known as ‘splice-insert signaling codes’. Receptor protein tyrosine phosphatase d (PTPd) can bidirectionally induce pre- and postsynaptic differentiation of neurons by trans-synaptically binding to interleukin-1 receptor accessory protein (IL-1RAcP) and IL-1RAcP-like-1 (IL1RAPL1) in a splicing-dependent manner. Here, we report crystal structures of PTPd in complex with IL1RAPL1 and IL-1RAcP. The ﬁrst immunoglobulin-like (Ig) domain of IL1RAPL1 directly recognizes the ﬁrst splice insert, which is critical for binding to IL1RAPL1. The second splice insert functions as an adjustable linker that positions the Ig2 and Ig3 domains of PTPd for simultaneously interacting with the Ig1 domain of IL1RAPL1 or IL-1RAcP. We further identiﬁed the IL1RAPL1-speciﬁc interaction, which appears coupled to the ﬁrst-splice-insert-mediated interaction. Our results thus reveal the decoding mechanism of splice-insert signaling codes for synaptic differentiation induced by trans-synaptic adhesion between PTPd and IL1RAPL1/IL-1RAcP. 1 Structural Biology Laboratory, Life Science Division, Synchrotron Radiation Research Organization and Institute of Molecular and Cellular Biosciences, The University of Tokyo, Tokyo 113-0032, Japan. 2 Department of Medical Genome Sciences, Graduate School of Frontier Sciences, The University of Tokyo, Chiba 277-8501, Japan. 3 CREST, JST, Saitama 332-0012, Japan. 4 Department of Molecular Neuroscience, Graduate School of Medicine and Pharmaceutical Sciences, University of Toyama, Toyama 930-0194, Japan. 5 Department of Molecular Neurobiology and Pharmacology, Graduate School of Medicine, The University of Tokyo, Tokyo 113-0033, Japan. 6 PRESTO, JST, Saitama 332-0012, Japan. 7 Department of Molecular and Cellular Physiology, Shinshu University School of Medicine, Nagano 390-8621, Japan. 8 Institute for Biomedical Sciences, Interdisciplinary Cluster for Cutting Edge Research, Shinshu University, Nagano 390-8621, Japan. 9 Brain Science Laboratory, The Research Organization of Science and Technology, Ritsumeikan University, Shiga 525-8577, Japan. Correspondence and requests for materials should be addressed to T.Y. (email: toyoshid@med.u-toyama.ac.jp) or to S.F. (email: fukai@iam.u-tokyo.ac.jp). NATURE COMMUNICATIONS \| 6:6926 \| DOI: 10.1038/ncomms7926 \| www.nature.com/naturecommunications 1 CATIONS \| 6:6926 \| DOI: 10.1038/ncomms7926 \| www.nature.com/nat & 2015 Macmillan Publishers Limited. All rights reserved. Results meB Ig1 Ig2 Ig3 Fn1 Fn2 Fn3 PTPδ-ECD Ig3 IL1RAPL1-ECD meA9 Ig1 Ig2 Ig3 Ig2 Ig1 PTPδ Ig1–Ig2 meA9 Ig1 Ig2 IL1RAPL1-ECD Ig domain Fn domain PTPδ Ig1 Ig2 Ig3 Fn1 Fn2 Fn3 Fn4 meB meA PTPδ Ig1–Ig2 IL1RAPL1 IL1RAPL1-ECD Ig1 Ig2 Ig3 TIR TM CT Q200 E188 S187 Y59 W34 R196 D37 D33 L185 L153 G58 PTPδ Ig2 meA9 IL1RAPL1 Ig1 T314 P270 Y273 I291 Y77 M75 P88 F91 PTPδ Ig3 IL1RAPL1 Ig1 M271 D A198 IL1RAPL1-ECD T314 P270 Y273 I291 Y77 M75 P88 F91 PTPδ Ig3 IL1RAPL1 Ig1 M271 F289 E291 D292 E337 R75 R98 R124 R95 PTPδ Ig1 IL1RAPL1 Ig3 Figure 1 \| Structure of the complex between PTPd and IL1RAPL1. (a) Domain organizations of PTPd and IL1RAPL1. The meA and meB insertion sites of PTPd are indicated by magenta and orange triangles, respectively. (b) Overall structures of the PTPd Ig1–Ig2IL1RAPL1-ECD and PTPd-ECDIL1RAPL1-ECD complexes. PTPd and IL1RAPL1 are coloured green and cyan, respectively. The meA9 and meB insertions are highlighted in magenta and orange, respectively. N-linked glycans are shown as sticks. (c) Interactions between the Ig1 domain of IL1RAPL1 and the Ig2 domain of PTPd. The colouring scheme is the same as that in b, except that IL-1RAcP/IL1RAPL1-Ig1-interacting b-strands of PTPd are coloured brown. Residues involved in the interactions are shown as sticks. Hydrogen bonds are indicated by dotted lines. (d) Hydrophobic interactions between the Ig1 domain of IL1RAPL1 and the Ig3 domain of PTPd. The representation scheme is the same as that in c. (e) Interactions between the Ig3 domain of IL1RAPL1 and the Ig1 domain of PTPd. The representation scheme is the same as that in c. form a loop connecting these two b-strands, whereas the last residue, Arg196, extends one of the two b-strands. Trp34 of IL1RAPL1 hydrophobically interacts with Leu153, Ala198 and Leu185 of PTPd (Fig. 1c). The afﬁnity of the IL1RAPL1 W34A mutant for PTPd could not be measured by surface-plasmon resonance (SPR) spectroscopy, due to substantially low signals, indicating that this hydrophobic interaction is essential for binding between PTPd and IL1RAPL1. The aliphatic portion of PTPd Arg196 also participates in this hydrophobic interaction, which enables the Ig2 domain of PTPd to more tightly interact with the Ig1 domain of IL1RAPL1 (Fig. 1c). The Arg196 side chain forms a hydrogen bond with Asp37 of IL1RAPL1, which is further stabilized by a hydrogen bond with Tyr59 of IL1RAPL1. Results The ﬁrst splicing site is located within Ig2, whereas the second one is in the junction between Ig2 and Ig3. These two splicing sites generate variants of the type-IIa RPTPs that lack or contain short-peptide inserts termed mini-exon peptides. A recent study has shown that Slitrk family members bind to all type-IIa RPTP members, depending on the second mini-exon peptide (meB)14. On the other hand, IL1RAPL1 and IL-1RAcP speciﬁcally bind to PTPd but not to LAR or PTPs. Furthermore, our previous cell-surface binding assays showed that both the ﬁrst mini-exon peptide (meA) and meB can alter their binding properties7,8. MeB comprises four residues, NATURE COMMUNICATIONS \| 6:6926 \| DOI: 10.1038/ncomms7926 \| www.nature.com/naturecommunications & 2015 Macmillan Publishers Limited. All rights reserved. 2 ARTICLE NATURE COMMUNICATIONS \| DOI: 10.1038/ncomms7926 meB Ig1 Ig2 Ig3 Fn1 Fn2 Fn3 PTPδ-ECD Ig3 IL1RAPL1-ECD meA9 Ig1 Ig2 Ig3 Ig2 Ig1 PTPδ Ig1–Ig2 meA9 Ig1 Ig2 IL1RAPL1-ECD Q200 E188 S187 Y59 W34 R196 D37 D33 L185 L153 G58 PTPδ Ig2 meA9 F289 E291 D292 E337 R75 R98 R124 R95 PTPδ Ig1 IL1RAPL1 Ig1 IL1RAPL1 Ig3 T314 P270 Y273 I291 Y77 M75 P88 F91 PTPδ Ig3 IL1RAPL1 Ig1 M271 Ig domain Fn domain PTPδ Ig1 Ig2 Ig3 Fn1 Fn2 Fn3 Fn4 TM D1 D2 Phosphatase domain meB meA PTPδ-ECD PTPδ Ig1–Ig2 IL1RAPL1 IL1RAPL1-ECD Ig1 Ig2 Ig3 TIR TM CT A198 Figure 1 \| Structure of the complex between PTPd and IL1RAPL1. (a) Domain organizations of PTPd and IL1RAPL1. The meA and meB insertion sites of PTPd are indicated by magenta and orange triangles, respectively. (b) Overall structures of the PTPd Ig1–Ig2IL1RAPL1-ECD and PTPd-ECDIL1RAPL1-ECD complexes. PTPd and IL1RAPL1 are coloured green and cyan, respectively. The meA9 and meB insertions are highlighted in magenta and orange, respectively. N-linked glycans are shown as sticks. (c) Interactions between the Ig1 domain of IL1RAPL1 and the Ig2 domain of PTPd. The colouring scheme is the same as that in b, except that IL-1RAcP/IL1RAPL1-Ig1-interacting b-strands of PTPd are coloured brown. Residues involved in the interactions are shown as sticks. Hydrogen bonds are indicated by dotted lines. (d) Hydrophobic interactions between the Ig1 domain of IL1RAPL1 and the Ig3 domain of PTPd. The representation scheme is the same as that in c. (e) Interactions between the Ig3 domain of IL1RAPL1 and the Ig1 domain of PTPd. The representation scheme is the same as that in c. & 2015 Macmillan Publishers Limited. All rights reserved. NATURE COMMUNICATIONS \| 6:6926 \| DOI: 10.1038/ncomms7926 \| www.nature.com/naturecommunications & 2015 Macmillan Publishers Limited. All rights reserved. Results Concomitantly, the side chain of Ser187 and the main-chain N atom of Glu188 in meA9 hydrogen bond with the main-chain O atoms of IL1RAPL1 Tyr59 and Gly58, respectively. The interac- tions mediated by Arg196 of PTPd are critically important: the R196A mutation of PTPd and the D37A mutation of region was not visible, likely due to the disorder. The length of the PTPd-ECDIL1RAPL1-ECD complex along the longest axis is 206 Å, which is equivalent to the average length of excitatory synaptic clefts, implying that PTPd-ECD is kinked at the junction between the Fn2 and Fn3 domains in synapses. Interactions between IL1RAPL1 Ig1 and PTPd Ig2 domains. The Ig1 domain of IL1RAPL1 interacts with both Ig2 and Ig3 domains of PTPd (Fig. 1b). The interface between the IL1RAPL Ig1 and PTPd Ig2 domains comprises both hydrophobic and hydrophilic interactions with a buried surface area of 699 Å2. These interactions occur on two b-strands comprising Arg181– Ser187 and Arg196–Glu202 in the Ig2 domain of PTPd, between which meA9 is inserted (Fig. 1c). These two b-strands are hereafter referred to as the IL1RAPL1/IL-1RAcP-Ig1- interacting b-strands, because they also mediate the interaction with the Ig1 domain of IL-1RAcP as described later. The ﬁrst eight residues of meA9 (188Glu-Ser-Ile-Gly-Gly-Thr-Pro-Ile195) 3 ARTICLE NATURE COMMUNICATIONS \| DOI: 10.1038/ncomms7926 Table 1 \| Data collection and reﬁnement statistics. Table 1 \| Data collection and reﬁnement statistics. PTPd Ig1–Ig2 IL1RAPL1-ECD PTPd-ECD IL1RAPL1-ECD PTPd Fn1–Fn2 PTPdA3B– Ig1–Fn2 PTPd Ig1–Fn2 IL-1RAcP-ECD IL1RAPL1-ECD Data collection Space group C2 I41 P212121 P1 P2 P42212 Cell dimensions a, b, c (Å) 163.0, 81.2, 91.5 286.3, 286.3, 70.2 51.9, 65.1, 134.0 70.5, 72.6, 94.4 77.5, 63.3, 169.4 102.5, 102.5, 222.7 a, b, g () 90.0, 91.2, 90.0 90.0, 90.0, 90.0 90.0, 90.0, 90.0 107.6, 94.4, 108.5 90.0, 94.2, 90.0 90.0, 90.0, 90.0 Resolution (Å) 50–2.70 (2.75–2.70) 50–4.40 (4.48–4.40) 50.0–1.97 (2.00–1.97) 50.0–3.50 (3.56–3.50) 50.0–3.25 (3.31–3.25) 50–3.00 (3.05–3.00) Rsym 0.12 (0.43) 0.07 (0.42) 0.12 (0.33) 0.19 (0.49) 0.12 (0.44) 0.12 (0.38) I/sI 12.2 (2.0) 17.7 (1.7) 26.6 (4.4) 4.8 (1.6) 10.7 (2.0) 10.3 (2.3) Completeness (%) 98.7 (95.9) 95.6 (86.0) 99.8 (99.7) 91.4 (84.2) 97.3 (92.5) 97.5 (93.4) Redundancy 5.1 (3.2) 4.2 (2.2) 10.7 (7.7) 2.5 (1.9) 4.7 (2.9) 7.1 (3.3) Reﬁnement Resolution (Å) 2.70 4.4 1.97 3.50 3.25 3.00 No. reﬂections 32,256 17,618 32,731 18,642 25,436 23,998 Rwork/Rfree 0.226/0.263 0.291/0.327 0.204/0.223 0.291/0.315 0.256/0.287 0.264/0.298 No. Results of atoms Protein 4,208 7,124 2,988 7,412 6,407 4,965 Sugar 143 117 56 98 325 Water 77 308 B-factors (Å2) Protein 66.90 147.10 28.90 138.50 107.90 64.70 Sugar 106.50 147.90 147.80 121.10 96.80 Water 62.90 34.70 R.m.s. deviations Bond lengths (Å) 0.005 0.006 0.006 0.005 0.005 0.004 Bond angles () 1.28 1.55 1.32 1.15 1.07 0.81 Values in parentheses are for highest-resolution shell. Table 3 \| Afﬁnities (KD) of PTPd or IL1RAPL1 mutants for binding between PTPd and IL1RAPL1. IL1RAPL1 (lM) PTPd WT 0.15±0.011 R75A 2.8±0.34 R196A 1.7±0.45 Y273A 0.86±0.11 meB GSSG 0.14±0.022 QLEQ 0.16±0.0026 ELREELRE 0.15±0.052 ELREELREELRE 0.42±0.038 PTPd (lM) IL1RAPL1 WT 0.15±0.011 W34A ND D37A 2.4±0.016 M75A/Y77A/P88A/F91A 2.8±0.48 D292A 2.2±0.075 ND, not detectable; WT, wild type. Data are presented as mean±s.d. Table 2 \| Afﬁnities (KD) between the splicing variants of PTPd and IL1RAPL1/IL-1RAcP. PTPd (meA/meB) IL1RAPL1 (lM) IL-1RAcP (lM) meA9/ þ 0.15±0.011 0.68±0.013 meA9/ 0.80±0.11 2.0±0.085 meA6/ þ 0.81±0.14 0.73±0.013 meA6/ ND 2.3±0.016 meA3/ þ ND 0.51±0.010 meA3/ ND 2.8±0.053 / þ ND 2.2±0.075 / ND 2.4±0.075 ND, not detectable. Data are presented as mean±s.d. Table 2 \| Afﬁnities (KD) between the splicing variants of PTPd and IL1RAPL1/IL-1RAcP. IL1RAPL1 decreased the afﬁnity 11- and 16-fold, respectively (Table 3). The meA6-containing PTPd variant also contains Arg196 but has a ﬁvefold reduced afﬁnity for IL1RAPL1 (Table 2). Therefore, this afﬁnity reduction of the meA6-con- taining variant might be due to a lack of the concomitant hydrogen bonding. The atomic-level meA-speciﬁc interactions described here illustrate how the meA insertion contributes to the afﬁnity between PTPd and IL1RAPL1. and Pro270, Met271, Ile291 and Thr314 of PTPd. The Y273A mutation of PTPd decreased the afﬁnity for IL1RAPL1 sixfold, whereas the M75A/Y77A/P88A/F91A quadruple mutation of IL1RAPL1 decreased the afﬁnity for PTPd 19-fold (Table 3). Relationship of meB with IL1RAPL1 Ig1–PTPd Ig3 interaction. On the opposite side to the IL1RAPL1 Ig1–PTPd Ig2 interface, the Ig1 domain of IL1RAPL1 interacts with the Ig3 domain of PTPd with a buried surface area of 694 Å2. The IL1RAPL1 Ig1– PTPd Ig3 interface is stabilized by hydrophobic interactions (Fig. 1d). The side chains of IL1RAPL1 Tyr77 and PTPd Tyr273 are aligned in an antiparallel manner to form a hydrophobic core, which is surrounded by Met75, Pro88 and Phe91 of IL1RAPL1 y The meB insertion of PTPd is important for binding between PTPd and IL1RAPL1. NATURE COMMUNICATIONS \| 6:6926 \| DOI: 10.1038/ncomms7926 \| www.nature.com/naturecommunications & 2015 Macmillan Publishers Limited. All rights reserved. NATURE COMMUNICATIONS \| 6:6926 \| DOI: 10.1038/ncomms7926 \| www.nature.com/naturecommunications Results The lack of meB reduces the afﬁnity ﬁvefold (Table 2). In the present PTPd-ECDIL1RAPL1-ECD complex, the meB insertion appears to adjust the relative spacing 4 4 ARTICLE NATURE COMMUNICATIONS \| DOI: 10.1038/ncomms7926 and orientation between the Ig2 and Ig3 domains of PTPd (Fig. 1b). When the apo structure of PTPd Ig1-Fn2 containing meA3 and lacking meB (PTPdA3B ) (Supplementary Fig. 1f), which we determined at 3.5 Å resolution, is superposed onto the IL1RAPL1-ECD-bound structure of PTPd Ig1-Fn2 (Supplementary Fig. 1g) using the Ig1–Ig2 unit of PTPd as the reference, the Ig3 domain of the apo PTPdA3B is oriented in the opposite direction from that of the IL1RAPL1-ECD-bound PTPd and cannot interact with IL1RAPL1, even if structural ﬂexibility of the linker between the Ig2 and Ig3 domains of PTPd is assumed (Supplementary Fig. 1g). Replacement of the native meB (Glu-Leu-Arg-Glu) by two tandem units of meB or a Gly-Ser-Ser-Gly or Gln-Leu-Glu-Gln tetrapeptide hardly affected the afﬁnity, whereas that by three tandem units reduced the afﬁnity threefold (Table 3). Insertions of four or more residues at the meB position are sufﬁcient for binding to IL1RAPL1, but the longer insertion prevents the efﬁcient binding. Therefore, the meB insertion likely acts as an adjustable linker to locate the Ig3 domain of PTPd in the appropriate position for interacting with the Ig1 domain of IL1RAPL1, representing how the meB insertion contributes to the afﬁnity between PTPd and IL1RAPL1. Interactions between IL1RAPL1 Ig3 and PTPd Ig1 domains. The Ig3 domain of IL1RAPL1 interacts with the Ig1 domain of PTPd with a buried surface area of 433 Å2. The IL1RAPL1 Ig3– Ig domain Fn domain PTPδ Ig1 Ig2 Ig3 Fn1 Fn2 Fn3 Fn4 TM D1 D2 Phosphatase domain meB meA PTPδ Ig1-Fn2 IL-1RAcP IL-1RAcP-ECD Ig1 Ig2 Ig3 TIR TM CT PTPδ Ig1-Fn2 IL-1RAcP-ECD IL-1RAcP Ig1 IL-1RAcP Ig1 PTPδ Ig2 PTPδ Ig3 PTPδ Ig3 IL-1RAcP Ig1 T314 M271 Y273 I291 P270 Y71 W70 F85 I69 P82 meA9 meB Ig1 Ig2 Ig3 Fn1 Fn2 Ig3 Ig2 Ig1 W27 F53 C24 D26 E286 D287 D288 Y58 T67 K275 R86 K94 E188 L185 L153 Q200 M289 A198 meA9 IL-1RAcP-ECD Figure 2 \| Structure of the complex between PTPd and IL-1RAcP. (a) Domain organizations of PTPd and IL-1RAcP. The meA and meB insertion sites of PTPd are indicated by magenta and orange triangles, respectively. (b) Overall structure of the PTPd Ig1–Fn2IL-1RAcP-ECD complex. PTPd and IL-1RAcP are coloured green and blue, respectively. Results The meA9 and meB insertions are highlighted in magenta and orange, respectively. N-linked glycans are shown as sticks. (c) Interactions between the Ig1 domain of IL-1RAcP and the Ig2 domain of PTPd. The colouring scheme is the same as that in b, except that IL1RAPL1/IL-1RAcP-Ig1-interacting b-strands of PTPd are coloured brown. Residues involved in the interactions are shown as sticks. Hydrogen bonds are indicated by dotted lines. (d) Hydrophobic interactions between the Ig1 domain of IL-1RAcP and the Ig3 domain of PTPd. The representation scheme is the same as that in c. (e) Hydrophilic interactions between the Ig1 domain of IL-1RAcP and the Ig3 domain of PTPd. The representation scheme is the same as that in c. PTPδ Ig1-Fn2 IL-1RAcP-ECD meA9 meB Ig1 Ig2 Ig3 Fn1 Fn2 Ig3 Ig2 Ig1 Fn1 IL-1RAcP-ECD Figure 2 \| Structure of the complex between PTPd and IL-1RAcP. (a) Domain organizations of PTPd and IL-1RAcP. The meA and meB insertion sites of PTPd are indicated by magenta and orange triangles, respectively. (b) Overall structure of the PTPd Ig1–Fn2IL-1RAcP-ECD complex. PTPd and IL-1RAcP are coloured green and blue, respectively. The meA9 and meB insertions are highlighted in magenta and orange, respectively. N-linked glycans are shown as sticks. (c) Interactions between the Ig1 domain of IL-1RAcP and the Ig2 domain of PTPd. The colouring scheme is the same as that in b, except that IL1RAPL1/IL-1RAcP-Ig1-interacting b-strands of PTPd are coloured brown. Residues involved in the interactions are shown as sticks. Hydrogen bonds are indicated by dotted lines. (d) Hydrophobic interactions between the Ig1 domain of IL-1RAcP and the Ig3 domain of PTPd. The representation scheme is the same as that in c. (e) Hydrophilic interactions between the Ig1 domain of IL-1RAcP and the Ig3 domain of PTPd. The representation scheme is the same as that in c. NATURE COMMUNICATIONS \| 6:6926 \| DOI: 10.1038/ncomms7926 \| www.nature.com/naturecommunications NATURE COMMUNICATIONS \| 6:6926 \| DOI: 10.1038/ncomms7926 \| www.nature.com/naturecommunications 5 NATURE COMMUNICATIONS \| 6:6926 \| DOI: 10.1038/ncomms7926 \| www.nature.com/naturecommunications & 2015 Macmillan Publishers Limited. All rights reserved. & 2015 Macmillan Publishers Limited. All rights reserved. ARTICLE ARTICLE NATURE COMMUNICATIONS \| DOI: 10.1038/ncomms7926 the Ig1 domain of PTPd is not conserved in IL-1RAcP (Supplementary Fig. 2e). the Ig1 domain of PTPd is not conserved in IL-1RAcP (Supplementary Fig. 2e). the Ig1 domain of PTPd is not conserved in IL-1RAcP (Supplementary Fig. 2e). PTPd Ig1 interface is overlapped with the putative binding pocket for glycosaminoglycan chains of heparan sulfate and chondroitin sulfate proteoglycans (Supplementary Fig. 1h), which mediate axonal growth control though the interaction with the type-IIa RPTPs19. This third interface is formed primarily by hydrophilic interactions (Fig. 1e): a positively charged cluster comprising Arg75, Arg95 and Arg98 of PTPd forms a hydrogen bond network with a negatively charged cluster comprising Glu291 and Asp292 of IL1RAPL1. This network is stabilized by a cation-p interaction between Phe289 of IL1RAPL1 and Arg98 of PTPd. In addition, Arg124 of PTPd forms a hydrogen bond with Glu337 of IL1RAPL1. To conﬁrm the importance of this interface, IL1RAPL1 Asp292–PTPd Arg75 interacting pair was disrupted by Ala replacement. The D292A mutation of IL1RAPL1 and the R75A mutation of PTPd reduced the afﬁnity between IL1RAPL1 and PTPd 15- and 19-fold, respectively (Table 3). When comparing the amino-acid sequences between IL1RAPL1 and IL-1RAcP, the Ig3 domain is the least conserved domain (23% identity) among three Ig domains. As described below, the Ig3 domain of IL-1RAcP does not interact with PTPd in the PTPdIL-1RAcP complex. Therefore, the negatively charged cluster of the IL1RAPL1 Ig3 domain and the positively charged cluster of the PTPd Ig1 domain are another determinant for selective binding between PTPd and IL1RAPL1. Interactions between IL-1RAcP Ig1 and PTPd Ig2. Similarly to the PTPdIL1RAPL1 complex, interactions between the Ig1 domain of IL-1RAcP and the Ig2 domain of PTPd occur on the IL1RAPL1/IL-1RAcP-Ig1-interacting b-strands (Fig. 2c). The interface between the IL-1RAcP Ig1 and PTPd Ig2 domains comprises both hydrophobic and hydrophilic interactions with a buried surface area of 726 Å2. Trp27 of IL-1RAcP hydro- phobically interacts with Leu153, Ala198 and Leu185 of PTPd (Fig. 2c). The W27A mutation of IL-1RAcP reduced the afﬁnity sevenfold (Table 4). Although Arg196 of PTPd is positioned in close proximity to Asp30 of IL-1RAcP, no observed electron density of its side chain guanidino group indicates that the interaction between Arg196 of PTPd and Asp30 of IL-1RAcP is substantially weak (Supplementary Fig. 2f). Thus, the D30A mutation of IL-1RAcP and the R196A mutation of PTPd hardly affected the afﬁnity (Table 4). ARTICLE Basically, meA does not extensively interact with IL-1RAcP in the PTPd Ig1-Fn2IL-1RAcP-ECD complex, consistent with the ﬁnding that the variation of meA does not affect the afﬁnity for IL-1RAcP (Table 2). Only one meA-mediated interaction is a hydrogen bond between the main-chain N atom of PTPd Glu188 and the main-chain O atom of IL-1RAcP Phe53, which could be formed in the shorter mini-exon variants, meA6 and meA3, but not in the variant lacking meA. The complete deletion of meA shortens the IL1RAPL1/IL-1RAcP-Ig1-interacting b-strands, which might disturb and weaken the hydrophobic interaction of IL-1RAcP Trp27 with Leu153, Ala198 and Leu185 of PTPd, resulting in the threefold reduced afﬁnity to IL-1RAcP (Table 2; compare meA9/ þ with / þ ). These structural features may reﬂect that the meA insertion increases the afﬁnity between PTPd and IL-1RAcP, independently of its variation. Structure of the complex between PTPd and IL-1RAcP. IL-1RAcP, a member of the same IL-1R family as IL1RAPL1 (B30% sequence identity), also functions as a postsynaptic organizer for PTPd. However, the afﬁnity of IL-1RAcP for PTPd (Kd: 0.68 mM) is ﬁvefold lower than that of IL1RAPL1 (Table 2). The meA variation hardly affects the afﬁnity. In addition, IL-1RAcP can bind to the PTPd variant lacking both meA and meB (Kd: 2.4 mM) in contrast to IL1RAPL1. To structurally explain these differences between IL1RAPL1 and IL-1RAcP, we also determined the crystal structure of IL-1RAcP-ECD in com- plex with PTPd containing three Ig domains and two Fn domains (Ig1-Fn2) at 3.25 Å resolution (Fig. 2a,b, Table 1). We had con- ﬁrmed that PTPd Ig1-Fn2 efﬁciently binds to IL-1RAcP-ECD by testing various lengths of PTPd (Supplementary Fig. 2a). The Ig1–Ig2 unit of PTPd is insufﬁcient for binding to IL-1RAcP. Similarly to the PTPdIL1RAPL1 complex, the Ig1-Fn2 domains of PTPd bind to IL-1RAcP-ECD at a ratio of 1:1 (Fig. 2b). The asymmetric unit contains one PTPdIL-1RAcP complex (Supplementary Fig. 2b). The crystal packing patterns of the PTPdIL1RAPL1 and PTPdIL-1RAcP complexes in this study are totally different from each other and not suggestive of their lateral associations in synapses. Although it can be assumed that activations of PTPd and IL1RAPL1/IL-1RAcP are mediated by a ligand-induced monomer–dimer transition by analogy with those of receptor tyrosine kinases20–23 and Toll/IL-1-receptor families24, our crystallographic and multiangle light scattering analyses using the ECDs of PTPd, IL-1RAcP and IL1RAPL1 could not support this activation mechanism (Supplementary Fig. 1b, 2b,c). NATURE COMMUNICATIONS \| 6:6926 \| DOI: 10.1038/ncomms7926 \| www.nature.com/naturecommunications Discussion l Recently, crystal structures of two different trans-synaptic type-IIa RPTP complexes have been reported14,25. One is the complex between Ig1–Ig2 of PTPs and leucine-rich repeat (LRR)-Ig1of TrkC25, while the other is that between Ig1–Ig3 of PTPd and the ﬁrst LRR (LRR1) of Slitrk1 (ref. 14, Supplementary Fig. 3a). These and our type-IIa RPTP complex structures illustrate that the type-IIa RPTPs can be recognized by structurally distinct architectures. The TrkC- or Slitrk1- interacting surface of the RPTPs is separated from the IL1RAPL1- or IL-1RAcP-interacting surface, except that a part of the TrkC-interacting surface is overlapped with the IL1RAPL1- interacting surface. The conserved arginine residues in Ig1 of the RPTPs (that is, Arg95 and Arg98 of PTPd and Arg96 and Arg99 of PTPs) are recognized by both IL1RAPL1 and TrkC through similar hydrogen bond interactions (Supplementary Fig. 3b). Despite the separation of these binding surfaces of the RPTPs, pair-wise superposition between any two of these complex structures shows steric clashes between the bound postsynaptic partners, ruling out their simultaneous binding to the type-IIa RPTPs (Supplementary Fig. 3c). The IL1RAPL1- or IL-1RAcP- interacting residues of PTPd are mostly conserved in the type-IIa RPTPs. Their differences in the speciﬁcity to IL1RAPL1 or IL-1RAcP appear to be dependent exclusively on the variation of meA and meB. Our previous analysis of Ptprs and Ptprf cDNA from the developing mouse brain suggested that none of the PTPs variants contains the mini-exon peptide insertion at the meA site8 and that 90% of LAR lacks the peptide insertion at the meB site8. Therefore, IL1RAPL1 and IL-1RAcP induce synaptogenesis speciﬁcally through PTPd, at least in the developing brain. On the other hand, the meB insertion is sufﬁcient for binding to Slitrks. They may regulate synaptogenesis mainly through PTPs or PTPd in the developing brain13,14, although they can potentially bind to all the type-IIa RPTP variants containing meB. Both the regulation of meA and meB choice by alternative splicing of the type-IIa RPTPs and the mutually exclusive binding among postsynaptic ligands for the type-IIa RPTPs may contribute to sharpening target speciﬁcity of central synaptogenesis. Replacement of the native meB (Glu-Leu-Arg-Glu) by a Gly-Ser-Ser-Gly or Gln-Leu-Glu-Gln tetrapeptide hardly affected the afﬁnity for IL-1RAcP, whereas that by two or three tandem units reduced the afﬁnity two- or sixfold, respectively (Table 4). ARTICLE The Ig1-Fn2 domains of PTPd exhibit an elongated shape with a length of 160 Å along the longest axis. Although overall shape of the IL-1RAcP-bound PTPd Ig1-Fn2 resembles that of the IL1RAPL1-bound PTPd Ig1-Fn2, the position and/or orientation of each domain are somewhat different between them. The Ig1–Ig3 domains of IL-1RAcP are arranged in an L-shape similar to those of IL1RAPL1, except that the relative orientation between the Ig1 and Ig2–Ig3 domains is changed by 20 degree (Supplementary Fig. 2d). Correspondingly, the Ig3 domain of IL-1RAcP does not interact with PTPd in contrast to that of IL1RAPL1. Superposition between the Ig3 domains of IL1RAPL1 and IL-1RAcP shows that the binding surface of IL1RAPL1 for Relationship of meB with IL-1RAcP Ig1–PTPd Ig3 interaction. Relationship of meB with IL-1RAcP Ig1–PTPd Ig3 interaction. The position and orientation of the PTPd Ig3 domain relative to the IL-1RAcP Ig1 domain are similar to those relative to the IL1RAPL1 Ig1 domain (Figs 1b and 2b). The interface between Table 4 \| Afﬁnities (KD) of PTPd or IL-1RAcP mutants for binding between PTPd and IL-1RAcP. IL-1RAcP (lM) PTPd WT 0.68±0.013 R75A 1.1±0.023 R196A 0.89±0.17 Y273A ND E286A 5.3±0.52 meB GSSG 0.64±0.036 QLEQ 0.91±0.019 ELREELRE 1.5±0.038 ELREELREELRE 3.8±0.21 PTPd (lM) IL-1RAcP WT 0.68±0.013 W27A 4.6±0.076 D30A 0.62±0.033 K94A 8.2±0.054 I69A/Y71A/P82A/F85A ND ND, not detectable; WT, wild type. Data are presented as mean±s.d. Table 4 \| Afﬁnities (KD) of PTPd or IL-1RAcP mutants for binding between PTPd and IL-1RAcP. IL-1RAcP (lM) PTPd WT 0.68±0.013 R75A 1.1±0.023 R196A 0.89±0.17 Y273A ND E286A 5.3±0.52 meB GSSG 0.64±0.036 QLEQ 0.91±0.019 ELREELRE 1.5±0.038 ELREELREELRE 3.8±0.21 PTPd (lM) IL-1RAcP WT 0.68±0.013 W27A 4.6±0.076 D30A 0.62±0.033 K94A 8.2±0.054 I69A/Y71A/P82A/F85A ND ND, not detectable; WT, wild type. Data are presented as mean±s.d. 6 & 2015 Macmillan Publishers Limited. All rights reserved. ARTICLE NATURE COMMUNICATIONS \| DOI: 10.1038/ncomms7926 R75A mutation in the Ig1 domain of PTPd reduced the synaptogenic activity to B7%. This indicates that the interaction between the IL1RAPL1 Ig3 and PTPd Ig1 domains is critically important for inducing postsynaptic differentiation. In contrast to the afﬁnity of IL1RAPL1, that of IL-1RAcP is not so consistently correlated with synaptogenic activities of PTPd mutants, possibly due to a lack of the interaction between the IL-1RAcP Ig3 and PTPd Ig1 domains. ARTICLE The inside of the interface is stabilized by more tightly packed hydrophobic interactions than that of the IL1RAPL1 Ig1– PTPd Ig3 interface (Figs 1d and 2d). The side chains of IL-1RAcP Tyr71 and PTPd Tyr273 are stacked in an antiparallel manner to form a hydrophobic core, which is surrounded by the side chains of IL-1RAcP Phe85, Pro82 and Ile69 and PTPd Pro270 and Ile291 and the aliphatic portion of PTPd Thr314. The side chains of IL-1RAcP Tyr71 and PTPd Tyr273 are stabilized by hydrogen bonds with the main-chain O atoms of PTPd Met271 and IL-1RAcP Trp70, respectively. Furthermore, the Ig1 domain of IL-1RAcP recognizes a negatively charged loop comprising Glu286, Asp287 and Asp288 of PTPd. Tyr58 and Arg86 of IL-1RAcP hydrogen bond with Glu286 and Asp288 of PTPd, respectively, whereas Lys94 of IL-1RAcP hydrogen bonds with both Glu286 and Asp287 of PTPd (Fig. 2e). Conformation of this negatively charged loop is supported by Met289 of PTPd. In addition, Thr67 of IL-1RAcP hydrogen bonds with Lys275 of PTPd. Mutations of the IL-1RAcP Ig1–PTPd Ig3 interface impaired the binding afﬁnity more drastically than those of the IL1RAPL1 Ig1–PTPd Ig3 interface: the Y273A mutation of PTPd and the I69A/Y71A/P82A/F85A quadruple mutation of IL-1RAcP completely eliminated the binding ability, whereas the K94A mutation of IL-1RAcP and the E286A mutation of PTPd decreased the afﬁnity 12- and 8-fold, respectively (Table 4). Therefore, the interface mediated by PTPd Ig3 is more critically important for binding to IL-1RAcP than for binding to IL1RAPL1, consistent with the ﬁnding that PTPd Ig3 is essential for binding to IL-1RAcP but not to IL1RAPL1 (Supplementary Figs 1a and 2a). & 2015 Macmillan Publishers Limited. All rights reserved. ARTICLE Considering that IL-1RAcP knockout neurons retain nearly full activity of PTPd-induced postsynaptic differentiation, we speculate that the PTPdIL-1RAcP complex might function as a unidirectional organizer that is primarily responsible for inducing presynaptic differentiation in vivo, although IL-1RAcP can promote postsynaptic differentiation of cultured cortical neurons by binding to the meA3-containing PTPd variant, which is speciﬁc to IL-1RAcP7. The Ig1 domain of PTPd might be universally important for postsynaptic differentiation in vivo, which involves IL1RAPL1 and possibly other unidentiﬁed postsynaptic organizer(s). the IL-1RAcP Ig1 and PTPd Ig3 domains buries a surface area of 906 Å2. The inside of the interface is stabilized by more tightly packed hydrophobic interactions than that of the IL1RAPL1 Ig1– PTPd Ig3 interface (Figs 1d and 2d). The side chains of IL-1RAcP Tyr71 and PTPd Tyr273 are stacked in an antiparallel manner to form a hydrophobic core, which is surrounded by the side chains of IL-1RAcP Phe85, Pro82 and Ile69 and PTPd Pro270 and Ile291 and the aliphatic portion of PTPd Thr314. The side chains of IL-1RAcP Tyr71 and PTPd Tyr273 are stabilized by hydrogen bonds with the main-chain O atoms of PTPd Met271 and IL-1RAcP Trp70, respectively. Furthermore, the Ig1 domain of IL-1RAcP recognizes a negatively charged loop comprising Glu286, Asp287 and Asp288 of PTPd. Tyr58 and Arg86 of IL-1RAcP hydrogen bond with Glu286 and Asp288 of PTPd, respectively, whereas Lys94 of IL-1RAcP hydrogen bonds with both Glu286 and Asp287 of PTPd (Fig. 2e). Conformation of this negatively charged loop is supported by Met289 of PTPd. In addition, Thr67 of IL-1RAcP hydrogen bonds with Lys275 of PTPd. Mutations of the IL-1RAcP Ig1–PTPd Ig3 interface impaired the binding afﬁnity more drastically than those of the IL1RAPL1 Ig1–PTPd Ig3 interface: the Y273A mutation of PTPd and the I69A/Y71A/P82A/F85A quadruple mutation of IL-1RAcP completely eliminated the binding ability, whereas the K94A mutation of IL-1RAcP and the E286A mutation of PTPd decreased the afﬁnity 12- and 8-fold, respectively (Table 4). Therefore, the interface mediated by PTPd Ig3 is more critically important for binding to IL-1RAcP than for binding to IL1RAPL1, consistent with the ﬁnding that PTPd Ig3 is essential for binding to IL-1RAcP but not to IL1RAPL1 (Supplementary Figs 1a and 2a). the IL-1RAcP Ig1 and PTPd Ig3 domains buries a surface area of 906 Å2. NATURE COMMUNICATIONS \| 6:6926 \| DOI: 10.1038/ncomms7926 \| www.nature.com/naturecommunications NATURE COMMUNICATIONS \| 6:6926 \| DOI: 10.1038/ncomms7926 \| www.nature.com/naturecommunications & 2015 Macmillan Publishers Limited. All rights reserved. NATURE COMMUNICATIONS \| 6:6926 \| DOI: 10.1038/ncomms7926 \| www.nature.com/naturecommunications & 2015 Macmillan Publishers Limited. All rights reserved. Discussion l Remarkably, the 7 7 ARTICLE NATURE COMMUNICATIONS \| DOI: 10.1038/ncomms7926 I i i B i l d h j i b i l d i B d l % f i l k Control (EGFP) WT W27A D30A K94A I69A Y71A P82A F85A FLAG Bassoon Merge WT W34A D37A D292A M75A Y77A P88A F91A Control (EGFP) FLAG Bassoon Merge Bassoon signal intensity (a.u.) 120 100 20 0 40 60 80 WT W27A D30A K94A IYPF (69/71/82/85) Control (EGFP) ** Bassoon signal intensity (a.u.) 140 120 100 20 0 40 60 80 WT W34A D37A MYPF (75/77/88/91) D292A Control (EGFP) Shank2 signal intensity (a.u.) 140 120 100 20 0 40 60 80 * ** Control (Fc) WT R75A Y273A E286A R196A Y273A E286A Control (Fc) WT R75A R196A Beads Shank2 aptogenic activities of IL1RAPL1, IL-1RAcP and PTPd mutants. (a,c) Accumulation of Bassoon signals (red) of cortical neurons co-cultured cells expressing FLAG-tagged IL1RAPL1 mutants (green) (a) and IL-1RAcP mutants (green) (c). (b,d) Intensity of staining signals for Bassoon of HEK293T cells expressing FLAG-tagged IL1RAPL1 mutants (b) and FLAG-tagged IL-1RAcP mutants (d) co-cultured with cortical neurons ) Accumulation of Shank2 signals (red) of cortical neurons co-cultured with magnetic beads conjugated with ECDs of PTPd mutants. staining signals for Shank2 on the surface of magnetic beads conjugated with ECDs of PTPd mutants co-cultured with cortical neurons ll values represent mean±s.e.m. Po0.05 and Po0.01 compared with HEK293T cells expressing wild-type IL1RAPL1 (b), wild-type r wild-type PTPd-ECD-Fc coated beads (f); Tukey’s test. Scale bars, 10 mm (a,c) and 5 mm (e). E NATURE COMMUNICATIONS \| DOI: 10.1038/ncomms7926 * Bassoon signal intensity (a.u.) 140 120 100 20 0 40 60 80 WT W34A D37A MYPF (75/77/88/91) D292A Control (EGFP) WT W34A D37A D292A M75A Y77A P88A F91A Control (EGFP) FLAG Bassoon Merge Bassoon signal intensity (a.u.) 120 100 20 0 40 60 80 WT W27A D30A K94A IYPF (69/71/82/85) Control (EGFP) (75/7 Control (EGFP) WT W27A D30A K94A I69A Y71A P82A F85A FLAG Bassoon Merge (69 Shank2 signal intensity (a.u.) 140 120 100 20 0 40 60 80 * ** ontrol Fc) WT R75A 273A 286A 196A Y273A E286A Control (Fc) WT R75A R196A Beads Shank2 Y273A E286A Control (Fc) WT R75A R196A Beads Shank2 Figure 3 \| Synaptogenic activities of IL1RAPL1, IL-1RAcP and PTPd mutants. Discussion l These impacts on the afﬁnity for IL-1RAcP are similar to those for IL1RAPL1, supporting the notion that the meB insertion acts as an adjustable linker to control relative positions and orientations of the PTPd Ig2 and Ig3 domains for their simultaneous interactions with the Ig1 domain of IL-1RAcP. Synaptogenic activity. Finally, we investigated relationships between the structures and synaptogenic activity (that is, IL1RAPL1/IL-1RAcP-induced presynaptic differentiation and PTPd-induced postsynaptic differentiation). IL1RAPL1/IL- 1RAcP mutations were examined by ﬁbroblast-neuron mixed culture assays (Fig. 3): HEK293T cells expressing IL1RAPL1/IL- 1RAcP mutants were co-cultured with cortical neurons and then, accumulation of a presynaptic protein Bassoon in the cortical neurons was analysed (Fig. 3a–d). Because IL1RAPL1- and IL-1RAcP-induced presynaptic differentiation is completely abolished and decreased by B70% in cortical neurons prepared from PTPd-deﬁcient mice, respectively7,8, IL1RAPL1/IL-1RAcP- induced presynaptic differentiation is mostly mediated by the mixed, cognate PTPd variants in neurons. On the other hand, to examine postsynapse-inducing activity of PTPd mutants, magnetic beads conjugated with the ECDs of PTPd mutants were co-cultured with cortical neurons, and accumulation of a postsynaptic protein Shank2 around the beads was analysed (Fig. 3e,f). PTPd-stimulating postsynaptic differentiation is mediated by IL1RAPL1, IL-1RAcP and possibly other unidentiﬁed postsynaptic organizer(s) in neurons. The Ig–Ig interaction between cell-surface receptors is generally important for cell adhesion and communication in immune and neuronal systems. To our knowledge, binding between Ig-containing adhesion receptors is typically mediated by single, homotypic Ig–Ig interactions26–28 in neuronal systems. In contrast, binding between PTPd and IL1RAPL1/IL-1RAcP is mediated by multiple, heterotypic Ig–Ig interactions. The meB insertion of PTPd substantially contributes to these multiple, p y p g Synaptogenic activities of IL1RAPL1 and IL-1RAcP mutants are basically correlated to their afﬁnities to PTPd, which govern stimulation of the receptor activation and signal transmission inside cells. Synaptogenic activities of PTPd mutants are also correlated with their afﬁnities to IL1RAPL1. Discussion l (a,c) Accumulation of Bassoon signals (red) of cortical neurons co-cultured with HEK293Tcells expressing FLAG-tagged IL1RAPL1 mutants (green) (a) and IL-1RAcP mutants (green) (c). (b,d) Intensity of staining signals for Bassoon on the surface of HEK293T cells expressing FLAG-tagged IL1RAPL1 mutants (b) and FLAG-tagged IL-1RAcP mutants (d) co-cultured with cortical neurons (n ¼ 16–19). (e) Accumulation of Shank2 signals (red) of cortical neurons co-cultured with magnetic beads conjugated with ECDs of PTPd mutants. (f) Intensity of staining signals for Shank2 on the surface of magnetic beads conjugated with ECDs of PTPd mutants co-cultured with cortical neurons (n ¼ 27–38). All values represent mean±s.e.m. Po0.05 and *Po0.01 compared with HEK293T cells expressing wild-type IL1RAPL1 (b), wild-type IL-1RAcP (d) or wild-type PTPd-ECD-Fc coated beads (f); Tukey’s test. Scale bars, 10 mm (a,c) and 5 mm (e). heterotypic Ig–Ig interactions; meB is located at the junction between the Ig2 and Ig3 domain of PTPd and adjusts their relative spacing and orientation so that they can simultaneously interact with the Ig1 domain of IL-1RAcP or IL1RAPL1 (Fig. 4a). In the IL-1RAcP/IL1RAPL1-bound state, the Ig2 and Ig3 domains of PTPd exhibits a unique conformation that differs from a linear or V-shaped conformation, which is typically observed in a tandem repeat of Ig domains. Furthermore, this unique conformation of PTPd Ig2–Ig3 also differs from the conformation in a Slitrk-bound state14. Most PTPd variants expressed in the developing mouse heterotypic Ig–Ig interactions; meB is located at the junction between the Ig2 and Ig3 domain of PTPd and adjusts their relative spacing and orientation so that they can simultaneously interact with the Ig1 domain of IL-1RAcP or IL1RAPL1 (Fig. 4a). In the IL-1RAcP/IL1RAPL1-bound state, the Ig2 and Ig3 domains of PTPd exhibits a unique conformation that differs from a linear or V-shaped conformation, which is typically observed in a tandem repeat of Ig domains. Furthermore, this unique conformation of PTPd Ig2–Ig3 also differs from the conformation in a Slitrk-bound state14. Most PTPd variants expressed in the developing mouse brain at postnatal day 11 contain meB and only 4% of variants lack meB8, suggesting that the meB-containing PTPd variants should be mostly utilized for synaptogenesis in the brain. Therefore, the insertion of meB might contribute to allowing PTPd Ig2–Ig3 to form some distinct conformations, depending on the structurally different postsynaptic ligands. For binding of PTPd to IL1RAPL1, either meA9 or meA6 is essential. NATURE COMMUNICATIONS \| 6:6926 \| DOI: 10.1038/ncomms7926 \| www.nature.com/naturecommunications & 2015 Macmillan Publishers Limited. All rights reserved. NATURE COMMUNICATIONS \| 6:6926 \| DOI: 10.1038/ncomms7926 \| www.nature.com/naturecommunications Methods P t i Protein expression and puriﬁcation. Genes encoding mouse PTPd-ECD, PTPd Ig1-Fn4, PTPd Ig1-Fn3, PTPd Ig1-Fn2, PTPd Ig1-Fn1, PTPd Ig1–Ig3 and PTPd Ig1–Ig2 (residues 28–861, 28–699, 28–611, 28–518, 28–418, 28–325 and 28–237, respectively) were ampliﬁed from cDNA (accession No. NM_011211.3) by PCR and cloned into pEBMulti-Neo vector (Wako Pure Chemical Industries) with N-terminal signal sequences derived from pHLsec vector and C-terminal hexahistidine tag29. PTPdA3B Ig1-Fn2 gene was also cloned from cDNA8 in the same manner as the PTPd constructs. Genes encoding mouse IL-1RAcP-ECD (residues 21–351, gene accession No. NM_008364.2) and IL1RAPL1-ECD (residues 19–352, gene accession No. NM_001160403.1) were cloned into pEBMulti-Neo vector, with N-terminal Igk signal sequence and C-terminal hexahistidine tag7,8. All proteins were transiently expressed using Freestyle 293-F cells (Invitrogen). For crystallography, proteins were puriﬁed from 0.2 to 1 l of culture media by Ni-NTA (Qiagen) or Talon metal afﬁnity resin (Clontech) with a standard protocol and dialysed against 20 mM Tris-HCl buffer (pH 7.5) containing 150 mM NaCl. Proteins were concentrated at 5–10 gl 1, ﬂash-frozen in liquid N2 and stored at 80 C until use. A gene encoding PTPd (Fn1-Fn2) (residues 328–518) was cloned into pET21a vector (Novagen). PTPd (Fn1–Fn2) was overexpressed in Rosetta (DE3) Escherichia coli cells (Novagen) and puriﬁed by the Ni afﬁnity chromatography followed by the size exclusion chromatography with Superdex 200 16/60 (GE Healthcare). Pull-down assay. Binding abilities of serially deleted PTPd-ECDs were tested by pull-down assay with the C-terminally Fc-tagged IL-1RAcP- or IL1RAPL1-ECD (IL-1RAcP- or IL1RAPL1-Fc, respectively). IL-1RAcP- and IL1RAPL1-Fc were puriﬁed by Protein G Sepharose (GE Healthcare)7,8. The puriﬁed IL-1RAcP- or IL1RAPL1-Fc was mixed with the PTPd proteins at 1:1 molar ratio and then immobilized with Protein G Sepharose beads (GE Healthcare). After washing by phosphate-buffered saline (PBS), the bound protein complexes were eluted by SDS sample loading buffer, followed by SDS–PAGE analyses. g IL-1RAcP functions in both immune and neuronal systems, whereas IL1RAPL1 is exclusively expressed in neuronal systems, suggesting that IL1RAPL1 has evolved to be more specialized for neuronal systems than IL-1RAcP. Correspondingly, IL-1RAcP binds to PTPd weaker than IL1RAPL1 but has a broader speciﬁcity to the meA variants. In contrast, IL1RAPL1 has more strict speciﬁcity to meA and can bind to PTPd more tightly than IL-1RAcP. The afﬁnity difference generated by the splice-insert signaling codes is substantially high for binding between PTPd and IL1RAPL1 but modest for binding between PTPd and IL-1RAcP. Discussion l Thus, the interface including meA is likely the primary interaction site. Accordingly, the W34A mutation of IL1RAPL1 (at the interface with PTPd Ig2) eliminated binding to PTPd. NATURE COMMUNICATIONS \| 6:6926 \| DOI: 10.1038/ncomms7926 \| www.nature.com/naturecommunications & 2015 Macmillan Publishers Limited. All rights reserved. 8 & 2015 Macmillan Publishers Limited. All rights reserved ARTICLE NATURE COMMUNICATIONS \| DOI: 10.1038/ncomms7926 Ig1 Ig2 meA9 Ig3 Ig2 Ig1 PTPδ Ig1-Ig2 IL1RAPL1-ECD IL-1RAcP Ig1 W34 D292 E291 R98 R75 R95 Ig3 Ig2 Ig1 Ig3 Y273 meB meA Ig2 W34 Ig1 Ig3 Ig2 Ig1 Y273 meB meA Ig2 W27 Ig1 Ig3 IL1RAPL1 IL-1RAcP Fn1 Fn1 . . . . . . . . . . . . . . . . Postsynaptic membrane PTPδ PTPδ Figure 4 \| Mechanistic insights into trans-synaptic adhesion between PTPd and IL1RAPL1 or IL-1RAcP. (a) Schematic drawing of the mechanism for the meA- and meB-regulated binding of PTPd to IL1RAPL1 or IL-1RAcP. The colouring scheme is the same as those in Figs 1a and 2a. Red dots indicate the primary interaction site, whereas black dots indicate the secondary interaction site. Arrows indicate a model that the interaction at the primary site may be followed by the interaction at the secondary site. (b) Superposition of the Ig1–Ig2 unit in the IL-1RAcP-bound PTPd (yellow) on that in the IL1RAPL1- bound PTPd (green), using the Ig1 domain of IL1RAPL1/IL-1RAcP as the reference. Residues involved in the IL1RAPL1 Ig3–PTPd Ig1 interface and the corresponding IL-1RAcP residues are shown as sticks. Ig3 Ig2 Ig1 Ig3 Y273 meB meA Ig2 W34 Ig1 Ig3 Ig2 Ig1 Y273 meB meA Ig2 W27 Ig1 Ig3 IL1RAPL1 IL-1RAcP Fn1 Fn1 . . . . . . . . . . . . . . Postsynaptic membrane PTPδ PTPδ Ig1 Ig2 meA9 Ig3 Ig2 Ig1 PTPδ Ig1-Ig2 IL1RAPL1-ECD IL-1RAcP Ig1 W34 D292 E291 R98 R75 R95 Postsynaptic membrane Figure 4 \| Mechanistic insights into trans-synaptic adhesion between PTPd and IL1RAPL1 or IL-1RAcP. (a) Schematic drawing of the mechanism for the meA- and meB-regulated binding of PTPd to IL1RAPL1 or IL-1RAcP. The colouring scheme is the same as those in Figs 1a and 2a. Red dots indicate the primary interaction site, whereas black dots indicate the secondary interaction site. Arrows indicate a model that the interaction at the primary site may be followed by the interaction at the secondary site. Discussion l (b) Superposition of the Ig1–Ig2 unit in the IL-1RAcP-bound PTPd (yellow) on that in the IL1RAPL1- bound PTPd (green), using the Ig1 domain of IL1RAPL1/IL-1RAcP as the reference. Residues involved in the IL1RAPL1 Ig3–PTPd Ig1 interface and the corresponding IL-1RAcP residues are shown as sticks. Moreover, the interaction with meA appears coupled to the third interface involving PTPd Ig1 and IL1RAPL1 Ig3, which is critical for binding between IL1RAPL1 and PTPd and inducing bidirectional synaptic differentiation of cultured cortical neurons, as mentioned above. Comparison between the IL1RAPL1 and IL-1RAcP complexes exhibits a rotational difference in the orientations of the PTPd Ig1–Ig2 unit, relative to the Ig1 domains of IL1RAPL1 and IL-1RAcP (Fig. 4b). Trp34 of IL1RAPL1 is positioned at the pivot for this rotation, which allows PTPd Ig1 to interact with IL1RAPL1 Ig3. In this context, meA also contributes to the multiple, heterotypic Ig–Ig interactions between IL1RAPL1 and PTPd, besides meB, which assists in locating the Ig3 domain of PTPd in the appropriate position for interacting with the Ig1 domain of IL1RAPL1. Consistent with this, the impact of the lack of meB on binding between PTPd and IL1RAPL1 is comparable to that of the PTPd Y273A mutation (deﬁcient in the PTPd Ig3–IL1RAPL1 Ig1 interaction). In contrast, the impact of the lack of meB on binding between PTPd and IL-1RAcP is incompatible to that of the PTPd Y273A mutation, but rather is comparable to that of mutations at the IL-1RAcP Ig1–PTPd Ig2 interface. The W27A mutation of IL-1RAcP, which is equivalent to the W34A mutation of IL1RAPL1, reduced the afﬁnity sevenfold but did not eliminate the binding. Therefore, for binding to IL-1RAcP, the PTPd Ig3–IL-1RAcP Ig1 interface is likely the primary interaction site, and the meB insertion may serve as the adjustable linker to locate the Ig2 domain of PTPd in the appropriate position for interacting with the Ig1 domain of IL-1RAcP (Fig. 4a). with low afﬁnity ligands. Further functional studies including identiﬁcation and characterization of additional regulatory factors are awaited to discuss the activation and signaling mechanisms of IL1RAPL1/IL-1RAcP and PTPd for inducing synaptic differentiation. NATURE COMMUNICATIONS \| 6:6926 \| DOI: 10.1038/ncomms7926 \| www.nature.com/naturecommunications & 2015 Macmillan Publishers Limited. All rights reserved. & 2015 Macmillan Publishers Limited. All rights reserved. References IL1RAPL1-ECD-His6 was immobilized on a CM5 sensor chip by the amine-coupling method at a density of about 2,000 resonance units (RU), whereas IL-1RAcP-ECD-Fc was captured at a density of about 700 RU by anti-Fc antibodies that were immobilized on a CM5 sensor chip. Splicing variants or mutants of PTPd Ig1–Fn1 were injected as the analytes with the concentration range from 15.6 to 4,000 nM. The IL1RAPL1-immobilized sensor chip was regenerated by 10 mM NaOH. The sensorgrams are shown in Supplementary Fig. 4. 13. Yim, Y. S. et al. Slitrks control excitatory and inhibitory synapse formation with LAR receptor protein tyrosine phosphatases. Proc. Natl Acad. Sci. USA 110, 4057–4062 (2013). 14. Um, J. W. et al. Structural basis for LAR-RPTP/Slitrk complex-mediated synaptic adhesion. Nat. Commun. 5, 5423 (2014). 15. Uemura, T. et al. Trans-synaptic interaction of GluRd2 and Neurexin through Cbln1 mediates synapse formation in the cerebellum. Cell 141, 1068–1079 (2010). 16. Siddiqui, T. J., Pancaroglu, R., Kang, Y., Rooyakkers, A. & Craig, A. M. LRRTMs and neuroligins bind neurexins with a differential code to cooperate in glutamate synapse development. J. Neurosci. 30, 7495–7506 (2010). Cell cultures and co-culture assay. A central nervous system-restricted isoform of IL-1RAcP, IL-1RAcPb, was used for HEK293T cell-based analyses of IL-1RAcP mutants. Primary cortical cultures were prepared from mice at E18 (ref. 38). The cortical cells were placed on coverslips coated with 30 mg ml 1 poly-L-lysine and 10 mg ml 1 mouse laminin at the density of 5 105 cells per well for co-culture assay. The cells were cultured in Neurobasal-A supplemented with 2% B-27 supplement (Invitrogen), 5% FCS, 100 U ml 1 penicillin, 100 mg ml 1 streptomycin and 0.5 mM L-glutamine for 24 h, and then cultured in the same medium without FCS. Cultures of HEK293T cells were maintained in DMEM supplemented with 10% FCS38. Expression vectors for mutated forms of PTPd- ECD-Fc, FLAG-tagged IL-1RAcPb and FLAG-tagged IL1RAPL1 were generated by PCR-based mutagenesis using pEB6-PTPd-ECD-Fc, pFLAG-IL-1RAcPb and pFLAG-IL1RAPL1 (ref. 7) as templates, respectively. Expression vectors for FLAG- tagged IL1RAPL1 and IL-1RAcP mutants were transfected to HEK293T cells using NanoJuice transfection reagent (Novagen). After 2 days of culture, the transfected cells were washed with PBS containing 2 mM EDTA, and incubated with the same buffer at 37 C for 10 min. Dispersed cells were plated onto cortical neurons at days in vitro (DIV) 8. References 1. Carrie, A. et al. A new member of the IL-1 receptor family highly expressed in hippocampus and involved in X-linked mental retardation. Nat. Genet. 23, 25–31 (1999). 2. Jamain, S. et al. Mutations of the X-linked genes encoding neuroligins NLGN3 and NLGN4 are associated with autism. Nat. Genet. 34, 27–29 (2003). J , g g g and NLGN4 are associated with autism. Nat. Genet. 34, 27–29 (2003). 3. Sudhof, T. C. Neuroligins and neurexins link synaptic function to cognitive disease. Nature 455, 903–911 (2008). 3. Sudhof, T. C. Neuroligins and neurexins link synaptic function to cognitive disease. Nature 455, 903–911 (2008). 4. Pinto, D. et al. Functional impact of global rare copy number variation in autism spectrum disorders. Nature 466, 368–372 (2010). 5. Takahashi, H. & Craig, A. M. Protein tyrosine phosphatases PTPd, PTPs, and LAR: presynaptic hubs for synapse organization. Trends Neurosci. 36, 522–534 (2013). 6. Um, J. W. & Ko, J. LAR-RPTPs: synaptic adhesion molecules that shape synapse development. Trends Cell Biol. 23, 465–475 (2013). 7. Yoshida, T. et al. Interleukin-1 receptor accessory protein organizes neuronal synaptogenesis as a cell adhesion molecule. J. Neurosci. 32, 2588–2600 (2012). 8. Yoshida, T. et al. IL-1 receptor accessory protein-like 1 associated with mental retardation and autism mediates synapse formation by trans-synaptic interaction with protein tyrosine phosphatase d. J. Neurosci. 31, 13485–13499 (2011). 9. Takahashi, H. et al. Postsynaptic TrkC and presynaptic PTPs function as a bidirectional excitatory synaptic organizing complex. Neuron 69, 287–303 (2011). 10. Woo, J. et al. Trans-synaptic adhesion between NGL-3 and LAR regulates the formation of excitatory synapses. Nat. Neurosci. 12, 428–437 (2009). 11. Kwon, S. K., Woo, J., Kim, S. Y., Kim, H. & Kim, E. Trans-synaptic adhesions between netrin-G ligand-3 (NGL-3) and receptor tyrosine phosphatases LAR, protein-tyrosine phosphatase d (PTPd), and PTPs via speciﬁc domains regulate excitatory synapse formation. J. Biol. Chem. 285, 13966–13978 (2010). 12. Takahashi, H. et al. Selective control of inhibitory synapse development by Slitrk3-PTPd trans-synaptic interaction. Nat. Neurosci. 15, 389–398 S1–S2 (2012). SPR analysis. SPR equilibrium experiments were carried out by using Biacore T200 (GE Healthcare) at 25 C in 10 mM Hepes (pH 7.9) containing 150 mM NaCl, 0.05% Tween-20. ARTICLE NATURE COMMUNICATIONS \| DOI: 10.1038/ncomms7926 surface, 4.0–4.5 mm diameter; Spherotech). Beads coupled with Fc or Fc fusion proteins were added to cortical neurons (DIV 14). After 24 h, cultures were ﬁxed for immunostaining with rabbit anti-Shank2 antibody (Frontier Institute, 1:200). crystallization experiments except for PTPd Fn1–Fn2 were performed at 20 C. Temperature shift from 4 to 28 C drastically improved the crystal size of PTPd Fn1–Fn2. Crystals were ﬂash-frozen in liquid N2, followed by soaking in the reservoir solutions supplemented with the following cryo-protectants: 25% glycerol for PTPdA3B Ig1–Fn2 and the PTPd Ig1–Ig2IL1RAPL1-ECD complex; 25% ethylene glycol for PTPd Fn1–Fn2; 27% ethylene glycol for IL1RAPL1; 35% xylitol for the PTPd Ig1–Fn2IL-1RAcP-ECD complex; 20% ethylene glycol for the PTPd-ECDIL1RAPL1-ECD complex. crystallization experiments except for PTPd Fn1–Fn2 were performed at 20 C. Temperature shift from 4 to 28 C drastically improved the crystal size of PTPd Fn1–Fn2. Crystals were ﬂash-frozen in liquid N2, followed by soaking in the reservoir solutions supplemented with the following cryo-protectants: 25% glycerol for PTPdA3B Ig1–Fn2 and the PTPd Ig1–Ig2IL1RAPL1-ECD complex; 25% ethylene glycol for PTPd Fn1–Fn2; 27% ethylene glycol for IL1RAPL1; 35% xylitol for the PTPd Ig1–Fn2IL-1RAcP-ECD complex; 20% ethylene glycol for the PTPd-ECDIL1RAPL1-ECD complex. Image acquisition and quantiﬁcation. Images of ﬁbroblast–neuron or bead– neuron co-culture experiments were acquired with a confocal laser-scanning microscope TCS SP5II (Leica) under constant conditions as to laser power, iris, gain, z-steps and zoom setting throughout the experiments. The images were collected from at least two separate cell cultures. All quantitative measurements were performed with ImageJ 1.36b software39. The intensities of immunostaining signals for Bassoon or Shank2 were measured as the optical mean density within circles of 30- and 7-mm diameters enclosing transfected HEK293T cells and coated-beads, respectively. Statistical signiﬁcance was evaluated by one-way analysis of variance followed by post hoc Tukey’s test. Crystallography. All data were collected at 100 K at BL41XU in SPring-8, and processed with HKL2000 (ref. 30) and CCP4 program suite31. Data collection and reﬁnement statistics were summarized in Table 1. Resolutions were estimated, basically based on I/sI values (B2). To improve the quality of the atomic models, the resolutions of PTPd-ECDIL1RAPL1-ECD and PTPdA3B Ig1–Fn2 were set to be as high as possible. The resolution of PTPd Fn1–Fn2 was limited by the size of the detector. ARTICLE We ﬁrst determined a crystal structure of the PTPd Ig1–Ig2IL1RAPL1-ECD complex by the molecular replacement method using IL-1RAcP-ECD (PDB:4DEP, 3O4O)32,33 and PTPd Ig1–Ig2 (PDB:2YD6)19 as the search models with the program Molrep34. Model building and reﬁnement were carried out using the programs Coot35 and Phenix36, respectively. The ﬁnal model of the PTPd Ig1–Ig2IL1RAPL1-ECD complex was reﬁned at 2.7 Å to Rwork and Rfree values of 22.6 and 26.3%, respectively. Then, the PTPd-ECDIL1RAPL1-ECD complex structure was determined by the molecular replacement method using IL1RAPL1-ECD in the PTPd Ig1–Ig2-bound state, PTPs Ig3 (PDB:2YD9)19, LAR Fn2 (PDB:2DJU) and PTPd Fn2 (PDB:2DLH) as the search models. The ﬁnal model of the PTPd-ECDIL1RAPL1-ECD complex was reﬁned at 4.4 Å to Rwork and Rfree values of 29.1 and 32.7%, respectively. In addition, 2.0-Å-resolution structure of PTPd Fn1–Fn2 was solved by the molecular replacement method and reﬁned with the program Phenix to Rwork and Rfree values of 20.4 and 22.3%, respectively. The structure of IL1RAPL1-ECD was determined by the molecular replacement method and reﬁned at 3.0 Å to Rwork and Rfree values of 26.4 and 29.8%, respectively. The structure of PTPd Ig1–Fn2IL-1RAcP-ECD complex was determined by the molecular replacement method using IL-1RAcP-ECD (PDB:4DEP, 3O4O)32,33, PTPd Ig1–Fn2 (PDB:2YD6)19 and our reﬁned PTPd Fn1–Fn2 structure as the search models. The ﬁnal model of the PTPd Ig1–Fn2IL- 1RAcP-ECD complex was reﬁned at 3.25 Å to Rwork and Rfree values of 25.6 and 28.7%, respectively. The structure of PTPdA3B Ig1–Fn2 was determined by the molecular replacement method and reﬁned at 3.5 Å to Rwork and Rfree values of 29.1 and 31.5%, respectively. The buried surface area was calculated with the program PISA37. The stereochemistry of the ﬁnal model was assessed by the program Procheck. Percentages of residues in the most favoured regions of the Ramachandran plot are 87.9%, 81.2%, 86.7%, 89.4%, 88.1% and 89.1% for PTPd Ig1–Ig2IL1RAPL1-ECD, PTPd-ECDIL1RAPL1-ECD, PTPd Fn1–Fn2, PTPdA3B– Ig1–Fn2, PTPd Ig1–Fn2IL-1RAcP-ECD and IL1RAPL1-ECD, respectively. No residues are in the disallowed regions for all these structures. Structural ﬁgures were prepared with the program PyMol (Schro¨dinger, LLC). Methods P t i On the other hand, in our synaptogenic assays, the strength of the induced synaptic differentiation seems non-linearly correlated with the afﬁnity; rather, there seems to be the threshold of the afﬁnity for inducing the synaptic differentiation, which might contribute to controlling appropriate synaptic target selection to avoid misconnections to target cells Crystallization. PTPdA3B Ig1-Fn2, PTPd Fn1–Fn2, IL1RAPL1 and the PTPd Ig1–Fn2IL-1RAcP-ECD, PTPd-ECDIL1RAPL1-ECD and PTPd Ig1–Ig2 IL1RAPL1-ECD complexes were crystallized by the sitting drop vapour diffusion method. Protein solutions were mixed with the following reservoir solutions: 10% polyethylene glycol (PEG) 3350, 0.1 M ammonium iodide for PTPdA3B Ig1–Fn2; 20% PEG3350, 0.1 M MgCl2 and 0.1 M BisTris (pH 5.5) for PTPd Fn1–Fn2; 2.0–2.4 M ammonium sulfate for IL1RAPL1; 15% PEG3350, 0.1 M ammonium sulfate and 0.1 M tri-sodium citrate (pH 5.5) for the PTPd Ig1–Fn2IL-1RAcP-ECD complex; 12% PEG4000, 0.1 M lithium sulfate, 0.1 M ADA (pH 6.5) for the PTPd-ECDIL1RAPL1-ECD complex; 15% PEG4000, 0.1 M MES (pH 6.0) for the PTPd Ig1–Ig2IL1RAPL1-ECD complex. For the crystal- lization of the complexes, two protein samples were mixed at a molar ratio of 1:1 to a ﬁnal concentration of B6 g l 1 before crystallization experiments. All 9 Author contributions 29. Aricescu, A. R., Lu, W. & Jones, E. Y. A time- and cost-efﬁcient system for high-level protein production in mammalian cells. Acta Crystallogr. D 62, 1243–1250 (2006). T.Y., A.M., T.S. and A.Y. performed pull-down assays and crystallization. A.Y., S.F., Y.S. and S.G.-I. collected diffraction data. A.Y. and S.F. analysed the collected data and determined the structures. T.Y., T.S. and S.I.-O. performed cell biological experiments. A.Y., T.Y. and S.F. wrote the paper with editing by T.U., H.M. and M.M. T.Y., M.M. and S.F. designed and supervised the study. T.Y., A.M., T.S. and A.Y. performed pull-down assays and crystallization. A.Y., S.F., Y.S. and S.G.-I. collected diffraction data. A.Y. and S.F. analysed the collected data and T.Y., A.M., T.S. and A.Y. performed pull-down assays and crystallization. A.Y., S.F., Y.S. and S.G.-I. collected diffraction data. A.Y. and S.F. analysed the collected data and determined the structures. T.Y., T.S. and S.I.-O. performed cell biological experiments. A.Y., T.Y. and S.F. wrote the paper with editing by T.U., H.M. and M.M. T.Y., M.M. and S.F. designed and supervised the study. 30. Otwinowski, Z. & Minor, W. Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 276, 20 (1997). 31. Winn, M. D. et al. Overview of the CCP4 suite and current developments. Acta Crystallogr. D Biol. Crystallogr. 67, 8 (2011). Additional information 32. Thomas, C., Bazan, J. F. & Garcia, K. C. Structure of the activating IL-1 receptor signaling complex. Nat. Struct. Mol. Biol. 19, 455–457 (2012). Accession codes: Atomic coordinates and structure factors have been deposited with the accession codes 4YFC (PTPd Ig1–Ig2IL1RAPL1-ECD), 4YFD (PTPd Ig1–Fn2IL- 1RAcP-ECD), 4YFE (PTPd Fn1–Fn2), 4YFG (PTPdA3B Ig1–Fn2), 4YH6 (IL1RAPL1- ECD) and 4YH7 (PTPd-ECDIL1RAPL1-ECD). 33. Wang, D. et al. Structural insights into the assembly and activation of IL-1b with its receptors. Nat. Immunol. 11, 905–911 (2010). p 34. Vagin, A. T. A. MOLREP: an automated program for molecular replacement. J. Appl. Cryst. 30, 4 (1997). Supplementary Information accompanies this paper at http://www.nature.com/ naturecommunications 35. Emsley, P. & Cowtan, K. Coot: model-building tools for molecular graphics. Acta Crystallogr. D 60, 2126–2132 (2004). Competing ﬁnancial interests: The authors declare no competing ﬁnancial interests. 36. Adams, P. D. et al. The Phenix software for automated determination of macromolecular structures. Methods 55, 94–106 (2011). Reprints and permission information is available online at http://npg.nature.com/ reprintsandpermissions/ 37. Krissinel, E. & Henrick, K. Inference of macromolecular assemblies from crystalline state. J. Mol. Biol. 372, 774–797 (2007). 38. Uemura, T., Mori, H. & Mishina, M. Direct interaction of GluRd2 with Shank scaffold proteins in cerebellar Purkinje cells. Mol. Cell Neurosci. 26, 330–341 (2004). How to cite this article: Yamagata, A. et al. Mechanisms of splicing-dependent trans- synaptic adhesion by PTPd–IL1RAPL1/IL-1RAcP for synaptic differentiation. Nat. Commun. 6:6926 doi: 10.1038/ncomms7926 (2015). 39. Uemura, T. & Mishina, M. The amino-terminal domain of glutamate receptor d2 triggers presynaptic differentiation. Biochem. Biophys. Res. Commun. 377, 1315–1319 (2008). This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ ARTICLE NATURE COMMUNICATIONS \| DOI: 10.1038/ncomms7926 help during data collection. This work was supported by Grant-in-Aid for Scientiﬁc Research on Innovative Areas 22121003 (S.F.), 22123008 (M.M.), 25110708 (T.Y.), 25121710 (T.U.), Grant-in-Aid for Scientiﬁc Research (A) 21249012, 24249014 (M.M.), Grant-in-Aid for Scientiﬁc Research (B) 25293057 (T.Y), 25290021 (T.U.), Grant-in-Aid for Scientiﬁc Research (C) 24570126 (A.Y.), Grant-in-Aid for challenging Exploratory Research 26640038 (T.Y.), 26640039 (T.U.), PRESTO, JST (T.Y.) and CREST, JST (S.F. and T.U.). help during data collection. This work was supported by Grant-in-Aid for Scientiﬁc Research on Innovative Areas 22121003 (S.F.), 22123008 (M.M.), 25110708 (T.Y.), 25121710 (T.U.), Grant-in-Aid for Scientiﬁc Research (A) 21249012, 24249014 (M.M.), Grant-in-Aid for Scientiﬁc Research (B) 25293057 (T.Y), 25290021 (T.U.), Grant-in-Aid for Scientiﬁc Research (C) 24570126 (A.Y.), Grant-in-Aid for challenging Exploratory Research 26640038 (T.Y.), 26640039 (T.U.), PRESTO, JST (T.Y.) and CREST, JST (S.F. and T.U.). 25. Coles, C. H. et al. Structural basis for extracellular cis and trans RPTPs signal competition in synaptogenesis. Nat. Commun. 5, 5209 (2014). y g 26. Aricescu, A. R. & Jones, E. Y. Immunoglobulin superfamily cell adhesion molecules: zippers and signals. Curr. Opin. Cell Biol. 19, 543–550 (2007). 27. Wei, C. H. & Ryu, S. E. Homophilic interaction of the L1 family of cell adhesion molecules. Exp. Mol. Med. 44, 413–423 (2012). 28. Samanta, D. & Almo, S. C. Nectin family of cell-adhesion molecules: structural and molecular aspects of function and speciﬁcity. Cell Mol. Life Sci. 72, 645–658 (2015). NATURE COMMUNICATIONS \| 6:6926 \| DOI: 10.1038/ncomms7926 \| www.nature.com/naturecommunications & 2015 Macmillan Publishers Limited. All rights reserved. References After 24 h of co-culture, cells were ﬁxed for immunostaining with mouse anti-Bassoon (Stressgen, 1:300) and rabbit anti-FLAG (Sigma, 1:1,000) antibodies. Fc and ECDs of PTPd mutant proteins fused to Fc in HEK293T cell culture medium were bound to Protein A-conjugated magnetic particles (smooth 17. Koehnke, J. et al. Splice form dependence of b-neurexin/neuroligin binding interactions. Neuron 67, 61–74 (2010). 18. Wei, Z. & Zhang, M. A structural approach to decipher the neurexin and neuroligin splice isoform code. Neuron 67, 1–2 (2010). 19. Coles, C. H. et al. Proteoglycan-speciﬁc molecular switch for RPTPs clustering and neuronal extension. Science 332, 484–488 (2011). 19. Coles, C. H. et al. Proteoglycan-speciﬁc molecular switc and neuronal extension. Science 332, 484–488 (2011). 20. Tonks, N. K. Protein tyrosine phosphatases: from genes, to function, to disease. Nat. Rev. Mol. Cell Biol. 7, 833–846 (2006). 21. Groves, J. T. & Kuriyan, J. Molecular mechanisms in signal transduction at the membrane. Nat. Struct. Mol. Biol. 17, 659–665 (2010). 22. Lemmon, M. A. & Schlessinger, J. Cell signaling by receptor tyrosine kinases. Cell 141, 1117–1134 (2010). 23. Dabrowski, A. & Umemori, H. Orchestrating the synaptic network by tyrosine phosphorylation signalling. J. Biochem. 149, 641–653 (2011). 24. Ferrao, R., Li, J., Bergamin, E. & Wu, H. Structural insights into the assembly of large oligomeric signalosomes in the Toll-like receptor-interleukin-1 receptor superfamily. Sci. Signal. 5, re3 (2012). NATURE COMMUNICATIONS \| 6:6926 \| DOI: 10.1038/ncomms7926 \| www.nature.com/naturecommunications 10 NATURE COMMUNICATIONS \| 6:6926 \| DOI: 10.1038/ncomms7926 \| www.nature.com & 2015 Macmillan Publishers Limited. All rights reserved. ARTICLE Acknowledgements We thank the beam-line staff at NW12A, BL-1A, BL-5A and BL17A of Photon Factory (Tsukuba, Japan) and BL32XU and BL41XU of SPring-8 (Hyogo, Japan) for technical 11 11
https://openalex.org/W2914038922	https://journals.uni-lj.si/knjiznica/article/download/13993/12303	Slovene	null	Digital library and the Slovenian academic environment	Knjižnica	2,014	cc-by-sa	5,386	Ključne besede: digitalne knjižnice, visokošolske knjižnice, uporabniki, študenti, bibliotekarstvo Professional article Ključne besede: digitalne knjižnice, visokošolske knjižnice, uporabniki, študenti, bibliotekarstvo Professional article Professional article UDC 02:004.738.52(497.4) DIGITALNA KNJIŽNICA IN SLOVENSKI VISOKOŠOLSKI PROSTOR Blaž Lesnik Brina Jerič Tanja Curhalek Martina Kerec Oddano: 06.01.2001 – Sprejeto: 14.05.2001 Oddano: 06.01.2001 – Sprejeto: 14.05.2001 Oddano: 06.01.2001 – Sprejeto: 14.05.2001 Strokovni članek UDK 02:004.738.52(497.4) LESNIK, Blaž; Brina, JERIČ; Tanja, CURHALEK; Martina, KEREC: Digital library and the Slovenian academic environment. Knjižnica, Ljubljana, 45(2001)1-2, xx-xx Izvleček Digitalna knjižnica je termin za knjižnico sedanjosti in prihodnosti, ki je postala izziv tradicionalnim oblikam knjižničnega dela. Avtorje zanima predvsem slovenski digitalni prostor visokošolskih ustanov. V raziskavi so bile pregledane digitalne zbirke večine fakultet Univerze v Ljubljani ter na osnovi ankete pridobljeni podatki o uporabnikovih (študentovih) potrebah in seznanjenosti z digitalno knjižnico. Z ugotovitvami raziskave se pričakuje pozitiven vpliv digitalnih knjižnic na študij, iz rezultatov ankete pa spodbuda bibliotekarski stroki za nadaljnja raziskovanja. Anketa je bila izvedena konec leta 1999 na Filozofski fakulteti v Ljubljani (FF), v njej pa je sodelovalo 275 študentov FF. Rezultati so pokazali, da je študentom bibliotekarstva pojem digitalne knjižnice bližji in da ga pravilneje razumejo kot ostali študenti. Večina anketiranih študentov uporablja osebni računalnik pogosto in z veseljem. 71,1% študentov bibliotekarstva pozna izraz digitalna knjižnica, medtem ko je le-ta poznan samo 43,8% ostalih študentov. Večina anketiranih se je odločila za pravilno opredelitev pojma digitalne knjižnice (digitalna knjižnica je zbirka medsebojno povezanih sistemov in virov, dostopnih preko mreže), pri tem pa je po mnenju avtorjev šlo predvsem za ugibanje. V skladu z rezultati raziskave avtorji članka menijo, da se bo nadaljnji razvoj tudi v Sloveniji usmeril k izboljšavi izgradnje in k širši uporabi digitalnih knjižnic. e besede: digitalne knjižnice, visokošolske knjižnice, uporabniki, študenti, karstvo 1 Rezultati anket govorijo temu v prid. 2 Mednarodni simpozij BOBCATSSS, ki ga organizirajo študenti, in na katerem se srečujejo strokovnjaki s področja bibliotekarskih in informacijskih znanosti, je v letu 1999 izbral aktualno temo: Intelektualna lastnina vs. pravica do znanja? V januarju 2000 je potekal v Krakowu že osmi simpozij BOBCATSSS, na katerem smo avtorji članka s pomočjo mentorjev dr. Primoža Južniča in asist. Polone Vilar aktivno sodelovali z referatom. 1 Uvod Pisati o digitalnih knjižnicah v slovenskem prostoru pomeni soočiti se z novimi izzivi in trendi, ki spremljajo pojav digitalnih knjižnic. Slovenska knjižničarska stroka nima dolgoletne tradicije na področju razvoja, delovanja in tudi raziskovanja digitalnih knjižnic; še najbolje so se le-te uveljavile na področju specialnih in visokošolskih ustanov. Marsikatero podjetje ima že svojo digitalno knjižnico, ki pa ni dostopna zunanjim uporabnikom, zato lahko le predvidevamo število vseh digitalnih knjižnic specialnih ustanov v Sloveniji. Lažje dostopne so digitalne knjižnice visokošolskih ustanov oz. fakultet, ki so namenjene ciljni skupini študentov in profesorjev ter so zanimive za nas kot študente. Te danes delujoče digitalne knjižnice fakultet delujejo predvsem kot pripomoček pri študiju in seveda klasičnim knjižnicam. Eksponentna rast osebnih računalnikov ter Interneta bo približala digitalno knjižnico vsakemu študentu. Pravzaprav bo njena uporaba postala obvezen del študija. Ker se zavedamo njenih prednosti1, bo to predvsem pozitivna sprememba, ki bo dvignila kakovost študija, saj bodo imeli študenti dostop do študijske literature brez časovnih in prostorskih omejitev. V kolikor pa nas navdušujejo te prednosti digitalne knjižnice, se moramo ob izgradnji le-te zavedati problema avtorskih pravic, s katerim smo se soočili tudi sami. Nastali tekst je tako rezultat raziskovanja fenomena digitalne knjižnice, ki postaja vsakdan v slovenskem prostoru, predvsem na področju visokošolskega izobraževanja2. Abstract Digital library is a term for a library of the present and of the future challenging the traditional libraries. Authors are interested mostly in the digital space of Slovene academic instutions. In the research, the digital collections of most of the University of Ljubljana faculties are examined. A questionnaire was used to obtain information about the users' (students') needs and their acquaintance with the concept of the digital library. Two things were expected: the research was aimed at finding the positive effect of digital libraries on studies, and the questionnaire as an incentive in the library profession for further research. The questionnaire was made in the end of 1999 at the Faculty of Arts in Ljubljana (FF). It included 275 students of FF. The results showed that the students of librarianship had a better understanding of the term digital library than the students of other courses. A personal computer is used frequently and with pleasure by most questioned students. The term digital library is known to 71,1% of the students of librarianship, and only to 43,8% of others. Most of the students chose the correct definition of the digital library (the digital library is a collection of disparate systems and resources, accessible on the net), but that was, by the authors' opinion, mostly a lucky guess. According to the findings of the research, the authors believe that future development will improve and accelerate a wider use of digital libraries, in Slovenia as well. Key words: digital libraries, academic libraries, users, students, librarianship LESNIK, Blaž; Brina, JERIČ; Tanja, CURHALEK; Martina, KEREC: Digital library a Slovenian academic environment. Knjižnica, Ljubljana, 45(2001)1-2, xx-xx LESNIK, Blaž; Brina, JERIČ; Tanja, CURHALEK; Martina, KEREC: Digital library and the Slovenian academic environment. Knjižnica, Ljubljana, 45(2001)1-2, xx-xx 1 1 1 Rezultati anket govorijo temu v prid. 2 3 Vsi ti različni izrazi so v največjem številu zbrani v poročilu raziskovalne komisije Britanske knjižnice in Inovacijskega centra z naslovom Understanding digital libraries : towards a conceptual framework avtorjev Davida Bawdena in Iana Rowlandsa, 1999. 2.1 Opredelitev termina »digitalna knjižnica« 2 Pisati o digitalnih knjižnicah pomeni dvoje: Pisati o digitalnih knjižnicah pomeni dvoje: 1. soočiti se s terminološko neenotnostjo samega izraza (sinonimno se uporabljajo izrazi kot virtualna, elektronska, hibridna, gateway knjižnica, knjižnica brez sten ter knjižnica prihodnosti)3, 2. obstajajo različne opredelitve oz. definiranja koncepta digitalne knjižnice. Avtorji članka uporabljamo izraz digitalna knjižnica, saj je »digitalna knjižnica najbolj svež in predvsem zelo razširjen termin, ki se danes skoraj izključno uporablja na raznih konferencah, na internetu ali v literaturi« (Cleveland, 1999). Različnemu pojmovanju digitalne knjižnice botruje tudi dejstvo, da jo uporabljajo različni uporabniki, ki jo poimenujejo vsak po svoje. Digitalna knjižnica je torej: • za informacijsko službo velika baza podatkov, • za informacijsko službo velika baza podatkov, • za ljudi, ki delajo s hipertekstnimi tehnologijami, je digitalna knjižnica lahko določena aplikacija, • tistim, ki se ukvarjajo z internetom, je lahko digitalna knjižnica internet sam oz. svetovni splet, • za bibliotekarsko stroko pa je naslednji korak pri avtomatizaciji knjižnic, ki poteka že vrsto let. Pravzaprav je digitalna knjižnica vse zgoraj našteto. Vendarle pa pojmovanja ne smemo posploševati in razumeti digitalne knjižnice enostavno kot internet. Svetovni splet je zbirka tisočih in tisočih dokumentov. Veliko ljudi zato pojmuje to ogromno zbirko kot digitalno knjižnico, ker lahko v njej najdejo željene informacije, prav tako kot lahko npr. naročijo priljubljeno zgoščenko ali opravijo elektronsko bančno transakcijo. Vendar, ali je to res digitalna knjižnica? Clifford Lynch, eden vodilnih avtoritet na področju razvoja digitalnih knjižnic, pravi, da ni: »Včasih slišimo, da ljudje označijo internet kot svetovno knjižnico digitalne dobe. Vendar ta opis ne vzdrži niti čisto običajne presoje. Internet – in posebej še njegova zbirka multimedijskih virov – namreč ni bil osnovan za organizirano hranjenje in izposojanje dokumentov, kot to poznamo pri klasičnih knjižnicah. Lahko bi kvečjemu rekli, da gre za kaotično zbirko naključno objavljenih dokumentov. Torej na kratko: internet ni digitalna knjižnica« (cv: Cleveland, 1999). 3 3 Kaj je digitalna knjižnica? Zavedati se je treba, da so digitalne knjižnice knjižnice, za katere veljajo isti cilji in nameni kot za klasične knjižnice, to so zbiranje, obdelava, hranjenje in izposoja. V prvi vrsti so digitalne knjižnice še vedno knjižnice, za katere veljajo nekatere skupne značilnosti. Digitalne knjižnice so lahko samostojne zbirke in vsebujejo gradivo v elektronski obliki, medtem ko hibridne knjižnice lahko vsebujejo tako elektronsko kot papirno gradivo. 2.1 Opredelitev termina »digitalna knjižnica« Oboje nudijo jasen pregled vseh informacij in gradiva, ki jih ima knjižnica na voljo, vključujejo vse procese in službe, ki so hrbtenica tradicionalnih knjižnic, ter bodo še vedno služile določenim krogom, institucijam, skupnostim. Digitalna knjižnica torej mora, če naj logično nadgradi tradicionalno knjižnico, ohraniti njene značilnosti, vendar jih mora razvijati in razširjati ter vključevati vanjo nove tehnologije, procese in medije. Če želimo objektivno definirati digitalno knjižnico, moramo izbirati izmed vrste različnih opredelitev. Tako smo se odločili, da prevzamemo opredelitev Garyja Clevelanda, ki pojmuje digitalno knjižnico kot »zbirko različnih sistemov in virov, ki so povezani preko mreže in integrirani drug v drugega ter namenjeni specifičnim namenom in uporabnikom« (Cleveland, 1999). 2.2 Organizacijski vidiki digitalne knjižnice Digitalna knjižnica je v zadnjih letih pritegnila pozornost mnogih računalniških strokovnjakov, programerjev, raziskovalcev sistemov in strokovnjakov s področja bibliotekarstva, vanjo so usmerjeni mnogi veliki nacionalni projekti in druge raziskave. Toda digitalna knjižnica je pogostokrat raziskovana z močnim poudarkom na tehnologiji, medtem ko organizacijska vprašanja ne vzbujajo prav veliko pozornosti. To poglavje zato govori o določenih organizacijskih vidikih, ki morajo biti proučeni za boljše razumevanje razvoja digitalne knjižnice. Pri tem se opiramo na članek Boba Travica (1999). Raziskava podaja razmerje med trendi v organizaciji knjižnic in digitalno knjižnico, kar je konceptualizirano v obliki štirih osnovnih modelov, ki se razlikujejo med seboj po obsegu, v katerem se oddaljujejo od obstoječe knjižnice. Ti modeli so nekakšen ideal, ki ga po vsej verjetnosti v praksi ne najdemo v čisti obliki, to pa so: 1) Podsistemski model (Subsystem Model): digitalna knjižnica nastopi kot nov podsistem v obstoječi knjižnici in je le del celotne knjižnične organizacije. Ta model odseva dandanašnje prevladujoče mišljenje, ki je osredotočeno na tehnologijo. Primer tega modela je knjižnična uporaba računalniške mreže, ki omogoča dostop do gradiva različnih knjižnic. Zmožnosti računalniškega omrežja so torej pri tem modelu pomembne. 4 2) Notranje-organizacijski model (Interorganizational Model): digitalna knjižnica lahko gradi notranje-organizacijske odnose z namenom, da izmenjava bibliografske informacije in gradivo itd. Digitalna knjižnica je v okviru tega modela lahko v celoti motivirana, grajena in vzdrževana s pomočjo računalniške mreže. Primer tega modela je majhna knjižnica z omejeno knjižnično zbirko, ki pa vendarle nudi visokokakovostne usluge s pomočjo mrežne povezave z drugimi knjižnicami. 3) Sistemski model (System Model): tu nastopa digitalna knjižnica kot popolnoma nova organizacijska oblika, ki ima boljši dostop in dostavo gradiva, nudi močnejšo podporo uporabnikom. V tem modelu prevladuje digitalna/virtualna organizacija, za katero je značilno: dinamično vključevanje različnih organizacij v virtualno organizacijo, ki se pojavljajo kot ena sama organizacija; fizična oddaljenost med deli organizacije, ki so povezani preko elektronskih kanalov; in možnost ponudbe običajnih storitev. Te značilnosti kažejo, da digitalna knjižnica vključuje vidike mrežne organizacije. 3) Sistemski model (System Model): tu nastopa digitalna knjižnica kot popolnoma nova organizacijska oblika, ki ima boljši dostop in dostavo gradiva, nudi močnejšo podporo uporabnikom. V tem modelu prevladuje digitalna/virtualna organizacija, za katero je značilno: dinamično vključevanje različnih organizacij v virtualno organizacijo, ki se pojavljajo kot ena sama organizacija; fizična oddaljenost med deli organizacije, ki so povezani preko elektronskih kanalov; in možnost ponudbe običajnih storitev. Te značilnosti kažejo, da digitalna knjižnica vključuje vidike mrežne organizacije. 2.2 Organizacijski vidiki digitalne knjižnice 4) Neposredovalni model (Disintermediation Model): knjižnica je del kroga, v katerem nastopajo avtor, založnik, trgovski posrednik, knjigarna, knjižnica, komercialni informator in uporabnik. V tem krogu vladajo različni odnosi, ki temeljijo na pretoku publikacij. Založniki in posredniki so ključni pri dobavi publikacij, medtem ko knjižnica posreduje med založnikom in uporabnikom, med posrednikom in uporabnikom ter delno med komercialnim informatorjem in uporabnikom. Ta model se sčasoma spreminja, saj nastopajo novi odnosi – založnik npr. dobavi knjigo neposredno uporabniku, še bolj radikalno spremembo pa predstavlja neposreden pretok publikacije od avtorja do uporabnika, kjer se preskoči vse faze pretoka. Tudi v primeru mrežnih knjigarn knjižnica ni več posredovalec dostopa publikacij za domačo uporabo. Ti novi odnosi knjižnico izključujejo iz njene ustaljene vloge, kar postane prikladno drugim elementom kroga. Založnik, knjigarna in avtor postanejo posredovalci digitalne knjižnice. Podsistemski model je torej nova organizacija kot del že obstoječe knjižnice, medtem ko notranje-organizacijski model zahteva večje organizacijske spremembe knjižnice. Še radikalnejše spremembe nastopajo pri sistemskem in neposredovalnem modelu. Uporaba kateregakoli izmed teh modelov zahteva upoštevanje organizacijske razsežnosti strukture, kulture, politike, profesionalnih spretnosti, menedžmenta, tehnologije in notranje- organizacijskih odnosov. 5 5 Travicove raziskave so pokazale, da je v ameriških univerzitetnih knjižnicah, ki jih je naključno izbrala American Library Directory, najvišje zastopan podsistemski model, najmanj oz. sploh pa ni zastopan neposredovalni model. Digitalna knjižnica ni zastopana kot radikalna sprememba v organizaciji, ampak je splošno mnenje o njej le nekoliko oddaljeno od klasične knjižnice. Digitalna knjižnica je torej največkrat prisotna kot podsistem znotraj knjižnične organizacije, bolj kot zbirka CD-ROM-ov kakor pa radikalno nova organizacijska oblika, kot predlaga sistemski model. 3 Digitalne zbirke fakultet Univerze v Ljubljani Digitalne zbirke ljubljanskih fakultet smo pregledovali po določenem vzorcu: zastavili smo si pet ključnih vprašanj, ki se nanašajo na digitalne knjižnice, oz. njihovi pritrdilni odgovori predstavljajo glavne značilnosti visokošolske digitalne knjižnice. 1. Ali zbirka vsebuje seznam študijske literature? 2. Ali je študijska literatura navedena v polnem tekstu? 3. Ali zbirka vsebuje gradivo s predavanj, seminarske in diplomske naloge? 4. Ali so vključeni tudi primeri vprašanj za izpite in kolokvije? 5. Ocena oblikovne podobe zbirke. 1. Ali zbirka vsebuje seznam študijske literature? 4. Ali so vključeni tudi primeri vprašanj za izpite in kolokvije? 5. Ocena oblikovne podobe zbirke. Po pregledu večine ljubljanskih fakultet na svetovnem spletu smo ugotovili, da imajo vse fakultete že oblikovane predstavitvene strani. Nas so seveda zanimale predvsem digitalne knjižnice. V splošnem prevladujejo zbirke seznamov literature za študente, ponekod z abstrakti, lahko pa so prisotni tudi polni teksti predavanj, vendar je to bolj izjema kot pravilo. Približno pri tretjini pregledanih zbirk se nahajajo tudi vprašanja z izpitov in kolokvijev. Seznami seminarskih in diplomskih nalog so prisotni v manjši meri, opise vsebin predmetov pa najdemo skoraj povsod. Vse je v večji meri prepuščeno posamezni fakulteti in njenim interesom, ažurnosti ter zagnanosti študentov v tej smeri. Veliko je poskusov prav s strani študentov, ki zbirajo gradivo za digitalno knjižnico na lastno pobudo. Digitalne zbirke pa se tako po vsebinski zasnovi kot oblikovni podobi močno razlikujejo tudi znotraj fakultet, saj imajo različni oddelki popolnoma različen odnos do elektronske oblike študijskega gradiva in predstavitve svojega oddelka. Tako lahko povzamemo, da gre v večini primerov le za poskuse oblikovanja digitalnih zbirk, o katerih lahko marsikdaj govorimo le kot o razširitvi predstavitvene strani. Pri tistih digitalnih zbirkah, kjer bi že lahko govorili o pravih digitalnih knjižnicah, pa naletimo na problem nedostopnosti za javnost – uporabljajo jih lahko samo študenti tistih fakultet, ki so jim zbirke 6 namenjene, oz. s pomočjo fakultetnih računalnikov ali z uporabniškim geslom. Zato nam je bil vpogled v prave digitalne zbirke oz. knjižnice fakultet onemogočen. Od zbirk, ki so se najbolj približale pojmovanju digitalne knjižnice, velja omeniti zbirko elektronskih dokumentov (predvsem gre za online dostopne elektronske revije) »Virtualna knjižnica« (»Elektronska čitalnica«) Ekonomske fakultete4. Zbirka Medicinske fakultete »Digitalna knjižnica«5 nam nudi dostop do elektronskih revij v polnem tekstu (ISIS, JAMA) ali pa so dostopni abstrakti posameznih člankov (Zdravniški vestnik, Medicinski razgledi). Vanjo je vključena tudi zbirka medicinskih multimedijskih knjižnic. Prav tako vreden ogleda je dostop do elektronskih revij Centralne tehniške knjižnice6. 3 Digitalne zbirke fakultet Univerze v Ljubljani Posebej je potrebno omeniti še projekt »Virtualna univerza«7, ki je začel nastajati leta 1998 na Fakulteti za matematiko in fiziko. Gre za interaktivno računalniško bazo, v katero naj bi bile vključene vse fakultete Univerze v Ljubljani. V enotni obliki naj bi vsebovala študijske pripomočke kot so: kratki povzetki predavanj, seznami literature, primeri nalog z izpitov in kolokvijev, primeri vprašanj ustnih izpitov, obvestila v zvezi s posameznimi predmeti, urniki itd. Računalniška baza je že zasnovana, ključni korak pa je navezava stikov s profesorji in študenti vseh ljubljanskih fakultet, ki bi bili odgovorni za pošiljanje ažurnih informacij in gradiva. Vsekakor je velika prednost projekta enotna zbirka, ki bo dostopna na enem mestu in bo imela enotno obliko za vse fakultete. 4 URL: http://www.cek.ef.uni-lj.si/vk/index.html 5 6 URL: http://www.ctk.uni-lj.si/revije.html 7 7 URL: http://vstudent.fmf.uni-lj.si/index.htm 4 URL: http://www.cek.ef.uni-lj.si/vk/index.html 5 URL: http://www mf uni lj si/mf/d knjiznica/inde 4 URL: http://www.cek.ef.uni-lj.si/vk/index.html 5 URL: http://www.mf.uni-lj.si/mf/d-knjiznica/index.html 6 URL: http://www.ctk.uni-lj.si/revije.html 7 URL: http://vstudent.fmf.uni-lj.si/index.htm 4.1 Metodologija Osnova naše raziskave je bila anketa, ki smo jo izvedli konec novembra in v začetku decembra leta 1999. Anketo smo izvedli na Filozofski fakulteti v Ljubljani, ki združuje 22 oddelkov. Oddelek za bibliotekarstvo je eden najmlajših na fakulteti; ustanovljen je bil v študijskem letu 1987/1988 in je imel v času naše raziskave vpisanih 352 študentov. V anketi je sodelovalo skupno 275 študentov Filozofske fakultete v Ljubljani, kar je predstavljalo 5,2% študentov celotne fakultete. Od tega je bilo 83 (30,2%) študentov bibliotekarstva, ostali pa so bili z drugih smeri: največ je bilo študentov geografije (15,1%), 7 potem študentov slovenskega jezika in književnosti (13,5%), sledijo študenti filozofije (12%) in študenti umetnostne zgodovine (11,5%)8. Za obe skupini (študenti bibliotekarstva in ostali) je značilna prevlada ženskega spola: pri »nebibliotekarjih« je bilo anketiranih 73% študentk, bodočih bibliotekark pa celo 77%. Porazdelitev po letnikih je pri bibliotekarjih enakomernejša kot pri drugi skupini. Pri nebibliotekarjih je na vprašalnik odgovorilo kar 44,3% študentov prvih letnikov, sledili so drugi (19,3%), tretji in četrti letniki (vsak po 13%) ter absolventi (10,4%). Pri študentih bibliotekarstva prevladuje tretji letnik (33,7%), sledita mu prvi (22,9%) ter drugi (19,3%), nadalje četrti (15,7%) ter absolventi (8,4%). Pri obeh skupinah tudi prevladuje redni način študija (pri obeh je več kot 97% študentov rednih), kar je gotovo posledica anketiranja v času, ko na fakulteti ne potekajo predavanja za izredne študente. Ker se naša anketa dotika tudi uporabe računalnikov in interneta, je treba povedati, da je fakulteta dobila računalniško učilnico konec osemdesetih let. Do računalnikov je imel dostop vsak študent Filozofske fakultete, vendar pa je bil zaradi prostorske stiske (v isti učilnici so potekale tudi obvezne vaje) in uničevanja programske opreme dostop v času anketiranja onemogočen. Študenti tako niso imeli dostopa do interneta na fakulteti, ker je učilnica z računalniki zaklenjena in namenjena samo predavanjem in vajam. Po naši anketi 17% anketiranih uporablja Internet na fakulteti, vendar je treba opozoriti, da gre pri tej skupini verjetno za študente, ki imajo pri obveznih vajah dostop do Interneta, možno pa je tudi, da jih nekaj uporablja Internet na kateri od drugih fakultet (anketa namreč ne omogoča opredelitve fakultete). Tudi iz naše ankete je razvidno, da je bil dostop do računalnikov in interneta v času anketiranja zelo omejen, kar študentom ni bilo po godu (»Na fakulteti nikakor nimam dostopa do Interneta.«9). 4.1 Metodologija V času, ko smo anketo zaključili, so na FF znova uredili dostop do računalnikov in interneta, vendar v manjši učilnici in s slabšo opremo. Prav tako še vedno prihaja do problemov pri vzdrževanju programske opreme in nadzoru dostopnosti do računalnikov. Anketa za študente nebibliotekarje je obsegala 13 vsebinskih vprašanj, za študente bibliotekarstva pa smo tem dodali še 7 dodatnih vprašanj, ki se v glavnem nanašajo na uporabo obstoječih zbirk predavanj v elektronski obliki. 8 Gre za študente, ki študirajo enopredmetno; od dvopredmetnih prevladujejo študenti sociologije (13,5 %). 9 Citat iz odgovora na odprti tip vprašanja. 8 Gre za študente, ki študirajo enopredmetno; od dvopredmetnih prevladujejo študenti sociologije (13,5 % 9 Citat iz odgovora na odprti tip vprašanja. 4.2 Rezultati ankete Z anketo smo želeli raziskati morebitne razlike pri uporabi interneta in poznavanju pojma »digitalna knjižnica« med študenti bibliotekarstva in ostalimi študenti Filozofske fakultete. 8 8 Predpostavili smo, da je študentom bibliotekarstva izraz digitalna knjižnica bolj domač, saj ga spoznajo med študijem. Prav tako smo predvidevali, da so študenti bibliotekarstva računalniku in internetu v splošnem bolj naklonjeni kot ostali študenti. Predstavljeni rezultati prikazujejo razlike, ki so prišle najbolj do izraza. 1. vprašanje: Kako bi ocenil-a svoj odnos do osebnega računalnika? 1. vprašanje: Kako bi ocenil-a svoj odnos do osebnega računalnika? 0% 10% 20% 30% 40% 50% 60% Delo za računalnikom mi je povsem tuje. Nerad uporabljam osebni računalnik, vendar moram. Navdušen sem nad osebnim računalnikom, vendar nimam možnosti uporabe. Osebni računalnik uporabljam pogosto in z veseljem. Drugo. Študenti bibliotekarstva Ostali študenti Študenti bibliotekarstva Ostali študenti Grafikon 1: Odnos do osebnega računalnika Tako študenti nebibliotekarji kot ostali študenti v večini uporabljajo računalnik pogosto in z veseljem. Ta podatek je precej presenetljiv, saj gre za študente humanističnih znanosti, za katere bi predvidevali, da pri delu redkeje in z manjšim veseljem uporabljajo računalnik.10 Velike razlike se pokažejo med študenti različnih smeri. Vendar pa je za nas najbolj zanimiva primerjava med študenti bibliotekarstva in ostalimi študenti: kar 53% prvih uporablja računalnik pogosto in z veseljem, ostalih študentov v tej skupini je 42,7%. Osebni računalnik je prisiljenih uporabljati 26,5% študentov bibliotekarstva in 23,4% ostalih študentov. Vzroki za nujnost uporabe računalnika so različni - lahko je uporaba računalnika neobhodna pri samem 9 9 študiju ali pa je celo vir eksistenčnega zaslužka (»Brez računalnika ni hrane.«11). V naslednji skupini so študenti, ki so sicer nad računalnikom navdušeni, vendar pa nimajo možnosti uporabe: 8,5% takih je študentov bibliotekarstva, ostalih pa 12,5%. Manjšini je delo za računalnikom povsem tuje: 2,4% študentov bibliotekarstva in 6,3% ostalih študentov FF. Vzroke je iskati v njihovem neznanju (»Nimam dovolj znanja.«12). Na odprti tip vprašanja je odgovorilo 9,6% študentov bibliotekarstva in kar 15,1% ostalih študentov. Vse te odgovore lahko razvrstimo v tri skupine: takšne, ki kažejo izrazito pozitiven odnos do računalnika in ga uporabljajo z entuziazmom, nadalje v odgovore, ki vsebujejo indiferentno stališče do uporabe računalnika (»delo z računalnikom ni vedno prijetno«, »menim, da je dobro povezati prijetno s koristnim…«13) in nazadnje v odgovore, ki nakazujejo neuporabo računalnika. 13 Citata iz odgovora na odprti tip vprašanja. 10 V zadnjih letih se je odnos do računalnika spremenil pri študentih humanističnih znanosti 11 Cit t i d d ti ti š j 10 V zadnjih letih se je odnos do računalnika spremenil pri študentih humanističnih znanosti. 11 Citat iz odgovora na odprti tip vprašanja. 12 Citat iz odgovora na odprti tip vprašanja. 13 Citata iz odgovora na odprti tip vprašanja. j j 11 Citat iz odgovora na odprti tip vprašanja. 12 12 Citat iz odgovora na odprti tip vprašanja. 13 Citat iz odgovora na odprti tip vprašanja. 12 Citat iz odgovora na odprti tip vprašanja. 4.2 Rezultati ankete Strnemo lahko, da so si rezultati med študenti bibliotekarstva in ostalimi študenti Filozofske fakultete podobni, bistvena razlika je v večji osveščenosti študentov bibliotekarstva glede aplikativnih razsežnosti uporabe računalnika. Vzrok gre pripisati sami naravi študija bibliotekarstva na FF, ki daje precejšen poudarek delu z računalniki. 2. vprašanje: Ali poznaš izraz digitalna oz. virtualna knjižnica? 2. vprašanje: Ali poznaš izraz digitalna oz. virtualna knjižnica? 3. vprašanje: Kaj pojmuješ pod tem izrazom? 2. vprašanje: Ali poznaš izraz digitalna oz. virtualna knjižnica? 2. vprašanje: Ali poznaš izraz digitalna oz. virtualna knjižnica? 10 0% 10% 20% 30% 40% 50% 60% 70% 80% DA NE NI ODGOVORA Študenti bibliotekarstva Ostali študenti 0% 10% 20% 30% 40% 50% 60% 70% 80% DA NE NI ODGOVORA Študenti bibliotekarstva Ostali študenti Grafikon 2: Seznanjenost z izrazom digitalna oz. virtualna knjižnica DA NE Grafikon 2: Seznanjenost z izrazom digitalna oz. virtualna knjižnica S tem vprašanjem smo želeli dobiti anketirančevo samooceno poznavanja izraza digitalna knjižnica, medtem ko naj bi naslednje vprašanje (glej vprašanje št. 3) pokazalo, kaj anketiranci dejansko pojmujejo pod tem izrazom. S tem vprašanjem smo želeli dobiti anketirančevo samooceno poznavanja izraza digitalna knjižnica, medtem ko naj bi naslednje vprašanje (glej vprašanje št. 3) pokazalo, kaj anketiranci dejansko pojmujejo pod tem izrazom. Razlika med študenti bibliotekarstva in ostalimi študenti je pri odgovoru na vprašanje 2 očitna, saj kar 71,1% študentov bibliotekarstva po lastni opredelitvi že pozna izraz digitalna oz. virtualna knjižnica, medtem ko je ta izraz poznan »le« 43,8% ostalih študentov. Gotovo gre razloge za tako razliko iskati v dejstvu, da se med študijskim procesom študenti bibliotekarstva pri različnih predmetih srečujemo s tem izrazom. K poznavanju izraza gotovo pripomore tudi praktična uporaba, saj sta v času anketiranja dva profesorja Oddelka za bibliotekarstvo nudila svoja predavanja v elektronski obliki, dostopna na mreži. 3. vprašanje: Kaj pojmuješ pod tem izrazom? 11 0% 10% 20% 30% 40% 50% 60% 70% 80% Zbirka medsebojno povezanih sistemov in virov, dostopnih preko mreže. Internet. Online knjižnični katalog. Drugo. Brez odgovora. Študenti bibliotekarstva Ostali študenti Grafikon 3: Razumevanje izraza digitalna oz. virtualna knjižnica Brez odgovora. Online knjižnični katalog. Drugo. Internet. Grafikon 3: Razumevanje izraza digitalna oz. virtualna knjižnica Pri tem vprašanju je potrebno opozoriti na splošni terminološki problem in problem definiranja izraza digitalna knjižnica. V anketi smo ponudili tri različne odgovore, četrta možnost pa je bila, da je anketiranec sam napisal, kaj pojmuje pod tem izrazom. Skladno z našo opredelitvijo pojma digitalna knjižnica (glej teoretični del) smo za pravilnega vzeli prvi odgovor (digitalna knjižnica je zbirka medsebojno povezanih sistemov in virov, dostopnih preko mreže), za katerega se je odločilo največ anketiranih. Preostali možnosti sta po našem pojmovanju digitalne knjižnice zavajajoči. Predvidevamo, da je veliko anketirancev pri odgovarjanju ugibalo, zato je na odločitev za prvi ponujeni odgovor najbrž vplivala tudi strokovna formulacija odgovora, ki je anketirancem dajala vtis pravilnega odgovora. 2. vprašanje: Ali poznaš izraz digitalna oz. virtualna knjižnica? Dejstvo je, da je na kratko težko zajeti bistvo digitalne knjižnice, saj se le-to zaradi neuveljavljenosti, nedorečenosti in množice različnih pogledov in definicij pojma digitalne knjižnice pravzaprav še ni izoblikovalo. Navkljub tem problemom lahko podamo splošno sliko, ki spet kaže na boljše poznavanje in seznanjenost študentov bibliotekarstva z obravnavano tematiko. Kar 72,3% se jih je namreč odločilo za prvi odgovor, medtem ko je ostalih študentov v tej skupini 52,6%. 12 Zanimivo je, da se je kar 26 % študentov nebibliotekarjev odločilo za odgovor, da je digitalna knjižnica online knjižnični katalog (študentov bibliotekarstva je bilo v tej skupini 10,8%). Najbrž se jih je nekaj za to možnost odločilo zaradi povezave online kataloga z bibliotekarstvom, saj so anketiranci vedeli, da anketo opravljamo študenti bibliotekarstva. Verjetno pa gre poglavitni vzrok iskati v napačnem pojmovanju sicer zelo popularnega slovenskega online knjižničnega kataloga COBISS kot slovenske virtualne knjižnice14. Za pojmovanje interneta kot digitalne knjižnice se je odločilo najmanj študentov iz obeh skupin (študentov bibliotekarstva 2,4% in ostalih študentov 1,6%). Precej anket pa je bilo vrnjenih brez odgovora (ali z neveljavnim odgovorom) na to vprašanje (študentov bibliotekarstva 8,5%, ostalih pa kar 17,7%). Le malo anketiranih študentov (6% študentov bibliotekarstva in 2,1% ostalih študentov) se je odločilo »ugovarjati« tem trem opredelitvam digitalne knjižnice, napisali so svoj pogled in razumevanje (npr.: »Digitalna knjižnica predstavlja možnost, da na enem računalniku dobiš informacije o vseh knjižničnih gradivih v vseh knjižnicah v Sloveniji,« ali: »Digitalna knjižnica je knjižnica preko mreže.«15). Kot smo pričakovali, je študentom bibliotekarstva pojem digitalna knjižnica bolj domač in ga tudi pravilneje razumejo kot ostali študenti. g p p p j 16 O pomembnosti digitalnih knjižnic za izobraževanje glej članek Advantages and disadvantages of use of digital collections in the process of education avtorjev Vrane, Badurine in Goluba. 14 Avtorji članka se ne strinjamo s tem javnim poimenovanjem in predlagamo čimprejšnjo spremembo naziva s strani terminološke komisije. 15 Citata iz odgovorov na odprti tip vprašanja. j 15 Citata iz odgovorov na odprti tip vprašanja. 16 14 Avtorji članka se ne strinjamo s tem javnim poimenovanjem in predlagamo čimprejšnjo spremembo strani terminološke komisije. 15 14 Avtorji članka se ne strinjamo s tem javnim poimenovanjem in predlagamo čimprejšnjo spremembo naziva s strani terminološke komisije 5 Projekt vzpostavitve digitalne knjižnice na Oddelku za bibliotekarstvo Filozofske fakultete Pri pripravljanju ankete za študente smo se spraševali, kakšni bodo rezultati. Upali smo, da naši študenti poznajo digitalne knjižnice, predvsem pa, da jih uporabljajo. Naše želje je potrdil tudi rezultat odgovora na vprašanje v anketi za bibliotekarje: »Ali misliš, da študent bibliotekarstva danes ali v bližnji prihodnosti potrebuje digitalno knjižnico?«, kjer je 91,6% študentov odgovorilo, da digitalno knjižnico potrebujemo. Vse skupaj nas je tako spodbudilo, da smo se odločili v prihodnosti izpeljati projekt vzpostavitve Digitalne knjižnice na Oddelku za bibliotekarstvo Filozofske fakultete. Digitalna knjižnica, ki jo oblikujemo, je manjša in specializirana, vendar nujno potrebna v naši stroki, namenjena predvsem študentom in profesorjem16. 13 Načrtovanje tako zahtevnega projekta je za skupino študentov velika naloga, zato se bomo z začetnimi idejami in načrti obrnili na profesorje oddelka. Upamo, da bo pod njihovim mentorstvom in z vključitvijo projekta v študij, projekt, ki zahteva organizacijsko in vsebinsko kontinuirano delo, lažje izvedljiv, hkrati pa bomo dosegli, da bodo študenti bibliotekarstva dobili priložnost sodelovanja pri projektu, s katerim bodo teoretično znanje podkrepili s prakso. Vsebinska zasnova digitalne knjižnice je vzpostaviti zbirko, namenjeno študiju bibliotekarstva. Ob največji problem pri urejanju te zbirke bomo trčili pri zbiranju obvezne literature za izpite, saj so mnoga besedila že objavljena v pisni obliki in s tem podvržena avtorskim pravicam. Problemi, ki bodo ob tem nastajali, so rešljivi, vendar so odvisni od individualnega interesa avtorjev. Čeprav se zavedamo dileme med varovanjem pravic avtorjem in načelom splošne dostopnosti informacij, bi želeli, da bi v našem primeru prevladalo slednje in tako omogočilo nastanek dobre zbirke v izobraževalne namene. 6 Zaključek Paradigma digitalne knjižnice, o kateri govorimo, je odziv na hitre spremembe v visokem šolstvu in bibliotekarstvu, na katerega vplivajo tehnološki razvoj in njegove možnosti. Knjižnične usluge digitalnih knjižnic so enake uslugam klasičnih knjižnic, vendar so ustrezno prilagojene novim tehničnim možnostim. Te spremembe se nam kažejo v dveh vidikih: kot študentom (uporabnikom) in kot bodočim bibliotekarjem – informacijskim strokovnjakom, ki nudijo te usluge drugim uporabnikom. S svojo raziskavo smo lahko samo potrdili pričakovano vsesplošno razumevanje in pozitivno sprejemanje možnosti, ki jih nudijo digitalne knjižnice s strani študentov. Te rezultate smo namreč pričakovali, četudi morda ne v tolikšnem obsegu. Velika večina študentov že uporablja digitalno knjižnico v različnih oblikah in ima očitno že potrebne sposobnosti ter pozitiven pristop do nje. Nova oblika knjižnice se pojavi, ko je zbrana zadostna količina dokumentov v digitalni obliki, in je zadostno število uporabnikov bolj naklonjeno digitalnim kot dokumentom v klasični obliki. Prav slednje je prava prednost nove informacijske dobe knjižnic. Menimo, da govorijo rezultati naše raziskave temu v prid. Torej, prihodnost je videti svetla. Vendar se dobro zavedamo, da so za uspešno delovanje digitalne knjižnice poleg tehnične osnove in uporabnikovih potreb pomembni tudi finančni in organizacijski vidiki. Študentska 14 iniciativa je pomemben dejavnik, in upamo, da bo imel naš projekt dolgoročne učinke, predvsem pa pozitiven vpliv na našo profesionalnost. Vendar obstaja dilema, ali bodo današnji študenti na delovnih mestih imeli možnost graditve digitalnih knjižnic. Nedvomno bi več sredstev za organizacijo digitalnih knjižnic in izboljšan dostop do gradiva v tej obliki moralo slediti tu ugotovljenemu interesu in znanju študentov na naši univerzi. Avtorji smo prepričani, da se bo nadaljnji razvoj tudi v Sloveniji usmeril k izboljšavi izgradnje in k širši uporabi digitalnih knjižnic. Študenti so izrazili potrebo po čim večji količini digitalnega gradiva in po dostopu do digitalne knjižnice, čemur se mora prilagoditi tudi stroka. V študentih je torej dober potencial, vprašanje je le, ali jih bo stroka znala uporabiti. Vsekakor je digitalna knjižnica pomembna za celotno knjižničarsko stroko in ne samo za študij. Čas, ki prihaja, je čas sprememb predvsem na področju uporabe informacijskih tehnologij, kar je in mora biti izziv knjižničarski stroki. Le z aktivnim prizadevanjem v tej smeri bomo slovenski knjižničarji pridobili na statusu stroke. Blaž Lesnik, Brina Jerič, Tanja Curhalek in Martina Kerec so študentje bibliotekarstva na Oddelku za bibliotekarstvo, Filozofska fakulteta Univerze v Ljubljani. Naslov: Aškerčeva 2, Ljubljana Naslov elektronske pošte: blaz.lesnik@kiss.uni-lj.si, brina.jeric@kiss.uni-lj.si, tanja.curhalek@amis.net, martina.kerec@siol.net 7 Citirani viri 1. Bawden, D., & Rowlands, I. (1999). Understanding digital libraries: towards a conceptual framework. London: City University London. 2. Cleveland, G. Digital libraries: definitions, issues and challenges. (marec, 1998). Ottawa, CA: IFLA UDT Core Programme. Pridobljeno 9.9.1999 s spletne strani: http://www.ifla.org/udt/op/ 2. Cleveland, G. Digital libraries: definitions, issues and challenges. (marec, 1998). Ottawa, CA: IFLA UDT Core Programme. Pridobljeno 9.9.1999 s spletne strani: http://www.ifla.org/udt/op/ 2. Cleveland, G. Digital libraries: definitions, issues and challenges. (marec, 1998). Ottawa, CA: IFLA UDT Core Programme. Pridobljeno 9.9.1999 s spletne strani: http://www.ifla.org/udt/op/ 3. Digital library on the sunny side of the Alps (2000). V: Access 2000 : intellectual property vs. the right to knowledge? Proceedings of the 8th International BOBCATSSS Symposium on Library and Information Science (str. 103-113). Copenhagen: Royal School of Library and Information Science. 3. Digital library on the sunny side of the Alps (2000). V: Access 2000 : intellectual property vs. the right to knowledge? Proceedings of the 8th International BOBCATSSS Symposium on Library and Information Science (str. 103-113). Copenhagen: Royal School of Library and Information Science. 4. Južnič, P. (1995). Virtualna knjižnica. Knjižnica, 39(4), 31-38. 4. Južnič, P. (1995). Virtualna knjižnica. Knjižnica, 39(4), 31-38. 5. Travica, B. (1999). Organizational aspects of the virtual library: a survey of academic libraries. Library and Information Science Research, 21(2), 173-203. 6. Vrana, R., Badurina, B., & Golub, K. (1999). Advantages and disadvantages of use of digital collections in the process of education. V: Learning society, learning organization, lifelong learning. Proceedings of the 7th International BOBCATSSS Symposium (str. 484-495). Darmstadt: Fachhochschule Darmstadt, University of Applied Science, Department of Information Science, Stuttgart : Fachhochschule Stuttgart, Hochschule für Bibliotheks und Informationswesen. 15
https://openalex.org/W3101940296	https://link.springer.com/content/pdf/10.1007%2Fs00332-016-9345-2.pdf	English	null	Isotropy of Angular Frequencies and Weak Chimeras with Broken Symmetry	Journal of nonlinear science	2,016	cc-by	11,927	B Christian Bick c.bick@exeter.ac.uk Isotropy of Angular Frequencies and Weak Chimeras with Broken Symmetry Christian Bick1 Received: 20 November 2015 / Accepted: 18 October 2016 / Published online: 10 November 2016 © The Author(s) 2016. This article is published with open access at Springerlink.com Abstract The notion of a weak chimeras provides a tractable deﬁnition for chimera states in networks of ﬁnitely many phase oscillators. Here, we generalize the deﬁni- tion of a weak chimera to a more general class of equivariant dynamical systems by characterizing solutions in terms of the isotropy of their angular frequency vector—for coupled phase oscillators the angular frequency vector is given by the average of the vector ﬁeld along a trajectory. Symmetries of solutions automatically imply angular frequency synchronization. We show that the presence of such symmetries is not nec- essary by giving a result for the existence of weak chimeras without instantaneous or setwise symmetries for coupled phase oscillators. Moreover, we construct a coupling function that gives rise to chaotic weak chimeras without symmetry in weakly coupled populations of phase oscillators with generalized coupling. Keywords Oscillator networks · Phase oscillators · Weak chimera · Symmetry · Asymptotic average frequencies Mathematics Subject Classiﬁcation 34C15 · 37E45 · 37C80 Mathematics Subject Classiﬁcation 34C15 · 37E45 · 37C80 J Nonlinear Sci (2017) 27:605–626 DOI 10.1007/s00332-016-9345-2 J Nonlinear Sci (2017) 27:605–626 DOI 10.1007/s00332-016-9345-2 1 Centre for Systems, Dynamics and Control and Department of Mathematics, University of Exeter, Exeter EX4 4QF, UK 1 Introduction The emergence of collective dynamics in networks of coupled oscillatory units is a fas- cinating phenomenon observed in science and technology (Strogatz 2000; Pikovsky Communicated by Eusebius Doedel. 1 Centre for Systems, Dynamics and Control and Department of Mathematics, University of Exeter, Exeter EX4 4QF, UK 12 3 J Nonlinear Sci (2017) 27:605–626 606 et al. 2003). Symmetric phase oscillator networks provide paradigmatic models to understand collective dynamics in the weak coupling limit (Strogatz 2004; Acebrón et al. 2005; Tchistiakov 1996; Ashwin et al. 2016c). Such dynamical systems with symmetry are equivariant with respect to the action of a group (Golubitsky et al. 1988; Golubitsky and Stewart 2002; Field 2007), that is, the vector ﬁeld commutes with the group action on phase space. Equivariance implies that any solution of the system is mapped to another solution by the action of the symmetry group and it typically con- strains the dynamics, for example, by giving rise to dynamically invariant subspaces. The solutions themselves may (but do not have to) have nontrivial symmetry, that is, there may be nontrivial elements elements of the symmetry group that keep the solution ﬁxed, either pointwise or as a set. For example, for globally coupled identical oscillators, the solution corresponding to full synchrony, where the states of all oscil- lators are equal, has full symmetry itself. Of course, there may be other solution with less symmetry relative to the symmetries of the system. Recently, the observation of “symmetry breaking” in symmetrically coupled phase oscillator systems, i.e., the observation of solutions with localized synchronous dynamics coexisting with localized incoherence, has sparked a lot of interest. Such solutions, commonly known as chimera states—see Panaggio and Abrams (2015) for a recent review—were ﬁrst observed in symmetric rings of coupled phase oscilla- tors (Kuramoto and Battogtokh 2002; Abrams and Strogatz 2004). In the limit of inﬁnitely many oscillators, they correspond to stationary or periodic patterns of the phase density distribution (Abrams et al. 2008; Omel’chenko 2013). By contrast, it was not until recently that Ashwin and Burylko (2015) gave a testable mathematical deﬁnition for chimera states, a weak chimera, for networks of ﬁnitely many phase oscillators whose phases ϕk ∈T = R/2πZ, k = 1, . . . , n, evolve according to dϕk dt = ˙ϕk = ω + 1 n n j=1 Hkjg(ϕk −ϕ j). 1 Introduction (1) (1) Here, the Hkj determine the network topology (respecting a subgroup of the group Sn of permutations of n symbols acting transitively on the indices of the oscillators) and g:T →R is the generalized coupling (or phase interaction) function. Weak chimeras are deﬁned in terms of partial angular frequency synchronization on tra- jectories. More precisely, if ˆϕ is a continuous lift of a solution ϕ of (1) with initial condition ϕ0 to Rn deﬁne the asymptotic angular frequency of oscillator k as k ϕ0 = lim T →∞ ˆϕ(T ) T . (2) (2) (2) According to Ashwin and Burylko (2015), a compact, connected, chain-recurrent, and dynamically invariant set A ⊂Tn is a weak chimera if there are distinct oscillators j, k, ℓsuch that j ϕ0 = k ϕ0 ̸= ℓ ϕ0 for all ϕ0 ∈A. j Weak chimeras and angular frequency synchronization relate to symmetry. Assum- ing that all limits (2) exist, we have a frequency vector ϕ0 = 1 ϕ0 , . . . , n ϕ0 ∈Rn. 123 123 607 J Nonlinear Sci (2017) 27:605–626 The group Sn also acts on Rn by permuting indices. If A is a weak chimera as above and τkj ∈Sn denotes the transposition swapping indices k and j, then τkj ϕ0 = ϕ0 . Thatis,τkj isasymmetryoftheangularfrequencyvector ϕ0 .Whileweakchimeras have provided a suitable framework to derive for example existence results (Bick and Ashwin 2016), there are two shortcomings. First, while chimera states have also been reported in more general oscillator models (Sethia et al. 2013; Zakharova et al. 2014), the deﬁnition above applies to phase oscillators only. Second, the symmetries of the angular frequency vector may be different from the symmetries of the system. As a consequence, if A is a weak chimera, then τkj A may not be a weak chimera or even a solution of the system at all. Interestingly, while it has been argued that chimera states are relevant due to their nature of solutions with broken symmetry (Abrams and Strogatz 2004), their properties have never been phrased in terms of symmetries of the dynamical system. The group Sn also acts on Rn by permuting indices. 2.1 Quasi-Regular Points Let X be a compact differentiable manifold with a ﬂow t : X →X, t ∈R. A point x ∈X is quasi-regular if the limit lim T →∞ 1 T T 0 f ( t(x)) dt exists for all continuous functions f : X →R. exists for all continuous functions f : X →R. exists for all continuous functions f : X →R. Theorem 1 (Schwartzman 1957; Oxtoby 1952) The set of points which are not quasi- regular has zero measure with respect to every ﬁnite measure on X that is invariant under the ﬂow t. 1 Introduction If A is a weak chimera as above and τkj ∈Sn denotes the transposition swapping indices k and j, then τkj ϕ0 = ϕ0 . Thatis τkj isasymmetryoftheangularfrequencyvector ϕ0 Whileweakchimeras The contribution of this paper is twofold: First, we give a deﬁnition of a weak chimera in the language of equivariant dynamical systems and, second, we show that symmetries of the solution are not necessary for the occurrence of weak chimeras. More precisely, we deﬁne weak chimeras in terms of the isotropy of the angu- lar frequency vector which can be stated for more general oscillator systems. We observe that, in a suitable setup, asymptotic angular frequencies are averages of equivariant observables. Therefore, symmetries of solutions translate directly into symmetries of the angular frequencies. Thus, the presence of symmetries of solu- tions facilitates the emergence of weak chimeras and, in fact, most weak chimeras that have been constructed explicitly (Ashwin and Burylko 2015; Panaggio et al. 2016; Bick and Ashwin 2016) are solutions with (instantaneous) symmetries. Is it possible to construct weak chimeras without instantaneous or setwise symmetries for which the angular frequencies have symmetries that are not a property of the solution itself? This question motivates the second contribution. Extending recent persistence results (Bick and Ashwin 2016) that rely on constructing generalized coupling func- tions between oscillators, we prove a persistence result for weak chimeras with trivial symmetry in weakly coupled populations of phase oscillators. Moreover, we present an explicit example of a C∞coupling function that gives rise to a chaotic weak chimera without instantaneous or setwise symmetries in a nontrivially coupled sys- tem. This paper is organized as follows. In Sect. 2, we review some terminology on equivariant dynamics that is needed in the subsequent sections. In Sect. 3, we then apply these notions to general oscillator systems with symmetry which yields a new deﬁnition of a weak chimera in terms of symmetries of the angular frequency vector. As we show in Sect. 4, this deﬁnition is compatible with previous deﬁnitions. In Sect. 5, we prove a persistence result for weak chimeras without instantaneous or average symmetries. Finally, we present an explicit example of a coupling function which gives rise to chaotic weak chimeras with trivial symmetries in Sect. 6 and ﬁnish with some concluding remarks. 12 3 J Nonlinear Sci (2017) 27:605–626 608 2.2 Equivariant Dynamical Systems However, it does not necessarily hold for dynamically invariant sets of saddle type, sets that are attracting (or repelling) in a more general sense, or heteroclinic attractors. Remark 1 The same statement holds for repellers—dynamically invariant sets that are attractors when time is reversed. However, it does not necessarily hold for dynamically invariant sets of saddle type, sets that are attracting (or repelling) in a more general sense, or heteroclinic attractors. For δ > 0, let Bδ(A) denote an (open) δ-neighborhood of A. Corollary 1 Let ⊂O(n) be a ﬁnite subgroup and let A ⊂Rn be compact attractor for the ﬂow deﬁned by (4). If (A) = {id}, then there exists a δ > 0 such that (D) = {id} for any D ⊂Bδ(A). Corollary 1 Let ⊂O(n) be a ﬁnite subgroup and let A ⊂Rn be compact attractor for the ﬂow deﬁned by (4). If (A) = {id}, then there exists a δ > 0 such that (D) = {id} for any D ⊂Bδ(A). Proof Suppose that (A) = {id}. By Proposition 1, we have γ A ∩A = ∅for any γ ̸= id. Since A is closed, there exists a δ > 0 such that γ Bδ(A) ∩Bδ(A) = ∅. Therefore, (D) = {id} for any D ⊂Bδ(A). ⊓⊔ Proof Suppose that (A) = {id}. By Proposition 1, we have γ A ∩A = ∅for any γ ̸= id. Since A is closed, there exists a δ > 0 such that γ Bδ(A) ∩Bδ(A) = ∅. Therefore, (D) = {id} for any D ⊂Bδ(A). ⊓⊔ 2.2 Equivariant Dynamical Systems Let F : X →TX be a smooth vector ﬁeld on X where TX denotes the tangent bundle. Suppose that a group acts on X. The vector ﬁeld F is -equivariant if F(γ x) = ˆγ F(x) (3) (3) for all γ ∈ where ˆγ is the induced action on the tangent space. A -equivariant vector ﬁeld deﬁnes a -equivariant dynamical system ˙x = F(x) (4) ˙x = F(x) (4) on X (Golubitsky and Stewart 2002; Field 2007). For a set A ⊂X deﬁne the set of instantaneous symmetries T (A) = { γ ∈ \| γ x = x for all x ∈A} (5) (5) and the set of symmetries on average (or setwise symmetries) and the set of symmetries on average (or setwise symmetries) (A) = { γ ∈ \| γ A = A} . (6) (6) Clearly, T (A) ⊂(A). If x = { γ ∈ \| γ x = x } denotes the stabilizer or isotropy subgroup of x ∈X we have T (A) = x∈A x. x∈A Note that if γ A ∩A = ∅for all γ ∈ ∖{id}, then (A) = {id}. The converse holds only under additional assumptions (Ashwin 1995). Henceforth, let t : X →X, t ∈R, denote the ﬂow deﬁned by the differential Eq. (4). A set A ⊂X is (forward) ﬂow-invariant or dynamically invariant if t(A) ⊂A for all t ≥0. Moreover, A is stable if for every neighborhood U of A, there exists an open neighborhood V ⊂U of A such that t(V ) ⊂U for all t ≥0. A compact stable set A is an attractor if A = ω(x) is the ω-limit set of some point x ∈X. For attractors and the action of the orthogonal group O(n) on Rn, there is the following dichotomy (Melbourne et al. 1993) that characterizes the symmetries on average. 123 123 J Nonlinear Sci (2017) 27:605–626 609 Proposition 1 Let ⊂O(n) be a ﬁnite subgroup. For an attractor A ⊂Rn, we have for any γ ∈ either γ A = A or γ A ∩A = ∅. Proposition 1 Let ⊂O(n) be a ﬁnite subgroup. For an attractor A ⊂Rn, we have for any γ ∈ either γ A = A or γ A ∩A = ∅. Remark 1 The same statement holds for repellers—dynamically invariant sets that are attractors when time is reversed. 2.3 Equivariant Observables Suppose that acts on both X and Rm for some m ∈N ∖{0}. Deﬁnition 1 A continuous -equivariant map O : X →Rm is an observable. Deﬁnition 1 A continuous -equivariant map O : X →Rm is an observable Given a solution x(t) of (4) with initial condition x(0) = x0, the limit KO x0 = lim T →∞ 1 T T 0 O (x(t)) dt (7) (7) (if it exists) is an average of O along the trajectory x (integrate componentwise if m > 1). The limit exists in particular for every quasi-regular initial condition x0, and henceforth, we will always assume that x0 ∈X is quasi-regular when averages (7) are evaluated. (if it exists) is an average of O along the trajectory x (integrate componentwise if m > 1). The limit exists in particular for every quasi-regular initial condition x0, and henceforth, we will always assume that x0 ∈X is quasi-regular when averages (7) are evaluated. Suppose that A ⊂X is dynamically invariant and supports a t-invariant ergodic probability measure μ. Write K μ O(A) = A O(x)dμ. (8) (8) By the Birkhoff ergodic theorem (Katok and Hasselblatt 1995, Theorem 4.1.2), we have KO x0 = K μ O(A). (9) (9) for μ-almost every x0 ∈A. In particular, the limit (7) exists for μ-almost every x0 ∈A. For ease of notation, we will simply write KO(A) = K μ O(A) unless the choice of measure is important. Of course, not every ergodic invariant measure is “physically relevant” since μ may be singular with respect to the Lebesgue measure. If an attractor A supports a Sinai–Ruelle–Bowen (SRB) measure μ (Young 2002; 12 3 610 J Nonlinear Sci (2017) 27:605–626 Katok and Hasselblatt 1995), there is a neighborhood W of A such that (9) holds for Lebesgue-almost every x0 ∈W. Thus, the average (8) is observed for “typical” initial conditions with respect to the Lebesgue measure. Now, KO(A) has an isotropy group KO(A) and a simple calculation (Golubitsky and Stewart 2002) shows that (A) ⊂KO(A), (10) (10) that is, any symmetry on average is contained in the isotropy group of the observation The converse does not hold for general observables. Detectives (Golubitsky and Stewart 2002; Barany et al. 1993; Dellnitz et al. 1994; Ashwin and Nicol 1997) are an important class of observables for which the isotropy is generically equal to the symmetries on average. 3.1 Isotropy of Angular Frequencies and Weak Chimeras The symmetry point of view now allows to deﬁne weak chimeras in terms of their sym- metries as solutions relative to the symmetries of the system itself. Write i = √−1. Let ⊂Sn be a subgroup that acts transitively on Cn • := (C ∖{0})n by permuting coor- dinates and suppose that F : Cn • →Cn is -equivariant. The map F = (F1, . . . , Fn) determines a dynamical system on Cn • where the evolution of z = (z1, . . . , zk)1 is given by ˙zk = zk Fk(z). (11) (11) For k ∈{1, . . . , n} deﬁne Ck(z) = zk \|zk\|. (12) (12) In the following, we assume that F is such that (a) the dynamics of (11) are well deﬁned on Cn •, (b) we have ω(z) ⊂Cn • for all z ∈Cn • and (c) the derivative C′ k(z(t)) := d dt Ck(z(t)) exists for any trajectory z(t). These assumptions are easy to work with but can be relaxed as one typically only needs well-deﬁned dynamics on a neighborhood of Tn ⊂Cn In the following, we assume that F is such that (a) the dynamics of (11) are well deﬁned on Cn •, (b) we have ω(z) ⊂Cn • for all z ∈Cn • and (c) the derivative C′ k(z(t)) := d dt Ck(z(t)) exists for any trajectory z(t). These assumptions are easy to work with but In the following, we assume that F is such that (a) the dynamics of (11) are we deﬁned on Cn •, (b) we have ω(z) ⊂Cn • for all z ∈Cn • and (c) the derivative C′ k(z(t)) := d In the following, we assume that F is such that (a) the dynamics of (11) are well deﬁned on Cn •, (b) we have ω(z) ⊂Cn • for all z ∈Cn • and (c) the derivative C′ k(z(t)) := d dt Ck(z(t)) exists for any trajectory z(t). These assumptions are easy to work with but can be relaxed as one typically only needs well-deﬁned dynamics on a neighborhood of Tn ⊂Cn •. Note that Ck(z1, . . . , zn) projects onto the unit circle in the kth coordinate. Let γT denote the parametrized curve in Cn • determined by a solution z(t) of (11) for t ∈[0, T ]. 2.3 Equivariant Observables Given a suitably large m ∈N ∖{0}, an observable is an (ergodic) detective if for any ω-limit set A, there exists an open dense set of near- identity -equivariant diffeomorphisms ψ : Rm →Rm such that KO(ψ(A)) = (A). Hence, detectives are particular observables to “detect” the symmetries of attractors. 3.1 Isotropy of Angular Frequencies and Weak Chimeras The change in argument of zk along γT is given by arg Ck(γT ) = 1 i T 0 C′ k(z(t)) Ck(z(t))dt = T 0 Im(Fk(z(t)))dt. (13) (13) 1 One can think of each complex variables zk representing phase and amplitude of an oscillator. 123 J Nonlinear Sci (2017) 27:605–626 611 Thus, we obtain the average angular frequency in the kth coordinate (equivalent to the average winding number when multiplied by 2π) k z0 := lim T →∞ 1 T T 0 Im (Fk(z(t))) dt (14) (14) along a trajectory z(t) with initial condition z0. along a trajectory z(t) with initial condition z0. along a trajectory z(t) with initial condition z0. along a trajectory z(t) with initial condition z0. Deﬁnition 2 The vector Deﬁnition 2 The vector z0 = 1 z0 , . . . , n z0 is the angular frequency vector of the trajectory with initial condition z0 ∈Cn •. Since Sn also acts on Rn by permuting indices, Im(F) : Cn • →Rn is a -equivariant observable for (11) and z0 = KIm(F) z0 , that is, the angular frequency vector is the observation of Im(F) along a trajectory. For a compact and invariant set A ⊂Cn • with an unique ergodic invariant measure, we write (A) = KIm(F)(A) for the angular frequency vector of A. that is, the angular frequency vector is the observation of Im(F) along a trajectory. For a compact and invariant set A ⊂Cn • with an unique ergodic invariant measure, we write (A) = KIm(F)(A) for the angular frequency vector of A. The symmetries of the system (11) now allow to phrase angular frequency syn- chronization in terms of the isotropy of the angular frequency vector. An observation of Im(F) has isotropy subgroup (A) ⊂. This motivates a deﬁnition of a weak chimera (Ashwin and Burylko 2015)—originally limited to networks of phase oscillator—to more general oscillator systems (11). Deﬁnition 3 A compact, connected, chain-recurrent and dynamically invariant set A ⊂Cn • is a weak chimera for (11) if {id} ⊊(ϕ0) ⊊ for all ϕ0 ∈A. If a weak chimera A supports an SRB measure, then it is called observable and we have {id} ⊊(A) ⊊. 3.1 Isotropy of Angular Frequencies and Weak Chimeras ⊓⊔ Deﬁnition 3 is compatible with the action of the symmetry group on Cn •. Proposition 3 If A ⊂Tn is a weak chimera, so is γ A for any γ ∈. Proof The assertion follows directly from -equivariance of F. ⊓⊔ ⊓⊔ Proof The assertion follows directly from -equivariance of F. This implies in particular that the isotropy of the angular frequency vectors are conjugate if weak chimeras are related by symmetry. If γ ∈(A), then (γ A) = (A), that is, the angular frequency vectors (and therefore the isotropy) are identical. 3.1 Isotropy of Angular Frequencies and Weak Chimeras Remark 2 Asymptotic winding (or rotation) numbers can be deﬁned in a more general setting: they quantify how trajectories of a given ﬂow wind around a topological space X; cf. Schwartzman (1957), Walsh (1995) for details. These winding numbers are deﬁned for continuous maps f : X →S1 = { z ∈C \| \|z\| = 1}. For spaces with ﬁnitely generated homology, it sufﬁces to evaluate winding numbers for maps fk corresponding to a basis of the ﬁrst cohomology (Schwartzman 1957). Here, we have X = Cn • and the maps Ck deﬁned above correspond to the generators of the homology of Cn •. Since we consider ﬂows given by a -equivariant differential 12 3 612 J Nonlinear Sci (2017) 27:605–626 equation, we characterize weak chimeras by the symmetry properties of the asymptotic winding numbers. In the language of asymptotic cycles, these are solutions where for certain “directions,” the winding behavior is the same, while for other directions, it is distinct. This suggests that the notion can be further extended to equivariant dynamical systems on more general X with nontrivial homology. Note that the weak chimeras of Deﬁnition 3 are deﬁned solely in terms of the sym- metry properties of the system. Moreover, the deﬁnition extends beyond the weak coupling limit of interacting limit cycle oscillators (Ashwin and Swift 1992): Systems of the form (11) describe dynamical systems close to a Hopf bifurcation (Ashwin and Rodrigues 2016) or more general oscillator models where “amplitude-mediated chimeras” have been observed (Sethia et al. 2013). Moreover, the next proposi- tion asserts that the change in argument of zk along Cℓ(γT ) cannot be bounded to obtain nontrivial winding numbers; such dynamics are observed for “pure amplitude chimeras” (Zakharova et al. 2014), and thus, our deﬁnition is sufﬁciently general to provide a rigorous framework for such chimeras. Proposition 2 Suppose that z(t) is a solution of (11) such that there are j ̸= ℓ, M, R > 0 such that arg Cℓ(γT ) < M and arg C j(γT ) > RT for all T . Then, (A) ⊊. roof Immediate from k z0 = limT →∞1 T arg Ck(γT ). ⊓⊔ Proof Immediate from k z0 = limT →∞1 T arg Ck(γT ). ⊓⊔ Proof Immediate from k z0 = limT →∞1 T arg Ck(γT ). 123 3.2 Symmetries Imply Frequency Synchronization Intuitively speaking, a weak chimera A consists of solutions of (11) along which the average angular frequencies have some symmetries but not too many. Inclusion (10) implies {id} ⊂T (A) ⊂(A) ⊂(A) ⊂ ⊂Sn. (15) (15) Consequently, if a solution has nontrivial instantaneous symmetry, then the corre- sponding angular frequency vector has nontrivial isotropy. Similarly, the angular frequency vector of dynamically invariant sets with nontrivial setwise symmetry has nontrivial isotropy. For invariant sets with nontrivial (setwise or instantaneous) sym- metry, (15) implies that one condition of Deﬁnition 3 is automatically satisﬁed. In that sense, the presence of symmetries “facilitates” the occurrence of weak chimera states. More generally speaking, symmetries of the system give sufﬁcient conditions for angular frequency synchronization (Golubitsky et al. 2006). These are not necessary 123 J Nonlinear Sci (2017) 27:605–626 613 as there may be other dynamically invariant subspaces where oscillators are phase and frequency locked which are not induced by symmetry but rather by balanced polydiagonals of colored graphs (Antoneli and Stewart 2006). as there may be other dynamically invariant subspaces where oscillators are phase and frequency locked which are not induced by symmetry but rather by balanced polydiagonals of colored graphs (Antoneli and Stewart 2006). 4 Weak Chimeras for Networks of Phase Oscillators Deﬁnition 3 relates to the original deﬁnition of a weak chimera for networks of cou- pled phase oscillators (Ashwin and Burylko 2015). We will not restrict ourselves to systems (1) but consider a more general setup that may include, for example, non- pairwise interactions (Ashwin and Rodrigues 2016; Bick et al. 2016). More precisely, let X = Tn and let ⊂Sn act transitively on Tn by permuting indices. A smooth -equivariant vector ﬁeld Y : Tn →Rn now deﬁnes a -equivariant dynamical system ˙ϕ = Y(ϕ). (16) (16) that describes the evolution of n phase oscillators where the state of oscillator k is given by ϕk ∈T. that describes the evolution of n phase oscillators where the state of oscillator k is given by ϕk ∈T. Write zk = exp(iϕk) and identify initial conditions ϕ0 ∈Tn with z0 ∈Cn •. The dynamics of (16) can be embedded in Cn • as ˙zk = zk (iYk(ϕ)) . (17) (17) Therefore, k ϕ0 := KYk z0 = lim T →∞ 1 T T 0 Yk(ϕ(t))dt (18) (18) and if A ⊂Tn is compact, dynamically invariant supporting an SRB measure then ϕ0 = KY (A) is the angular frequency vector for A. Moreover, with (16), we have is the angular frequency vector for A. Moreover, with (16), we have T 0 Yk(ϕ(t))dt = T 0 ˙ϕk(t)dt = ˆϕk(T ) −ˆϕk(0) (19) (19) where ˆϕ is a continuous lift of the trajectory ϕ(t) to Rn. Thus, k ϕ0 = lim T →∞ ˆϕ(T ) T as given by (2). Note also that k(A) correspond to the average frequency deﬁned in(Golubitskyetal.2006)andrelatestotherotationvectorfortorusmaps(Misiurewicz and Ziemian 1989). as given by (2). Note also that k(A) correspond to the average frequency deﬁned in(Golubitskyetal.2006)andrelatestotherotationvectorfortorusmaps(Misiurewicz and Ziemian 1989). Compared to the original deﬁnition of a weak chimera in (Ashwin and Burylko 2015), Deﬁnition 3 is more restrictive. More precisely, for A, we require that frequency 12 3 614 J Nonlinear Sci (2017) 27:605–626 synchronization is only relevant for a weak chimera if the oscillators are related by symmetry. By contrast, the original deﬁnition considers the set (A) = { γ ∈Sn \| γ (A) = (A)} (20) (20) rather than the isotropy k(A). Note that (A) may be strictly larger than (A). 4 Weak Chimeras for Networks of Phase Oscillators For example if Zn = Z/nZ ⊂Sn denotes the cyclic group and X is Zn equivariant but not Sn-equivariant and ϕ0 1 = · · · = ϕ0 n [for example, a nonlocally coupled ring of phase oscillators (Kuramoto and Battogtokh 2002)] is a solution of ˙ϕ = X(ϕ) then ϕ0 = Sn ⊋Zn. 5 Persistence of Weak Chimeras Without Symmetry on Average for Diffusively Coupled Phase Oscillators The inclusions (15) in Sect. 3.2 imply that any (nontrivial) instantaneous or average symmetry of a dynamically invariant set gives nontrivial isotropy of the angular fre- quency vector. This is the case for the weak chimeras constructed in (Ashwin and Burylko 2015; Panaggio et al. 2016; Bick and Ashwin 2016). In this section, we con- struct weak chimeras with trivial average symmetries for systems consisting of two weakly interacting populations of phase oscillators. 2 We obtain (22) by setting Hkj = 1 for all k, j in (1). 5.1 Coupling Function Separability for Symmetric Diffusively Coupled Phase Oscillators Deﬁnition 4 Two sets A1, A2 ⊂C are coupling function separated if there are open intervals Q A1, Q A2 ⊂T with (Aℓ) ⊂Q Aℓ, ℓ= 1, 2, and Deﬁnition 4 Two sets A1, A2 ⊂C are coupling function separated if there are open intervals Q A1, Q A2 ⊂T with (Aℓ) ⊂Q Aℓ, ℓ= 1, 2, and Deﬁnition 4 Two sets A1, A2 ⊂C are coupling function separated if there are open intervals Q A1, Q A2 ⊂T with (Aℓ) ⊂Q Aℓ, ℓ= 1, 2, and Q A1 ∩Q A2 = ∅ Q A1 ∩Q A2 = ∅ where the bar denotes topological closure. where the bar denotes topological closure. where the bar denotes topological closure. For Q ⊂T deﬁne −1(Q) := { ϕ ∈T \| ({ϕ}) ⊂Q } and W (g)(Q) := min k∈{1,...,n} inf ϕ∈−1(Q) Y (g) k (ϕ), max k∈{1,...,n} sup ϕ∈−1(Q) Y (g) k (ϕ) . W (g)(Q) := min k∈{1,...,n} inf ϕ∈−1(Q) Y (g) k (ϕ), max k∈{1,...,n} sup ϕ∈−1(Q) Y (g) k (ϕ) . Lemma 1 1. If A ⊂C with (A) ⊂Q is dynamically invariant for the dynamics of (22) with coupling function g, then k ϕ0 ∈W (g)(Q). Lemma 1 1. If A ⊂C with (A) ⊂Q is dynamically invariant for the dynamics of (22) with coupling function g, then k ϕ0 ∈W (g)(Q). Lemma 1 1. If A ⊂C with (A) ⊂Q is dynamically invariant for the dynamics of (22) with coupling function g, then k ϕ0 ∈W (g)(Q). 2. Suppose that Aℓ⊂C, ℓ= 1, 2, are compact and coupling function separated with separating sets Q Aℓ. Then for any η ≥0, we can ﬁnd a coupling function ˆg such that 2. Suppose that Aℓ⊂C, ℓ= 1, 2, are compact and coupling function separated with separating sets Q Aℓ. Then for any η ≥0, we can ﬁnd a coupling function ˆg such that Bη W ( ˆg) Q A1 ∩Bη W ( ˆg) Q A2 = ∅. Bη W ( ˆg) Q A1 ∩Bη W ( ˆg) Q A2 = ∅. 3. Let Aℓbe as above and let A′ 1, A′ 2 ⊂C be dynamically invariant for the dynamics of (22) with (A′ ℓ) ⊂Q Aℓ. 5.1 Coupling Function Separability for Symmetric Diffusively Coupled Phase Oscillators or φ, ψ ∈Tn deﬁne X = (X1, . . . , Xn) by Xk(φ, ψ) := 1 n n j=1 g φk −ψ j . (21) (21) The dynamics of a fully symmetric network of n phase oscillators with coupling function g is given by the Sn-equivariant dynamical system on Tn where The dynamics of a fully symmetric network of n phase oscillators with coupling function g is given by the Sn-equivariant dynamical system on Tn where ˙ϕk = Yk(ϕ) := Xk(ϕ, ϕ) (22) (22) describes the evolution of the kth oscillator.2 We may assume g(0) = 0 by going to suitable co-rotating reference frame, ϕk →ϕk −ωt. If the choice of coupling func- tion g is important, we write Y (g) or X(g) to highlight the dependency. Reducing the continuous T symmetry of (22) allows to set ϕ1 = 0. Because of the Sn-equivariance, the canonical invariant region C := ϕ ∈Tn \| 0 = ϕ1 < · · · < ϕn < 2π (23) (23) 2 We obtain (22) by setting Hkj = 1 for all k, j in (1). 2 We obtain (22) by setting Hkj = 1 for all k, j in (1). 123 J Nonlinear Sci (2017) 27:605–626 615 is dynamically invariant. It is bounded by hypersurfaces corresponding to cluster states with ϕk = ϕk+1, and there is a residual Zn symmetry on C (Ashwin and Swift 1992; Ashwin et al. 2016a). is dynamically invariant. It is bounded by hypersurfaces corresponding to cluster states with ϕk = ϕk+1, and there is a residual Zn symmetry on C (Ashwin and Swift 1992; Ashwin et al. 2016a). ) For a compact ﬂow-invariant set A ⊂Tn deﬁne For a compact ﬂow-invariant set A ⊂Tn deﬁne (A) = k̸= j ϕk −ϕ j \| ϕ ∈A . (24) (24) Note that (γ A) = (A) for all γ ∈Sn. Note that (γ A) = (A) for all γ ∈Sn. Note that (γ A) = (A) for all γ ∈Sn. to obtain the desired coupling function ˆg. Note that replacing g by ˆg as above preserves dynamically invariant sets A with (A) ⊂Q Aℓ. Claim (3) now follows from (1) and (2) with η = 0. ⊓⊔ ⊓⊔ Remark 3 The notion of function coupling separability and Lemma 1 generalize to a ﬁnite number of sets A1, . . . , Ar. The function g −ˆg can typically be chosen to be C∞. 5.1 Coupling Function Separability for Symmetric Diffusively Coupled Phase Oscillators Then, there is a coupling function ˆg such that A′ 1, A′ 2 are dynamically invariant for the dynamics of (22) with ˆg and 3. Let Aℓbe as above and let A′ 1, A′ 2 ⊂C be dynamically invariant for the dynamics of (22) with (A′ ℓ) ⊂Q Aℓ. Then, there is a coupling function ˆg such that A′ 1, A′ 2 are dynamically invariant for the dynamics of (22) with ˆg and k ϕ0 A′ 1 ̸= j ϕ0 A′ 2 for all k, j and ϕ0 A′ ℓ∈A′ ℓ. for all k, j and ϕ0 A′ ℓ∈A′ ℓ. for all k, j and ϕ0 A′ ℓ∈A′ ℓ. Proof To prove (1), note ﬁrst that invariance of C implies that j ϕ0 = k ϕ0 for all k, j and ϕ0 ∈A. Standard integral estimate for (18) yield Proof To prove (1), note ﬁrst that invariance of C implies that j ϕ0 = k ϕ0 for all k, j and ϕ0 ∈A. Standard integral estimate for (18) yield k ϕ0 ∈ inf ϕ∈A Y (g) k (ϕ), sup ϕ∈A Y (g) k (ϕ) ⊂W (g)(Q). 123 3 616 J Nonlinear Sci (2017) 27:605–626 To prove (2), consider coupling functions ˆg with ˆg(φ) = g(φ)+aℓfor all φ ∈Q Aℓ, ℓ= 1, 2. Since Y ( ˆg) k (ϕ) = 1 n n j=1 g(φk −ψ j) + aℓ = aℓ+ Y (g) k (ϕ) for all ϕ with ({ϕ}) ⊂Q Aℓwe have W ( ˆg)(Q Aℓ) = inf W (g)(Q Aℓ) + aℓ, sup W (g)(Q Aℓ) + aℓ . for all ϕ with ({ϕ}) ⊂Q Aℓwe have for all ϕ with ({ϕ}) ⊂Q Aℓwe have for all ϕ with ({ϕ}) ⊂Q Aℓwe have W ( ˆg)(Q Aℓ) = inf W (g)(Q Aℓ) + aℓ, sup W (g)(Q Aℓ) + aℓ . For a given η ≥0, choose a1, a2 such that For a given η ≥0, choose a1, a2 such that Bη W (g)(Q A1) ∩Bη W (g)(Q A2) = ∅ to obtain the desired coupling function ˆg. 5.2 Relative Equilibria with Trivial Symmetry Proof It sufﬁces to show that ({ϕ⋆}) = {id}. Assume that γ ∈({ϕ⋆}) with γ ̸= id. Then, there exists a τ ≥0 such that Proof It sufﬁces to show that ({ϕ⋆}) = {id}. Assume that γ ∈({ϕ⋆}) with γ ̸= id. Then, there exists a τ ≥0 such that αγ k = αk + τω⋆ mod 2π for all k. Recall that 0 = α1 < · · · < αn < 2π. Since γ ̸= id, γ permutes some indices. Assume without loss of generality that αγ 1 > α1 and αγ 2 < α2. We have αγ 1 −α1 = αγ 2 −α2 = τω⋆mod 2π. But since αγ 1 −α1 > 0 and αγ 2 −α2 < 0, there has to be an m > 0 such that αγ 1−α1−αγ 2+α2 = 2mπ. This is a contradiction since αγ 1 −α1 −αγ 2 + α2 ≤4αn < 2π. ⊓⊔ ⊓⊔ 5.3 Weak Chimeras with (A) = {id} in Weakly Coupled Populations of Phase Oscillators 5.2 Relative Equilibria with Trivial Symmetry We now show that choosing the coupling function appropriately in an arbitrarily small neighborhood of zero gives rise to asymptotically stable relative equilibria with trivial symmetry for (22). y y Let 0 = α1 < · · · < αn < 2π. The function ϕ⋆(t) = α1 + tω⋆, . . . , αn + tω⋆ ∈C (25) ϕ⋆(t) = α1 + tω⋆, . . . , αn + tω⋆ ∈C (25) (25) withω⋆= 1 n n j=2 g(−α j)isarelativeequilibriumof (22)foranycouplingfunction g such that withω⋆= 1 n n j=2 g(−α j)isarelativeequilibriumof (22)foranycouplingfunction g such that 1 n j̸=k g(αk −α j) −g(−α j) = 0 (26) (26) for all k = 2, . . . , n. For a relative equilibrium, we have for all k = 2, . . . , n. For a relative equilibrium, we have ϕ⋆ = k̸= j αk −α j ⊂[−2αn, 2αn] ⊂T. (27) (27) In particular, we have a relative equilibrium if the coupling function g vanishes on ({ϕ⋆}). Since αn can be chosen arbitrarily small, the relative equilibrium can be chosen arbitrarily close to the fully synchronized solution ϕ1 = · · · = ϕn. We have ({ϕ⋆}) = (ω⋆, . . . , ω⋆). 123 123 J Nonlinear Sci (2017) 27:605–626 617 Stability of the relative equilibrium is determined by the linearization Stability of the relative equilibrium is determined by the linearization ∂Yk ∂ϕ j = 1 n l̸=k g′(αk −αl) if k = j, −1 n g′(αk −α j) otherwise. (28) (28) By choosing the coupling function appropriately on (A), the relative equilibrium will be asymptotically stable. For example, if g′(φ) = 0 for φ < 0 and g′(φ) < 0 for φ > 0, we have a lower triangular matrix with negative values on the diagonal (apart from one zero eigenvalue) implying that ϕ⋆is asymptotically stable. By choosing the coupling function appropriately on (A), the relative equilibrium will be asymptotically stable. For example, if g′(φ) = 0 for φ < 0 and g′(φ) < 0 for φ > 0, we have a lower triangular matrix with negative values on the diagonal (apart from one zero eigenvalue) implying that ϕ⋆is asymptotically stable. Lemma 2 Let ϕ⋆(t) be a relative equilibrium of (22) as deﬁned in (25). If \|αn\| < π 2 then T ({ϕ⋆}) = ({ϕ⋆}) = {id}. 5.3 Weak Chimeras with (A) = {id} in Weakly Coupled Populations of Phase Oscillators Chaotic weak chimeras have many features associated with classical chimera states including positive maximal Lyapunov exponents. Hence, rather than using a hyper- bolicity argument to construct nonchaotic weak chimeras as in (Ashwin and Burylko 2015), we aim to construct weak chimeras with (A) = {id} in a more general setup which allows for positive maximal Lyapunov exponents. To this end, we extend recent results from (Bick and Ashwin 2016) with respect to the instantaneous and setwise symmetries of the constructed sets. Coupling two populations of n oscillators, whose uncoupled dynamics are given by (22), deﬁnes a dynamical system on T2n. More explicitly, write ϕ = (ϕ1, ϕ2) ∈ Tn × Tn = T2n, ϕℓ= (ϕℓ,1, . . . , ϕℓ,n) and consider the product system ˙ϕ1 = Y(g,ε) 1 (ϕ1, ϕ2) = Y (g)(ϕ1) + εX(g)(ϕ1, ϕ2), ˙ϕ2 = Y(g,ε) 2 (ϕ1, ϕ2) = Y (g)(ϕ2) + εX(g)(ϕ2, ϕ1), (29) (29) with Y (g), X(g) as in (22), (21). Observe that for ε = 0, the system decouples into two identical groups of n oscillators—both of which with nontrivial dynamics (22). For ϕ0 ∈T2n, we denote the asymptotic angular frequency of the oscillator with phase ϕℓ,k by ℓ,k ϕ0 = (g,ε) ℓ,k ϕ0 . 12 618 J Nonlinear Sci (2017) 27:605–626 Let = Sn ≀S2 where ≀is the wreath product. The system (29) is - equivariant (Dionne et al. 1996); we have = Sn ≀S2 = (Sn)2 ⋊S2 where the elements of Sn permute the oscillators within each group of n oscillators and the action of S2 permutes the two groups. Observe that this is only a semidirect prod- uct ⋊as the two sets of permutations do not necessarily commute. The oscillators are indistinguishable as this group acts transitively on the oscillators. Weak chimeras in the product system persist for weak coupling 0 ≤ε ≪1. As in (Bick and Ashwin 2016), we call a dynamically invariant set A is sufﬁciently stable if there is an open neighborhood of A on which a Lyapunov function is deﬁned. The persistence theorem for weak chimeras (Bick and Ashwin 2016, Theorem 4) generalizes to coupling function separated sets that are sufﬁciently stable. Theorem 2 Suppose that g is a coupling function such that A1, A2 ⊂C are compact, forward invariant, coupling function separated and sufﬁciently stable sets for the dynamics of (22) with Y (g). 5.3 Weak Chimeras with (A) = {id} in Weakly Coupled Populations of Phase Oscillators Then, for any sufﬁciently small δ > 0, there exist a smooth coupling function ˆg and ε0 > 0 such that for any 0 ≤ε < ε0, the weakly coupled product system (29) with g replaced by ˆg has a sufﬁciently stable weak chimera A(ε) with A(ε) ⊂Bδ(A1 × A2). Proof First, consider the coupling function separated sets A1, A2 ⊂Tn as dynam- ically invariant sets for (22), a factor of the uncoupled system. Suppose that Q A1, Q A2 are the separating sets and for a coupling function ˜g deﬁne M( ˜g) := max(ϕ1,ϕ2)∈T2n X( ˜g)(ϕ1, ϕ2) < ∞. Now, choose a coupling function ˆg according to Lemma 1(2) for η = 1 and ﬁx ε1 := M( ˆg)−1. For any 0 ≤ε < ε1, we have BεM( ˆg) W ( ˆg)(Q A1) ∩BεM( ˆg) W ( ˆg)(Q A2) = ∅. (30) (30) since εM( ˆg) < ε1M( ˆg) = 1. Now, consider the product system (29). For any sufﬁciently small δ > 0, we obtain ε2 > 0 and compact invariant sets A(ε) ⊂Bδ(A1 × A2) for all 0 ≤ε < ε2 as in (Bick and Ashwin 2016). Set ε0 < min {ε1, ε2}. By restricting δ appropriately, the sets A(ε) are weak chimeras. To show that (ϕ0) ̸= {id}, assume that δ is so small that Bδ(A1 × A2) ⊂C2. This implies that the phase ordering within each population is preserved. Hence, for given ℓ= 1, 2, we have ( ˆg,ε) ℓ,k ϕ0 = ( ˆg,ε) ℓ, j ϕ0 =: ( ˆg,ε) ℓ,∗ ϕ0 for all k, j and any ϕ0 ∈Aε. Thus, (ϕ0) ̸= {id}. for all k, j and any ϕ0 ∈Aε. Thus, (ϕ0) ̸= {id}. for all k, j and any ϕ0 ∈Aε. Thus, (ϕ0) ̸= {id}. It remains to be shown that (ϕ0) ̸= . Let A(ε) ℓ denote the projection of A(ε) onto ϕℓ. Now, assume that δ is sufﬁciently small such that A(ε) ⊂Bδ(A1 × A2) implies that A(ε) ℓ ⊂Q Aℓfor all 0 ≤ε < ε0. Since Y( ˆg,ε) 1 (ϕ1, ϕ2) = Y ( ˆg)(ϕ1) + εX(ϕ1, ϕ2) integral estimates as in Lemma 1(1) imply that It remains to be shown that (ϕ0) ̸= . Let A(ε) ℓ denote the projection of A(ε) onto ϕℓ. 5.3 Weak Chimeras with (A) = {id} in Weakly Coupled Populations of Phase Oscillators Now, assume that δ is sufﬁciently small such that A(ε) ⊂Bδ(A1 × A2) implies that A(ε) ℓ ⊂Q Aℓfor all 0 ≤ε < ε0. Since Y( ˆg,ε) 1 (ϕ1, ϕ2) = Y ( ˆg)(ϕ1) + εX(ϕ1, ϕ2) integral estimates as in Lemma 1(1) imply that ( ˆg,ε) ℓ,∗ ∈Bε0M W ( ˆg)(Q Aℓ) 12 123 J Nonlinear Sci (2017) 27:605–626 619 for all 0 ≤ε < ε0. By choice of ˆg, Eq. (30) now implies ( ˆg,ε) 1,∗ ϕ0 ̸= ( ˆg,ε) 2,∗ ϕ0 for all ϕ0 ∈A(ε). Thus, (ϕ0) ̸= and A(ε) is a weak chimera. ⊓⊔ for all 0 ≤ε < ε0. By choice of ˆg, Eq. (30) now implies ( ˆg,ε) 1,∗ ϕ0 ̸= ( ˆg,ε) 2,∗ ϕ0 for all ϕ0 ∈A(ε). Thus, (ϕ0) ̸= and A(ε) is a weak chimera. ⊓⊔ ⊓⊔ The following statement asserts that trivial symmetries in the factors carry over to the product dynamics. Lemma 3 Let A1, A2 ⊂Tn be coupling function separated attractors for (22) with (A1) = (A2) = {id}. Then (A1 × A2) = {id} for the product system (29). Proof Write S2 = {id, τ}. For any γ ∈(Sn)2, we have (γ, id)(A1× A2)∩(A1× A2) = ∅in T2n by assumption. Since A1 and A2 are coupling function separated, we have A1 ∩A2 = ∅in Tn. Write V = γ ∈ γ V , V ∈{A1, A2}. The fact that (γ A1) = (A1) implies A1 ∩A2 = ∅in Tn. Since (γ, τ)(A1 × A2) ⊂A2 × A1 for any γ ∈(Sn)2, we have (γ, τ)(A1 × A2)∩(A1 × A2) = ∅and by Proposition 1, the claim follows. ⊓⊔ Proof Write S2 = {id, τ}. For any γ ∈(Sn)2, we have (γ, id)(A1× A2)∩(A1× A2) = ∅in T2n by assumption. Since A1 and A2 are coupling function separated, we have A1 ∩A2 = ∅in Tn. Write V = γ ∈ γ V , V ∈{A1, A2}. The fact that (γ A1) = (A1) implies A1 ∩A2 = ∅in Tn. Since (γ, τ)(A1 × A2) ⊂A2 × A1 for any γ ∈(Sn)2, we have (γ, τ)(A1 × A2)∩(A1 × A2) = ∅and by Proposition 1, the claim follows. ⊓⊔ ⊓⊔ Combining the perturbation result Theorem 2 with the symmetry considerations, we can now state the main theorem of this section. 5.3 Weak Chimeras with (A) = {id} in Weakly Coupled Populations of Phase Oscillators In order to apply Theorem 2, we make a slightly stronger assumption concerning stability of the relative periodic orbits. Theorem 3 Suppose that g is a coupling function such that for the Sn-equivariant dynamics of (22) with Y (g), the set Acoh = {ϕ⋆} is a sufﬁciently stable relative equilib- rium and Ainc ⊂Tn is sufﬁciently stable attractor that are coupling function separated and (Acoh) = (Ainc) = {id}. Then, for any sufﬁciently small δ > 0, there is a coupling function ˆg and ε0 > 0 such that for any 0 ≤ε < ε0, there is a weak chimera A(ε) ⊂Bδ(Acoh × Ainc) with (A(ε)) = {id} for the Sn ≀S2-equivariant dynamics of (29) with ˆg. Proof Suppose that (Acoh) = (Ainc) = {id} ⊂Sn. By Lemma 3, we have (Acoh × Ainc) = {id} ⊂Sn ≀S2. Since Acoh × Ainc is assumed to be an attrac- tor, Corollary 1 implies that there exists a δ0 > 0 such that for any invariant set A ⊂Bδ0(Acoh × Ainc), we have (A) = {id} ⊂Sn ≀S2. Proof Suppose that (Acoh) = (Ainc) = {id} ⊂Sn. By Lemma 3, we have (Acoh × Ainc) = {id} ⊂Sn ≀S2. Since Acoh × Ainc is assumed to be an attrac- tor, Corollary 1 implies that there exists a δ0 > 0 such that for any invariant set A ⊂Bδ0(Acoh × Ainc), we have (A) = {id} ⊂Sn ≀S2. 0 For sufﬁciently small 0 < δ < δ0, Theorem 2 yields an ε0 > 0 and weak chimeras A(ε) ⊂Bδ(Acoh × Ainc) for all 0 ≤ε < ε0. Since δ < δ0, the argument above implies that (A(ε)) = {id} for all such ε. ⊓⊔ 0 For sufﬁciently small 0 < δ < δ0, Theorem 2 yields an ε0 > 0 and weak chimeras A(ε) ⊂Bδ(Acoh × Ainc) for all 0 ≤ε < ε0. Since δ < δ0, the argument above implies that (A(ε)) = {id} for all such ε. ⊓⊔ For sufﬁciently small 0 < δ < δ0, Theorem 2 yields an ε0 > 0 and weak chimeras A(ε) ⊂Bδ(Acoh × Ainc) for all 0 ≤ε < ε0. Since δ < δ0, the argument above implies that (A(ε)) = {id} for all such ε. ⊓⊔ Remark 4 1. In fact, the condition that Ainc is a relative equilibrium is not necessary. 5.3 Weak Chimeras with (A) = {id} in Weakly Coupled Populations of Phase Oscillators Theorem 3 holds for any compact, sufﬁciently stable attractor Acoh ⊂C that is coupling function separated from Ainc. Moreover, the same statement holds for (sufﬁciently unstable) repellers. 2. Even if the weak chimera A(0) = Acoh × Ainc is observable, extra assumptions on the persistence of SRB measures are needed to prove that A(ε) is an observable weak chimera. 6 A Numerical Example of a Chaotic Weak Chimera with Trivial Symmetry We now give an explicit example of a coupling function such that the dynamics of the product system (29) give rise to a chaotic weak chimera with trivial symmetry for 3 620 J Nonlinear Sci (2017) 27:605–626 ε > 0 following the construction described in the previous section. In contrast to the examples in (Bick and Ashwin 2016), the main focus here is on the symmetries of the weak chimeras which we calculate explicitly. ε > 0 following the construction described in the previous section. In contrast to the examples in (Bick and Ashwin 2016), the main focus here is on the symmetries of the weak chimeras which we calculate explicitly. Recall that the dynamics of (22) for n = 4 oscillators give rise to chaotic attractors A with (A) = {id} (Bick et al. 2011; Bick 2012). Deﬁne g(φ) = 4 r=0 cr cos (rφ + ξr) (31) (31) with c1 = −2, c2 = −2, c3 = −1, and c4 = −0.88. For ξ1 = η1, ξ2 = −η1, ξ3 = η1 +η2, and ξ4 = η1 +η2 with η1 = 0.138, η2 = 0.057511 the dynamics of (22) with this particular choice of coupling function g give rise to a chaotic attracting set Ainc ⊂C with positive maximal Lyapunov exponents and (Ainc) = {id}. For Ainc, we have (Ainc) ⊂[0.4, 2π −0.4] as shown in Fig. 1. A suitable local perturbation of the coupling function g yields bistability between Ainc and a relative equilibrium with trivial symmetry in the system deﬁned by (22). Let 24 ˜g(φ) = 24 r=6 ar cos(rφ + ζr) (32) (32) with parameters ar, ζr as given in Appendix A. Moreover, deﬁne β(x) := exp − 1 1−x2 if −1 < x < 1, 0 otherwise and let a ∈R, b ∈(0, π) be parameters. Now, deﬁne βab(φ) := aβ φ b with φ taken modulo 2π with values in (−π, π] is a 2π-periodic “bump function.” Fix a = 2.5, b = 0.25. Deﬁne the C∞function ˆg := g + ˜gβab. (33) (33) We have ˆg(φ) = g(φ) for all φ ∈[b, 2π −b]. Since (Ainc) ⊂[b, 2π −b], we have Y ( ˆg)(ϕ) = Y (g)(ϕ) for all ϕ ∈Ainc. Thus, Ainc ⊂C is also a chaotic attracting We have ˆg(φ) = g(φ) for all φ ∈[b, 2π −b]. 6 A Numerical Example of a Chaotic Weak Chimera with Trivial Symmetry Since (Ainc) ⊂[b, 2π −b], we have Y ( ˆg)(ϕ) = Y (g)(ϕ) for all ϕ ∈Ainc. Thus, Ainc ⊂C is also a chaotic attracting Fig. 1 Attracting sets Ainc and Acoh of the dynamics given by (22) with coupling function ˆg are coupling function separated. The coupling function ˆg as deﬁned in (33) is depicted by a gray line. The values of g on (Ainc) are indicated by ﬁlled circles and on (Acoh) by 12 hollow circles Fig. 1 Attracting sets Ainc and Acoh of the dynamics given by (22) with coupling function ˆg are coupling function separated. The coupling function ˆg as deﬁned in (33) is depicted by a gray line. The values of g on (Ainc) are indicated by ﬁlled circles and on (Acoh) by 12 hollow circles Fig. 1 Attracting sets Ainc and Acoh of the dynamics given by (22) with coupling function ˆg are coupling function separated. The coupling function ˆg as deﬁned in (33) is depicted by a gray line. The values of g on (Ainc) are indicated by ﬁlled circles and on (Acoh) by 12 hollow circles 123 621 J Nonlinear Sci (2017) 27:605–626 (a) (b) Fig. 2 Chaotic weak chimeras with trivial setwise symmetries appear in the S4 ≀S2-equivariant system (29) with two populations of n = 4 oscillators for ε = 0.01. a The phase evolution: The phase of the oscillators (periodic color scale, ϕℓ,k(t) = 0 in black and ϕℓ,k(t) = π in white) in a co-rotating frame at the speed of the ﬁrst oscillator is shown at the top, the instantaneous frequencies ˙ϕℓ,k(t) in the middle and convergence of the maximal Lyapunov exponent at the bottom. b The dynamics on the attracting set for each population in the Z4-equivariant projections yℓ= (sin(ϕℓ,3 −ϕℓ,1), sin(ϕℓ,4 −ϕℓ,2)) where Z4 are the permutations within populations that preserve the phase ordering, a phase evolution and b projected dynamics of each population (a) (a) (b) Fig. 2 Chaotic weak chimeras with trivial setwise symmetries appear in the S4 ≀S2-equivariant system (29) with two populations of n = 4 oscillators for ε = 0.01. 3 IntegrationwascarriedoutinMATLABusingthevariableorderscheme ode113withaadaptivetimestep t ≤10−1 subject to conservative relative and absolute error tolerances of 10−9 and 10−11, respectively. 6 A Numerical Example of a Chaotic Weak Chimera with Trivial Symmetry The shaded regions show the intervals [mink,t ˙ϕℓ,k(t), maxk,t ˙ϕℓ,k(t)] for ℓ= 1 (dark gray) and ℓ= 2 (light gray)—where these do not overlap, there is no frequency synchronization between the two populations and hence a weak chimera, a attractor in the S4-equivariant projection yℓ, b maximal Lyapunov exponents and symmetries with varying ε (a) (b) Fig. 3 Increasing ε yields chaotic weak chimeras that undergo symmetry increasing bifurcations. a trajectory for ε = 0.1 converging to an attractor A(ε) with (A(ε)) ̸= {id}. b Maximal Lyapnuov exponen obtained by integrating (29) from a ﬁxed initial condition on A(0) for T = 2 × 105 time units. The mark Fig. 3 Increasing ε yields chaotic weak chimeras that undergo symmetry increasing bifurcations. a A trajectory for ε = 0.1 converging to an attractor A(ε) with (A(ε)) ̸= {id}. b Maximal Lyapnuov exponents obtained by integrating (29) from a ﬁxed initial condition on A(0) for T = 2 × 105 time units. The marker indicates the symmetry of the attractor A(ε) ⊂T2n: “•” for (A(ε)) = {id}, “⋄” if (A(ε)) ?= {id}, and “◦” if (A(ε)) ̸= {id}. The shaded regions show the intervals [mink,t ˙ϕℓ,k(t), maxk,t ˙ϕℓ,k(t)] for ℓ= 1 (dark gray) and ℓ= 2 (light gray)—where these do not overlap, there is no frequency synchronization between the two populations and hence a weak chimera, a attractor in the S4-equivariant projection yℓ, b maximal Lyapunov exponents and symmetries with varying ε For increasing coupling parameter ε (while keeping the initial condition ﬁxed), the symmetries of the attracting chaotic weak chimeras A(ε) change; cf. Fig. 3. We integrated the system for T = 2 × 105 time units to calculate both the maximal Lya- punov exponents and detect the presence of nontrivial symmetries. For A(ε) ⊂C2, we have to check for permutations of oscillators that preserve the ordering of the phases within each population to determine the symmetry of the attractor. To this end, we calculated the ergodic average Sℓ(ε) = T 0 sin(ϕℓ,3(t) −ϕℓ,1(t))dt along the trajectory which converges zero if (A(ε)) ̸= {id}. Note that if symmetric copies of attractors merge in a symmetry increasing bifurcation (Chossat and Golubitsky 1988), these ergodic averages may converge very slowly. 6 A Numerical Example of a Chaotic Weak Chimera with Trivial Symmetry a The phase evolution: The phase of the oscillators (periodic color scale, ϕℓ,k(t) = 0 in black and ϕℓ,k(t) = π in white) in a co-rotating frame at the speed of the ﬁrst oscillator is shown at the top, the instantaneous frequencies ˙ϕℓ,k(t) in the middle and convergence of the maximal Lyapunov exponent at the bottom. b The dynamics on the attracting set for each population in the Z4-equivariant projections yℓ= (sin(ϕℓ,3 −ϕℓ,1), sin(ϕℓ,4 −ϕℓ,2)) where Z4 are the permutations within populations that preserve the phase ordering, a phase evolution and b projected dynamics of each population set for the dynamics of (22) with coupling function ˆg. In addition, there is a stable relative periodic orbit ϕ⋆(t) ≈(tω⋆, 0.0975 + tω⋆, 0.1253 + tω⋆, 0.2247 + tω⋆). For Acoh = { ϕ⋆(t) \| t ≥0} we have (Acoh) ⊂[−0.3, 0.3]. Therefore, the sets Ainc and Acoh are coupling function separated; see Fig. 1. Now, consider two weakly coupled populations (29) of n = 4 oscillators. Since (Acoh) = (Ainc) = {id}, we have that (Acoh × Ainc) = {id} for ε = 0 and we expect dynamically invariant sets with trivial symmetry for small ε > 0. We integrated system (29) numerically3 and calculated the maximal Lyapunov exponent from the variational equations. The attracting set A(ε) for ε = 0.01 close to Acoh× Ainc with trivial setwise symmetries and positive maximal Lyapunov exponent is shown in Fig. 2; the absolute value of the local order parameter Rℓ(t) = 1 4 4 j=1 exp(iϕℓ, j) gives information about the synchronization of each population: It is equal to one if all oscillators within the populations are phase-synchronized. 12 3 J Nonlinear Sci (2017) 27:605–626 622 (a) (a) (b) (a) (b) Fig. 3 Increasing ε yields chaotic weak chimeras that undergo symmetry increasing bifurcations. a A trajectory for ε = 0.1 converging to an attractor A(ε) with (A(ε)) ̸= {id}. b Maximal Lyapnuov exponents obtained by integrating (29) from a ﬁxed initial condition on A(0) for T = 2 × 105 time units. The marker indicates the symmetry of the attractor A(ε) ⊂T2n: “•” for (A(ε)) = {id}, “⋄” if (A(ε)) ?= {id}, and “◦” if (A(ε)) ̸= {id}. 123 7 Discussion If a dynamical system has permutational symmetry , what are the symmetry prop- erties of the asymptotic angular frequencies which describe how trajectories wind around phase space? For the dynamical systems on Cn • considered in Sect. 3, the asymptotic angular frequencies are given by averages of -equivariant observables. This observation yields a natural reformulation of the notion of a weak chimera in terms of the isotropy of the vector of asymptotic angular frequencies. Our deﬁnition is not only compatible with the action of but also goes beyond phase oscillators in the weak coupling limit: It applies to more general oscillator models where chimera states have been reported (Sethia et al. 2013; Zakharova et al. 2014). With a rigorous deﬁnition in place, it would be desirable to prove the existence of weak chimeras in such systems and show that the dynamics observed are persistent phenomena. These ideas equally apply to more general spaces X with a symmetry group acting on it; here, asymptotic winding numbers of asymptotic cycles describe the rotation of a trajectory with respect to topological properties of X (Schwartzman 1957; Fried 1982; Walsh 1995). Precisifying the notion of a weak chimera for general topological spaces X with symmetry is beyond the scope of the current paper and will be addressed in future work. Using coupling functions that give rise to relative equilibria with trivial symmetry, we showed that for symmetric phase oscillator systems that there are indeed weak chimeras that have symmetries in the frequencies that are not present in the solutions. This motivates some further symmetry-related questions. For example, what are the possible isotropy groups of the angular frequency vector for a -equivariant system that do not arise from the symmetries of the solutions themselves? (These are obviously restricted to subgroups of the symmetry group.) Which symmetry increasing bifurca- tions happen as the inter-population coupling ε is increased (Fig. 3)? While chaotic dynamics do persist up to ε ≈0.1 [and for other choices of coupling function even up to ε ≈0.3 (Bick and Ashwin 2016)], chaotic weak chimeras with trivial symmetry only persist for values of ε close to zero. Thus, are there chaotic weak chimeras with trivial symmetry for “strongly coupled” populations of phase oscillators? Moreover, in general, there will be more than one ergodic invariant measure supported on a weak chimera. 6 A Numerical Example of a Chaotic Weak Chimera with Trivial Symmetry Previous numerical inves- tigations of the chaotic attractor in the uncoupled system (Bick 2012) showed that attractors with trivial symmetry are conﬁned to one quadrant under the projection yℓ= (sin(ϕℓ,3 −ϕℓ,1), sin(ϕℓ,4 −ϕℓ,2)). Thus, the number of quadrants Qℓ(ε) that the projected trajectory enters being greater than one indicates that a symmetry increas- ing bifurcation may have occurred; compare also Figs 2b and 3a. Consequently, we conclude (Aε) = {id} if \|S2(ε)\| > 10−1—but we write (Aε) ?= {id} if Q2(ε) > 1 at the same time to indicate that a symmetry increasing bifurcation may have happened already—and (Aε) ̸= {id} otherwise. Further numerical investigation shows that there is multistability for ε ≥0; the attracting sets A(ε) for ε > 0 may coexist with other attracting solutions (not shown). 123 J Nonlinear Sci (2017) 27:605–626 623 7 Discussion Thus, clarifying the relationship between clas- sical chimera states on rings and the symmetry of the system further—also with respect to the symmetries of the system that describes the continuum limit—provides exciting directions for future research. Acknowledgements The author would like to thank the anonymous referees for their feedback which signiﬁcantly helped to improve the presentation of the results. Moreover, the author would like to thank Peter Ashwin, Erik Martens and Oleh Omel’chenko for stimulating discussions and critical feedback on the manuscript. The research leading to these results has received funding from the People Programme (Marie Curie Actions) of the European Union’s Seventh Framework Programme (FP7/2007–2013) under REA Grant Agreement No. 626111. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 Interna- tional License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. 7 Discussion For each of these measures, we obtain asymptotic angular frequency vectors that potentially have different isotropy. While the set of all measures supported on the invariant set of interest (Jenkinson 2006) yields bounds of the asymptotic angular frequencies (see also Bick and Ashwin 2016), a more detailed understanding what the speciﬁc isotropy subgroups for the invariant measures are and how they bifurcate would be desirable. It is also worth noting that asymptotic angular frequencies as averages and their isotropy may still be well deﬁned if the permutational symmetry of the system is broken due to a (small) perturbation. However, care has to be taken to extend the notion of a weak chimera to nearly symmetric systems since symmetry breaking can have drastic effects on frequency synchronization (Ashwin et al. 2006b). “Classical” chimera states were ﬁrst observed on rings of nonlocally coupled phase oscillators (Kuramoto and Battogtokh 2002). A ﬁnite-dimensional approxi- mation yields a dynamical system that is equivariant with respect to the action of 12 3 624 J Nonlinear Sci (2017) 27:605–626 the dihedral group (Ashwin and Swift 1992). Roughly speaking, classical chimeras on ﬁnite-dimensional rings are trajectories that show characteristic angular frequency synchronization for some ﬁnite time as they exhibit pseudo-random drift along the ring before converging to the fully synchronized state (Omel’chenko et al. 2010; Wolfrum and Omel’chenko 2011). These are not weak chimeras in the sense deﬁned above. the dihedral group (Ashwin and Swift 1992). Roughly speaking, classical chimeras on ﬁnite-dimensional rings are trajectories that show characteristic angular frequency synchronization for some ﬁnite time as they exhibit pseudo-random drift along the ring before converging to the fully synchronized state (Omel’chenko et al. 2010; Wolfrum and Omel’chenko 2011). These are not weak chimeras in the sense deﬁned above. By contrast, initial conditions in the (dynamically invariant) ﬁxed point spaces of a reﬂection symmetry yield symmetric solutions that eventually converge to the fully synchronized state (Omel’chenko 2015), resembling a transient weak chimera. Inter- estingly, the chaotic weak chimeras constructed here share an important feature with these “classical” chimera states: The isotropy of the angular frequency vector may be larger than the symmetry of the solution itself as the oscillators in the “coherent” region are never perfectly phase synchronized. Appendix: A Trigonometric Polynomial Coupling Function The Fourier coefﬁcients of (32) in Sect. 6 are given by a6 = −0.676135392447403 ζ6 = 0.846647746060342 a8 = 0.844660333390606 ζ8 = 0.954985847962987 a10 = 0.087624615584542 ζ10 = 0.212748482509925 a12 = −0.644961491438887 ζ12 = 0.025296512718163 a14 = −0.459724407978054 ζ14 = 0.180050952622569 a16 = −1.175355598611419 ζ16 = 0.835173783095831 a18 = 0.799302873723814 ζ18 = 0.850732311209280 a20 = −1.303832930713680 ζ20 = 0.863697094160152 a22 = 0.094742514998172 ζ22 = 0.355260772731067 a24 = −2.293749915528502 ζ24 = 0.463364388737488 and ar = 0, ζr = 0 for all other r ∈N. 1 a6 = −0.676135392447403 ζ6 = 0.846647746060342 a8 = 0.844660333390606 ζ8 = 0.954985847962987 a10 = 0.087624615584542 ζ10 = 0.212748482509925 a12 = −0.644961491438887 ζ12 = 0.025296512718163 a14 = −0.459724407978054 ζ14 = 0.180050952622569 a16 = −1.175355598611419 ζ16 = 0.835173783095831 a18 = 0.799302873723814 ζ18 = 0.850732311209280 a20 = −1.303832930713680 ζ20 = 0.863697094160152 a22 = 0.094742514998172 ζ22 = 0.355260772731067 a24 = −2.293749915528502 ζ24 = 0.463364388737488 and ar = 0, ζr = 0 for all other r ∈N. J Nonlinear Sci (2017) 27:605–626 625 References Abrams, D.M., Strogatz, S.H.: Chimera states for coupled oscillators. Phys. Rev. Lett. 93(17), 17410 (2004) Abrams, D.M., Mirollo, R.E., Strogatz, S.H., Wiley, D.A.: Solvable model for chimera states of couple oscillators. Phys. Rev. Lett. 101(8), 084103 (2008) Acebrón, J., Bonilla, L., Vicente, C.P., Ritort, F., Spigler, R.: The Kuramoto model: a simple paradigm for synchronization phenomena. Rev. Mod. Phys. 77(1), 137–185 (2005) Antoneli, F., Stewart, I.: Symmetry and synchrony in coupled cell networks 1: ﬁxed-point spaces. Int. Bifurc. Chaos 16(03), 559–577 (2006) Ashwin, P.: Attractors stuck on to invariant subspaces. Phys. Lett. A 209(5–6), 338–344 (1995) Ashwin, P., Burylko, O.: Weak chimeras in minimal networks of coupled phase oscillators. Chaos 2 013106 (2015) Ashwin, P., Nicol, M.: Detection of symmetry of attractors from observations I. Theory. Phys. D 100(1), 58–70 (1997) Ashwin, P., Rodrigues, A.: Hopf normal form with S_N symmetry and reduction to systems of nonlinearly coupled phase oscillators. Phys. D 325, 14–24 (2016) Ashwin, P., Swift, J.W.: The dynamics of n weakly coupled identical oscillators. J. Nonlinear Sci. 2(1), 69–108 (1992) Ashwin, P., Bick, C., Burylko, O.: Identical phase oscillator networks: bifurcations, symmetry and reversibil- ity for generalized coupling. Front. Appl. Math. Stat. (2016a). doi:10.3389/fams.2016.00007 Ashwin, P., Burylko, O., Maistrenko, Y.L., Popovych, O.V.: Extreme sensitivity to detuning for global coupled phase oscillators. Phys. Rev. Lett. 96(5), 054102 (2006b) Ashwin, P., Coombes, S., Nicks, R.: Mathematical frameworks for oscillatory network dynamics in neur science. J. Math. Neurosci. 6(1), 2 (2016c) Barany, E., Dellnitz, M., Golubitsky, M.: Detecting the symmetry of attractors. Phys. D 67(1–3), 66–87 (1993) ( ) Bick, C.: Chaos and chaos control in network dynamical systems. Ph.D. Dissertation. Georg-August- Universität Göttingen (2012) Bick, C., Ashwin, P.: Chaotic weak chimeras and their persistence in coupled populations of phase oscilla- tors. Nonlinearity 29(5), 1468–1486 (2016) Bick, C., Timme, M., Paulikat, D., Rathlev, D., Ashwin, P.: Chaos in symmetric phase oscillator networks. Phys. Rev. Lett. 107(24), 244101 (2011) Bick, C., Ashwin, P., Rodrigues, A.: Chaos in generically coupled phase oscillator networks with nonpair- wise interactions. Chaos 26(9), 094814 (2016) hossat, P., Golubitsky, M.: Symmetry-increasing bifurcation of chaotic attractors. Phys. D 32(3), 423–436 (1988) Chossat, P., Golubitsky, M.: Symmetry-increasing bifurcation of chaotic attractors. Phys. D 32(3), 423–436 (1988) Dellnitz, M., Golubitsky, M., Nicol, M.: Symmetry of attractors and the Karhunen–Loève decomposition. In: Sirovich, L. (ed.) Trends and Perspectives in Applied Mathematics, pp. 73–108. Springer, New York (1994). References (Chapter 4) Dionne, B., Golubitsky, M., Stewart, I.: Coupled cells with internal symmetry: I. Wreath products. Nonlin- earity 9(2), 559–574 (1996) Fried, D.: The geometry of cross sections to ﬂows. Topology 21(4), 353–371 (1982) Field, M.J.: Dynamics and Symmetry, Volume 3 of ICP Advanced Texts in Mathematics. Imperial College Press, London (2007) Golubitsky, M., Stewart, I.: The Symmetry Perspective, Volume 200 of Progress in Mathematics. Birkhäuser, Basel (2002) Golubitsky,M.,Stewart,I.,Schaeffer,D.G.:SingularitiesandGroupsinBifurcationTheory,vol.II.Springer, New York (1988) olubitsky, M., Josic, K., Shea-Brown, E.: Winding numbers and average frequencies in phase oscillator networks. J. Nonlinear Sci. 16(3), 201–231 (2006) Jenkinson, O.: Ergodic optimization. Discret. Contin. Dyn. Syst. 15(1), 197–224 (2006) Katok, A., Hasselblatt, B.: Introduction to the Modern Theory of Dynamical Systems, Volume 54 of Ency- clopedia of Mathematics and its Applications. Cambridge University Press, Cambridge (1995) Kuramoto, Y., Battogtokh, D.: Coexistence of coherence and incoherence in nonlocally coupled phase oscillators. Nonlinear Phenom. Complex Syst. 5(4), 380–385 (2002) 12 3 J Nonlinear Sci (2017) 27:605–626 626 Melbourne, I., Dellnitz, M., Golubitsky, M.: The structure of symmetric attractors. Arch. Ration. Mec Anal. 123(1), 75–98 (1993) ( ), ( ) Misiurewicz, M., Ziemian, K.: Rotation sets for maps of tori. J. Lond. Math. Soc. s2–40(3), 490–506 (1989) Misiurewicz, M., Ziemian, K.: Rotation sets for maps of tori. J. Lond. Math. Soc. s2–40(3), 490–506 (198 Misiurewicz, M., Ziemian, K.: Rotation sets for maps of tori. J. Lond. Math. Soc. s2–40(3), 490–506 (1989) Omel’chenko, O.E.: Coherence-incoherence patterns in a ring of non-locally coupled phase oscillators. Nonlinearity 26(9), 2469–2498 (2013) Omel’chenko, O.E.: Coherence-incoherence patterns in a ring of non-locally coupled phase oscillators. Nonlinearity 26(9), 2469–2498 (2013) Omel’chenko, O.E.: Private communication (2015) Omel’chenko, O.E., Wolfrum, M., Maistrenko, Y.L.: Chimera states as chaotic spatiotemporal patterns. Phys. Rev. E 81(6), 3–6 (2010) Oxtoby, J.C.: Ergodic sets. Bull. Am. Math. Soc. 58(2), 116–137 (1952) Panaggio, M.J., Abrams, D.M.: Chimera states: coexistence of coherence and incoherence in networks coupled oscillators. Nonlinearity 28(3), R67–R87 (2015) p y Panaggio, M.J., Abrams, D.M., Ashwin, P., Laing, C.R.: Chimera states in networks of phase oscillators: the case of two small populations. Phys. Rev. E 93(1), 012218 (2016) Pikovsky, A., Rosenblum, M., Kurths, J.: Synchronization: A Universal Concept in Nonlinear Science Cambridge University Press, Cambridge (2003) g y g Schwartzman, S.: Asymptotic cycles. Ann. Math. 66(2), 270–284 (1957) Sethia, G.C., Sen, A., Johnston, G.L.: Amplitude-mediated chimera states. Phys. Rev. References E 88(4), 04291 (2013) Strogatz, S.H.: From Kuramoto to Crawford: exploring the onset of synchronization in populations of coupled oscillators. Phys. D 143(1–4), 1–20 (2000) p y Strogatz, S.H.: Sync: The Emerging Science of Spontaneous Order. Penguin, Harmondsworth (2004) Tchistiakov, V.: Detecting symmetry breaking bifurcations in the system describing the dynamics of coupl arrays of Josephson junctions. Phys. D 91(1–2), 67–85 (1996) Tchistiakov, V.: Detecting symmetry breaking bifurcations in the system des arrays of Josephson junctions. Phys. D 91(1–2), 67–85 (1996) Walsh, J.A.: Rotation vectors for toral maps and ﬂows: a tutorial. Int. J. Bifurc. Chaos 05(02), 321–348 (1995) Wolfrum, M., Omel’chenko, O.E.: Chimera states are chaotic transients. Phys. Rev. E 84(1), 015201 (2011) Wolfrum, M., Omel’chenko, O.E.: Chimera states are chaotic transients. Phys. Rev. E 84(1), 015201 (2011) Young L -S : What are SRB measures and which dynamical systems have them? J Stat Phys 108(5/6) Wolfrum, M., Omel’chenko, O.E.: Chimera states are chaotic transients. Phys. Rev. E 84(1), 015201 (2011) Young, L.-S.: What are SRB measures, and which dynamical systems have them? J. Stat. Phys. 108(5/6), 733–754 (2002) Young, L.-S.: What are SRB measures, and which dynamical systems have them? J. Stat. Phys. 108(5/6), 733–754 (2002) Zakharova, A., Kapeller, M., Schöll, E.: Chimera death: symmetry breaking in dynamical networks. Phy Rev. Lett. 112(15), 154101 (2014) 12 123
https://openalex.org/W3040490636	https://www.frontiersin.org/articles/10.3389/fvets.2020.00324/pdf	English	null	A McAb-Based Direct Competitive ELISA to Detect O:9 Salmonella Infection in Chicken	Frontiers in veterinary science	2,020	cc-by	4,797	Edited by: Guillermo Tellez, University of Arkansas, United States Salmonella enteritidis and Salmonella pullorum belonging to Group O9 Salmonella are major causative agents of infectious diseases in chicken. O9 antigen as a part of lipopolysaccharide (LPS) is a predominant detected target for Salmonella infection. To identify the infection, an anti-O9 monoclonal antibody (McAb)-based direct competitive enzyme-linked assay (O9 Dc-ELISA) was developed after constraints were optimized; the establishment and application of O9 Dc-ELISA, compared to two commercial kits and plate agglutination test (PAT), showed that O9 Dc-ELISA could screen out more positive samples than the PAT method could and produce the same agreement rates with commercial kits in terms of sensitivity in addition to strong speciﬁcity to clinical serum samples. Reviewed by: Daniel Hernandez-Patlan, Universidad Nacional Autonóma de México, Mexico Ruben Merino-Guzman, National Autonomous University of Mexico, Mexico Correspondence: Xin’an Jiao jiao@yzu.edu.cn Shizhong Geng gszhong@yzu.edu.cn †These authors have contributed equally to this work †These authors have contributed equally to this work A McAb-Based Direct Competitive ELISA to Detect O:9 Salmonella Infection in Chicken Zemiao Xia 1,2,3,4†, Haopeng Geng 1,2,3,4†, Yuan Cai 1,2,3,4, Yaonan Wang 1,2,3,4, Daquan Sun 1,2,3,4, Jian Zhang 1,2,3,4, Zhiming Pan 1,2,3,4, Xin’an Jiao 1,2,3,4 and Shizhong Geng 1,2,3,4* 1 Jiangsu Co-innovation Center for Prevention and Control of Important Animal Infectious Diseases and Zoonoses, Yangzhou University, Yangzhou, China, 2 Jiangsu Key Laboratory of Zoonosis, Yangzhou University, Yangzhou, China, 3 Key Laboratory of Prevention and Control of Biological Hazard Factors (Animal Origin) for Agrifood Safety and Quality, Ministry of Agriculture of China, Yangzhou University, Yangzhou, China, 4 Joint International Research Laboratory of Agriculture and Agri-Product Safety of the Ministry of Education, Yangzhou University, Yangzhou, China ORIGINAL RESEARCH published: 03 July 2020 doi: 10.3389/fvets.2020.00324 Specialty section: Specialty section: This article was submitted to Veterinary Infectious Diseases, a section of the journal Frontiers in Veterinary Science Salmonella, an important zoonotic pathogen, is one of the major causative agents of food-borne infectious diseases worldwide (1). Consumption of foods such as egg, chicken, pork, beef, and dairy products contaminated with Salmonella can cause salmonellosis in humans (2–4). This pathogen not only brings huge economic loss in the animal industry but also impacts human health, even death (5–8). Because of these disease harms and public health hazards, eﬃcient surveillance is very important to reduce the prevalence of Salmonella and the risk of transmission to humans. Received: 06 March 2020 Accepted: 11 May 2020 Published: 03 July 2020 Salmonella enteritidis and Salmonella pullorum, which are important members in group O:9 Salmonella, are the main pathogens found in modern large-scale chicken farms in China (2, 9–11). In addition to their high morbidity and mortality in young broilers, they cause non-apparent infections in adult chickens without obvious clinical symptoms. Thus, it is diﬃcult to ﬁnd Salmonella infection in adult chickens. If Salmonella-infected chicken is not found on time, it may be a source of infection causing unlimited spread in chicken, even to humans, because of horizontal transmission and vertical transmission. It is necessary to carry out a seroepidemiological survey on Salmonella for healthy breeding and food safety. Keywords: O:9 Salmonella, McAb, O9 Dc-ELISA, speciﬁcity, PAT Primary Quantity of Coated LPS and HRP-Labeled O9 McAb for Competitive Setting Up of the Cutoff Value and Comparison With Commercial Kits and PAT One hundred serum samples from artiﬁcially infected chickens at diﬀerent time points as positive control and 100 serum samples from SPF chickens as negative control were detected using the France ID.vet Salmonella kit, and these 200 serum samples were detected by O9 Dc-ELISA; the receiver operating characteristic (ROC) curve was made according to the inhibition rate. Based on these results, the cutoﬀvalue which was the value of negative samples + 3SD as a negative/positive judgment boundary was set up. Primary quantities of LPS and HRP-labeled O9 McAb were conﬁrmed by chessboard titration to develop a direct ELISA following conventional ELISA protocol. Horizontal gradient dilution of HRP-labeled O:9 McAb and vertical gradient dilution of coating antigens were performed. The ﬁnal concentration of a tested positive serum was diluted 1:10. According to the serum inhibition rate [inhibition rate = (1-detected serum OD value/blank control OD) × 100%], optimal balanced concentrations were selected. Fifty random clinical serum samples were tested using a double blind test by O9 Dc-ELISA and compared to IDEXX ELISA kit (IDEXX USA, 99-0002040) and ID.vet ELISA kit (ID.vet France, SALSGPD-5P) to judge the accuracy of O9 Dc- ELISA in clinical application. Citation: Xia Z, Geng H, Cai Y, Wang Y, Sun D, Zhang J, Pan Z, Jiao X and Geng S (2020) A McAb-Based Direct Competitive ELISA to Detect O:9 Salmonella Infection in Chicken. Front. Vet. Sci. 7:324. doi: 10.3389/fvets.2020.00324 July 2020 \| Volume 7 \| Article 324 Frontiers in Veterinary Science \| www.frontiersin.org O9 Dc-ELISA for O:9 Salmonella Detection Xia et al. Xia et al. was used as a competitor of positive serum and subjected to a competition ELISA. The serum dilution at the highest inhibition rate was selected as the serum dilution of the competitive ELISA method. Currently, plate agglutination test (PAT) is the main detection method used during Salmonella surveillance for its easy operation and low cost, but its sensitivity and speciﬁcity are poor and can easily cause false results because of antigen detection, visual observation, and subjective judgment. Time of HRP-labeled McAb Binding With LPS and Reacting With TMB The present study was conducted under the approval of Laboratory Animal Ethics Committee of Yangzhou University (Jiangsu province, China) in accordance with Laboratory Animal Guidelines for ethical review of animal welfare (GB/T 35892- 2018, National Standards of the People’s Republic of China). Competitive ELISA was performed with the LPS coating concentration and HRP-labeled McAb and serum dilution, which were optimized in the previous steps. To ensure McAb to bind with coated LPS as possible, the incubation time of the HRP- labeled McAb was set to 1.0, 1.5, 2.0, 2.5, and 3.0 h, respectively, for analysis based on the N/P value. Strain, Hybridoma Cell Line, and Animals Salmonella enteritidis (C50041) and Salmonella pullorum (S06004) were stored by our laboratory. A 3-47-0 hybridoma cell line secreting anti-O9 McAb was developed and preserved by our laboratory. Thirty 10-week BALB/c female mice were purchased for ascites from Comparative Medical Center of Yangzhou University. After optimization time of HRP-labeled McAb binding with LPS, the incubation time of HRP-labeled McAb to react with substrate 3,3′,5,5′-tetramethylbenzidine (TMB) was also optimized. The hydrolysis time for TMB substrate by HRP was set to 3, 5, and 10 min. Quantity of Coated LPS and HRP-Labeled O:9 McAb LPS is the main antigen found on the Salmonella surface and the primary target for the immune system (12). After Salmonella infection, LPS can induce and keep a high level of antibody from early stages. Serotyping using serum/antibodies to the O-antigen of Salmonella lipopolysaccharide (LPS) (13, 14) is a critical basis of current Salmonella surveillance programs. Routine serotyping helps in monitoring public health response to the global challenge of salmonellosis and the eﬀectiveness of control measures (9, 15– 17). Based on the previous ELISA, LPS were divided into four groups, 480, 320, 190, and 160 ng/mL. By comparing the N/P values (negative serum OD value/positive serum OD value), the coating concentration at which the N/P value was the largest was chosen out as the optimal concentration. Similarly, the antibody was divided into six groups, 56.8, 52.0, 48.0, 44.6, 41.6, and 39.1 ng/mL, to optimize the concentration of HRP-labeled McAb. The N/P values were compared (negative serum OD value/positive serum OD value) with the concentration of the HRP-labeled O9 McAb. Therefore, the development of readily available detection systems of the Salmonella antibody in chicken is important for mass-scale laboratory diagnosis. In this study, we developed an anti-O:9 Salmonella McAb-based direct competitive ELISA method to meet the requirements of accurate Salmonella surveillance. Time of LPS Being Coated Onto a Plate Three ELISA plates were coated at 100 µL/well at an optimized coating concentration. The coating time of the three ELISA plates was 16, 24, and 36 h, respectively. Statistical Analysis FIGURE 1 \| Standard curve of sugar concentration determined by the anthracene-ketone method. y Statistical analysis was performed using GraphPad Prism 5 (GraphPad Software, USA). One-way ANOVA followed by Dunnett’s multiple-comparison tests was used to determine the statistical diﬀerences between multiple experimental groups. All data are expressed as mean ± standard error of the mean (SEM) unless otherwise speciﬁed. P < 0.05 was considered statistically signiﬁcant. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. Constraint Optimizations for O9 Dc-ELISA Tested Positive Serum Dilution Two Salmonella pullorum-positive sera, two S. enteritidis-positive sera, and two negative sera from speciﬁc pathogen-free (SPF) chicken were diluted 1:4, 1:8, 1:16, 1:32, 1:64, 1:128, and 1:256. Based on the previous direct ELISA, each tested serum dilution The coincidence rate was calculated by the following formula: number of [(+,+)+(-,-)]/total number %. July 2020 \| Volume 7 \| Article 324 Frontiers in Veterinary Science \| www.frontiersin.org 2 O9 Dc-ELISA for O:9 Salmonella Detection Xia et al. FIGURE 1 \| Standard curve of sugar concentration determined by the anthracene-ketone method. FIGURE 2 \| Curves of the HRP-labeled McAb binging to coated LPS. FIGURE 3 \| Concentration optimization of coated LPS in O9 Dc-ELISA. FIGURE 1 \| Standard curve of sugar concentration determined by the anthracene-ketone method. FIGURE 2 \| Curves of the HRP-labeled McAb binging to coated LPS. Preparation of Coated LPS and HRP-Labeled O9 McAb Preparation of Coated LPS and HRP-Labeled O9 McAb FIGURE 1 \| Standard curve of sugar concentration determined by the anthracene-ketone method. In this study, LPS were puriﬁed by the hot phenol–water method (18) and its concentration was calculated according to the standard sugar curve made by the anthracene ketone method. According to the measured OD620nm value and standard curve (Figure 1), the ﬁnal concentration of LPS was 484.31 µg/mL. McAb against O9 LPS (O9 McAb) was puriﬁed from hybridoma supernatants by caprylic/ammonium sulfate precipitation (19) and labeled with horseradish peroxidase (HRP) (20). The titer of HRP-labeled O9 McAb (HRP-O9 McAb) was up to 51,200 by indirect ELISA. FIGURE 2 \| Curves of the HRP-labeled McAb binging to coated LPS. Setting Up of the Cutoff Value According to the results using the France ID.vet Salmonella kit as a reference of positive and negative chicken sera, and the percent inhibition (PI) values by O9 Dc-ELISA which were calculated using the formula PI (%) = (1−OD450 of test serum/OD450 of blank control) × 100%, the cutoﬀbased on the ROC curve was 38% (Figure 10). Under PI of 38%, the speciﬁcity of O9 Dc- ELISA reached up to 99.7% and the sensitivity reached up to 96.2% in ROC. The distribution of 100 positive serum samples and the 100 negative serum samples determined by O9 Dc-ELISA showed that 38% of inhibiting rate was indeed a threshold which could distinguish positive serum and negative serum (Figure 11). Comparison Among O9 Dc-ELISA, Three Commercial Kits, and PAT FIGURE 6 \| The negative/positive OD value of different LPS coating time in O9 Dc-ELISA. **p < 0.0001. FIGURE 6 \| The negative/positive OD value of different LPS coating time in O9 Dc-ELISA. p < 0.0001. To validate the test ability of O9 Dc-ELISA, we randomly collected 50 serum samples for comparison to the results using O9 Dc-ELISA, PAT, IDEXX ELISA kit, and ID.vet ELISA kit; the results revealed that their coincidence rates were 88% (44/50, Table 1), 98% (49/50, Table 2), and 98% (49/50, Table 3), respectively. Although four samples negative with PAT were positive with O9 Dc-ELISA and two commercial ELISA kits, and 1 sample negative with O9 Dc-ELISA, IDEXX, and ID.vet ELISA kit was positive with PAT, there was no statistical diﬀerence among 4 methods. The results showed that O9 Dc-ELISA could screen out more positive samples than the PAT method could and produced the same agreement rates with two commercial kits in terms of sensitivity in addition to strong speciﬁcity. FIGURE 7 \| Time effect of HRP-labeled O9 McAb binding with coated LPS in O9 Dc-ELISA based on the negative/positive OD value. p < 0.001, *p < 0.0001. Constraint Determination of Competitive ELISA (SE, SE2: serum by Salmonella enteritidis 1, 2; SP1, SP2: serum by Salmonella pullorum 1, 2; SPF1, SPF2: serum from SPF chickens). FIGURE 6 \| The negative/positive OD value of different LPS coating time in O9 Dc-ELISA. p < 0.0001. FIGURE 4 \| Concentration optimization of HRP-labeled O9 McAb in O9 Dc-ELISA. FIGURE 4 \| Concentration optimization of HRP-labeled O9 McAb in O9 Dc-ELISA. FIGURE 5 \| Optimization of positive serum dilution. (SE, SE2: serum by Salmonella enteritidis 1, 2; SP1, SP2: serum by Salmonella pullorum 1, 2; SPF1, SPF2: serum from SPF chickens). FIGURE 5 \| Optimization of positive serum dilution. (SE, SE2: serum by Salmonella enteritidis 1, 2; SP1, SP2: serum by Salmonella pullorum 1, 2; SPF1, SPF2: serum from SPF chickens). FIGURE 6 \| The negative/positive OD value of different LPS coating time in O9 Dc-ELISA. p < 0.0001. FIGURE 7 \| Time effect of HRP-labeled O9 McAb binding with coated LPS in O9 Dc-ELISA based on the negative/positive OD value. p < 0.001, *p < 0.0001. July 2020 \| Volume 7 \| Article 324 FIGURE 5 \| Optimization of positive serum dilution. (SE, SE2: serum by Salmonella enteritidis 1, 2; SP1, SP2: serum by Salmonella pullorum 1, 2; SPF1, SPF2: serum from SPF chickens). FIGURE 4 \| Concentration optimization of HRP-labeled O9 McAb in O9 Dc-ELISA. FIGURE 4 \| Concentration optimization of HRP-labeled O9 McAb in O9 Dc-ELISA. FIGURE 6 \| The negative/positive OD value of different LPS coating time in O9 Dc-ELISA. p < 0.0001. Constraint Determination of Competitive ELISA ELISA A series of dilutions of LPS, HRP-labeled O9 McAb, and positive sera were prepared for chessboard titration and optimization (Figure 2). After optimization assay, 320 ng/ml LPS for coating (Figure 3), 41.6 ng/ml HRP-labeled O9 McAb (Figure 4), and positive serum dilution of 1:4 (Figure 5) were selected for developing O9 Dc-ELISA. The inhibition rate of positive serum by Salmonella pullorum and Salmonella enteritidis was up to 94 and 89%, respectively. On this basis, a standard operating procedure was formulated, after 96-well plates (Bioﬁl company, Canada, FEP101896) were coated with ∼100 µL puriﬁed LPS (320 ng/ml) in carbonate bicarbonate buﬀer (CBS, pH 9.4) at 4◦C for 24 h (Figure 6) and washed with PBST (0.05% Tween 20 in phosphate-buﬀered saline) two times; 200 µL/well 2% BSA PBS solution was added again for blocking for 3 h at 37◦C, then 50 µL 1:2 diluted chicken serum (PBS for blank control) and 50 µL HRP-labeled O9 McAb of 41.6 ng/ml were added at the same time. After incubation at 37◦C for 2 h (Figure 7), all unbound materials were removed by washing with PBST six times. 100 µL of TMB chromogenic substrate was added to each well and incubated at 37◦C for 3 min (Figure 8). After the color development was completed, 50 µL of 2 M H2SO4 was added to each well to terminate the color development, and the OD450nm absorption value was read. FIGURE 2 \| Curves of the HRP-labeled McAb binging to coated LPS. FIGURE 3 \| Concentration optimization of coated LPS in O9 Dc-ELISA. Speciﬁcity Analysis of O9 Dc-ELISA We prepared the tested sera from chicken infected by Escherichia coli, Proteus mirabilis, non-O9 Salmonella [Salmonella typhimurium (O:4)], the negative sera from SPF chickens, and the positive sera from chickens infected with Salmonella pullorum and Salmonella enteritidis; the results showed that O9 Dc-ELISA could not check out the sera against non-Salmonella and non-O9 Salmonella. The value of negative sera was more than 1.0 whereas the OD value of positive sera was less than 0.25 as a control (Figure 9) based on P/N≥2.1. July 2020 \| Volume 7 \| Article 324 Frontiers in Veterinary Science \| www.frontiersin.org 3 O9 Dc-ELISA for O:9 Salmonella Detection Xia et al. Xia et al. FIGURE 5 \| Optimization of positive serum dilution. DISCUSSION Salmonella enteritidis and Salmonella pullorum are two of the most important Salmonella spp. that threaten the poultry industry, and humans are infected by directly or indirectly FIGURE 7 \| Time effect of HRP-labeled O9 McAb binding with coated LPS in O9 Dc-ELISA based on the negative/positive OD value. p < 0.001, *p < 0.0001. July 2020 \| Volume 7 \| Article 324 Frontiers in Veterinary Science \| www.frontiersin.org 4 O9 Dc-ELISA for O:9 Salmonella Detection Xia et al. FIGURE 8 \| The negative/positive OD value of different reactive times of HRP labeled on O9 McAb with substrate TMB. p < 0.01. FIGURE 9 \| The OD value of chicken sera against different pathogens for speciﬁcity analysis. eating contaminated water and food, which causes great hazard to human public health security (21, 22). In our study, a FIGURE 10 \| The distribution of inhibiting rate by Dc-ELISA and setting up of the cut-off value. FIGURE 11 \| ROC analysis curve for competitive ELISA detecting chicken serum samples. Salmonella pullorum and Salmonella enteritidis could cause signiﬁcant inhibition. FIGURE 8 \| The negative/positive OD value of different reactive times of HRP labeled on O9 McAb with substrate TMB. p < 0.01. FIGURE 10 \| The distribution of inhibiting rate by Dc-ELISA and setting up of the cut-off value. FIGURE 8 \| The negative/positive OD value of different reactive times of HRP labeled on O9 McAb with substrate TMB. p < 0.01. FIGURE 9 \| The OD value of chicken sera against different pathogens for speciﬁcity analysis. FIGURE 10 \| The distribution of inhibiting rate by Dc-ELISA and setting up of the cut-off value. FIGURE 10 \| The distribution of inhibiting rate by Dc-ELISA and setting up of the cut-off value. FIGURE 11 \| ROC analysis curve for competitive ELISA detecting chicken serum samples. FIGURE 11 \| ROC analysis curve for competitive ELISA detecting chicken serum samples. Salmonella pullorum and Salmonella enteritidis could cause signiﬁcant inhibition. eating contaminated water and food, which causes great hazard to human public health security (21, 22). In our study, a McAb-based competitive ELISA was established to detect O:9 Salmonella infection in chicken. In order to achieve a better reaction system, we explored various conditions, including concentration of LPS coating and HRP-labeled O9 McAb, serum dilution, LPS coating time, and reaction time of HRP- labeled McAb. Frontiers in Veterinary Science \| www.frontiersin.org CONCLUSION O9 Dc-ELISA has good ability in O9 antibody detection and had better performance than the PAT method and agreement rates with commercial kits in terms of sensitivity during the detection of clinical chicken serum samples. It must play an important role in O:9 Salmonella detection for Salmonella clearance in China in the future. FUNDING This work was supported by the National Key Research and Development Program Special Project (2016YFD0501607), the Special Fund for Agroscientiﬁc Research in the Public Interest (201403054), and the Natural Science Foundation of Jiangsu Province of China (BK20151306). ETHICS STATEMENT The animal study was reviewed and approved by the Animal Welfare and Ethics Committees of Yangzhou University. DATA AVAILABILITY STATEMENT All datasets generated for this study are included in the article/supplementary material. AUTHOR CONTRIBUTIONS SG, XJ, and HG designed the paper. HG, ZX, and DS performed the experiments. YC, JZ, and YW provided help during experiments. ZP and XJ made critical revisions to the paper and contributed to paper writing. All authors contributed to the article and approved the submitted version. rate with PAT was 88%. The above results indicated that this O9 Dc-ELISA has a good detection eﬀect on the O9 antibody and had better performance than the PAT method based on more positive samples being checked out and the same agreement rates with commercial kits in terms of sensitivity in addition to strong speciﬁcity in the detection of clinical samples. This kit oﬀered a good base as a ﬁrst-generation product; it will be further evaluated and optimized according to clinical detection rate with PAT was 88%. The above results indicated that this O9 Dc-ELISA has a good detection eﬀect on the O9 antibody and had better performance than the PAT method based on more positive samples being checked out and the same agreement rates with commercial kits in terms of sensitivity in addition to strong speciﬁcity in the detection of clinical samples. This kit oﬀered a good base as a ﬁrst-generation product; it will be further evaluated and optimized according to clinical detection DISCUSSION By testing 100 artiﬁcial positive samples and 100 negative serum samples from SPF chickens and 50 random clinical serum samples, the sensitivity and speciﬁcity at diﬀerent thresholds were compared, and the ﬁnal selected inhibition rate was 38% as the critical value of the competition ELISA kit. According to ROC, the speciﬁcity of O9 Dc-ELISA was 99.7%, and the sensitivity was 96.2%. This O9 Dc-ELISA was compared with PAT, IDEXX ELISA kit, and ID.vet ELISA kit, respectively. The results showed that the coincidence rate of the O9 Dc-ELISA kit and ID.vet ELISA kit was 98%; the coincidence rate with the Salmonella enteritidis test kit was 98%; and the coincidence In order to conﬁrm that the established O9 Dc-ELISA did not cause a cross-reaction, we used O9 Dc-ELISA to test Escherichia coli, Proteus mirabilis, Salmonella typhimurium (O:4), and negative sera from SPF chicken, Salmonella pullorum, and Salmonella enteritidis. The tests showed that only sera from July 2020 \| Volume 7 \| Article 324 Frontiers in Veterinary Science \| www.frontiersin.org 5 O9 Dc-ELISA for O:9 Salmonella Detection Xia et al. TABLE 1 \| Comparison of the results between O9 Dc-ELISA and PAT. PAT Total + - O9 Dc-ELISA + 1 4 5 - 2 43 45 Total 3 47 50 TABLE 2 \| Comparison of the results between O9 Dc-ELISA and IDEXX ELISA kit. IDEXX ELISA Kit Total + - O9 Dc-ELISA + 4 1 5 - 0 45 45 Total 4 46 50 TABLE 3 \| Comparison of the results between O9 Dc-ELISA and ID.vet ELISA kit. ID.vet ELISA Kit Total + - O9 Dc-ELISA + 4 1 5 - 0 45 45 Total 4 46 50 TABLE 1 \| Comparison of the results between O9 Dc-ELISA and PAT. performance based on more serum samples to develop a second- generation kit in the future. performance based on more serum samples to develop a second- generation kit in the future. REFERENCES 5. Schrader KN, Fernandez-Castro A, Cheung WKW, Crandall CM, Abbott SL. Evaluation of commercial Antisera for Salmonella serotyping. J Clin Microbiol. (2008) 46:685–8. doi: 10.1128/JCM.01808-07 5. Schrader KN, Fernandez-Castro A, Cheung WKW, Crandall CM, Abbott SL. Evaluation of commercial Antisera for Salmonella serotyping. J Clin Microbiol. (2008) 46:685–8. doi: 10.1128/JCM.01808-07 1. Fei X, He X, Guo R, Yin C, Geng H, Wu K et al. Analysis of prevalence and CRISPR typing reveals persistent antimicrobial- resistant Salmonella infection across chicken breeder farm production stages. Food Control. (2017) 77:102–9. doi: 10.1016/j.foodcont.2017. 01.023 6. Hendriksen RS, Mikoleit M, Carlson VP, Karlsmose S, Vieira AR, Jensen AB, et al. WHO Global Salm-Surv external quality assurance system for serotyping of Salmonella isolates from 2000 to 2007. J Clin Microbiol. (2009) 47:2729–36. doi: 10.1128/JCM.02437-08 2. Geimba MP, Tondo EC, de Oliveira, Canal CW, Brandelli A. Serological characterization and prevalence of spvR genes in Salmonella isolated from foods involved in outbreaks in Brazil. J Food Prot. (2004) 67:1229– 33. doi: 10.4315/0362-028X-67.6.1229 7. Hoelzer K, Moreno Switt Al, Wiedmann M. Animal contact as a source of human non-typhoidal salmonellosis. Vet Res. (2011) 42:34. doi: 10.1186/1297-9716-42-34 8. Mather AE, Reid SW, Maskell DJ, Parkhill J, Fookes MC, Harris SR, et al. Distinguishable epidemics within diﬀerent hosts of the multidrug- resistant zoonotic pathogen Salmonella Typhimurium DT104. Science. (2013) 341:1514. doi: 10.1126/science.1240578 3. Meemken D, Tangemann AH, Meermeier D,Gundlach S, Mischok D, Greiner M, et al. Establishment of serological herd proﬁles for zoonoses and production diseases in pigs by “meat juice multi-serology”. Prev Vet Med. (2014) 113.4:589–98. doi: 10.1016/j.prevetmed.2013. 12.006 3. Meemken D, Tangemann AH, Meermeier D,Gundlach S, Mischok D, Greiner M, et al. Establishment of serological herd proﬁles for zoonoses and production diseases in pigs by “meat juice multi-serology”. Prev Vet Med. (2014) 113.4:589–98. doi: 10.1016/j.prevetmed.2013. 12.006 9. Herikstad H, Motarjemi Y, Tauxe RV. Salmonella surveillance: a global survey of public health serotyping. Epidemiol Infect. (2002) 129:1– 8. doi: 10.1017/S0950268802006842 9. Herikstad H, Motarjemi Y, Tauxe RV. Salmonella surveillance: a global survey of public health serotyping. Epidemiol Infect. (2002) 129:1– 8. doi: 10.1017/S0950268802006842 4. Vo TH, Le NH, Cao TT, Nuorti JP, Minh NN. An outbreak of food-borne salmonellosis linked to a bread takeaway shop in Ben Tre City Vietnam. Int J Infect Dis. (2014) 26:128–31. doi: 10.1016/j.ijid.2014.05.023 4. Vo TH, Le NH, Cao TT, Nuorti JP, Minh NN. 8. Mather AE, Reid SW, Maskell DJ, Parkhill J, Fookes MC, Harris SR, et al. Distinguishable epidemics within diﬀerent hosts of the multidrug- resistant zoonotic pathogen Salmonella Typhimurium DT104. Science. (2013) 341:1514. doi: 10.1126/science.1240578 5. Schrader KN, Fernandez-Castro A, Cheung WKW, Crandall CM, Abbott SL. Evaluation of commercial Antisera for Salmonella serotyping. J Clin Microbiol. (2008) 46:685–8. doi: 10.1128/JCM.01808-07 REFERENCES An outbreak of food-borne salmonellosis linked to a bread takeaway shop in Ben Tre City Vietnam. Int J Infect Dis. (2014) 26:128–31. doi: 10.1016/j.ijid.2014.05.023 10. Singh V. Salmonella serovars and their host speciﬁcity. J Vet Sci & Anim Husb. (2013) 1:301. doi: 10.15744/2348-9790.1.301 10. Singh V. Salmonella serovars and their host speciﬁcity. J Vet Sci & Anim Husb. (2013) 1:301. doi: 10.15744/2348-9790.1.301 July 2020 \| Volume 7 \| Article 324 Frontiers in Veterinary Science \| www.frontiersin.org 6 O9 Dc-ELISA for O:9 Salmonella Detection Xia et al. 11. Gong J, Zhang J, Xu M, Zhu C, Yu Y, Liu X, et al. Prevalence and ﬁmbrial genotype distribution of poultry Salmonella isolates in China (2006 to 2012). Appl Environ Microbiol. (2014) 80:687–93. doi: 10.1128/AEM.03223-13 19. Fearnley E, Raupach J, Lagala F, Cameron S. Salmonella in chicken meat, eggs and humans; Adelaide, South Australia, 2008. Int J Food Microbiol. (2011) 146:219–27. doi: 10.1016/j.ijfoodmicro.2011.02.004 12. Ko HJ, Yang JY, Shim DH, Yang H, Park SM, Curtiss R III, et al. Innate immunity mediated by MyD88 signal is not essential for induction of lipopolysaccharide-speciﬁc B cell responses but is indispensable for protection against Salmonella enterica serovar Typhimurium infection. J Immunol. (2009) 182:2305–12. doi: 10.4049/jimmunol.0801980 20. Almeida C, Cerqueira L, Azevedo NF, Vieira MJ. Detection of Salmonella enterica serovar enteritidis using real time PCR, immunocapture assay, PNA FISH and standard culture methods in diﬀerent types of food samples. Int J Food Microbiol. (2013) 161:16–22. doi: 10.1016/j.ijfoodmicro.2012.11.014 21. Saeed AFUH, Ling S, Yuan J, Wang S. The Preparation and identiﬁcation of a monoclonal antibody against domoic acid and establishment of detection by indirect competitive ELISA. Toxins (Basel). (2017) 9:250 doi: 10.3390/toxins9080250 13. Ronholm J, Zhang Z, Cao X, Lin M. Monoclonal antibodies to lipopolysaccharide antigens of Salmonella enterica serotype Typhimurium DT104. Hybridoma. (2011) 30:43–52. doi: 10.1089/hyb.2010.0066 22. Li YS, Zhou Y, Meng XY, Zhang YY, Liu JQ, Zhang Y, et al. Enzyme- antibody dual labeled gold nanoparticles probe for ultrasensitive detection of κ-casein in bovine milk samples. Biosens Bioelectron. (2014) 61:241– 4. doi: 10.1016/j.bios.2014.05.032 14. Eguchi M, Kikuchi Y. Binding of Salmonella-speciﬁc antibody facilitates speciﬁc T cell responses via augmentation of bacterial uptake and induction of apoptosis in macrophages. J Infect Dis. (2010) 201:62–70. doi: 10.1086/648615 15. Wahlström H, Sternberg Lewerin S, Sundström, K, Ivarsson S. Frontiers in Veterinary Science \| www.frontiersin.org July 2020 \| Volume 7 \| Article 324 REFERENCES Estimation of the expected change in domestic human Salmonella cases in Sweden in 2010, given a hypothetical relaxation of the current Salmonella control programme. PLoS ONE. (2014) 9:e89833. doi: 10.1371/journal.pone.0089833 Conﬂict of Interest: The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. 16. Wattiau P, Boland C, Bertrand S. Methodologies for Salmonella enterica subsp. enterica subtyping: gold standards and alternatives. Appl Environ Microbiol. (2011) 77:7877–88. doi: 10.1128/AEM.05527-11 Copyright © 2020 Xia, Geng, Cai, Wang, Sun, Zhang, Pan, Jiao and Geng. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. 17. Jang YH, Lee SJ, Lim JG, Lee HS, Kim TJ, Park JH, et al. The rate of Salmonella spp. infection in zoo animals at Seoul Grand Park Korea. J Vet Sci. (2008) 9:177–81. doi: 10.4142/jvs.2008.9.2.177 18. Westphal O, Jann JK. Bacterial lipopolysaccharide extraction with phenol- water and further applications of the procedure. Methods Carbohydr. Chem. (1965) 5:83–92. 18. Westphal O, Jann JK. Bacterial lipopolysaccharide extraction with phenol- water and further applications of the procedure. Methods Carbohydr. Chem. (1965) 5:83–92. July 2020 \| Volume 7 \| Article 324 Frontiers in Veterinary Science \| www.frontiersin.org 7
https://openalex.org/W4307869407	https://www.frontiersin.org/articles/10.3389/fonc.2022.984770/pdf	English	null	Thymic cyst: Is attenuation artifactually increased on contrast-enhanced CT?	Frontiers in oncology	2,022	cc-by	4,810	OPEN ACCESS OPEN ACCESS EDITED BY Mohamed Rahouma, NewYork-Presbyterian, United States REVIEWED BY Hilal Ozakinci, Mofﬁtt Cancer Center, United States Wang Shuai, Fudan University, China CORRESPONDENCE Ling Hu hulingsmart@163.com Chao Wang wangchaosmart@zju.edu.cn SPECIALTY SECTION This article was submitted to Thoracic Oncology, a section of the journal Frontiers in Oncology RECEIVED 02 July 2022 ACCEPTED 19 October 2022 PUBLISHED 31 October 2022 CITATION Wang C, Mao J, Yang S, Xie H, Wang S and Hu L (2022) Thymic cyst: Is attenuation artifactually increased on contrast-enhanced CT? Front. Oncol. 12:984770. doi: 10.3389/fonc.2022.984770 EDITED BY Mohamed Rahouma, NewYork-Presbyterian, United States REVIEWED BY Hilal Ozakinci, Mofﬁtt Cancer Center, United States Wang Shuai, Fudan University, China Chao Wang 1, Jin Mao 1, Siyu Yang 1, Huanhuan Xie 1, Shan Wang 1 and Ling Hu 2* Chao Wang 1, Jin Mao 1, Siyu Yang 1, Huanhuan Xie 1, Shan Wang 1 and Ling Hu 2 1Department of Radiology, The Second Afﬁliated Hospital, Zhejiang University School of Medicine, Hangzhou, Zhejiang, China, 2Department of Ultrasound, Hangzhou Women’s Hospital, Hangzhou, Zhejiang, China Background: Thymic cysts are often misinterpreted as thymomas or lymph nodes, then leading to unnecessary thymectomy. The purpose of this study was to investigate how the adjacent large vessels artifactually affected attenuation values of thymic cysts on contrast–enhanced CT (CE-CT). Methods: In this retrospective study, a total of 84 patients were included with pathological diagnosis of thymic cysts and preoperative CE-CT. Quantitative measurements of the size, CT attenuation of thymic cysts and CT attenuation of adjacent large vessels were performed on preoperative CE-CT. According to the absolute change in attenuation of the cysts between contrast-enhanced and nonenhanced CT, the patients were classiﬁed into the groups of artifactual hyper-density, unchanged density, and artifactual hypo-density. CT characteristics were compared between the three groups. Furthermore, multivariable logistic regression analysis was performed to determine the independent factors for artifactual hyper-density. Wang C, Mao J, Yang S, Xie H, Wang S and Hu L (2022) Thymic cyst: Is attenuation artifactually increased on contrast-enhanced CT? Front. Oncol. 12:984770. doi: 10.3389/fonc.2022.984770 COPYRIGHT © 2022 Wang, Mao, Yang, Xie, Wang and Hu. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. TYPE Brief Research Report PUBLISHED 31 October 2022 DOI 10.3389/fonc.2022.984770 TYPE Brief Research Report PUBLISHED 31 October 2022 DOI 10.3389/fonc.2022.984770 thymic cysts, CT, pseudoenhancement, artifacts, misdiagnosis KEYWORDS Introduction after 60 seconds of delay. Then, axial images of 5 mm thickness were reconstructed. Finally, the reconstructed images were transferred to the picture archiving communication system (PACS) workstation. after 60 seconds of delay. Then, axial images of 5 mm thickness were reconstructed. Finally, the reconstructed images were transferred to the picture archiving communication system (PACS) workstation. With the widespread use of computed tomography (CT) in thoracic imaging, detection of incidental mediastinal lesions has become very common. However, CT showed a low diagnostic accuracy in differentiating thymic cysts from thymic tumors (1). Despite advances in CT techniques, the evaluation of the solitary thymic cyst, especially the cyst of non-water attenuation, is still a diagnostic challenge for radiologists (2). A previous study reported there were > 25% unnecessary thymectomy rates (3). Thymic cysts are often misinterpreted as thymomas or lymph nodes because of attenuation values greater than that of water with its “solid” appearance (3, 4). Furthermore, one of the important CT features applied to distinguish thymic cysts from thymomas or lymph nodes is an increase in the attenuation of the mass after intravenous contrast medium administration. Pseudo-enhancement phenomenon of renal cysts on contrast–enhanced CT (CE-CT) has been widely observed and fully acknowledged to be artifactually increased by marked enhancement of peripheral renal parenchyma (5, 6). However, little is known about how the dense contrast medium passing through mediastinal large vessels affects attenuation values of thymic cysts on contrast-enhanced CT. The purpose of this study was to investigate how the adjacent large vessels artifactually affected attenuation values of thymic cysts and its inﬂuencing factors on CE-CT. Quantitative measurements of the size and CT attenuation of the thymic cysts were performed on the PACS by a board- certiﬁed radiologist (SW) with 6 years of experience in chest imaging. This radiologist independently measured the sizes and attenuation of the cysts, and she was blinded to any information about the purpose of the study to reduce a potential bias in measurement. The longest long diameter and the longest short diameter perpendicular to the long diameter of the lesion were obtained at the same largest cross-sectional area (Figure 1). CT attenuation values of cysts were measured by placing an oval region of interest (ROI) within the cysts. The ROIs used for the pre- and postcontrast evaluation of a given cyst were obtained at the same cross-sectional location. The mean CT attenuation of each cyst were recorded. Introduction The change in attenuation of the cysts were calculated as the absolute difference between the mean contrast-enhanced and nonenhanced attenuation values. Postcontrast CT attenuation values of the adjacent large vessels in both arterial and venous phase were measured by placing an oval ROI within the adjacent large vessels. Then, according to the absolute change in attenuation of the cysts between contrast-enhanced and nonenhanced CT, the patients were classiﬁed into three groups, including the groups of > 5 HU (n=27), -5~5 HU (n=38), and <-5 HU (n=19). OPEN ACCESS No use, distribution or reproduction is permitted which does not comply with these terms. Results: The group of artifactual hyper-density had smaller short diameter of the cysts, higher postcontrast attenuation values and lower nonenhanced attenuation values of the adjacent large vessel. Furthermore, the multivariable logistic analysis showed that artifactual hyper-density of thymic cysts was negatively associated with nonenhanced attenuation of adjacent large vessel, and positively associated with postcontrast attenuation of adjacent large vessel and postcontrast attenuation of cysts. Conclusions: Most cases with >20 HU nonenhanced CT attenuation in surgically resected cases. Artifactual hyper-density─pseudo-enhancement phenomenon of thymic cysts was more apparent in higher increasing attenuation of adjacent large vessels on CE-CT. A well understanding of this phenomenon can help reduce preoperative misdiagnosis and unnecessary thymectomy. thymic cysts, CT, pseudoenhancement, artifacts, misdiagnosis Frontiers in Oncology 01 frontiersin.org frontiersin.org Wang et al. 10.3389/fonc.2022.984770 10.3389/fonc.2022.984770 Frontiers in Oncology frontiersin.org Methods Statistical analyses were performed using IBM SPSS Statistics 23 software (IBM Corporation, New York). The normality of the data was tested using Kolmogorov‐Smirnov tests. Normally distributed data were assessed using parametric one‐way analysis of variance (ANOVA) or t test. Post hoc tests were performed after ANOVA. If the assumption of normality was not met, non-parametric Kruskal‐Wallis H tests or Mann–Whitney U tests were used to assess between- group differences. In addition, chi-square tests were used to evaluate categorical variables. Post-hoc analysis with the least signiﬁcant difference method was performed to further identify where the difference came from. Moreover, Pearson’s correlation analysis was performed to assess correlations between nonenhanced CT attenuation of the cysts and cysts size, and between postcontrast CT attenuation change of thymic cysts and nonenhanced/postcontrast attenuation values of adjacent large vessel. Furthermore, the factors with p < 0.05 were included in the multivariable binary logistic regression analysis to determine the independent factors to differentiate artifactual hyper-density from unchanged- and artifactual hypodensity density. The accuracy odds ratio (OR) and 95% conﬁdence interval (CI) were recorded. All tests were two-tailed and results were considered signiﬁcant at p<0.05. This retrospective study was approved by the ethics commission of the Second Afﬁliated Hospital of Zhejiang University School of Medicine, and our institutional ethics commission waived the requirement for informed consent. We searched our retrospectively the pathology patient database for consecutive patients who underwent preoperative CE-CT between January 2010 and November 2021, which resulted in a total of 84 patients with pathological diagnosis of thymic cysts and preoperative CE-CT. Chest CE-CT was performed on inspiration with the standard chest CT protocol. In general, patients were scanned in the supine position from the cranial to caudal direction from the clavicles to the adrenal gland, using multidetector scanners with 16 or 64 detector rows, with 120 kVp and 120–250 mAs (automatic dose modulation). For CE-CT imaging, about 1.2 mL/kg of iodinated contrast material (Iohexol, GE Healthcare, USA; Ultravist, Bayer, Germany; Iopamidol, Bracco Diagnostics, Italy) was injected intravenously at a rate of 3.0 mL/s. A total of 56 patients were performed arterial and venous phase scans, and 28 patients were performed only arterial phase scans. Arterial phase was obtained after 30 seconds of delay after intravenous injection of contrast material, and venous phase was performed Frontiers in Oncology frontiersin.org 02 Wang et al. Methods 10.3389/fonc.2022.984770 FIGURE 1 (A-D) Images of a thymic cyst with increased postcontrast attenuation values on CE-CT in 62-year-old woman (age in 2018). Imaging shows (A) a thymic cyst measuring 11.6×8.8 mm, (B) 12.9HU with aortic arch measuring 51.3HU on nonenhanced CT, (C) 28.2HU with aortic arch measuring 337.7HU on arterial phase, and (D) 24.1HU with aortic arch measuring 197.0HU on venous phase of CE-CT. (E-H) Images of a thymic cyst with stable postcontrast attenuation values on CE-CT in 65-year-old man (age in 2017). Imaging shows (E) a thymic cyst measuring 78.3×34 mm, (F) 9.0HU with main pulmonary artery measuring 40.5HU on nonenhanced CT, (G) 9.0HU with main pulmonary artery measuring 233.8HU on arterial phase, and (H) 10.0HU with main pulmonary artery measuring 105.6HU on venous phase of CE-CT. (I-L) Images of a thymic cyst with decreased postcontrast attenuation values on CE-CT in 53-year-old woman (age in 2018). Imaging shows (I) a thymic cyst measuring 20.1×17.0 mm, (J) 45.4HU with brachiocephalic trunk measuring 54.1HU on nonenhanced CT, (K) 29.0HU with brachiocephalic trunk measuring 288.5HU on arterial phase, and (L) 38.8HU with brachiocephalic trunk measuring 148.6HU on venous phase of CE-CT. FIGURE 1 (A-D) Images of a thymic cyst with increased postcontrast attenuation values on CE-CT in 62-year-old woman (age in 2018). Imaging shows (A) a thymic cyst measuring 11.6×8.8 mm, (B) 12.9HU with aortic arch measuring 51.3HU on nonenhanced CT, (C) 28.2HU with aortic arch measuring 337.7HU on arterial phase, and (D) 24.1HU with aortic arch measuring 197.0HU on venous phase of CE-CT. (E-H) Images of a thymic cyst with stable postcontrast attenuation values on CE-CT in 65-year-old man (age in 2017). Imaging shows (E) a thymic cyst measuring 78.3×34 mm, (F) 9.0HU with main pulmonary artery measuring 40.5HU on nonenhanced CT, (G) 9.0HU with main pulmonary artery measuring 233.8HU on arterial phase, and (H) 10.0HU with main pulmonary artery measuring 105.6HU on venous phase of CE-CT. (I-L) Images of a thymic cyst with decreased postcontrast attenuation values on CE-CT in 53-year-old woman (age in 2018). Imaging shows (I) a thymic cyst measuring 20.1×17.0 mm, (J) 45.4HU with brachiocephalic trunk measuring 54.1HU on nonenhanced CT, (K) 29.0HU with brachiocephalic trunk measuring 288.5HU on arterial phase, and (L) 38.8HU with brachiocephalic trunk measuring 148.6HU on venous phase of CE-CT. Frontiers in Oncology frontiersin.org Results of Measurement Value Cyst size (mm) Long diameter 84 31.8 ± 23.4 (24.2) [5.9 to 135.2]* Short diameter 84 20.3 ± 13.2 (15.9) [5.3 to 72.4]* CT attenuation of cysts (HU) Nonenhanced CT 84 33.0 ± 17.5 (36.2) [0.6 to 69.2]* Contrast-enhanced CT in arterial phase 84 34.1 ± 19.9 (36.1) [-17.8 to 74.1]* Contrast-enhanced CT in venous phase 56 37.0 ± 22.9 (38.9) [-2.3 to 97]* Contrast-enhanced attenuation change in arterial phase 84 1.1 ± 11.3 (0.3) [-27.4 to 39.2]* Contrast-enhanced attenuation change in venous phase 56 4.2 ± 10.8 (1.8) [-14.0 to 44.3]* Contrast-enhanced attenuation change in arterial and venous phase† 140 2.3 ± 11.2 (0.5) [-27.4 to 44.3]* CT attenuation of adjacent large vessel (HU) Nonenhanced CT 84 46.8 ± 6.2 (46.9) [19.0 to 57.9]* Contrast-enhanced CT in arterial phase 84 286.2 ± 58.8 (289.9) [140.5 to 492.3]* Contrast-enhanced CT in venous phase 56 150.6 ± 24.7 (150.8) [94.3 to 210.3]* Numbers are means ± standard deviations, with medians in parentheses and ranges in brackets. †The absolute difference between the mean contrast-enhanced and nonenhanced attenuation values of cysts in both arterial and venous phase. Total No. of Measurement Numbers are means ± standard deviations, with medians in parentheses and ranges in brackets. †The absolute difference between the mean contrast-enhanced and nonenhanced attenuation values of cysts in both arterial and venous phase. the group of unchanged density (p<0.001) and the group of artifactual hypo-density (p=0.001) (Figure 2B); the group of artifactual hyper-density had higher postcontrast attenuation values of adjacent large vessel than the group of unchanged density (p<0.001) (Figure 2C); and the group of artifactual hyper-density had higher postcontrast attenuation values of attenuation values of adjacent large vessel (p<0.001), postcontrast attenuation values of adjacent large vessel (p<0.001) and postcontrast attenuation values of cysts (p<0.001) among the three groups. Then, post hoc tests showed the group of artifactual hyper-density had lower nonenhanced attenuation values of adjacent large vessel than TABLE 2 Comparison of clinical characteristics and CT ﬁndings between the groups with different absolute difference between the mean contrast-enhanced and nonenhanced attenuation values of thymic cysts. Parameter Group of Artifactual Hyper-density Group of Unchanged density Group of Artifactual Hypo-density P Value Total No. Results p<0.001) and short diameter (r=-0.396, p<0.001). The change in attenuation of the cysts between contrast-enhanced and nonenhanced CT ranged from -27.4 to 44.3 HU (mean, 2.3 ± 11.2 HU). The study group consisted of 84 patients (33 men, 51 women; mean age, 54 ± 11 years; range, 33–82 years). The diagnosis-treatment interval time ranged from 1 to 17 days (mean, 5.2 ± 3.2 days). Nonenhanced CT attenuation of the cysts ranged from 0.6 to 69.2 HU (mean, 33.0 ± 17.5 HU) (Table 1). In 62 out of 84 patients (73.8%), the nonenhanced CT attenuation was >20 HU, the threshold usually used to differentiate ﬂuid from soft tissue (Figure S1). The thymic cysts ranged from 5.9 to 135.2 mm in long diameter (mean, 31.8 ± 23.4 mm), and from 5.3 to 72.4 mm in short diameter (mean, 20.3 ± 13.2 mm). Nonenhanced CT attenuation of cysts was negatively correlated with the long diameter (r=-0.400, In addition, according to the change in attenuation of the cysts between contrast-enhanced and nonenhanced CT, the patients were classiﬁed into three groups, including the groups of > 5 HU (artifactual hyper-density), -5~5 HU (unchanged density), and <-5 HU (artifactual hypo-density) (Table 2). There was signiﬁcant difference of short diameter among these three groups (p=0.025). Then, post hoc tests showed the group of artifactual hyper-density had smaller short diameter than the group of unchanged density (p=0.008) (Figure 2A). Furthermore, there were signiﬁcant differences of nonenhanced Frontiers in Oncology 03 frontiersin.org frontiersin.org Wang et al. 10.3389/fonc.2022.984770 TABLE 1 Cyst size, attenuation of thymic cysts and attenuation of the adjacent large vessel. Parameter Total No. According to the absolute difference in attenuation of the cysts between contrast-enhanced and nonenhanced CT, the patients were classiﬁed into the three groups of artifactual hyper- density (> 5 HU), unchanged density (-5~5 HU), and artifactual hypo-density (<-5 HU). Results of patients (n) 27 38 19 Age (years) 55.9 ± 9.1 51.0 ± 10.6 59.0 ± 12.5 0.023 Sex 0.025 Men 5 18 10 Women 22 20 9 Diagnosis-treatment interval time 5.2 ± 2.5 5.3 ± 3.7 4.9 ± 3.0 0.833 Cyst Size Total No. of cysts (n) 27 38 19 Long diameter (mm) 23.7 ± 12.6 36.4 ± 24.7 34.3 ± 29.7 0.086 Short diameter (mm) 15.5 ± 8.2 24.3 ± 15.4 19.2 ± 11.9 0.025 Nonenhanced CT attenuation Total No. of plain scans (n) 27 38 19 CT attenuation of cysts (HU) 30.7 ± 16.9 32.1 ± 17.6 38.2 ± 17.9 0.322 CT attenuation of adjacent large vessel (HU) 42.8 ± 7.0 48.7 ± 5.1 48.5 ± 3.9 <0.001 Contrast-enhanced CT attenuation of both arterial and venous phase Total No. of arterial and venous phase scans (n) 47 68 25 CT attenuation of cysts (HU) 46.2 ± 21.7 30.9 ± 18.0 26.6 ± 20.4 <0.001 CT attenuation of adjacent large vessel (HU) 280.3 ± 68.8 222.5 ± 78.88 249.6 ± 74.2 <0.001 According to the absolute difference in attenuation of the cysts between contrast-enhanced and nonenhanced CT, the patients were classiﬁed into the three groups of artifactual hyper- density (> 5 HU), unchanged density (-5~5 HU), and artifactual hypo-density (<-5 HU). he groups with different absolute difference between the mean Parameter According to the absolute difference in attenuation of the cysts between contrast-enhanced and nonenhanced CT, the patients were classiﬁed into the three groups of artifactual hyper- density (> 5 HU), unchanged density (-5~5 HU), and artifactual hypo-density (<-5 HU). 04 Frontiers in Oncology frontiersin.org Wang et al. 10.3389/fonc.2022.984770 B C D A FIGURE 2 The group difference between the groups of artifactual hyper-density, unchanged density, and artifactual hypo-density according to the CT attenuation change of thymic cysts between contrast-enhanced and nonenhanced CT. (A) The group of > 5 HU had smaller short diameter (p=0.008), (B) lower nonenhanced attenuation values of adjacent large vessel than the group of -5~5 HU (p < 0.001) and the group of <-5 HU (p=0.001), (C) higher postcontrast attenuation values of adjacent large vessel than the group of -5~5 HU (p < 0.001), and (D) higher postcontrast attenuation values of cysts than the group of -5~5 HU (p < 0.001) and the group of <-5 HU (p < 0.001). * indicates signiﬁcant differences. B C D A D FIGURE 2 The group difference between the groups of artifactual hyper-density, unchanged density, and artifactual hypo-density according to the CT attenuation change of thymic cysts between contrast-enhanced and nonenhanced CT. (A) The group of > 5 HU had smaller short diameter (p=0.008), (B) lower nonenhanced attenuation values of adjacent large vessel than the group of -5~5 HU (p < 0.001) and the group of <-5 HU (p=0.001), (C) higher postcontrast attenuation values of adjacent large vessel than the group of -5~5 HU (p < 0.001), and (D) higher postcontrast attenuation values of cysts than the group of -5~5 HU (p < 0.001) and the group of <-5 HU (p < 0.001). * indicates signiﬁcant differences. showed female predominant with 51 females and 33 males, which is similar to the previously reported gender predominant (4). cysts than the group of unchanged density (p<0.001) and the group of artifactual hypo-density (p<0.001) (Figure 2D). In addition, the multivariable logistic analysis showed that nonenhanced attenuation of adjacent large vessel (OR: 0.731, 95% CI: 0.611, 0.875), postcontrast attenuation of adjacent large vessel (OR: 1.015, 95% CI: 1.003, 1.027) and postcontrast attenuation of cysts (OR: 1.048, 95% CI: 1.006, 1.092) were identiﬁed as independent predictors of artifactual hyper-density (Table S1). Parameter Correlation analysis showed postcontrast attenuation change of thymic cysts was negatively correlated with nonenhanced attenuation values of adjacent large vessel (r=-0.299, p=0.006), and postcontrast attenuation change of thymic cysts was positively correlated with postcontrast attenuation values of cysts (r=0.488, p<0.001). In this study cohort, 73.8% of patients showed nonenhanced CT attenuation was >20 HU, with a mean attenuation of 33 HU range from 0.6 to 69.2 HU. On nonenhanced CT, CT attenuation is much higher than “water/ﬂuid attenuation” which is often used to describe the characteristic appearance of cysts (11). In this study, we found CT attenuation of most thymic cysts were comparable to that of soft tissue lesions rather that water/ﬂuid, which is consistent with the ﬁnding of another previous study (4). Among thymic lesions, thymic epithelial tumors and other thymic malignancy is typically considered CT attenuation of soft tissue lesions (>20 HU). As a result, a thymic lesion with a high CT attenuation may be suspicious for solid tumors than cysts by an inexperienced radiologist or thoracic surgeon, then resulting in unnecessary thymectomy. Frontiers in Oncology Discussion Both hyper- and hypodense artifacts are common CT artifacts after contrast media injection in chest imaging. In this study, we found that both hyper- and hypodense artifacts affected the attenuation values of the thymic cysts. This study showed that artifactual hyper-density ─the pseudo- enhancement of thymic cysts was associated with smaller short diameter of the cysts, higher postcontrast attenuation values, and lower nonenhanced attenuation values of the adjacent large vessel. A host of technical factors such as partial volume effect and beam hardening effect due to marked enhancement of adjacent tissue can cause the attenuation change of lesions and result in pseudo-enhancement on CE-CT (5, 12). Maki et al. found renal cyst attenuation increased consistently with increasing background attenuation (12). Similarly, our results ﬁrstly conﬁrmed that thymic cysts attenuation increased consistently with increasing attenuation of the adjacent large vessel. Furthermore, our results showed postcontrast attenuation change of thymic cysts was negatively correlated with nonenhanced attenuation values of adjacent large vessel. This study provided the clinical, radiological features of surgically resected and histopathologically conﬁrmed thymic cysts. Moreover, this study investigated the inﬂuencing factors of artifactually affecting attenuation values of thymic cysts on CE-CT. This study showed most cases (73.8%) with >20 HU nonenhanced CT attenuation, the threshold usually used to differentiate ﬂuid from soft tissue. Nonenhanced CT attenuation of cysts was negatively correlated with cysts size. In addition, the results showed that artifactual hyper-density of thymic cysts was negatively associated with nonenhanced attenuation of adjacent large vessel, and positively associated with postcontrast attenuation of adjacent large vessel and postcontrast attenuation of cysts. In this study, patients showed a mean age of 54 years with a range of 33 to 82 years, which is similar to the previously reported age range of patients with thymic tumors (7–10). Therefore, age is not a reliable diagnostic factor for differentiating thymic cysts from thymic tumors. This cohort 05 frontiersin.org Wang et al. 10.3389/fonc.2022.984770 10.3389/fonc.2022.984770 Hospital of Zhejiang University School of Medicine. The ethics committee waived the requirement of written informed consent for participation. Hospital of Zhejiang University School of Medicine. The ethics committee waived the requirement of written informed consent for participation. Additionally, Bae et al. found pseudo-enhancement phenomenon was more apparent in small renal cysts than in large renal cysts (5). Funding This research was supported by Zhejiang Provincial Natural Science Foundation of China under Grant No. LY21H180003. Conclusion We would like to thank all colleagues for helping us during the current study and all the volunteers who participated in the study. This study showed most cases (73.8%) with >20 HU nonenhanced CT attenuation in surgically resected cases, the threshold usually used to differentiate ﬂuid from soft tissue. Nonenhanced CT attenuation of cysts was negatively correlated with cysts size. In addition, artifactual hyper-density─pseudo- enhancement phenomenon of thymic cysts was more apparent in higher increasing attenuation of adjacent large vessels on CE- CT. Pseudo-enhancement phenomenon of thymic cysts is still not well understood. A well understanding of the pseudo- enhancement phenomenon can help reduce preoperative misdiagnosis and reduce unnecessary thymectomy. If pseudo- enhancement phenomenon of thymic cysts is suspected, further MRI or regular CT follow-up should be recommended. Publisher’s note All claims expressed in this article are solely those of the authors and do not necessarily represent those of their afﬁliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Data availability statement The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding authors. Conﬂict of interest The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Discussion Although the multivariable logistic analysis showed small short diameter was not an independent predictor of artifactual hyper-density, ANOVA analysis between groups found the group of artifactual hyper-density had smaller short diameter than the group of unchanged density. The long axis of thymic cysts is usually appressed against adjacent mediastinal large vessels. So, the smaller the diameter of thymic cysts, the closer it is to the adjacent large vessels. Thus, pseudo- enhancement phenomenon of thymic cysts may be more apparent in small short diameter. Author contributions CW and LH conceived and designed this paper. CW wrote the draft. JM and SW performed the study and analyzed the data. SY contributed to the literature research. SY and HX revised the draft. All the authors read and approved the ﬁnal manuscript. As a study limitation, as this study focused on the population of thymic cysts that were histopathologically conﬁrmed after thymectomy, the study population is certainly skewed and represented one end of the spectrum of thymic cysts where the imaging diagnosis suspected to be solid tumors rather than cysts. However, this study was designed to focus on the pathologically conﬁrmed cases, as there has been no specialized study focusing on the CT artifacts ﬁndings of histopathologically conﬁrmed thymic cysts. Ethics statement The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/ fonc.2022.984770/full#supplementary-material The studies involving human participants were reviewed and approved by The ethics committee of the Second Afﬁliated frontiersin.org Frontiers in Oncology 06 frontiersin.org 10.3389/fonc.2022.984770 10.3389/fonc.2022.984770 Wang et al. References 7. Jeong YJ, Lee KS, Kim J, Shim YM, Han J, Kwon OJ. Does CT of thymic epithelial tumors enable us to differentiate histologic subtypes and predict prognosis? AJR Am J Roentgenol (2004) 183(2):283–9. doi: 10.2214/ajr.183.2.1830283 1. Tomiyama N, Honda O, Tsubamoto M, Inoue A, Sumikawa H, Kuriyama K, et al. Anterior mediastinal tumors: diagnostic accuracy of CT and MRI. Eur J Radiol (2009) 69(2):280–8. doi: 10.1016/j.ejrad.2007.10.002 1. Tomiyama N, Honda O, Tsubamoto M, Inoue A, Sumikawa H, Kuriyama K, et al. Anterior mediastinal tumors: diagnostic accuracy of CT and MRI. Eur J Radiol (2009) 69(2):280–8. doi: 10.1016/j.ejrad.2007.10.002 2. Ackman JB, Chintanapakdee W, Mendoza DP, Price MC, Lanuti M, Shepard JO. Longitudinal CT and MRI characteristics of unilocular thymic cysts. Radiology (2021) 301(2):443–54. doi: 10.1148/radiol.2021203593 8. McErlean A, Huang J, Zabor EC, Moskowitz CS, Ginsberg MS. Distinguishing benign thymic lesions from early-stage thymic malignancies on computed tomography. J Thorac Oncol (2013) 8(7):967–73. doi: 10.1097/ JTO.0b013e3182904bc2 8. McErlean A, Huang J, Zabor EC, Moskowitz CS, Ginsberg MS. Distinguishing benign thymic lesions from early-stage thymic malignancies on computed tomography. J Thorac Oncol (2013) 8(7):967–73. doi: 10.1097/ JTO.0b013e3182904bc2 3. Ackman JB, Verzosa S, Kovach AE, Louissaint A Jr., Lanuti M, Wright CD, et al. High rate of unnecessary thymectomy and its cause. can computed tomography distinguish thymoma, lymphoma, thymic hyperplasia, and thymic cysts? Eur J Radiol (2015) 84(3):524–33. doi: 10.1016/j.ejrad.2014.11.042 9. Engels EA, Pfeiffer RM. Malignant thymoma in the United States: demographic patterns in incidence and associations with subsequent malignancies. Int J Cancer (2003) 105(4):546–51. doi: 10.1002/ijc.11099 9. Engels EA, Pfeiffer RM. Malignant thymoma in the United States: demographic patterns in incidence and associations with subsequent malignancies. Int J Cancer (2003) 105(4):546–51. doi: 10.1002/ijc.11099 4. Araki T, Sholl LM, Gerbaudo VH, Hatabu H, Nishino M. Intrathymic cyst: Clinical and radiological features in surgically resected cases. Clin Radiol (2014) 69 (7):732–8. doi: 10.1016/j.crad.2014.03.002 10. Sadohara J, Fujimoto K, Müller NL, Kato S, Takamori S, Ohkuma K, et al. Thymic epithelial tumors: comparison of CT and MR imaging ﬁndings of low-risk thymomas, high-risk thymomas, and thymic carcinomas. Eur J Radiol (2006) 60 (1):70–9. doi: 10.1016/j.ejrad.2006.05.003 5. Bae KT, Heiken JP, Siegel CL, Bennett HF. Renal cysts: Is attenuation artifactually increased on contrast-enhanced CT images? Radiology (2000) 216 (3):792–6. doi: 10.1148/radiology.216.3.r00se14792 11. Jeung MY, Gasser B, Gangi A, Bogorin A, Charneau D, Wihlm JM, et al. Imaging of cystic masses of the mediastinum. Radiographics (2002) 22:S79–93. 5. Bae KT, Heiken JP, Siegel CL, Bennett HF. Renal cysts: Is attenuation artifactually increased on contrast-enhanced CT images? Radiology (2000) 216 (3):792–6. doi: 10.1148/radiology.216.3.r00se14792 6. Mileto A, Nelson RC, Samei E, Jaffe TA, Paulson EK, Barina A, et al. Impact of dual-energy multi-detector row CT with virtual monochromatic imaging on renal cyst pseudoenhancement: In vitro and in vivo study. Radiology (2014) 272 (3):767–76. doi: 10.1148/radiol.14132856 12. Maki DD, Birnbaum BA, Chakraborty DP, Jacobs JE, Carvalho BM, Herman GT. Renal cyst pseudoenhancement: Beam-hardening effects on CT numbers. Radiology (1999) 213(2):468–72. doi: 10.1148/radiology.213.2.r99nv33468 9. Engels EA, Pfeiffer RM. Malignant thymoma in the United States: demographic patterns in incidence and associations with subsequent malignancies. Int J Cancer (2003) 105(4):546–51. doi: 10.1002/ijc.11099 10. Sadohara J, Fujimoto K, Müller NL, Kato S, Takamori S, Ohkuma K, et al. Thymic epithelial tumors: comparison of CT and MR imaging ﬁndings of low-risk thymomas, high-risk thymomas, and thymic carcinomas. Eur J Radiol (2006) 60 (1):70–9. doi: 10.1016/j.ejrad.2006.05.003 References doi: 10.1148/radiographics.22.suppl_1.g02oc09s79 6. Mileto A, Nelson RC, Samei E, Jaffe TA, Paulson EK, Barina A, et al. Impact of dual-energy multi-detector row CT with virtual monochromatic imaging on renal cyst pseudoenhancement: In vitro and in vivo study. Radiology (2014) 272 (3):767–76. doi: 10.1148/radiol.14132856 12. Maki DD, Birnbaum BA, Chakraborty DP, Jacobs JE, Carvalho BM, Herman GT. Renal cyst pseudoenhancement: Beam-hardening effects on CT numbers. Radiology (1999) 213(2):468–72. doi: 10.1148/radiology.213.2.r99nv33468 Frontiers in Oncology frontiersin.org 07 Frontiers in Oncology
https://openalex.org/W4280535879	https://bmcendocrdisord.biomedcentral.com/track/pdf/10.1186/s12902-022-01043-1	English	null	Associations between serum vitamin D3, atherogenic indices of plasma and cardiometabolic biomarkers among patients with diabetes in the KERCADR study	BMC endocrine disorders	2,022	cc-by	10,990	Mahmoodi and Najafipour BMC Endocrine Disorders (2022) 22:126 https://doi.org/10.1186/s12902-022-01043-1 Mahmoodi and Najafipour BMC Endocrine Disorders (2022) 22:126 https://doi.org/10.1186/s12902-022-01043-1 Abstract Background: We sought the association between serum 25-hydroxyvitamin D3 (25(OH) D3) levels and atherogenic indices of plasma as novel predictive biomarkers of cardiometabolic disease risk in patients with type 2 diabetes mel- litus (T2DM). Methods: The present study was a nested case-control study conducted on 252 participants with T2DM and controls from the second phase of the KERCADR cohort study. The participants with a mean (±SD) age of 49.79 ± 5.85 years were randomly selected and allocated into case and control groups. Independent t-test, Hierarchical Linear Regres- sion, Univariate ANOVA, and partial correlation were used for analysis the data. Atherogenic indices of plasma include Castelli Risk Index I (CRI I), Castelli Risk Index II (CRI II), and the novel Atherogenic Index of Plasma (AIP), and Athero- genic Coefficient (AC). Results: There was a significant difference among case and control groups for AIP in males and females (P < 0.001 and P = 0.007, respectively). The levels of AIP, CRI I, and AC significantly decreased (P = 0.017, P = 0.029, and P = 0.029, respectively) with improved serum vitamin D status only in control male participants. The main effect of BMI and vita- min D status on AIP, CRI I, and AC, and the main effect of BMI on CRI I, CRI II, and AC were significant in control males and females, respectively. Conclusion: We conclude that there is a reverse significant association between AIP and serum vitamin D among healthy males. Low serum level of vitamin D is associated with atherogenic dyslipidemia. Therefore, improving vitamin D status as an important indicator may alleviate AIP as a surrogate marker for predicting the risk of CVD events in healthy men and women with normal BMI. Keywords: 25-hydroxyvitamin D3, Atherogenic indices of plasma, Biomarkers of cardiometabolic ydroxyvitamin D3, Atherogenic indices of plasma, Biomarkers of cardiometabolic risk, Type 2 diabetes Associations between serum vitamin D3, atherogenic indices of plasma and cardiometabolic biomarkers among patients with diabetes in the KERCADR study Mohammad Reza Mahmoodi1,2* and Hamid Najafipour1,3 © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativeco mmons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Introduction Low serum 25-hydroxyvitamin D3 [25(OH) D3] levels are significantly associated with a higher risk of devel- oping prediabetes and type 2 diabetes mellitus (T2DM) in individuals [1, 2]. Epidemiologic studies reveal an association between low serum 25(OH) D3 level and an Correspondence: mahmoodimr@yahoo.com; mr_mahmoodi@kmu.ac.ir 2 Department of Nutrition, Faculty of Public Health, Kerman University of Medical Sciences, Kerman, Iran. Haft Bagh‑E‑Alavi Highway, Kerman 7616913555, Iran , Full list of author information is available at the end of the article The serum vitamin D levels were inversely associated with HbA1c, FPG, TG, TC, and non-HDL-C [7].hi increased risk of metabolic syndrome (MS) and T2DM [3]. In addition, observational studies have shown that vitamin D adequacy can reduce the severity of T2DM, insulin resistance, prediabetes, and MS. However, there is a lack of convincing evidence from randomized con- trol clinical trials that these complications are prevented following optimization of serum levels of 25(OH) D3 [4]. Although, the prevalence of hypovitaminosis D do not differ significantly in healthy adults; the mean of 25(OH) D3 decrease with an increasing number of cardiometa- bolic risk factors such as central obesity, hypertension, increased atherogenic risk, and insulin resistance [5]. The prevalence of low serum 25(OH) D3 levels is considerably high in patients with cardiovascular disease (CVD) risk factors. These patients present significantly higher values for cardiometabolic biomarkers such as fasting plasma glucose (FPG), hemoglobin A1c (HbA1c), total choles- terol (TC), triglyceride (TG), body mass index (BMI), waist circumference (WC), and atherogenic indices (Cas- telli Risk Index I (CRI I), Castelli Risk Index II (CRI II), and Atherogenic index of plasma (AIP)) [6]. Serum vita- min D correlated negatively with glycemia, HbA1C, TG, atherogenic indices, BMI, and hypertension [6]. There is a significant negative correlation of serum vitamin D with lipid markers and atherogenic variables in poor glycemic control diabetic patients. The serum vitamin D levels were inversely associated with HbA1c, FPG, TG, TC, and non-HDL-C [7].hi visceral adiposity index in both males and females [15]. In another study, the serum 25(OH) D levels are closely associated with the serum lipids and AIP. Vitamin D deficiency is associated with an increased risk of dyslipi- demias, especially in men. Accordingly, the association between vitamin D status and AIP varies by gender [16]. 25(OH) D and AIP are significantly different between control and T2DM groups. Serum 25(OH) D showed a significant negative correlation with AIP among total study subjects. The association between 25(OH) D and various CVD risk markers suggests that 25(OH) D might help in the prediction of CVD risk [17]. A progressive decrease in TC, LDL-C, and non-HDL-C is revealed as the serum vitamin D level increased. There is a negative linear association between 25(OH) D and TC, LDL-C, and non-HDL-C in obese patients [18]. Materials and methods There is a positive and significant correlation between AIP and cardiometabolic risk factors such as BMI, TG, WC, TC, low-density lipoprotein cholesterol (LDL-C), HbA1c, FPG, systolic blood pressure (SBP), diastolic blood pressure (DBP) in many studies [8–10]. Therefore, AIP can be used as a surrogate marker both for predict- ing the risk of CVD risk factors and, additionally, it has been shown that AIP is associated with subclinical ath- erosclerosis and CVD events in both healthy women and CVD postmenopausal women [8–11].ii Considering the limited studies regarding the associa- tion between serum 25(OH) D3 levels and atherogenic indices of plasma in patients with DM, this study aims to assess vitamin D status in participants and to find out whether there is an association between serum 25(OH) D3 levels and atherogenic indices of plasma as novel sur- rogate markers as well as biomarkers of cardiometabolic disease risk in patients with DM and a healthy popula- tion based on genders in KERCADR study as an Iranian community. © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativeco mmons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Mahmoodi and Najafipour BMC Endocrine Disorders (2022) 22:126 Page 2 of 12 increased risk of metabolic syndrome (MS) and T2DM [3]. In addition, observational studies have shown that vitamin D adequacy can reduce the severity of T2DM, insulin resistance, prediabetes, and MS. However, there is a lack of convincing evidence from randomized con- trol clinical trials that these complications are prevented following optimization of serum levels of 25(OH) D3 [4]. Although, the prevalence of hypovitaminosis D do not differ significantly in healthy adults; the mean of 25(OH) D3 decrease with an increasing number of cardiometa- bolic risk factors such as central obesity, hypertension, increased atherogenic risk, and insulin resistance [5]. The prevalence of low serum 25(OH) D3 levels is considerably high in patients with cardiovascular disease (CVD) risk factors. These patients present significantly higher values for cardiometabolic biomarkers such as fasting plasma glucose (FPG), hemoglobin A1c (HbA1c), total choles- terol (TC), triglyceride (TG), body mass index (BMI), waist circumference (WC), and atherogenic indices (Cas- telli Risk Index I (CRI I), Castelli Risk Index II (CRI II), and Atherogenic index of plasma (AIP)) [6]. Serum vita- min D correlated negatively with glycemia, HbA1C, TG, atherogenic indices, BMI, and hypertension [6]. There is a significant negative correlation of serum vitamin D with lipid markers and atherogenic variables in poor glycemic control diabetic patients. Anthropometry assessment Weight and BMI (the weight in kilograms divided by the square of the height in meters) were measured and recorded in questionnaires. WC was measured at the umbilical level using a non-stretchable measuring tape, without any pressure to the body surface. The protocol was approved by review panels and eth- ics committees (Approval ID: IR.KMU.REC. 1399.405) of the Vice-chancellor for Research of Kerman University of Medical Sciences. Atherogenic indices of plasma The criteria for eligibility of participants with diabetes mellitus were 1) willingness to participate in the study and sign the informed consent, 2) the presence of type 2 diabe- tes at least for one year 3) patients receive either diet ther- apy or diet therapy with a combination of oral anti-diabetic medications, 4) no history of myocardial infarction, stroke, cardiovascular disease, active cancer, liver, kidney, and thy- roid dysfunction, and infectious diseases, 5) no history of high blood pressure, 6) BMI lower than 30 and from both genders. The criteria for eligibility of control participants were 1) willingness to participate in the study and sign the informed consent, 2) no history of diabetes mellitus, 3) no history of myocardial infarction, stroke, cardiovascular disease, active cancer, liver, kidney, and thyroid dysfunc- tion, and infectious diseases, 4) no history of high blood pressure, 5) BMI lower than 30 and from both of genders that matched with case participants.h Atherogenic indices of plasma include Castelli Risk Index I, Castelli Risk Index II, and the novel Atherogenic Index of Plasma, and Atherogenic Coefficient. AIP or TG/high- density lipoprotein cholesterol (TG/HDL-C) ratio is a logarithmic transformation of the ratio of molar con- centrations of TG to HDL-C. CRI I is the ratio of TC to HDL-C, and CRI II is the ratio of LDL-C to HDL-C. AC is the ratio of non-HDL-C to HDL-C. These parameters are being applied for assessing cardiovascular risk. Determination of 25‑hydroxyvitamin D3 Blood samples were drawn into tubes after a 12-14 h fast and immediately stored at − 80 °C until an assay for 25-hydroxyvitamin D3 could be performed. Serum 25-hydroxyvitamin D3 was measured by an enzyme- linked immune sorbent assay (ELISA) (Monobind 25-OH Vit D [Direct]) with the use of an automated analyzer with a sensitivity of ng/ml through the protocol of the ELISA kit. The existence of any of the exclusion criteria among participants may profoundly affect plasma atherogenic indices and the other cardiovascular biomarkers. There- fore, compliance with all of these criteria would result in greater transparency of the association between vitamin D levels and plasma atherogenic indices. Participants eligibility and study designh The present study is a nested case-control study con- ducted on participants with type 2 diabetes and controls from the second phase of the KERCADR cohort study. For each case, a healthy matched control was selected from among participants in the KERCADR cohort study. The second phase of KERCADR is a cohort study on over 10,000 individuals aged 15-75 years old who were recruited in the household survey on Kerman province residences. The baseline protocol, the sampling method, and the recruitment have been previously described in detail [19,20]. Kerman province is one of the 31 provinces of Iran. Kerman is in the southeast of Iran with its admin- istrative center in the city of Kerman. Vitamin D deficiency as a modifiable risk factor is asso- ciated with a worse cardiometabolic risk profile. A posi- tive and significant association between AIP and higher HbA1c, CRI I, and lower HDL-C are seen in people with plasma 25(OH) D3 less than 25 nmol/L [12]. Serum lev- els of HDL-C, 25(OH) D3, free vitamin D and bioavail- able vitamin D are significantly lower in diabetic patients than in non-diabetic patients while TG and remnant cho- lesterol levels are found to be significantly higher [13]. In a study, the majority of Korean adults with prediabetes have a serum 25(OH) D3 less than 20 ng/ml, and the pro- portion of adults having low HDL-C is the highest among the vitamin D deficiency group [14]. Partial correlations adjusting for age and sex show that vitamin D concentra- tions are significantly inversely associated with AIP and Two hundred fifty-two participants (136 males and 116 females) were randomly selected and the total number of participants with diabetes was 124 (69 males and 55 females) and controls was 128 (67 males and 61 females) from the KERCADR study. The management of con- founding variables in study design and to ensure that the study groups did not differ concerning effective con- founders as inclusion criteria, the cases to controls ratio became 1:1. Limiting the study to participants in effective confounders was a simple technique of ensuring that all participants have the same level of the confounder. Page 3 of 12 Mahmoodi and Najafipour BMC Endocrine Disorders (2022) 22:126 Mahmoodi and Najafipour BMC Endocrine Disorders (2022) 22:126 Mahmoodi and Najafipour BMC Endocrine Disorders (2022) 22:126 Mahmoodi and Najafipour BMC Endocrine Disorders (2022) 22:126 Atherogenic indices of plasma Statistical analysis l l (cm) 91.4 ± 0.8 88.7 ± 0.9 0.027 93.9 ± 0.9 90.4 ± 1.1 0.015 88.2 ± 1.3 86.8 ± 1.4 0.480 Hip Circum. (cm) 98.8 ± 0.63 99.1 ± 0.7 0.708 99.2 ± 0.8 98.5 ± 0.9 0.559 98.2 ± 1.0 99.8 ± 0.9 0.245 Waist to Hip ratio 0.93 ± 0.01 0.89 ± 0.01 0.000 0.95 ± 0.01 0.92 ± 0.01 0.000 0.90 ± 0.01 0.87 ± 0.01 0.039 Fasting Blood Sugar (mg/dl) 181.3 ± 6.1 90.1 ± 1.0 0.000 177.5 ± 8.5 90.4 ± 1.2 0.000 186.1 ± 8.8 89.7 ± 1.5 0.000 HbA1c 8.4 ± 0.3 – – 8.5 ± 0.3 – – 8.2 ± 0.4 – – Serum triglycer- ide (mg/dl) 184.7 ± 9.9 120.7 ± 5.6 0.000 181.9 ± 13.9 127.8 ± 8.0 0.001 188.3 ± 14.1 112.9 ± 7.8 0.000 Total Cholesterol (mg/dl) 193.1 ± 3.9 183.0 ± 3.0 0.043 191.7 ± 5.1 182.7 ± 4.5 0.191 194.8 ± 6.2 183.3 ± 4.0 0.120 LDL-C (mg/dl) 111.8 ± 3.7 114.4 ± 2.5 0.549 111.8 ± 5.0 114.0 ± 3.4 0.720 111.7 ± 5.4 114.9 ± 3.8 0.619 HDL-C (mg/dl) 45.2 ± 0.9 44.4 ± 0.9 0.562 44.5 ± 1.0 43.1 ± 1.3 0.411 46.1 ± 1.6 45.9 ± 1.3 0.922 Systolic blood pressure (mmHg) 112.7 ± 1.1 110.4 ± 1.1 0.151 113.1 ± 1.5 113.4 ± 1.4 0.971 112.0 ± 1.6 107.1 ± 1.8 0.044 Diastolic blood pressure (mmHg) 74.0 ± 0.8 71.7 ± 0.9 0.056 74.0 ± 1.03 73.8 ± 1.2 0.887 73.9 ± 1.1 69.3 ± 1.4 0.013 Table 1 Mean ± SE of Cardiometabolic Biomarkers of Participants in Case Control Study according to Gender Table 2 Mean ± SE and Interquartile Rangea of Serum 25(OH) Vit. Statistical analysis l l D and Atherogenic Indicesb of Plasma of Participants according to Gender a Interquartile range is in bracket b Atherogenic Index of Plasma = AIP; Castelli Risk Index I=CRI I; Castelli Risk Index II=CRI II; Atherogenic Coefficient = AC c Based on ng/ml Total Participants (n = 252) P value Male (n = 136) P value Female (n = 116) P value Case Control Case (n = 69) Control (n = 67) Case (n = 55) Control (i61) Serum 25(OH) Dc 26.6 ± 1.1 (18.0) 25.8 ± 1.1 (17.75) 0.656 26.4 ± 1.3 (15.0) 26.1 ± 1.2 (13.0) 0.677 26.8 ± 2.1 (23.0) 25.6 ± 1.9 (23.0) 0.860 AIP 0.56 ± 0.03 (0.41) 0.39 ± 0.02 (0.35) 0.000 0.56 ± 0.03 (0.35) 0.43 ± 0.03 (0.37) 0.000 0.57 ± 0.04 (0.49) 0.35 ± 0.03 (0.34) 0.007 CRI I 4.45 ± 0.12 (1.58) 4.26 ± 0.08 (1.33) 0.200 4.44 ± 0.14 (1.75) 4.37 ± 0.12 (1.17) 0.174 4.47 ± 0.21 (1.60) 4.14 ± 0.12 (1.55) 0.706 CRI II 2.50 ± 0.09 (1.09) 2.68 ± 0.07 (1.17) 0.109 2.49 ± 0.10 (1.14) 2.74 ± 0.09 (0.93) 0.557 2.50 ± 0.15 (1.18) 2.61 ± 0.10 (1.27) 0.079 AC 3.45 ± 0.12 (1.58) 3.26 ± 0.08 (1.33) 0.200 3.44 ± 0.14 (1.75) 3.37 ± 0.12 (1.17) 0.174 3.47 ± 1.59 (1.60) 3.14 ± 0.12 (1.55) 0.706 Table 1 Mean ± SE of Cardiometabolic Biomarkers of Participants in Case Control Study according to Gender Total Participants (n = 252) P value Male (n = 136) P value Female (n = 116) P value Case (n = 124) Control (n = 128) Case (n = 69) Control (n = 67) Case (n = 55) Control (n = 61) Age (Year) 51.3 ± 0.5 50.1 ± 0.5 0.086 49.0 ± 0.6 49.8 ± 0.7 0.355 45.7 ± 0.7 46.0 ± 0.7 0.726 Weight (Kg) 70.6 ± 1.0 68.1 ± 0.9 0.061 74.3 ± 1.2 71.4 ± 1.3 0.105 65.9 ± 1.2 64.5 ± 1.2 0.411 Height (m) 164.2 ± 0.8 164.7 ± 0.8 0.689 169.6 ± 0.8 170.9 ± 0.8 0.257 157.4 ± 0.8 157.8 ± 0.7 0.677 Body Mass Index (Kg/m2) 26.1 ± 0.3 25.1 ± 0.3 0.013 25.8 ± 0.3 24.5 ± 0.4 0.014 26.5 ± 0.4 25.9 ± 0.4 0.250 Waist Circum. Statistical analysis l l Statistical analysis was performed using IBM SPSS Sta- tistics software, version 22.0. The normal distribution of biomarkers was investigated by the Kolmogorov- Smirnov test. Significance was assumed at P < 0.05. The mean differences of the cardiometabolic biomarkers, the levels of atherogenic indices of plasma, and serum vita- min D3 between the case and the control participants in each gender group were compared by independent t-test (Tables 1 & 2). The one-way analysis of variance (ANOVA) in which all pairwise comparisons among the three vitamin D statuses (vitamin D deficiency with serum vitamin D ≤ 20 ng/ml, vitamin D insufficiency with serum vitamin D > 20-30 ng/ml, and vitamin D satisfac- tory with serum vitamin D > 30 ng/ml) were performed with the use of Tukey’s HSD honestly significant differ- ence procedure (Table 3). Hierarchical Linear Regression was applied to investigate whether adding atherogenic indices of plasma as independent variables (predictors) significantly improves a model’s ability to predict serum 25(OH) vitamin D3 as a dependent variable and/or to investigate a moderating effect of a variable (Table 4).h Blood samples were drawn into EDTA tubes after a 12-14 h fast at the study baseline. Plasma samples were stored at − 80 °C until a final assay for glycemic and lipo- protein biomarkers could be performed. Cardiometabolic biomarkers include FBS, HbA1c, TC, HDL-C, LDL-C, TG, SBP, DBP, WC, hip circumference (HC), waist to hip ratio (WHR), weight, and BMI. As previously described in the other studies [19, 20], all measurements were performed according to the standard protocol. The patients fasted for 12-14 h before admis- sion. FBS (KIMIA Kit, Code 890410, Iran) was measured using the glucose oxidase-peroxidase method. HDL-C (PARS Kit, Code 89022, Iran) and TG (KIMIA Kit, Code 890201, Iran), were measured by standard enzymatic pro- cedures. BP was recorded using an automated oscillomet- ric BP monitor (standard mercury manometer–Model RIESTER, Germany) after at least 10 min of rest in a chair and arm supported at heart level. TC (KIMIA Kit, Code 890303, Iran) and LDL-C were calculated based on the Friedewald formula [LDL-C = TC – (HDL-C + TG/5)]. HbA1C (NYCOCARD Kit, Code 1042184, Austria) was determined based on Bio-rad Variant High-Performance Liquid Chromatography [HPLC] assay. Statistical analysis l l f Then, general linear models (Univariate ANOVA) were used to find out whether the interaction between Mahmoodi and Najafipour BMC Endocrine Disorders (2022) 22:126 Page 4 of 12 two independent variables such as vitamin D concen- tration and BMI on atherogenic indices of plasma is significant. As previously described, serum vitamin D status divided into deficiency, insufficiency, and satisfac- tory statuses. BMI dichotomized into BMI equal or less than 26 00 and greater than 26 00 in the case males and control females and dichotomized into BMI equal or less than 26.50 and greater than 26.50 in the case females; and dichotomized into BMI equal or less than 25.00 and greater than 25.00 in the control males (Table 5). Because the mean and median of the independent variable BMI (less than or equal and greater than median BMI) have Table 1 Mean ± SE of Cardiometabolic Biomarkers of Participants in Case Control Study according to Gender Total Participants (n = 252) P value Male (n = 136) P value Female (n = 116) P value Case (n = 124) Control (n = 128) Case (n = 69) Control (n = 67) Case (n = 55) Control (n = 61) Age (Year) 51.3 ± 0.5 50.1 ± 0.5 0.086 49.0 ± 0.6 49.8 ± 0.7 0.355 45.7 ± 0.7 46.0 ± 0.7 0.726 Weight (Kg) 70.6 ± 1.0 68.1 ± 0.9 0.061 74.3 ± 1.2 71.4 ± 1.3 0.105 65.9 ± 1.2 64.5 ± 1.2 0.411 Height (m) 164.2 ± 0.8 164.7 ± 0.8 0.689 169.6 ± 0.8 170.9 ± 0.8 0.257 157.4 ± 0.8 157.8 ± 0.7 0.677 Body Mass Index (Kg/m2) 26.1 ± 0.3 25.1 ± 0.3 0.013 25.8 ± 0.3 24.5 ± 0.4 0.014 26.5 ± 0.4 25.9 ± 0.4 0.250 Waist Circum. (cm) 91.4 ± 0.8 88.7 ± 0.9 0.027 93.9 ± 0.9 90.4 ± 1.1 0.015 88.2 ± 1.3 86.8 ± 1.4 0.480 Hip Circum. Statistical analysis l l (cm) 98.8 ± 0.63 99.1 ± 0.7 0.708 99.2 ± 0.8 98.5 ± 0.9 0.559 98.2 ± 1.0 99.8 ± 0.9 0.245 Waist to Hip ratio 0.93 ± 0.01 0.89 ± 0.01 0.000 0.95 ± 0.01 0.92 ± 0.01 0.000 0.90 ± 0.01 0.87 ± 0.01 0.039 Fasting Blood Sugar (mg/dl) 181.3 ± 6.1 90.1 ± 1.0 0.000 177.5 ± 8.5 90.4 ± 1.2 0.000 186.1 ± 8.8 89.7 ± 1.5 0.000 HbA1c 8.4 ± 0.3 – – 8.5 ± 0.3 – – 8.2 ± 0.4 – – Serum triglycer- ide (mg/dl) 184.7 ± 9.9 120.7 ± 5.6 0.000 181.9 ± 13.9 127.8 ± 8.0 0.001 188.3 ± 14.1 112.9 ± 7.8 0.000 Total Cholesterol (mg/dl) 193.1 ± 3.9 183.0 ± 3.0 0.043 191.7 ± 5.1 182.7 ± 4.5 0.191 194.8 ± 6.2 183.3 ± 4.0 0.120 LDL-C (mg/dl) 111.8 ± 3.7 114.4 ± 2.5 0.549 111.8 ± 5.0 114.0 ± 3.4 0.720 111.7 ± 5.4 114.9 ± 3.8 0.619 HDL-C (mg/dl) 45.2 ± 0.9 44.4 ± 0.9 0.562 44.5 ± 1.0 43.1 ± 1.3 0.411 46.1 ± 1.6 45.9 ± 1.3 0.922 Systolic blood pressure (mmHg) 112.7 ± 1.1 110.4 ± 1.1 0.151 113.1 ± 1.5 113.4 ± 1.4 0.971 112.0 ± 1.6 107.1 ± 1.8 0.044 Diastolic blood pressure (mmHg) 74.0 ± 0.8 71.7 ± 0.9 0.056 74.0 ± 1.03 73.8 ± 1.2 0.887 73.9 ± 1.1 69.3 ± 1.4 0.013 Table 2 Mean ± SE and Interquartile Rangea of Serum 25(OH) Vit. Statistical analysis l l D and Atherogenic Indicesb of Plasma of Participants according to Gender a Interquartile range is in bracket b Atherogenic Index of Plasma = AIP; Castelli Risk Index I=CRI I; Castelli Risk Index II=CRI II; Atherogenic Coefficient = AC c Based on ng/ml Total Participants (n = 252) P value Male (n = 136) P value Female (n = 116) P value Case Control Case (n = 69) Control (n = 67) Case (n = 55) Control (i61) Serum 25(OH) Dc 26.6 ± 1.1 (18.0) 25.8 ± 1.1 (17.75) 0.656 26.4 ± 1.3 (15.0) 26.1 ± 1.2 (13.0) 0.677 26.8 ± 2.1 (23.0) 25.6 ± 1.9 (23.0) 0.860 AIP 0.56 ± 0.03 (0.41) 0.39 ± 0.02 (0.35) 0.000 0.56 ± 0.03 (0.35) 0.43 ± 0.03 (0.37) 0.000 0.57 ± 0.04 (0.49) 0.35 ± 0.03 (0.34) 0.007 CRI I 4.45 ± 0.12 (1.58) 4.26 ± 0.08 (1.33) 0.200 4.44 ± 0.14 (1.75) 4.37 ± 0.12 (1.17) 0.174 4.47 ± 0.21 (1.60) 4.14 ± 0.12 (1.55) 0.706 CRI II 2.50 ± 0.09 (1.09) 2.68 ± 0.07 (1.17) 0.109 2.49 ± 0.10 (1.14) 2.74 ± 0.09 (0.93) 0.557 2.50 ± 0.15 (1.18) 2.61 ± 0.10 (1.27) 0.079 AC 3.45 ± 0.12 (1.58) 3.26 ± 0.08 (1.33) 0.200 3.44 ± 0.14 (1.75) 3.37 ± 0.12 (1.17) 0.174 3.47 ± 1.59 (1.60) 3.14 ± 0.12 (1.55) 0.706 q g b Atherogenic Index of Plasma = AIP; Castelli Risk Index I=CRI I; Castelli Risk Index II=CRI II; Atherogenic Coefficient = AC c Based on ng/ml control females and dichotomized into BMI equal or less than 26.50 and greater than 26.50 in the case females; and dichotomized into BMI equal or less than 25.00 and greater than 25.00 in the control males (Table 5). Because the mean and median of the independent variable BMI (less than or equal and greater than median BMI) have two independent variables such as vitamin D concen- tration and BMI on atherogenic indices of plasma is significant. As previously described, serum vitamin D status divided into deficiency, insufficiency, and satisfac- tory statuses. Statistical analysis l l There were no significant differences in serum 25(OH) D3 and the other atherogenic indices of plasma of par- ticipants among the case and control groups according to gender (Table 2). closely corresponded, this variable was dichotomized. The assumptions for ANOVA and Univariate ANOVA principally the assumption of homogeneity of variances have been met. The Pearson correlation coefficient was calculated to examine the relationship between atherogenic indices as well as between serum vitamin D and lipid and lipopro- tein profiles (Tables 6 & 7). Characteristics of participantsh The mean (±SD) age of participants was 49.79 ± 5.85 years (cases, 47.54 ± 5.49; controls, 48.05 ± 5.80). Fifty-four per- cent of the participants were male. Cardiometabolic biomarkers of participantsh The Cardiometabolic biomarkers of participants in case and control groups according to gender are shown in Table 1. There were significant differences among case and control males for BMI, WC, WHR, FBS, and TG (P = 0.014, P = 0.015, P < 0.001, P < 0.001, and P = 0.001, respectively) and among females for WHR, FBS, TG, SBP, and DBP (P = 0.039, P < 0.001, P < 0.001, P = 0.044 and P = 0.013, respectively) (Table 1). Atherogenic indices according to vitamin D status of participants Table 3 indicates the mean (±SE) of atherogenic indices of plasma according to the vitamin D status of partici- pants. Results of one-way ANOVA showed that the levels of AIP, CRI I, and AC significantly decreased (P = 0.017, P = 0.029, and P = 0.029, respectively) with improved serum vitamin D status only in control male participants. The levels of atherogenic indices of plasma non-signifi- cantly decreased with improved serum vitamin D status only in case male participants. Statistical analysis l l BMI dichotomized into BMI equal or less than 26.00 and greater than 26.00 in the case males and Mahmoodi and Najafipour BMC Endocrine Disorders (2022) 22:126 Page 5 of 12 Table 3 Mean ± SEa of Atherogenic Indicesb of Plasma according to Vitamin D Statusc of Participants Atherogenic Indices Vitamin D Status (Male) P value Vitamin D Status (Female) Table 3 Mean ± SEa of Atherogenic Indicesb of Plasma according to Vitamin D Statusc of Participants a One-way ANOVA analyzed the differences (M ± SE) atherogenic indices between three vitamin D statuses for each gender group in case and control groups b Atherogenic Index of Plasma = AIP; Castelli Risk Index I=CRI I; Castelli Risk Index II=CRI II; Atherogenic Coefficient = AC c Vitamin D status divided into vitamin D deficiency (serum vitamin D ≤ 20 ng/ml), vitamin D insufficiency (serum vitamin D > 20-30 ng/ml), and vitamin D satisfactory (serum vitamin D > 30 ng/ml) d There are statistically significant differences between vitamin D deficiency and vitamin D satisfactory Atherogenic Indices Vitamin D Status (Male) P value Vitamin D Status (Female) P value Vitamin D Deficiency Vitamin D Insufficiency Vitamin D Satisfactory Vitamin D Deficiency Vitamin D Insufficiency Vitamin D Satisfactory Case group (n = 19) (n = 29) (n = 21) (n = 23) (i10) (i22) AIP 0.62 ± 0.06 0.52 ± 0.05 0.55 ± 0.06 0.523 0.60 ± 0.06 0.44 ± 0.09 0.59 ± 0.07 0.310 CRI I 4.71 ± 0.20 4.34 ± 0.23 4.37 ± 0.32 0.556 4.31 ± 0.27 4.04 ± 0.26 4.84 ± 0.43 0.351 CRI II 2.72 ± 0.16 2.41 ± 0.16 2.43 ± 0.22 0.447 2.28 ± 0.17 2.38 ± 0.16 2.82 ± 0.34 0.275 AC 3.71 ± 0.20 3.34 ± 0.23 3.37 ± 0.32 0.556 3.31 ± 0.27 3.04 ± 0.26 3.84 ± 0.43 0.351 Control group (i21) (n = 23) (n = 23) (n = 27) (n = 13) (n = 21) AIP 0.55 ± 0.06d 0.39 ± 0.05 0.36 ± 0.04 0.017 0.35 ± 0.06 0.34 ± 0.06 0.36 ± 0.04 0.977 CRI I 4.76 ± 0.25d 4.37 ± 0.19 4.01 ± 0.13 0.029 4.10 ± 0.19 4.11 ± 0.28 4.26 ± 0.21 0.792 CRI II 2.93 ± 0.21 2.79 ± 0.15 2.50 ± 0.11 0.166 2.49 ± 0.13 2.62 ± 0.25 2.75 ± 0.20 0.553 AC 3.76 ± 0.25d 3.37 ± 0.19 3.01 ± 0.13 0.029 3.10 ± 0.19 3.11 ± 0.28 3.26 ± 0.21 0.792 Table 3 Mean ± SEa of Atherogenic Indicesb of Plasma according to Vitamin D Statusc of Participants There were no significant differences in serum 25(OH) D3 and the other atherogenic indices of plasma of par- ticipants among the case and control groups according to gender (Table 2). Results from hierarchical linear regression To investigate whether adding atherogenic indices of plasma as independent variables significantly improve a model’s ability to predict serum 25(OH) vitamin D3 as a dependent variable, Hierarchical Linear Regression was applied. Table 4 quantifies the relationship between predictor variables and serum vitamin D. Results identi- fies none of the atherogenic indices of plasma impact on serum vitamin D of case participants. However, results showed that CRI I and CRI II in models 3 and 4 (P = 0.040 and 0.041; and 0.046 and 0.040 respectively) as well as CRI II and AC in model 5 (P = 0.040 and 0.046) have a slight impact on serum vitamin D of control partici- pants. It seems that models 3 and 4 are the right models Serum 25(OH) D3 and atherogenic indices of participants Serum 25(OH) D3 and atherogenic indices of participants Table 2 indicates the mean (±SE) and interquartile range of serum 25(OH) D3 and atherogenic indices of plasma of participants according to gender. There was a signifi- cant difference among case and control groups for AIP in males and females (P < 0.001 and P = 0.007, respectively). Mahmoodi and Najafipour BMC Endocrine Disorders (2022) 22:126 Page 6 of 12 Table 4 Hierarchical Linear Regression for Serum Vitamin Da and Atherogenic Indices of Plasma of Case and Control Participants a Dependent variable: Serum 25(OH) vitamin D3 b Model 1. Predictors: (constant), Atherogenic Index of Plasma Model 2 Predictors: (constant) Castelli Risk Index I Atherogenic Index of Plasma Unstandardized Coefficients Standardized Coefficients t Sig. 95.0% Confidence Interval for B Modelb/Case B Std. a Dependent variable: Serum 25(OH) vitamin D3 b Model 1. Predictors: (constant), Atherogenic Index of Plasma a Dependent variable: Serum 25(OH) vitamin D3 Serum 25(OH) D3 and atherogenic indices of participants Error Beta Lower Bound Upper Bound 1 (Constant) 27.496 2.580 10.656 0.000 22.388 32.604 AIP −1.685 4.085 −0.037 − 0.412 0.681 −9.773 6.402 2 (Constant) 24.013 4.010 5.988 0.000 16.073 31.953 AIP −6.338 5.787 −0.140 −1.095 0.276 −17.796 5.120 CRI I 1.367 1.206 0.145 1.134 0.259 −1.020 3.755 3 (Constant) 24.216 4.198 5.769 0.000 15.900 32.531 AIP −10.293 8.514 −0.212 −1.209 0.229 −27.159 6.574 CRI I 2.073 3.098 0.213 0.669 0.505 −4.065 8.210 CRI II −0.595 3.241 −0.047 − 0.184 0.855 −7.016 5.825 4 (Constant) 28.026 4.815 5.821 0.000 18.487 37.564 AIP −12.551 8.578 −0.258 −1.463 0.146 −29.545 4.444 CRI I 3.092 3.145 0.317 0.983 0.328 −3.138 9.322 CRI II 0.713 3.324 0.056 0.214 0.831 −5.873 7.299 Non-HDL −0.071 0.045 −0.233 −1.583 0.116 −0.159 0.018 5 (Constant) 31.118 4.823 6.452 0.000 21.562 40.673 AIP −12.551 8.578 −0.258 −1.463 0.146 −29.545 4.444 CRI II 0.713 3.324 0.056 0.214 0.813 −5.873 7.299 Non-HDL −0.071 0.045 −0.233 −1.583 0.116 −0.159 0.018 AC 3.092 3.145 0.317 0.983 0.328 −3.138 9.322 Model/Control 1 (Constant) 28.336 2.105 13.464 0.000 24.171 32.500 AIP −6.353 4.553 −0.123 −1.395 0.165 −15.362 2.657 2 (Constant) 29.641 5.465 5.424 0.000 18.826 40.547 AIP −5.275 6.180 −0.102 −0.857 0.395 −17.505 6.956 CRI I −0.406 1.567 −0.031 −0.259 0.796 −3.508 2.696 3 (Constant) 47.017 10.005 4.699 0.000 27.214 66.820 AIP 20.133 13.749 0.391 1.464 0.146 −7.080 47.346 CRI I −17.627 8.493 −1.350 −2.075 0.040 −34.438 −0.816 CRI II 17.184 8.333 1.067 2.062 0.041 0.690 33.678 4 (Constant) 48.017 10.269 4.676 0.000 27.690 68.344 AIP 19,877 13.804 0.386 1.440 0.152 −7.447 47.201 CRI I −17.250 8.560 −1.321 −2.015 0.046 −34.193 −0.307 CRI II 17.405 8.374 1.081 2.079 0.040 0.830 33.980 Non-HDL −0.022 0.048 −0.056 − 0.461 0.646 − 0.118 0.074 5 (Constant) 30.767 5.013 6.137 0.000 20.843 40.690 AIP 19.877 13.804 0.386 1.440 0.152 −7.447 47.201 CRI II 17.405 8.374 1.081 2.079 0.040 0.830 33.980 Non-HDL −0.022 0.048 −0.056 − 0.461 0.646 − 0.118 0.074 AC −17.250 8.560 −1.321 −2.051 0.046 −34.193 −0.307 Table 4 Hierarchical Linear Regression for Serum Vitamin Da and Atherogenic Indices of Plasma of Case and Control Participants Regression for Serum Vitamin Da and Atherogenic Indices of Plasma of Case and Control Participants Mahmoodi and Najafipour BMC Endocrine Disorders (2022) 22:126 Page 7 of 12 Table 5 Univariate Analysis of Variancesa of Two Independent Vitamin D Statusesb and Body Mass Indexc on Atherogenic Indices of Plasma a Univariate Analysis of Variances analyzed the differences (Mean ± SE) atherogenic indices between serum vitamin D concentrations based on three vitamin D statuses and body mass index based on median for each gender group in case and control groups b Vitamin D status divided into vitamin D deficiency (serum vitamin D ≤ 20 ng/ml), vitamin D insufficiency (serum vitamin D > 20-30 ng/ml), and vitamin D satisfactory (serum vitamin D > 30 ng/ml) c Body mass index dichotomized into body mass index ≤25.99 and > 26.00 in male case group and dichotomized into body mass index ≤26.49 and > 26.50 in female case group; and dichotomized into body mass index ≤24.99 and > 25.00 in male control group and dichotomized into body mass index ≤25.99 and > 26.00 in female control group Vitamin D Deficiency Vitamin D Insufficiency Vitamin D Satisfactory Sig. Serum 25(OH) D3 and atherogenic indices of participants Vitamin D Sig. BMI Sig. Serum 25(OH) D3 and atherogenic indices of participants Vit DaBMI ≤ median BMI > median BMI ≤ median BMI > median BMI ≤ median BMI > median BMI Male Atherogenic Index of Plasma Case 0.54 ± 0.11 0.66 ± 0.08 0.46 ± 0.08 0.56 ± 0.07 0.55 ± 0.08 0.56 ± 0.11 0.596 0.290 0.846 Control 0.45 ± 0.07 0.63 ± 0.06 0.29 ± 0.05 0.59 ± 0.07 0.35 ± 0.06 0.38 ± 0.06 0.025 0.002 0.112 Castelli Risk Index I Case 4.40 ± 0.46 4.90 ± 0.35 4.12 ± 0.35 4.50 ± 0.30 4.58 ± 0.33 3.95 ± 0.46 0.583 0.787 0.288 Control 4.34 ± 0.29 5.08 ± 0.25 4.10 ± 0.23 4.88 ± 0.31 3.85 ± 0.26 4.16 ± 0.25 0.029 0.007 0.622 Castelli Risk Index II Case 2.61 ± 0.32 2.79 ± 0.26 2.44 ± 0.25 2.39 ± 0.21 2.58 ± 0.24 2.16 ± 0.32 0.461 0.665 0.584 Control 2.67 ± 0.24 3.13 ± 0.21 2.67 ± 0.19 3.01 ± 0.26 2.36 ± 0.22 2.65 ± 0.21 0.165 0.052 0.923 Atherogenic Coefficient Case 3.40 ± 0.46 3.90 ± 0.35 3.12 ± 0.35 3.50 ± 0.30 3.58 ± 0.33 2.95 ± 0.46 0.583 0.787 0.288 Control 3.34 ± 0.29 4.10 ± 0.25 3.10 ± 0.23 3.87 ± 0.31 2.85 ± 0.26 3.16 ± 0.25 0.029 0.007 0.622 Female Atherogenic Index of Plasma Case 0.58 ± 0.09 0.62 ± 0.09 0.49 ± 0.18 0.42 ± 0.12 0.61 ± 0.10 0.58 ± 0.09 0.441 0.834 0.873 Control 0.30 ± 0.07 0.40 ± 0.07 0.29 ± 0.12 0.36 ± 0.08 0.27 ± 0.09 0.40 ± 0.07 0.948 0.138 0.939 Castelli Risk Index I Case 4.28 ± 0.47 4.35 ± 0.49 4.05 ± 0.94 4.04 ± 0.61 4.60 ± 0.51 5.04 ± 0.47 0.416 0.735 0.908 Control 3.68 ± 0.26 4.43 ± 0.25 3.60 ± 0.47 4.34 ± 0.32 3.95 ± 0.36 4.41 ± 0.25 0.823 0.019 0.868 Castelli Risk Index II Case 2.14 ± 0.34 2.41 ± 0.34 2.24 ± 0.65 2.44 ± 0.43 2.52 ± 0.38 3.06 ± 0.34 0.317 0.340 0.908 Control 2.24 ± 0.22 2.75 ± 0.21 2.19 ± 0.40 2.81 ± 0.27 2.58 ± 0.30 2.84 ± 0.21 0.625 0.043 0.800 Atherogenic Coefficient Case 3.28 ± 0.47 3.35 ± 0.49 3.05 ± 0.94 3.04 ± 0.61 3.60 ± 0.51 4.04 ± 0.47 0.416 0.735 0.908 Control 2.68 ± 0.26 3.43 ± 0.25 2.60 ± 0.47 3.34 ± 0.32 2.95 ± 0.36 3.41 ± 0.25 0.823 0.019 0.868 Table 5 Univariate Analysis of Variancesa of Two Independent Vitamin D Statusesb and Body Mass Indexc on Atherogenic Indices of l indicated interaction results between two independ- ent variables on atherogenic indices of participants. Serum 25(OH) D3 and atherogenic indices of participants The interaction between BMI and vitamin D status on athero- genic indices were not significant in case and control par- ticipants. However, the main effect of BMI and vitamin D status on AIP, CRI I, and AC were significant in control males. The main effect of BMI on CRI I, CRI II, and AC were significant in control females. to determine variables associated with serum vitamin D as a dependent variable for control participants (Table 4). However, due to the number of participants studied in each gender, the gender effect was not investigated in the model, separately. Therefore, caution should be taken to consider the analysis of the genders in the models. 1 FBS, and TG among control males and for WHR, FBS, TG, SBP, and DBP among control females. Gender dif- ferences in the cardiometabolic biomarkers have been revealed similar to other studies [20, 21]. In the present study, there were no significant differences in serum 25(OH) D between case and control participants. Unlike, our current findings, in a study, the prevalence of low serum 25(OH) D3 level considerably was high in patients with CVD risk factors. These patients presented signifi- cantly higher values for cardiometabolic biomarkers [6]. Multiple Regression Model showed that for an individual to maintain metabolic parameters, at least at borderline values, serum 25(OH) D3 level should be 37.64 nmol/L [6]. A positive and significant association between AIP and higher HbA1c and lower HDL-C were seen in peo- ple with plasma 25(OH) D3 less than 25 nmol/L [12]. This finding was parallel to our findings (Data not shown). Moreover, we found a significant increase in AIP for case males and females than control males and females. Although, the other atherogenic indices of plasma increased nearly in cases than controls; there were no sig- nificant differences between case and control males and females. Therefore, the results of our study were parallel to the other studies [9–11] that revealed AIP compared to CRI I, CRI II, and AC can be used as a novel surro- gate marker; however, the other studies have been shown that AIP is associated with subclinical atherosclerosis and CVD events in participants with DM [8–11]. In a study, the serum 25(OH) D levels were negatively associated males (0.701-0.717) and females (0.615-0.727) in both case and controls (P < 0.001). However, very strong sig- nificant relationships were found between CRI I with CRI II (0.920-0.957), CRI II with AC (0.920-0.957), and CRI I with AC (1.000) for control males and females and case females (P < 0.001) (Table 6). A weak significant negative relationship was found between serum vitamin D with TC for case males (r = − 0.280, P < 0.020). However, the partial correla- tion adjusting for BMI and WHR did not change the significance of the relationship between the biomarkers (Table 7). 1 Figure 1 indicates the bivariate (Pearson) correlation of AIP with BMI (controls: R = 0.281, cases: R = 0.089), WC (controls: R = 0.289, cases: R = 0.134), WHR (controls: R = 0.258, cases: R = 0.144), TG (controls: R = 0.894, cases: R = 0.919), TC (controls: R = 0.161, cases: R = 0.100), and HDL-C (controls: R = 0.509, cases: R = 0.695). The correlations of AIP with BMI, WC, WHR, TG, and TC are direct relationship, however, the correla- tion of AIP with HDL-C is inverse relationship. Results from bivariate and partial correlation To find out whether the interaction between two inde- pendent variables such as three vitamin D statuses and BMI on atherogenic indices of plasma is significant, Uni- variate ANOVA was applied. The Univariate ANOVA Table 6 indicates the Pearson correlation coefficient between atherogenic indices. The strong significant rela- tionships were found between AIP with CRI I and AC for Mahmoodi and Najafipour BMC Endocrine Disorders (2022) 22:126 Page 8 of 12 Table 6 Pearson Correlation Coefficients for the Relationship between Atherogenic Indices of Plasma according to gender Atherogenic Indices Atherogenic Index of Plasma Castelli Risk Index I Castelli Risk Index II Atherogenic Coefficient Atherogenic Index of Plasma Castelli Risk Index I Castelli Risk Index II Atherogenic Coefficient Case Male Female Atherogenic Index of Plasma 1 0.701 < 0.001 −0.003 0.979 0.701 < 0.001 1 0.727 < 0.001 0.440 < 0.001 0.727 < 0.001 Castelli Risk Index I 1 0.686 < 0.001 1.000 < 0.001 1 0.957 < 0.001 1.000 < 0.001 Castelli Risk Index II 1 0.686 < 0.001 1 0.957 < 0.001 Atherogenic Coefficient 1 1 Control Male Female Atherogenic Index of Plasma 1 0.717 < 0.001 0.429 < 0.001 0.717 < 0.001 1 0.615 < 0.001 0.294 < 0.001 0.615 < 0.001 Castelli Risk Index I 1 0.929 < 0.001 1.000 < 0.001 1 0.920 < 0.001 1.000 < 0.001 Castelli Risk Index II 1 0.929 < 0.001 1 0.920 < 0.001 Atherogenic Coefficient 1 1 Discussion We found that cardiometabolic biomarkers significantly decreased for BMI, WC, WHR, Page 9 of 12 Mahmoodi and Najafipour BMC Endocrine Disorders (2022) 22:126 Table 7 Pearson Correlation Coefficients for the Relationship between Serum Vitamin D and the Lipid and Lipoprotein Profiles according to gender Biomarkers Serum Vitamin D Triglyceride Total Cholesterol LDL-Chol HDL-Chol Serum Vitamin D Triglyceride Total Cholesterol LDL-Chol HDL-Chol Case Male Female Serum Vitamin D 1 −0.147 0.237 −0.280 0.020 − 0.204 0.101 0.053 0.671 1 −0.070 0.628 0.148 0.301 0.217 0.125 −0.057 0.694 Triglyceride 1 0.126 0.313 −0.270 0.029 −0.430 < 0.001 1 0.260 0.065 0.042 0.772 −0.401 0.004 Total Cholesterol 1 0.901 < 0.001 0.346 0.004 1 0.948 < 0.001 0.394 < 0.001 LDL-Chol 1 0.389 0.001 1 0.306 0.029 HDL-Chol 1 1 Control Male Female Serum Vitamin D 1 −0.238 0.053 −0.152 0.221 −0.134 0.281 0.110 0.376 1 −0.088 0.500 0.015 0.907 0.059 0.654 −0.017 0.896 Triglyceride 1 0.448 < 0.001 0.164 0.184 −0.097 0.435 1 0.198 0.127 −0.126 0.332 −0.237 0.066 Total Cholesterol 1 0.914 < 0.001 0.532 < 0.001 1 0.896 < 0.001 0.217 0.093 LDL-Chol 1 0.363 0.003 1 −0.016 0.904 HDL-Chol 1 1 Table 7 Pearson Correlation Coefficients for the Relationship between Serum Vitamin D and the Lipid and Lipoprotein Profiles according to gender Biomarkers Serum Vitamin D Triglyceride Total Cholesterol LDL-Chol HDL-Chol Serum Vitamin D Triglyceride Total Cholesterol LDL-Chol HDL-Chol Case Male Female Serum Vitamin D 1 −0.147 0.237 −0.280 0.020 − 0.204 0.101 0.053 0.671 1 −0.070 0.628 0.148 0.301 0.217 0.125 −0.057 0.694 Triglyceride 1 0.126 0.313 −0.270 0.029 −0.430 < 0.001 1 0.260 0.065 0.042 0.772 −0.401 0.004 Total Cholesterol 1 0.901 < 0.001 0.346 0.004 1 0.948 < 0.001 0.394 < 0.001 LDL-Chol 1 0.389 0.001 1 0.306 0.029 HDL-Chol 1 1 Control Male Female Serum Vitamin D 1 −0.238 0.053 −0.152 0.221 −0.134 0.281 0.110 0.376 1 −0.088 0.500 0.015 0.907 0.059 0.654 −0.017 0.896 Triglyceride 1 0.448 < 0.001 0.164 0.184 −0.097 0.435 1 0.198 0.127 −0.126 0.332 −0.237 0.066 Total Cholesterol 1 0.914 < 0.001 0.532 < 0.001 1 0.896 < 0.001 0.217 0.093 LDL-Chol 1 0.363 0.003 1 −0.016 0.904 HDL-Chol 1 1 Female Mahmoodi and Najafipour BMC Endocrine Disorders (2022) 22:126 Page 10 of 12 Fig. 1 Correlation of AIP with A) BMI, B) WC, C) WHR, D) TG, E) TC, F) HDL-C. Discussion We sought the association between serum 25(OH) D3 levels and atherogenic indices of plasma as novel surro- gate markers of cardiometabolic disease risk in patients with DM in a population-based KERCADR cohort study as an Iranian community. Discussion Control: circle; Case: square; Control regression line: solid line; Case regression line: dashed line Fig. 1 Correlation of AIP with A) BMI, B) WC, C) WHR, D) TG, E) TC, F) HDL-C. * Control: circle; Case: square; Control regression line: solid line; Case regression line: dashed line with AIP in men but not in women. In addition, vitamin D deficient men had higher AIP values than vitamin D sufficient men [16]. Deficient serum 25(OH) D was asso- ciated with higher TC, LDL-C, and TG in middle-aged and elderly Chinese individuals. This finding suggested that low 25(OH) D levels was a marker for elevated ath- erogenic lipoproteins [22]. In another study, the research- ers found that AIP was an independent predictor of CAD [23]. One of the most important reasons for the lack of significant differences in these biomarkers may be the lack of significant differences between the cardiometa- bolic biomarkers that make up these novel biomarkers. The other reasons for the lack of significant differences may be the selection of stringent exclusion criteria as well as the correlation coefficients between AIP and the other biomarkers (Table 6). Interestingly, the levels of AIP, CRI, and AC significantly decreased with improved serum vitamin D status only in control males. Although, the trends of the levels of atherogenic indices were irregular in case and control females with improving vitamin D status; the levels of these indices in case males decreased non-significantly with improving vitamin D status. Vita- min D deficiency in poor glycemic control is likely to develop dyslipidemia as compared to vitamin D insuffi- cient and sufficient groups [7]. A Univariate ANOVA to determine whether the interaction between two inde- pendent variables such as three vitamin D statuses and with AIP in men but not in women. In addition, vitamin D deficient men had higher AIP values than vitamin D sufficient men [16]. Deficient serum 25(OH) D was asso- ciated with higher TC, LDL-C, and TG in middle-aged and elderly Chinese individuals. This finding suggested that low 25(OH) D levels was a marker for elevated ath- erogenic lipoproteins [22]. In another study, the research- ers found that AIP was an independent predictor of CAD [23]. One of the most important reasons for the lack of significant differences in these biomarkers may be the lack of significant differences between the cardiometa- bolic biomarkers that make up these novel biomarkers. Discussion The results of our study revealed that the level of AIP among the other atherogenic indices of plasma could a surrogate markers for the incidence of T2DM and CAD in participants with CVD. As the level of AIP was positively associated with CRI I and AC as two novel atherogenic risk markers, they can be used as predic- tive surrogate markers for CAD/CVD in populations. Although the correlation between CRI II and CRI I was very strong; however, CRI II could not be used as an alternative to AIP for predicting CVD risk in our study. The correlation of AIP with the other cardiometabolic biomarkers was also similar to the other studies [9, 25]. A strong negative correlation between low vitamin D status (serum 25(OH) D < 15 ng/mL) and the three identified biomarkers of atherogenic dyslipidemia: high serum levels of small density LDL-C, TG, and VLDL- C in middle-aged adults without CVD [26]. The find- ings of our study parallel to the other studies [11, 27] revealed that AIP can be recommended as a novel sur- rogate marker in the diagnosis of CVD and progression of atherosclerosis in healthy and diabetic participants. other lipid values were within the normal range [25]. The results of our study revealed that the level of AIP among the other atherogenic indices of plasma could a surrogate markers for the incidence of T2DM and CAD in participants with CVD. As the level of AIP was positively associated with CRI I and AC as two novel atherogenic risk markers, they can be used as predic- tive surrogate markers for CAD/CVD in populations. Although the correlation between CRI II and CRI I was very strong; however, CRI II could not be used as an alternative to AIP for predicting CVD risk in our study. The correlation of AIP with the other cardiometabolic biomarkers was also similar to the other studies [9, 25]. A strong negative correlation between low vitamin D status (serum 25(OH) D < 15 ng/mL) and the three identified biomarkers of atherogenic dyslipidemia: high serum levels of small density LDL-C, TG, and VLDL- C in middle-aged adults without CVD [26]. The find- ings of our study parallel to the other studies [11, 27] revealed that AIP can be recommended as a novel sur- rogate marker in the diagnosis of CVD and progression of atherosclerosis in healthy and diabetic participants. Strengths and limitationh There are several strengths in the present study. The par- ticipants in this nested case-control study were selected from the second phase of a large KERCADR cohort study as an Iranian community. Participants with diabetes and controls were randomly selected and matched by some factors. Exception for glycemic indices and lipid and lipo- protein profiles, case and control groups were matched by the other risk factors such as high blood pressure and BMI equal or greater than 30 and known potential con- founders. All possible analyzes were performed between participants with diabetes and controls by gender. One of the most important limitations of the present study was the non-entry of a number of patients with diabetes who had the other risk factors and did not enroll in our inves- tigation. Therefore, with the management of confounding in study design, the cases to controls ratio became 1:1. Funding h This research was supported by grant from Vice chancellor for Research of Kerman University of Medical Sciences. Acknowledgements Acknowledgements The authors are grateful to Vice chancellor for Research of Kerman University of Medical Sciences in this study. The authors are also grateful to participants that contributed in this study. Author details 1 Physiology Research Center, Institute of Neuropharmacology, Kerman University of Medical Sciences, Kerman, Iran. 2 Department of Nutrition, Faculty of Public Health, Kerman University of Medical Sciences, Kerman, Iran. Haft Bagh‑E‑Alavi Highway, Kerman 7616913555, Iran. 3 Department of Physiology and Pharmacology, Afzalipour Faculty of Medicine, Kerman University of Medi- cal Sciences, Kerman, Iran. Discussion other lipid values were within the normal range [25]. The results of our study revealed that the level of AIP among the other atherogenic indices of plasma could a surrogate markers for the incidence of T2DM and CAD in participants with CVD. As the level of AIP was positively associated with CRI I and AC as two novel atherogenic risk markers, they can be used as predic- tive surrogate markers for CAD/CVD in populations. Although the correlation between CRI II and CRI I was very strong; however, CRI II could not be used as an alternative to AIP for predicting CVD risk in our study. The correlation of AIP with the other cardiometabolic biomarkers was also similar to the other studies [9, 25]. biomarker for the incidence of T2DM and CAD in par- ticipants with CVD. In general, the level of atherogenic indices of plasma in the controls with normal BMI is lower than the controls with high BMI. This effect is pre- dominant with improving serum vitamin D. Conclusions Received: 14 January 2021 Accepted: 4 May 2022 We conclude that there is a reverse significant associa- tion between AIP and serum vitamin D among healthy males. Low serum level of vitamin D is associated with atherogenic dyslipidemia. Therefore, improving vitamin D status as an important indicator may alleviate AIP as a surrogate marker for predicting the risk of CVD events in healthy men and women with normal BMI. The effect of decreasing BMI on significant decrease of AIP level for healthy participants with improving serum vitamin D is more effective. The level of AIP among the other ath- erogenic indices of plasma could a potential predictive Authors’ contributions MRM contributed to the conception of the original idea, conducting the study design, analysis and interpretation of the data, drafting and revising the draft, approval of the final version of the manuscript, and agreed to all aspects of the work. HN is the executor of the population-based KERCADR cohort study. HN contributed to the conception of primary study (KERCADRS), the acquisi- tion of biochemical data, approval of the final version of the manuscript, and agreed to all aspects of the work. A strong negative correlation between low vitamin D status (serum 25(OH) D < 15 ng/mL) and the three identified biomarkers of atherogenic dyslipidemia: high serum levels of small density LDL-C, TG, and VLDL- C in middle-aged adults without CVD [26]. The find- ings of our study parallel to the other studies [11, 27] revealed that AIP can be recommended as a novel sur- rogate marker in the diagnosis of CVD and progression of atherosclerosis in healthy and diabetic participants. Availability of data and materialsi The data that support the findings of this study are available from the Head of Physiology Research Center, but restrictions apply to the availability of these data, which were used under license for the current study, and so are not pub- licly available. Data are however available from the authors upon reasonable request and with permission of the Head of Physiology Research Center. Competing interests No potential conflict of interest relevant to this article was reported. p g No potential conflict of interest relevant to this article was reported. 1. Gao Y, Zheng T, Ran X, Ren Y, Chen T, Zhong L, et al. Vitamin D and incidence of prediabetes or type 2 diabetes: a four-year follow-up community-based study. Dis Markers. 2018:1926308. 2. Akter S, Kuwahara K, Matsushita Y, Nakagawa T, Konishi M, Honda T, et al. Serum 25-hydroxyvitamin D3 and risk of type 2 diabetes among Japanese adults: the Hitachi Health Study. Clin Nutr. 2019;S0261-5614(19):30221–3. 3. Lips P, Eekhoff M, van Schoor N, Oosterwerff M, de Jongh R, Krul-Poel Y, et al. Vitamin D and type 2 diabetes. J Steroid Biochem Mol Biol. 2017 Oct;173:280–5. Ethics approval and consent to participate The protocol was also approved by review panels and ethics committees (Approval ID: IR.KMU.REC. 1399.405). All methods were performed in accordance with the relevant guidelines and regulations of Biomed Central. “Informed” consent was obtained from all participants that took part in KERCADR cohort study. Discussion The other reasons for the lack of significant differences may be the selection of stringent exclusion criteria as well as the correlation coefficients between AIP and the other biomarkers (Table 6). Interestingly, the levels of AIP, CRI, and AC significantly decreased with improved serum vitamin D status only in control males. Although, the trends of the levels of atherogenic indices were irregular in case and control females with improving vitamin D status; the levels of these indices in case males decreased non-significantly with improving vitamin D status. Vita- min D deficiency in poor glycemic control is likely to develop dyslipidemia as compared to vitamin D insuffi- cient and sufficient groups [7]. A Univariate ANOVA to determine whether the interaction between two inde- pendent variables such as three vitamin D statuses and BMI on atherogenic indices of plasma as the dependent variable is significant. We revealed that there was a posi- tive significant association between BMI and the level of AIP with increasing serum vitamin D in control males and females. In other words, the effect of decreasing BMI on significant decrease the level of AIP for control participants with improving serum vitamin D was more effective. An adequate serum vitamin D level might have possible beneficial effects on the level of AIP in normal BMI. Generaly, the mean difference AIP between the case and control females was significant. The researchers observed a negative correlation between 25(OH)D levels and the atherogenic profile in obese patients [18].h i The correlation analysis showed a negative linear association between serum 25(OH) D and TG, TC, and LDL-C and a positive linear association between serum 25(OH) D and HDL-C in case and control males but none with females. This current finding was parallel to another study [18]. The results of a meta-analysis showed that lipid and lipoprotein profiles indicate the risk of T2DM; however, the level of AIP might be more closely associated with the risk of T2DM and be used as a predictive indica- tor in evaluating the risk of T2DM [24]. The findings of another study showed that increasing in AIP was asso- ciated with the other CVD risk factors and AIP can be used as a sensitive and regular index of CVD when the Mahmoodi and Najafipour BMC Endocrine Disorders (2022) 22:126 Page 11 of 12 other lipid values were within the normal range [25]. Publisher’s Note S N Horm Metab Res. 2021. 18. Curvello-Silva KL, Oliveira NA, Silva TSS, Sousa CD, Daltro C. Association between cardiovascular risk factors and 25(OH) D levels in obese patients. Metab Syndr Relat Disord. 2020;18(7):328–32. 19. Najafipour H, Mirzazadeh A, Haghdoost A, Shadkam M, Afshari M, Moa- zenzadeh M, et al. Coronary artery disease risk factors in an urban and Peri-urban setting, Kerman, southeastern Iran (KERCADR study): method- ology and preliminary report. Iran J Public Health. 2012;41(9):86–92. 20. Mahmoodi MR, Najafipour H, Mohsenpour MA, Amiri M. The relationship between food insecurity with cardiovascular risk markers and metabolic syndrome components in diabetic patients: a population based study from KERCADRS. J Res Med Sci. 2017;22:118. 21. Mahmoodi MR, Abadi AR, Kimiagar SM. Sex differences in myocardial infarction events between patients with and without conventional risk factors: the Modares heart study. Am Heart Hosp J. 2007;5:228–35. • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: y 22. References Page 12 of 12 Page 12 of 12 Mahmoodi and Najafipour BMC Endocrine Disorders (2022) 22:126 4. Wimalawansa SJ. Associations of vitamin D with insulin resistance, obesity, type 2 diabetes, and metabolic syndrome. J Steroid Biochem Mol Biol. 2018 Jan;175:177–89. 26. Han YY, Hsu SHJ, Su TC. Association between vitamin D deficiency and high serum levels of small dense LDL in middle-aged adults. Biomedi- cines. 2021;9(5):464. 27. Yildiz G, Duman A, Aydin H, Yilmaz A, Hür E, Mağden K, et al. Evaluation of association between atherogenic index of plasma and intima-media thickness of the carotid artery for subclinic atherosclerosis in patients on maintenance hemodialysis. Hemodial Int. 2013;17(3):397–405. 27. Yildiz G, Duman A, Aydin H, Yilmaz A, Hür E, Mağden K, et al. Evaluation of association between atherogenic index of plasma and intima-media thickness of the carotid artery for subclinic atherosclerosis in patients on maintenance hemodialysis. Hemodial Int. 2013;17(3):397–405. 5. Krivošíková Z, Gajdoš M, Šebeková K. Vitamin D levels decline with rising number of Cardiometabolic risk factors in healthy adults: association with Adipokines, inflammation, oxidative stress and advanced glycation mark- ers. PLoS One. 2015;10(6):e0131753. 6. Barbalho SM, Tofano RJ, de Campos AL, Rodrigues AS, Quesada K, Bechara MD, et al. Association between vitamin D status and metabolic syndrome risk factors. Diabetes Metab Syndr. 2018;12(4):501–7. Mahmoodi and Najafipour BMC Endocrine Disorders (2022) 22:126 Publisher’s Note S N Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. y y 7. Pokhrel S, Giri N, Pokhrel R, Dev Pardhe B, Lamichhane A, Chaudhary A, et al. Vitamin D deficiency and cardiovascular risk in type 2 diabetes population. Open Life Sci. 2021;16(1):464–74. 8. Nansseu JR, Moor VJ, Nouaga ME, Zing-Awona B, Tchanana G, Ketcha A. Atherogenic index of plasma and risk of cardiovascular disease among Cameroonian postmenopausal women. Lipids Health Dis. 2016 Mar;9(15):49. 9. Bo MS, Cheah WL, Lwin S, Moe Nwe T, Win TT, Aung M. Understanding the relationship between Atherogenic index of plasma and cardiovas- cular disease risk factors among staff of a University in Malaysia. J Nutr Metab. 2018;4(2018):7027624. 10. Barua L, Faruque M, Banik PC, Ali L. Atherogenic index of plasma and its association with cardiovascular disease risk factors among postmenopau- sal rural women of Bangladesh. Indian Heart J. 2019;71(2):155–60. 11. Fernández-Macías JC, Ochoa-Martínez AC, Varela-Silva JA, Pérez-Mal- donado IN. Atherogenic index of plasma: novel predictive biomarker for cardiovascular illnesses. Arch Med Res. 2019;50(5):285–94. 12. Wang EW, Pang MYC, Siu PMF, Lai CKY, Woo J, Collins AR, et al. Vitamin D status and cardiometabolic risk factors in young adults in Hong Kong: associations and implications. Asia Pac J Clin Nutr. 2018;27(1):231–7. 13. Turkes GF, Uysal S, Demir T, Demiral Y, Pamuk BO, Yılmaz H, et al. Asso- ciations between bioavailable vitamin D and remnant cholesterol in patients with type 2 diabetes mellitus. Cureus. 2021;13(2):e13248. 14. Kwon HN, Lim H. Relationship between serum vitamin D status and metabolic risk factors among Korean adults with prediabetes. PLoS One. 2016;11(10):e0165324. 15. Izadi A, Aliasghari F, Gargari BP, Ebrahimi S. Strong association between serum vitamin D and Vaspin levels, AIP, VAI and liver enzymes in NAFLD patients. Int J Vitam Nutr Res. 2019;1:1–8. 16. Wang Y, Si S, Liu J, Wang Z, Jia H, Feng K, et al. The associations of serum lipids with vitamin D status. PLoS One. 2016;11(10):e0165157. 17. Dhas Y, Banerjee J, Damle G, Mishra N. Serum 25(OH) D concentration and cardiovascular disease risk markers among middle-aged healthy and type 2 diabetic subjects. Horm Metab Res. 2021. 17. Dhas Y, Banerjee J, Damle G, Mishra N. Serum 25(OH) D concentration and cardiovascular disease risk markers among middle-aged healthy and type 2 diabetic subjects. • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: 26. Han YY, Hsu SHJ, Su TC. Association between vitamin D deficiency and high serum levels of small dense LDL in middle-aged adults. Biomedi- cines. 2021;9(5):464. Publisher’s Note S N Guan C, Fu S, Zhen D, Li X, Niu J, Cheng J, et al. Correlation of serum vita- min D with lipid profiles in middle-aged and elderly Chinese individuals. Asia Pac J Clin Nutr. 2020;29(4):839–45. 23. Wu TT, Gao Y, Zheng YY, Ma YT, Xie X. Atherogenic index of plasma (AIP): a novel predictive indicator for the coronary artery disease in postmeno- pausal women. Lipids Health Dis. 2018;17(1):197. 24. Zhu XW, Deng FY, Lei SF. Meta-analysis of Atherogenic index of plasma and other lipid parameters in relation to risk of type 2 diabetes mellitus. Prim Care Diabetes. 2015 Feb;9(1):60–7. 25. Niroumand S, Khajedaluee M, Khadem-Rezaiyan M, Abrishami M, Juya M, Khodaee G, et al. Atherogenic index of plasma (AIP): a marker of cardio- vascular disease. Med J Islam Repub Iran. 2015 Jul;25(29):240.
https://openalex.org/W4393103483	https://www.scielo.br/j/bioet/a/hVQr9QLJ4nPv7tt3mCfbVJq/?lang=pt&format=pdf	Portuguese	null	Bioética aplicada aos cuidados paliativos: questão de saúde pública	Revista Bioética	2,023	cc-by	9,673	Revista Bioética Print version ISSN 1983-8042 \| On-line version ISSN 1983-8034 Revista Bioética Print version ISSN 1983-8042 \| On-line version ISSN 1983-8034 Rev. Bioét. vol.31 Brasília 2023 Resumo Cuidados paliativos são um conjunto de ações que visam melhorar a qualidade de vida do paciente e de sua família quando a doença já não responde a tratamentos curativos. Abrangem cuidados físicos, psicológicos, espirituais e sociais, entendendo a morte como um processo natural, não acelerando nem retardando seu desfecho. Esta revisão integrativa qualitativa selecionou 131 artigos sobre bioética e cuidados paliativos publicados nos últimos cinco anos, analisando 10 deles. Esses estudos destacam a importância da bioética no contexto dos cuidados paliativos, abordando temas como definição, morte, final de vida e a necessi- dade de equipe multiprofissional interdisciplinar. A espiritualidade também desempenha papel relevante, com o paciente e a família no centro das decisões, baseadas em uma comunicação eficaz. Cuidados paliativos buscam proporcionar conforto, dignidade e suporte integral para pacientes em fase avançada de doenças, permitindo que tenham o máximo de qualidade de vida possível em seus últimos momentos. Palavras-chave: Cuidados paliativos. Cuidados paliativos na terminalidade da vida. Bioética. Saúde pública. Palavras-chave: Cuidados paliativos. Cuidados paliativos na terminalidade da vida. Bioética. Saúde pública. Bioética aplicada aos cuidados paliativos: questão de saúde pública Mario Angelo Cenedesi Júnior Universidad de Ciencias Empresariales y Sociales, Buenos Aires, Argentina. Mario Angelo Cenedesi Júnior Universidad de Ciencias Empresariales y Sociales, Buenos Aires, Argentina. Bioética aplicada a los cuidados paliativos: una cuestión de salud públicaii Los cuidados paliativos constituyen acciones para mejorar la calidad de vida de los pacientes y sus familias cuando la enfermedad ya no responde a los tratamientos curativos. Abarcan la atención física, psicológica, espiritual y social, considerando la muerte como un proceso natural y sin acelerar ni retrasar su desenlace. Esta revisión integradora cualitativa seleccionó 131 artículos sobre bioética y cuidados paliativos publicados en los últimos cinco años, y analizó diez. Los estudios destacan la importancia de la bioética en los cuidados paliativos, abordando cuestiones como la definición, la muerte, el final de la vida y la necesidad de un equipo interdisciplinar multiprofesional. La espiritualidad también desempeña un papel importante, con el paciente y la familia en el centro de las decisiones basadas en una comuni- cación eficaz. Estos cuidados proporcionan confort, dignidad y apoyo integral a los pacientes terminales permitiéndoles una mayor calidad de vida posible en sus últimos momentos. Palabras clave: Cuidados paliativos. Cuidados paliativos al final de la vida. Bioética. Salud pública. Bioética aplicada aos cuidados paliativos: questão de saúde pública implica que esta pode sofrer alterações que a levem de um estado inicial A para um estado final B, de forma que se torne impossível regres- sar ao estado inicial, mesmo se forem alteradas as causas da transição inicial 4. Quando uma equipe médica multiprofissio- nal e interdisciplinar se depara com um paciente terminal, ela deve mobilizar ações que busquem não preservar e manter a vida, mas abrandá-la e resguardá-la. Consciente de que não será mais possível optar pela preservação e manutenção da vida, a equipe precisa, de forma uníssona, empenhar-se para que o sofrimento seja ameni- zado e a qualidade de vida mantida até o findar 1. Esses conceitos são definidos com base em dados objetivos e subjetivos, colhidos e/ou identifi- cados por uma equipe multiprofissional, composta pelas áreas de medicina, enfermagem, fisioterapia, nutrição, fonoaudiologia, terapia ocupacional, psicologia e neuropsicopedagogia, com suporte de um serviço social eficaz, associado à equipe de humanização, por exemplo. O objetivo é esta- belecer diagnóstico ou prognóstico e, assim, definir a estratégia terapêutica e de assistência ao paciente 5. Vislumbrando reflexões sobre o agir humano de maneira a assegurar o bem-estar e a sobre- vivência da humanidade, a bioética, ciência relacionada à sobrevivência humana, busca defender a melhoria das condições de vida com base nos princípios fundamentais da autonomia, beneficência, não maleficência e justiça. Na esteira da bioética, admite-se que todo avanço no campo das ciências biomédicas deve estar a serviço da humanidade, com um posicionamento ético consciente e em constante evolução que busque respostas equilibradas diante dos conflitos atuais 2. De acordo com a Organização Mundial da Saúde (OMS) 6, cuidados paliativos representam um conjunto de ações de cuidados físicos, psicoló- gicos, espirituais e sociais prestados ao indivíduo cuja enfermidade não responde mais aos cuidados curativos. O objetivo é melhorar a qualidade de vida do paciente e de sua família por meio da iden- tificação e do alívio da dor, entendendo a morte como um processo natural, sem acelerá-lo ou retardá-lo. Nesse caso, a morte se torna um des- fecho esperado e não deve ser combatida, o que, contudo, não significa que condutas não possam ser aplicadas, pois o paciente e sua família devem receber medidas terapêuticas específicas. Bioethics applied to palliative care: a public health issueiii Palliative care is a set of actions aimed at improving patients’ and family members’ quality of life when no curative treatment is available. It encompasses physical, psychological, spiritual and social care, understanding death as a natural process whose outcome should be accelerated or delayed. Of the 131 articles on bioethics and palliative care published in the last five years selected, this integrative review analyzes 10. These studies highlight the importance of bioethics for palliative care, addressing themes such as definitions, death, end of life and the need for a multi-professional interdisciplinary team. Spirituality also plays a relevant role, putting the patient and family members as central to decisions made based on effective communication. Palliative care aims to provide comfort, dignity and comprehensive support for patients with advanced illnesses, allowing them the maximum quality of life possible. Keywords: Palliative care Hospice care Bioethics Public health Keywords: Palliative care. Hospice care. Bioethics. Public health. Declara não haver conflito de interesse. Declara não haver conflito de interesse. http://dx.doi.org/10.1590/1983-803420233532PT Rev. bioét. 2023; 31: e3532PT 1-14 1 Bioética aplicada aos cuidados paliativos: questão de saúde pública Bioética aplicada aos cuidados paliativos: questão de saúde pública A bioética enquanto campo nasceu nos Estados Unidos entre o final dos anos 1960 e o início dos anos 1970, quando uma série de fatores his- tóricos e culturais chamaram a atenção para a ética aplicada. O pesquisador em oncologia Van Rensselaer Potter, em 1971, criou e proclamou o novo termo “bioética” em seu livro Bioética: uma ponte para o futuro, como aponta Mori 3. Nesse cenário, os cuidados paliativos surgem como uma modalidade ética no cuidado de pacientes que cursarão a terminalidade da vida, devido a doença crônica. Gutierrez e Barros 7 advertem que cuidados paliativos devem ser ministrados a pacientes cujas doenças apresentam pouca chance de rever- são com o uso de terapêutica curativa desde o momento do diagnóstico, e não apenas nas últi- mas horas de vida. O cuidado paliativo ministrado em fase inicial tem como prerrogativa uma atenção diferenciada, individualizada, considerando as particularidades do indivíduo e suas necessida- des como pessoa em condição de dependência. Assim, o paciente pode receber controle da dor e alívio do sofrimento nas dimensões físicas, psíquicas, sociais e espirituais 5. Antes de discorrer a respeito dessa moda- lidade, é preciso rememorar o entendimento sobre alguns conceitos coadjuvantes: terminali- dade e irreversibilidade. Terminalidade, ou estado terminal, deve ser compreendida como o momento em que, inevitavelmente, a evolução de determi- nada doença resultará em morte, mesmo com a adoção de medidas de intervenção terapêutica 4. Pesquisa Já a irreversibilidade, aquilo cujo sentido não se pode inverter, deve ser compreendida num processo evolutivo referente ao estado do paciente diante da atuação e dos desdobramen- tos da doença que o acomete. A condição de irre- versibilidade de uma propriedade de um sistema Segundo Wittmann-Vieira e Goldim 8, mesmo no sentido biológico, preservar ou salvar a vida não é mais o foco da assistência ao paciente, pois, no que tange às relações entre pessoas, o viver continua sendo o tema fundamental. Oferecer cuidados Rev. bioét. 2023; 31: e3532PT 1-14 http://dx.doi.org/10.1590/1983-803420233532PT Bioética aplicada aos cuidados paliativos: questão de saúde pública paliativos não significa que não há mais nada a fazer pelo paciente ou familiares, mas que o diagnóstico é de doença grave, que evoluirá com ameaça à vida, e que uma equipe, juntamente com os profissionais especialistas na enfermidade, cuidará de quem está doente e daqueles que o cercam. Ou seja, “há muito a fazer” pelo paciente e por sua família. Método O método aplicado na construção deste artigo foi o de revisão integrativa da literatura, que sintetiza inúmeros artigos no intuito de condensar resultados de estudos acerca do mesmo tema 12. A abordagem é qualitativa por lidar com um nível de realidade que não pode ser quantificado, isto é, um universo de significados, motivações, aspirações, crenças, valores e atitudes, o que corresponde a um espaço mais profundo dentro das relações 13. Ainda, utilizaram-se outros artigos impor- tantes sobre o tema como suporte teórico para enriquecimento e sustentação das discussões. Foi dispensado o termo de consen- timento livre e esclarecido (TCLE) por se tra- tar de revisão integrativa, sem envolvimento de seres humanos. Bioética aplicada aos cuidados paliativos: questão de saúde pública publicados, pré-estabelecida da seguinte maneira: 1) elaboração da questão de pesquisa; 2) definição dos critérios de inclusão e exclusão; 3) definição da amostragem; 4) avaliação dos estudos incluídos; 5) interpretação dos resultados; e 6) apresentação da síntese 13. A busca foi realizada nas bases de dados SciELO e LILACS por meio dos descritores em ciências da saúde (DeCS) “bioética” e “cuidados paliativos”, e o recorte temporal abrangeu os últimos cinco anos (2018-2022). Identificaram-se 131 artigos, cujos resumos foram lidos integralmente. Segundo dados da OMS, a cada ano cerca de 56 milhões de pessoas em todo o mundo deman- dam cuidados paliativos 9 e, de acordo com Santos e colaboradores 10, somente no Brasil a projeção é de mais de 1.166.279 pacientes em 2040). Com isso, fica evidente que a temática é uma importante questão de saúde pública 11, sendo necessária sua discussão à luz da bioética, a fim de que haja diálogo, sensibilização e compreensão do tema por parte de todos os envolvidos. Foram incluídos artigos com resumo completo nos idiomas português, inglês ou espanhol, que abordassem a temática bioética e cuida- dos paliativos e estivessem em acesso aberto, e excluídos aqueles com resumos incompletos, com mais de cinco anos de publicação e sem os descritores definidos. Do total que atendeu aos critérios de inclusão, 27 foram excluídos por estarem duplicados nas plataformas e, ao final, restaram 10 artigos. Resultados Optou-se pela revisão integrativa diante da busca de manuscritos que evidenciassem infor- mações sobre o tema, a partir de estudos já O Quadro 1 condensa as principais impressões contidas nos artigos pesquisados. http://dx.doi.org/10.1590/1983-803420233532PT Rev. bioét. 2023; 31: e3532PT 1-14 3 Bioética aplicada aos cuidados paliativos: questão de saúde pública dro 1. Síntese dos artigos incluídos na revisão integrativa Autores; ano Periódico Título Objetivo Método Conclusão Chaves e colaboradores; 2021 14 Revista Bioética “Cuidados paliativos: conhecimento de pacientes oncológicos e seus cuidadores” Verificar a percepção sobre cuidados paliativos, diretivas antecipadas de vontade e ordem de não reanimar de pacientes oncológicos e seus cuidadores, bem como a relação destes com os profissionais de saúde. Pesquisa descritiva quantitativa, realizada entre 2018 e 2019 no centro de alta complexidade em oncologia de um hospital universitário brasileiro. A amostra contou com 200 participantes (100 pacientes oncológicos e 100 cuidadores informais). Desconhecimento por parte dos participantes relacionado a questões ligadas ao findar da vida, bem como um conflito entre distanásia e reanimação obstinada. Os resultados também relatam a importância dos profissionais de saúde estarem capacitados ao exercício da função. Lima, Manchola- Castillo; 2021 15 Revista Bioética “Bioética, cuidados paliativos e libertação: contribuição ao ‘bem morrer’” Refletir sobre o morrer e as contribuições que a bioética tem dado ao assunto, seja por meio dos princípios tradicionais de autonomia e dignidade, seja por meio da defesa de uma nova categoria: a libertação, proposta pela bioética de intervenção com base em Paulo Freire. Pesquisa qualitativa, de abordagem hermenêutica, reflexiva, sociocrítica e analítica, cujo objetivo é demonstrar que a libertação pode contribuir para formar profissionais e pacientes mais críticos, comprometidos e livres, capazes de enfrentar um momento de tanta vulnerabilidade como é o da morte. O artigo defende que a adoção do conceito de libertação na reflexão bioética sobre cuidados paliativos pode contribuir no processo de “morrer bem”. Floriani; 2021 16 Cadernos de Saúde Pública “Considerações bioéticas sobre os modelos de assistência no fim da vida” Discutir os três campos assistenciais institucionalizados nas sociedades contemporâneas para cuidados no fim da vida e seus respectivos modelos de morte: a eutanásia/ suicídio assistido; a futilidade médica; e a calotanásia, fundamento do moderno movimento hospice. Ensaio que analisa de que modo os modelos propostos impactam a vida dos pacientes, bem como a fragilidade conceitual de algumas das bandeiras tradicionalmente utilizadas, como a da dignidade humana. É proposto um debate bioético acerca da necessidade de serem repensados alguns postulados, especialmente aqueles referentes à eutanásia. http://dx.doi.org/10.1590/1983-803420233532PT Apresenta e analisa a fundamentação ética e filosófica da calotanásia e suas implicações para a organização de boas práticas de cuidados no fim da vida. Esperandio, Leget; 2021 17 Revista Bioética “Espiritualidade nos cuidados paliativos: questão de saúde pública?” Apresentar o estado da arte sobre a temática, trazer breves orientações sobre como identificar necessidades espirituais e descrever quatro ferramentas úteis para esse cuidado. Revisão integrativa com estudos indexados no Portal de Periódicos Capes, na Biblioteca Virtual em Saúde, na SciELO e no PubMed, utilizando os termos “cuidado espiritual and cuidados paliativos”, “espiritualidade and cuidados paliativos” e “bioética”. Foram encontrados 18 estudos empíricos publicados entre 2003 e 2018: quatro dissertações, duas teses e 12 artigos. Reflete sobre os campos da bioética e da teologia pública e a correlação desses com a saúde pública. O texto apresenta recomendações para subsidiar políticas públicas voltadas à implementação do cuidado espiritual na assistência paliativa no Brasil. continua... Quadro 1. Síntese dos artigos incluídos na revisão integrativa Bioética aplicada aos cuidados paliativos: questão de saúde pública Maingué e colaboradores; 2020 18 Revista Bioética “Discussão bioética sobre o paciente em cuidados de fim de vida” Identificar fatores que influenciam a tomada de decisões de profissionais de saúde diante de pacientes em cuidados de fim de vida internados em unidades de terapia intensiva. Trata-se de pesquisa quantitativa realizada em dois hospitais paranaenses, entre março e maio de 2018, com amostra de 45 integrantes de equipe multiprofissional. O artigo deixa evidenciada a preocupação dos entrevistados a respeito da autonomia, proteção da dignidade e preservação da qualidade de vida dos pacientes e familiares por meio da decisão compartilhada. Porém, a tendência de obstinação terapêutica para cumprir o dever profissional mostrou necessidade de mais discussões e formação em cuidados paliativos para minimizar conflitos éticos Autores; ano Periódico Título Objetivo Método Conclusão E Maingué e colaboradores; 2020 18 Revista Bioética “Discussão bioética sobre o paciente em cuidados de fim de vida” Identificar fatores que influenciam a tomada de decisões de profissionais de saúde diante de pacientes em cuidados de fim de vida internados em unidades de terapia intensiva. Trata-se de pesquisa quantitativa realizada em dois hospitais paranaenses, entre março e maio de 2018, com amostra de 45 integrantes de equipe multiprofissional. O artigo deixa evidenciada a preocupação dos entrevistados a respeito da autonomia, proteção da dignidade e preservação da qualidade de vida dos pacientes e familiares por meio da decisão compartilhada. http://dx.doi.org/10.1590/1983-803420233532PT Porém, a tendência de obstinação terapêutica para cumprir o dever profissional mostrou necessidade de mais discussões e formação em cuidados paliativos para minimizar conflitos éticos. F Fusculim e colaboradores; 2022 19 Revista Bioética “Diretivas antecipadas de vontade: amparo bioético às questões éticas em saúde” Analisar o contexto atual de implementação das diretivas antecipadas de vontade no Brasil a partir do conhecimento de profissionais da área de medicina e enfermagem, bem como identificar potenciais contribuições da bioética para sua implementação. Trata-se de estudo exploratório, transversal, de abordagem quantitativa, com participação de 143 médicos e enfermeiros. Foi aplicado questionário eletrônico em plataforma on-line, objetivando analisar o contexto atual de implementação das diretivas antecipadas de vontade no Brasil a partir da percepção dos participantes. De acordo com o artigo, saber o que são as diretivas facilita o processo de deliberação com o paciente, sendo os conhecimentos em bioética fundamentais para embasar a decisão de profissionais no momento de escolher a melhor conduta a ser adotada. G Cecconello, Erbs, Geisler; 2022 5 Revista Bioética “Condutas éticas e o cuidado ao paciente terminal” Analisar a abordagem direcionada ao paciente terminal, uma vez que ética e cuidado precisam estar presentes em uma relação cautelosa, que suscita discussões e diferentes interpretações no meio médico. Trata-se de revisão integrativa de bibliografia realizada a partir de análise e integração de trabalhos encontrados em múltiplas bases de dados on-line, como SciELO, PubMed, portal do Conselho Federal de Medicina (CFM) e Código de Ética Médica (CEM), além de periódicos, plataformas e obras impressas. A busca incluiu publicações de 2001 a 2020, sendo selecionados 23 artigos. A abordagem das condutas e dos cuidados em pacientes terminais é de extrema importância, haja vista as possibilidades de e as discordâncias que estas podem causar. O artigo conclui que os cuidados paliativos devem ser prestados por equipe multiprofissional, de modo a promover o cuidado integral ao paciente e seus familiares, respeitando aspectos biopsicossociais e espirituais.i para minimizar conflitos éticos. F Fusculim e colaboradores; 2022 19 Revista Bioética “Diretivas antecipadas de vontade: amparo bioético às questões éticas em saúde” Analisar o contexto atual de implementação das diretivas antecipadas de vontade no Brasil a partir do conhecimento de profissionais da área de medicina e enfermagem, bem como identificar potenciais contribuições da bioética para sua implementação. Trata-se de estudo exploratório, transversal, de abordagem quantitativa, com participação de 143 médicos e enfermeiros. http://dx.doi.org/10.1590/1983-803420233532PT O artigo deixa explícito que é imprescindível a discussão dos dilemas apontados de modo que o processo de deliberação na alocação de recursos em saúde seja feito de maneira justa, como direciona o princípio da equidade, podendo utilizar referenciais bioéticos para dar maior transparência às decisões, a fim de que a população possa entendê-las como justas e justificadas.i O artigo estabelece que Autores; ano Periódico Título Objetivo Método Conclusão H Espíndola e colaboradores; 2018 20 Revista Bioética “Relações familiares no contexto dos cuidados paliativos” Revisão narrativa da literatura, ou seja, trata-se de levantamento não sistemático que não especifica palavras- chave, descritores, fontes e período de publicação. O artigo objetiva explorar possíveis implicações às relações familiares no fim da vida por meio de uma revisão narrativa de literatura. O artigo conclui que as mudanças e perdas no processo de adoecimento acometem o doente e também seus familiares, o que justifica a necessidade de assistência que oferte cuidados a esses indivíduos e dê suporte aos sofrimentos físico, psicossocial e espiritual a que estão sujeitos. I Souza e colaboradores; 2022 21 Revista Bioética “Dilemas éticos no direito de acesso aos cuidados paliativos na pandemia da covid-19” Elencar alguns dilemas éticos relacionados ao direito de acesso aos cuidados paliativos no contexto da pandemia da covid-19 e fomentar a discussão destes, visando a alocação de recursos de forma equitativa, segundo pressupostos doutrinários da legislação do Sistema Único de Saúde. Trata-se de reflexão teórica, utilizando como referencial a abordagem de responsabilização pela razoabilidade de Norman Daniels. O artigo deixa explícito que é imprescindível a discussão dos dilemas apontados de modo que o processo de deliberação na alocação de recursos em saúde seja feito de maneira justa, como direciona o princípio da equidade, podendo utilizar referenciais bioéticos para dar maior transparência às decisões, a fim de que a população possa entendê-las como justas e justificadas. Trata-se de pesquisa qualitativa, descritiva e exploratória envolvendo O artigo estabelece que a boa relação entre as três partes, por meio do Quadro 1. Continuação tores; ano Periódico Título Objetivo Método Conclusão dola e oradores; 20 Revista Bioética “Relações familiares no contexto dos cuidados paliativos” Revisão narrativa da literatura, ou seja, trata-se de levantamento não sistemático que não especifica palavras- chave, descritores, fontes e período de publicação. O artigo objetiva explorar possíveis implicações às relações familiares no fim da vida por meio de uma revisão narrativa de literatura. http://dx.doi.org/10.1590/1983-803420233532PT Foi aplicado questionário eletrônico em plataforma on-line, objetivando analisar o contexto atual de implementação das diretivas antecipadas de vontade no Brasil a partir da percepção dos participantes. De acordo com o artigo, saber o que são as diretivas facilita o processo de deliberação com o paciente, sendo os conhecimentos em bioética fundamentais para embasar a decisão de profissionais no momento de escolher a melhor conduta a ser adotada. G Cecconello, Erbs, Geisler; 2022 5 Revista Bioética “Condutas éticas e o cuidado ao paciente terminal” Analisar a abordagem direcionada ao paciente terminal, uma vez que ética e cuidado precisam estar presentes em uma relação cautelosa, que suscita discussões e diferentes interpretações no meio médico. Trata-se de revisão integrativa de bibliografia realizada a partir de análise e integração de trabalhos encontrados em múltiplas bases de dados on-line, como SciELO, PubMed, portal do Conselho Federal de Medicina (CFM) e Código de Ética Médica (CEM), além de periódicos, plataformas e obras impressas. A busca incluiu publicações de 2001 a 2020, sendo selecionados 23 artigos. A abordagem das condutas e dos cuidados em pacientes terminais é de extrema importância, haja vista as possibilidades de e as discordâncias que estas podem causar. O artigo conclui que os cuidados paliativos devem ser prestados por equipe multiprofissional, de modo a promover o cuidado integral ao paciente e seus familiares, respeitando aspectos biopsicossociais e espirituais.i Pesquisa Quadro 1. Continuação Quadro 1. Continuação F http://dx.doi.org/10.1590/1983-803420233532PT Bioética aplicada aos cuidados paliativos: questão de saúde pública 2018 20i dos cuidados paliativos” q pi p chave, descritores, fontes e período de publicação.i p revisão narrativa de literatura. oferte cuidados a esses indivíduos e dê suporte aos sofrimentos físico, psicossocial e espiritual a que estão sujeitos. I Souza e colaboradores; 2022 21 Revista Bioética “Dilemas éticos no direito de acesso aos cuidados paliativos na pandemia da covid-19” Elencar alguns dilemas éticos relacionados ao direito de acesso aos cuidados paliativos no contexto da pandemia da covid-19 e fomentar a discussão destes, visando a alocação de recursos de forma equitativa, segundo pressupostos doutrinários da legislação do Sistema Único de Saúde. Trata-se de reflexão teórica, utilizando como referencial a abordagem de responsabilização pela razoabilidade de Norman Daniels. Ideia de morte como um processo natural e inevitável De acordo com o artigo A 14, os avanços tecnoló- gicos e científicos da era pós-Revolução Industrial foram responsáveis por uma mudança no padrão das doenças populacionais, um aumento da expectativa de vida e o consequente crescimento da parcela idosa da população. Esse cenário de transição demográfica e epidemiológica está asso- ciado à diminuição das doenças infectocontagiosas e ao aumento da incidência de doenças crônico- -degenerativas, que hoje correspondem a 70% de todas as mortes, totalizando 41 milhões de mortes por ano no mundo todo. Essas mortes estão rela- cionadas a um grupo de doenças cardiovasculares, endocrinológicas, osteoarticulares e neoplásicas. Nesse contexto as pendências do paciente estariam sanadas e haveria boa relação entre o paciente e sua família junto aos profissionais de saúde. Tais ações, contudo, exigem uma obser- vação sensível das condições em que o paciente lida com sua morte, de forma a evitar expectati- vas frustradas, e dos aspectos culturais em que o paciente está inserido, especialmente quando se trata de sociedades pluralistas, nas quais há dife- rentes concepções do que seja uma boa morte 26. Em tese, os aparatos tecnológicos que consa- graram o progresso e mitigaram os índices de mor- talidade continuam desempenhando importante atuação nos diagnósticos e tratamentos precoces, além de travarem disputa acirrada para intervenção nas fases terminais da vida humana. Assim, diante da tecnologia, a morte tem sido entendida como um acidente indesejável, causado por uma doença que não foi detectada a tempo ou, ainda, por inca- pacidade na atuação dos médicos, e não como parte do ciclo natural da vida 22. Cuidados paliativos: conceitos e definiçõesi O artigo A 14 mostra que o conceito de cuida- dos paliativos surge na década de 1960 por meio da atuação da médica, assistente social e enfer- meira Cicely Saunders. Saunders se dedicou à construção de um trabalho voltado não apenas à assistência, mas também ao ensino e à pesquisa, resultando na criação do St. Christopher’s Hospice, em 1967, na cidade de Londres, o que inaugurou uma importante fase de expansão do movimento paliativista 25. Por se tratar de uma experiência sobre a qual poucos se dispõem a dialogar, o tema morte sempre se apresenta como um desafio. Entretanto, ao ser abordada e nomeada, a morte desperta curiosidade, desconforto, dúvidas, emoções e reflexões. Discussão profissionais, que se veem impotentes diante da pálida e gelada presença da morte. O olhar com naturalidade – defendido por Cicely Saunder, conforme será abordado no pró- ximo tópico –, despojado de preconceitos e medos, poderia redirecionar o modo como pensamos e dia- logamos sobre a morte, entendendo que, apesar de inevitável, ela pode ser uma boa morte. Nesse sentido, o artigo C 16 aborda os cuidados ao termo vida por meio de uma morte sem dor, em que os desejos do paciente são respeitados, com a possi- bilidade de morrer em casa, na presença de fami- liares e amigos, e sem sofrimento para o paciente, família e cuidadores. http://dx.doi.org/10.1590/1983-803420233532PT O artigo conclui que as mudanças e perdas no processo de adoecimento acometem o doente e também seus familiares, o que justifica a necessidade de assistência que oferte cuidados a esses indivíduos e dê suporte aos sofrimentos físico, psicossocial e espiritual a que estão sujeitos. a e oradores; 21 Revista Bioética “Dilemas éticos no direito de acesso aos cuidados paliativos na pandemia da covid-19” Elencar alguns dilemas éticos relacionados ao direito de acesso aos cuidados paliativos no contexto da pandemia da covid-19 e fomentar a discussão destes, visando a alocação de recursos de forma equitativa, segundo pressupostos doutrinários da legislação do Sistema Único de Saúde. Trata-se de reflexão teórica, utilizando como referencial a abordagem de responsabilização pela razoabilidade de Norman Daniels. O artigo deixa explícito que é imprescindível a discussão dos dilemas apontados de modo que o processo de deliberação na alocação de recursos em saúde seja feito de maneira justa, como direciona o princípio da equidade, podendo utilizar referenciais bioéticos para dar maior transparência às decisões, a fim de que a população possa entendê-las como justas e justificadas. pos, Silva , 2019 22 Revista Bioética “Comunicação em cuidados paliativos: equipe, paciente e família” Avaliar a comunicação na assistência paliativa e sua influência na relação entre equipe, paciente e família. Trata-se de pesquisa qualitativa, descritiva e exploratória envolvendo seis participantes abordados em instituição hospitalar de saúde pública da cidade de Suzano, no estado de São Paulo. Dados foram coletados em entrevistas semiestruturadas e avaliados com a técnica de análise de conteúdo. O artigo estabelece que a boa relação entre as três partes, por meio do diálogo, é fundamental, sendo, no entanto, necessário identificar outros fenômenos que estão além das habilidades comunicativas dos profissionais. Na perspectiva da bioética, a comunicação se destaca na assistência e cria vínculo que possibilita decisões compartilhadas. Continuação Título Quadro 1. Continuação Rev. bioét. 2023; 31: e3532PT 1-14 Bioética aplicada aos cuidados paliativos: questão de saúde pública Ideia de morte como um processo natural e inevitável Kovács 24 afirma que as emoções geradas no contexto da morte são as responsáveis por calar a curiosidade gerada e, como consequência, a maioria das pes- soas não conseguem experienciar tais sentimentos, fugindo deles e evitando o assunto. Diante da busca pela preservação da digni- dade dos pacientes e concessão de sustentação para o paciente e seus familiares na etapa ter- minal da vida, o artigo B 15 apresenta dois pila- res considerados fundamentais nos cuidados paliativos: a eficiência no controle da dor e de outros sintomas que surgem no estágio final de uma doença e a extensão do cuidado a aspectos psicológicos, sociais e espirituais do paciente e de seus familiares 27. O historiador Philippe Ariès 25 colabora com o tema ao afirmar que a morte traz consigo uma censura, uma ideia de proibido, fazendo com que se evite falar a respeito dela, pois é considerada malogro e ruína. No campo da saúde, há uma sensação de poder relacionado ao tema morte por parte dos profissionais, que em suas forma- ções são forjados para superá-la, combatê-la, vencê-la. Mas, de acordo com Kovács 24, essa sen- sação se desfaz na fragilidade humana desses Em um primeiro momento, tais cuidados foram direcionados especificamente a pacientes onco- lógicos, porém, na década de 2000, o conceito Rev. bioét. 2023; 31: e3532PT 1-14 7 http://dx.doi.org/10.1590/1983-803420233532PT Bioética aplicada aos cuidados paliativos: questão de saúde pública considerada promoção de óbito e contraria a ideia de prolongar a vida diminuindo o sofri- mento, ou seja, não se enquadra na beneficência nem na não maleficência. foi expandido, agregando enfermidades neuroló- gicas, cardíacas, renais e outras nas quais a vida possa ser ameaçada. Como versa o artigo G 5, por meio da inter- -relação profissional, com participação de equipe multidisciplinar e processo terapêutico, cuidados paliativos suscitam uma abordagem de cuidado integral ao paciente e seus familiares. Por con- seguinte, vários princípios devem ser observa- dos para que a assistência paliativa seja efetiva, como estabeleceu a OMS em 2002: Em síntese, fundamentados no direito à auto- nomia, à preservação da identidade social e à dig- nidade da vida e da morte, os cuidados paliativos fornecem assistência integral ao paciente sem proposta terapêutica de cura. Dilemas bioéticos em cuidados paliativosi A bioética emerge após a segunda metade do século XX diante dos extraordinários avanços tecnológicos nas áreas da biomedicina, genética, biologia molecular, transplantes e cuidados paliativos. Com a evolução da tecnologia e da infor- mática, expandiram-se as possibilidades de inter- venção nas condições de saúde do ser humano e, pouco a pouco, surgiram questões éticas deriva- das da aplicabilidade dessas ciências, que aumen- taram o poder de intervenção sobre a vida e a natureza 32,33. Em contrapartida, o artigo D 17 mostra que, infelizmente, a maioria dos pacientes conti- nua a receber assistência inadequada, focada apenas no propósito de cura. A assistência dis- pensada nos cuidados paliativos visa extin- guir o hiato entre conhecimento científico e humanístico, a fim de resgatar a dignidade da vida e a possibilidade de morrer como se deseja 27. No entanto, dados da OMS mostram que apenas 14% daqueles que têm indicação de trata- mento paliativo efetivamente o recebem 6,29. A bioética preocupa-se com o uso ético das novas tecnologias na área das ciências médicas e com a solução adequada dos dilemas morais por elas apresentados 34, possibilitando a aplicação dos princípios éticos aos problemas relativos à prática médico-assistencial 35. Assim busca entender e resolver os conflitos morais com implicações em práticas no âmbito do viver e da saúde, sempre respeitando os valores de uma sociedade demo- crática e secular 2.l Rememorando o artigo G 5, não se deve esque- cer o entendimento de que cuidados paliativos não buscam aligeirar a morte com a eutanásia e muito menos prolongar o sofrimento às portas da morte, mas sim possibilitar a ortotanásia, que signi- fica a boa morte, sem sofrimentos desnecessários naquele momento e/ou condição do paciente 30. Pesquisa A distanásia, conhecida como “morte ruim”, lenta e com intenso sofrimento, é pouco discutida no meio médico e refere-se a submeter o paciente a situação de sofrimento ou tortura, visando salvar sua vida por meio de tratamento sem efetividade. Já a eutanásia, que seria a indução à morte para que não haja sofrimento, é assunto de discussão acalorada em muitos países, como no Brasil, onde não é permitida, sendo condenada na prática médica por ferir os princípios da ética e da moral. Ideia de morte como um processo natural e inevitável Assim, controlam efetivamente a dor e proporcionam qualidade de vida física, emocional, psicológica e familiar, respeitando a autonomia e a autodeterminação do paciente em íntima relação com a bioética, principalmente no cuidado a pessoas idosas e em condição de dependência no término da vida 31. Cuidados paliativos consistem na assistência pro- movida por uma equipe multidisciplinar, que obje- tiva a melhoria da qualidade de vida do paciente e seus familiares, diante de uma doença que ameace a vida, por meio da prevenção e alívio do sofrimento, por meio de identificação precoce, avaliação impecável e tratamento de dor e demais sintomas físicos, psicológicos e espirituais 28. Dilemas bioéticos em cuidados paliativosi Nesse sentido, a eutanásia é Apropriar-se dos conflitos e dilemas morais existentes nas relações humanas, entendendo o ser humano como indivíduo possuidor de compe- tência cognitiva e moral, capaz de atuar de forma livre e se responsabilizar pelos seus atos, é o cerne da bioética. Portanto, uma investigação biomédica requer reflexão orientada pela ética, na medida em que a bioética, enquanto novo conheci- mento, combina características como humildade, responsabilidade e competência interdisciplinar e intercultural, além de potencializar o senso de Rev. bioét. 2023; 31: e3532PT 1-14 http://dx.doi.org/10.1590/1983-803420233532PT 8 Bioética aplicada aos cuidados paliativos: questão de saúde pública humanidade 8. A dignidade do ser humano se edi- fica à medida que princípios éticos são observa- dos nas tomadas de decisões e intervenções e nas relações interpessoais de todos os segmentos e pessoas envolvidas 36. devem ser tecnicamente habilitados na compe- tência científica e atualizados constantemente nos conceitos da bioética e humanização, a fim de pro- mover acolhimento e cuidado adequados.i Cuidados paliativos não se referem exclusi- vamente a atos médicos diante do fim da vida e discorrer sobre esse tema vai além de defender uma boa morte com uma equipe multiprofissio- nal humana e capacitada. Esse tema relaciona-se com vida, a como se vive e como se pode viver quando preservar ou salvar a vida no sentido bio- lógico não é mais possível 8.ii Dentre as várias tendências da bioética, destaca-se a bioética principialista, cujo foco é preo- cupação dos seres humanos participantes das pes- quisas para a área clínico-assistencial. Proposto por Beauchamp e Childress 37, esse modelo aplica o sistema de princípios, como a beneficência, a não maleficência, a autonomia e a justiça, proporcio- nando uma nova forma de dialogar com os profis- sionais da área da saúde. Por fim, o artigo E 18 defende que a equipe multidisciplinar deve reavaliar continuamente o quadro clínico do paciente, redefinindo objeti- vos do tratamento e considerando a assistência paliativa, principalmente quando há limita- ções da terapêutica modificadora da doença. Quando a enfermidade se encontra em fase avançada, com sinais de que a morte está próxima, define-se essa fase como a de cuidados de fim de vida. O princípio da beneficência refere-se ao atendimento dos interesses importantes e legí- timos dos indivíduos por meio da utilização de conhecimentos e habilidades técnicas que minimizem riscos e maximizem benefícios aos pacientes 31. Dilemas bioéticos em cuidados paliativosi Já a não maleficência estabelece que qualquer intervenção profissional deve evitar ou minimizar riscos e danos, o que implica não fazer o mal, em nenhuma hipótese. Este é consi- derado princípio fundamental da tradição hipo- crática da ética médica, que preconiza: cria o hábito de duas coisas: socorrer (ajudar) ou, ao menos, não causar danos 36. Relação entre família e cuidadoresii O artigo H 20 evidencia que a investigação sobre relações familiares no fim da vida envolve muita reflexão sobre conceitos, princípios e concep- ções acerca da instituição familiar. As famílias são o núcleo de uma sociedade, as primeiras redes de apoio dos indivíduos e as responsáveis pela formação, desenvolvimento e socialização dos sujeitos. Em seu cerne, são regidas por um con- junto de normas que orientam as relações estabe- lecidas entre os membros, organizando padrões de interação, bem como papéis e atribuições de cada indivíduo. A autonomia faz referência à liberdade de ação, considerando que as pessoas, possuindo razoável maturidade e consciência, são capazes de escolher e agir de acordo com seus próprios anseios, e o respeito a essa autonomia é indispensável, desde que não resulte em danos aos demais 33. Já o prin- cípio de justiça diz respeito à distribuição coe- rente e adequada de deveres e benefícios sociais, apoiando-se na equidade e salientando que situa- ções idênticas devem ser tratadas igualmente, e as que não são iguais, de forma diferente 34. Diante de adoecimento ou morte iminente, ocorrem mudanças na organização familiar e, consequentemente, nos papéis desempenhados pelos familiares, o que justifica a inclusão des- sas pessoas na assistência prestada pelas equi- pes de cuidados paliativos. A intervenção da equipe com a assistência paliativa auxilia tanto o paciente quanto os familiares a enfrentar a situação de terminalidade da vida. Isso contribui para amenizar o sofrimento físico, psicossocial e espiritual, uma vez que a família já passou por um longo trajeto de perdas simbólicas, como a dos papéis sociais, da autonomia e da identidade, além da perda real, ou seja, o óbito do enfermo. Bioética aplicada aos cuidados paliativos: questão de saúde pública Nesse contexto, os cuidados devem incluir uma comunicação honesta e clara, evitando omissões e conversas paralelas entre os profissionais na frente dos familiares. É preciso manter, sempre, a autono- mia e a dignidade dos doentes e seus familiares. componente físico, sofrimento psicológico, social e espiritual 42. A busca de sentido para o sofrimento se situa no contexto da história de vida da pessoa. Do mesmo modo, decisões éticas não são iso- ladas do contexto relacional e social da unidade de tratamento (paciente e familiares), como apontam Muldoon e King 43. Nas narrativas do enfermo estão presentes seus valores e crenças, aquilo que o ajuda a encontrar sentido e embasa suas decisões. O artigo G 5 afirma que o paciente e sua famí- lia são unidade fundamental de cuidado e que os familiares devem receber auxílio em todo o processo de enfermidade do paciente, evitando, assim, promoção de sofrimento que possa ter impactos negativos no curso da doença. A partici- pação da família, amigos e parceiros é de grande valia para a equipe de cuidados paliativos, visto que podem auxiliar nas necessidades, especificidades, angústias e vontades do paciente por conhecê-lo melhor que os profissionais. É esperado que a família sofra e adoeça junto com o paciente e, assim, esse sofrimento deve ser levado em conta, acolhido e incluído no tratamento 39. Ainda de acordo com o artigo E 18, a espirituali- dade é tida pela OMS como um dos componentes intrínsecos das boas práticas em assistência paliativa. Conforme a Resolução 41/2018, do Ministério da Saúde 44, que trata da implementação dos cuida- dos paliativos na atenção primária, essa reflexão é primordial. Com base nela, é possível integrar efetivamente a dimensão espiritual a esse tipo de assistência, melhorando a efetividade do cuidado tanto dos pacientes quanto daqueles que os acom- panham. Diferentes áreas são convidadas a desen- volver estudos sobre o tema, produzindo evidências que levem à transformação das políticas públicas e da práxis de uma bioética no cuidado, de modo que este seja integral e centrado na pessoa 43. Os familiares e/ou responsáveis também devem ser avisados sobre a não possibilidade de cura, com o consentimento do paciente, para que o apoio familiar, imprescindível, seja efetivo. Esse tipo de notícia traz complicações e, desde a primeira conversa até a fase de luto familiar, o suporte psicossocial torna-se fundamental. Equipe multiprofissional e interdisciplinari No artigo E 18, entende-se que a interdiscipli- naridade é absolutamente necessária na práxis do cuidado paliativo, ou seja, é preciso haver uma equipe multiprofissional que trabalha de maneira interligada na discussão e condução dos casos. O escopo de cuidados e o planejamento terapêutico devem envolver toda a equipe em busca da melhor qualidade de vida do enfermo e de seus familiares 37,38. Os profissionais envolvidos Rev. bioét. 2023; 31: e3532PT 1-14 9 http://dx.doi.org/10.1590/1983-803420233532PT Bioética aplicada aos cuidados paliativos: questão de saúde pública Comunicação com o paciente e a família Comunicação com o paciente e a família manifestações de vontade prévia que terão efeito quando o paciente não conseguir se expressar. Dividem-se em seis tipos: testamento vital, mandato duradouro, ordens de não reanimação, diretivas antecipadas psiquiátricas, diretivas para demência e plano de parto. Por fim, o artigo J 22 menciona um estudo quali- tativo sobre a importância da comunicação clara e direta entre a equipe e, sobretudo, com o paciente e sua família, evitando ao máximo termos técnicos e desconhecidos para o paciente. No âmbito da bioética, o diálogo com empatia e compaixão é estratégia e habilidade essencial para um bom trabalho em equipe, pois é necessário compreen- der as angústias e o sofrimento do paciente. Ainda, o trabalho mostra que a aceitação do diag- nóstico e prognóstico e a adesão ao tratamento são visivelmente influenciadas pela relação que se estabelece entre equipe e paciente e pela forma como os profissionais norteiam a comunicação. Entretanto, como já mencionado neste artigo, é utilizada a expressão DAV para se referir àquilo que se compreende como testa- mento vital, de acordo com a perspectiva de Dadalto 47. Sob esse aspecto, as DAV propiciam ao paciente o respeito à possibilidade de ter acolhida e respeitada sua vontade em seus momentos finais. Ao honrar a autodeterminação da pessoa enferma, assegura-se-lhe o direito de ser protago- nista de seu fim com dignidade quando, no uso da sua liberdade de dispor de seu bem-estar, escolher a suspensão dos tratamentos que somente prolonguem seu sofrer. Pode haver muitas orientações, porém o pri- meiro passo é sempre a escuta do paciente, visto que ela permite avaliar a melhor forma de passar informações em cada caso, diante do que se conhece do paciente e/ou da família. Essa ati- tude pode diminuir conflitos, angústias e fantasias originários tanto daquilo que o paciente e seus familiares imaginam sobre sua condição de saúde, quanto do próprio profissional, que faz uma leitura do outro a partir de seu mundo particular 49. No contexto brasileiro, o entendimento de que a aplicação das DAV ocorre principalmente no fim da vida justifica a associação do conhecimento em cuidados paliativos àquele sobre as DAV, estatisticamente significativa no estudo I 21. O res- peito à vontade do paciente, a suas determina- ções e anseios, inclui reconhecer e responder às necessidades dele e dos familiares com uma visão ampla e transdisciplinar. Comunicação com o paciente e a família Mais do que falar, saber ouvir direciona a comunicação à necessidade verossímil do paciente, permitindo-lhe expressar seus desejos. Isso demonstra respeito pelo outro, ratificando sua percepção sobre a própria saúde e a prerro- gativa de estabelecer os limites daquilo de que deseja estar ciente. É certo que todo paciente tem o direito de saber, mas nem todos têm essa necessidade, e quem deixará claro até onde o profissional deve ir na comunicação é o próprio paciente 47-49. Reconhecem-se, aqui, as conquistas da tecno- logia médica, porém com a ressalva de que deve haver uma transição gradual e equilibrada entre tentativas legítimas de manter a vida, quando há chances reais de recuperação, e a abordagem paliativa, controle de sintomas, sem nunca descon- siderar a dimensão da finitude humana 48. Ainda, o artigo I 21 considera que o direito sanitário, além de objetivar a promoção, prevenção e recuperação da saúde dos indivíduos por meio de um conjunto de normas jurídicas, também abarca a preocupação ética com os temas que interessam à saúde. Zelar pela ética na saúde requer constantes discussões sobre critérios da alocação de recursos em todas as esferas públicas e o estabelecimento de políticas que possam impactar na saúde indi- vidual ou coletiva. As considerações éticas devem ser consonantes à ética universal e aos valores da sociedade, visando à conciliação de interesses e buscando primordialmente a manutenção da dig- nidade da pessoa humana. Além disso, há a relevância de respeitar, na inter- locução, o tempo do paciente e da família, a fim de que possam entender o diagnóstico, o prognóstico e o cuidado proposto. Espiritualidade diante dos cuidados paliativosi Segundo o artigo G 5, os cuidados paliativos sur- giram no Brasil por volta de 1980, mas foi somente a partir do final da década de 1990 que passaram por considerável crescimento. Gradativamente, a prá- tica paliativa ganhou espaço nos cenários de saúde brasileira, mas esse tipo de cuidado ainda é bas- tante mistificado e desconhecido por grande parte da população, inclusive por profissionais de saúde. Versa o artigo E 18 que, no contexto dos cuidados paliativos, a espiritualidade é fonte de sentido para a experiência da doença, produzindo sensação de bem-estar e qualidade de vida, sendo recurso de coping (enfrentamento), suporte para enfermos e familiares e meio de desenvolvimento e ressig- nificação da vida. Entretanto, destaca-se que pro- fissionais da saúde têm dificuldade em identificar necessidades espirituais e atendê-las, e um estudo identificou que menos de 15% dos pacientes inter- nados tiveram tais demandas atendidas ou recebe- ram apoio psicológico 40,41. Cabe destacar, ainda, que o Brasil não dispõe de normativas legais específicas voltadas aos cui- dados paliativos. Apesar disso, o Código de Ética Médica (CEM), elaborado pelo Conselho Federal de Medicina (CFM) 45, assegura alguns pontos, como a legitimidade da ortotanásia, as diretivas anteci- padas de vontade (DAV) e o reconhecimento da medicina paliativa como campo de atuação 46. Pesquisa É importante ressaltar que, no contexto dos cuidados paliativos, o paciente enfrenta um sofri- mento multifatorial que extrapola os limites do corpo físico, manifestando-se em crises e con- flitos espirituais não tratáveis com medicamen- tos e que podem agravar a percepção da dor. Cicely Saunders nomeou essa experiência como “dor total”, argumentando que engloba, além do Outra questão importante são as DAV, discutidas no artigo F 19, nas quais são estabelecidos limites terapêuticos que devem ser respeitados caso o paciente não possa mais se expressar em algum momento. As DAV não se referem apenas a desejos de fim de vida, pois são também entendidas como Rev. bioét. 2023; 31: e3532PT 1-14 http://dx.doi.org/10.1590/1983-803420233532PT 10 Bioética aplicada aos cuidados paliativos: questão de saúde pública Bioética aplicada aos cuidados paliativos: questão de saúde pública responder por si –, familiares e equipe de saúde multiprofissional, de maneira interdisciplinar. A fase final da vida é conhecida como aquela em que o processo de morte ocorre de forma irre- versível e o prognóstico de morte pode ser definido em dias, semanas, meses ou mesmo anos. Nesse caminho, os cuidados paliativos se tornam indispen- sáveis e complexos em razão do objetivo de suprir a demanda de atenção específica e contínua ao doente e sua família, prevenindo e/ou reduzindo o sofrimento, com o intuito de ampliar a quali- dade de vida e assegurar uma morte digna. A con- dução de cada caso sempre deve ser discutida e acordada entre paciente – quando em condições de Este artigo mostra, de maneira resumida, a bioética aplicada a cuidados paliativos, modalidade que está cada vez mais em voga, sendo entendida como uma necessidade real e crescente. Assim, entender os conceitos teóricos desse tipo de cuidado e sobretudo sua práxis faz com que ele seja aplicado da melhor maneira possível, sem ferir os princípios éticos que correspondem ao cuidado, deixando o paciente (e seus familiares) em evidência quanto ao empode- ramento e protagonismo de sua própria história. Referências 1. Santos CE, Campos LS, Barros N, Serafim JA, Klug D, Cruz RP. Palliative care in Brasil: present and future. Rev Assoc Med Bras [Internet]. 2019 [acesso 18 out 2023];65(6):796-800. DOI: 10.1590/1806-9282.65.6.796 1. Santos CE, Campos LS, Barros N, Serafim JA, Klug D, Cruz RP. Palliative care in Brasil: present and future. Rev Assoc Med Bras [Internet]. 2019 [acesso 18 out 2023];65(6):796-800. DOI: 10.1590/1806-9282.65.6.796 2. Freitas EEC, Schramm FR. Argumentos morais sobre inclusão/exclusão de idosos na atenção à saúde. Rev. bioét. (Impr.) [Internet]. 2013 [acesso 18 out 2023];21(2):318-27. Disponível: https://bit.ly/49kLMSe 3. Mori M. A bioética: sua natureza e história: capacitação para comitês de ética em pesquisa [Internet]. Vol. 1. Brasília: Ministério da Saúde; 2006 [acesso 18 out 2023]. Disponível: https://bit.ly/40pqMWi 4. Moritz RD, Lago PM, Souza RP, Silva NB, Meneses FA, Othero JCB. Terminalidade e cuidados paliativos na unidade de terapia intensiva. Rev Bras Ter Intensiva [Internet]. 2008 [acesso 18 out 2023];20(4):422-8. DOI: 10.1590/S0103-507X2008000400016 4. Moritz RD, Lago PM, Souza RP, Silva NB, Meneses FA, Othero JCB. Terminalidade e cuidados paliativos na unidade de terapia intensiva. Rev Bras Ter Intensiva [Internet]. 2008 [acesso 18 out 2023];20(4):422-8. DOI: 10.1590/S0103-507X2008000400016 5. Cecconello L, Erbs EG, Geisler L. Condutas éticas e o cuidado ao paciente terminal. Rev. bioét. (Impr.) [Internet]. 2022 [acesso 18 out 2023];30(2):405-12. DOI: 10.1590/1983-80422022302536PT 5. Cecconello L, Erbs EG, Geisler L. Condutas éticas e o cuidado ao paciente terminal. Rev. bioét. (Impr.) [Internet]. 2022 [acesso 18 out 2023];30(2):405-12. DOI: 10.1590/1983-80422022302536PT 6. World Health Organization. National cancer control programmes: policies and managerial guidelines [Internet]. 2ª ed. Geneva: World Health Organization; 2002 [acesso 18 out 2023]. Disponível: https://bit.ly/3wRitEj 6. World Health Organization. National cancer control programmes: policies and managerial guidelines [Internet]. 2ª ed. Geneva: World Health Organization; 2002 [acesso 18 out 2023]. Disponível: https://bit.ly/3wRitEj 7. Gutierrez BAO, Barros TC. O despertar das competências profissionais de acompanhantes de idosos em cuidados paliativos. Rev Temát Kairós Gerontol [Internet]. 2012 [acesso 18 out 2023];15(4):239-58. DOI: 10.23925/2176-901X.2012v15iEspecial12p239-258 7. Gutierrez BAO, Barros TC. O despertar das competências profissionais de acompanhantes de idosos em cuidados paliativos. Rev Temát Kairós Gerontol [Internet]. 2012 [acesso 18 out 2023];15(4):239-58. DOI: 10.23925/2176-901X.2012v15iEspecial12p239-258 8. Wittmann-Vieira RW, Goldim JR. Bioética e cuidados paliativos: tomada de decisões e qualidade de vida. Acta Paul Enferm [Internet]. 2012 [acesso 18 out 2023];25(3):334-9. DOI: 10.1590/S0103-21002012000300003li 8. Wittmann-Vieira RW, Goldim JR. Bioética e cuidados paliativos: tomada de decisões e qualidade de vida. Acta Paul Enferm [Internet]. Considerações finais Cuidados paliativos são indicados a todos os pacientes cuja doença ameace a continuidade de vida por qualquer diagnóstico (não apenas oncológico), com prognóstico de terminalidade, seja qual for a idade e a qualquer momento em que esses pacientes tenham expectativas ou necessidades. Rev. bioét. 2023; 31: e3532PT 1-14 11 http://dx.doi.org/10.1590/1983-803420233532PT Bioética aplicada aos cuidados paliativos: questão de saúde pública Referências 2012 [acesso 18 out 2023];25(3):334-9. DOI: 10.1590/S0103-21002012000300003 9. OMS divulga recursos para lidar com flagrante escassez de serviços de cuidados paliativos de qualidade. Organização Pan-Americana da Saúde [Internet]. 5 out 2021 [acesso 29 nov 2023]. Disponível: https://bit.ly/46VKr1F 9. OMS divulga recursos para lidar com flagrante escassez de serviços de cuidados paliativos de qualidade. Organização Pan-Americana da Saúde [Internet]. 5 out 2021 [acesso 29 nov 2023]. Disponível: https://bit.ly/46VKr1F Pesquisa 10. Santos CE, Caldas JMP, Serafim JA, Barros N, Pereira AC, Capra MEZ et al. Palliative care in Brazil: with a view to future needs? Int Arch Med. [Internet] 2017 [acesso 18 out 2023]; 10(148): 1-9. DOI: 10.3823/2418 10. Santos CE, Caldas JMP, Serafim JA, Barros N, Pereira AC, Capra MEZ et al. Palliative care in Brazil: with a view to future needs? Int Arch Med. [Internet] 2017 [acesso 18 out 2023]; 10(148): 1-9. DOI: 10.3823/2418 11. Burlá C, Py L. Cuidados paliativos: ciência e proteção ao fim da vida. Cad Saúde Pública [Internet]. 2014 [acesso 18 out 2023];30(6):1-3. DOI: 10.1590/0102-311XPE020614 12. Mendes KS, Silveira RCCP, Galvão CM. Revisão integrativa: método de pesquisa para a incorporação de evidências na saúde e na enfermagem. Texto Contexto Enferm [Internet]. 2008 [acesso 18 out 2023];17(4):758-64. DOI: 10.1590/S0104-07072008000400018 13. Minayo MCS. Pesquisa social: teoria, método e criatividade. Petrópolis: Vozes; 2002. 14. Chaves JHB, Angelo Neto LM, Tavares VMC, Tuller LPS, Santos CT, Coelho JAPM. Cuidados paliativos: conhecimento de pacientes oncológicos e seus cuidadores. Rev. bioét. (Impr.) [Internet]. 2021 [acesso 18 out 2023];29(3):519-33. DOI: 10.1590/1983-80422021293488 Rev. bioét. 2023; 31: e3532PT 1-14 http://dx.doi.org/10.1590/1983-803420233532PT 12 Bioética aplicada aos cuidados paliativos: questão de saúde pública 15. Lima MA, Manchola-Castillo C. Bioética, cuidados paliativos e libertação: contribuição ao “bem morrer”. Rev. bioét. (Impr.) [Internet]. 2021 [acesso 18 out 2023];29(2):268-78. DOI: 10.1590/1983-80422021292464ii 16. Floriani CA. Considerações bioéticas sobre os modelos de assistência no fim da vida. Cad Saúde Pública [Internet]. 2021 [acesso 18 out 2023];37(9):e00264320. DOI: 10.1590/0102-311X00264320i 17. Esperandio ME, Leget C. Espiritualidade nos cuidados paliativos: questão de saúde pública? Rev. bioét. (Impr.) [Internet]. 2020 [acesso 18 out 2023];28(3):543-53. DOI: 10.1590/1983-80422020283419i 18. Maingué PCPM, Sganzerla A, Guirro UBP, Perini CC. Discussão bioética sobre o paciente em cuidados de fim de vida. Rev. bioét. (Impr.) [Internet]. 2020 [acesso 18 out 2023];28(1):135-46. DOI: 10.1590/1983- 80422020281376 19. Fusculim ARB, Guirro UBP, Souza W, Corradi-Perini C. Diretivas antecipadas de vontade: amparo bioético às questões éticas em saúde. Rev. bioét. (Impr.) [Internet]. 2022 [acesso 18 out 2023];30(3):589-97. Referências DOI: 10.1590/1983-80422022303552PT 20. Espíndola AV, Quintana AM, Farias CP, München MAB. Relações familiares no contexto dos cuidados paliativos. Rev. bioét. (Impr.) [Internet]. 2018 [acesso 18 out 2023];26(3):371-7. DOI: 10.1590/1983-80422018263256 21. Souza MT, Martins TCR, Moreira AS, Costa EF, Pessalacia JDR. Dilemas éticos no direito de acesso aos cuidados paliativos na pandemia da covid-19. Rev Bioét Derecho [Internet]. 2022 [acesso 18 out 2023];55:123-45. DOI 10.1344/rbd2022.55.36164 22. Campos VF, Silva JM, Silva JJ. Comunicação em cuidados paliativos: equipe, paciente e família. Rev. bioét. (Impr.) [Internet]. 2019 [acesso 18 out 2023];27(4):711-8. DOI: 10.1590/1983-80422019274354 23. Pereira RA, Alves-Souza RA, Vale JS. O processo de transição epidemiológica no Brasil: uma revisão de literatura. Revista Científica FAEMA [Internet]. 2015 [acesso 18 out 2023];6(1):99-108. DOI: 10.31072/ rcf.v6i1.322 24. Kovács MJ. Bioethics oncerning life and death. Psicol USP [Internet]. 2003 [acesso 18 out 2023];14(2):115-67. DOI: 10.1590/S0103-65642003000200008 25. Ariès P. História da morte no Ocidente. São Paulo: Saraiva; 2003. 25. Ariès P. História da morte no Ocidente. São Paulo: Saraiva; 2003. 26. Walter T. Historical and cultural variants on the good death. BMJ [Internet]. 2003 [acesso 18 out 2023];327:218-20. DOI: 10.1136/bmj.327.7408.218 27. Rocha AR, Buonicore GP, Silva AC, Pithan LH, Feijó AGS. Declaração prévia de vontade do paciente terminal: reflexão bioética. Rev. bioét. (Impr.) [Internet]. 2013 [acesso 18 out 2023];21(1):84-95. DOI: 10.1590/ S1983-80422013000100010 28. Camargos MG, Paiva CE, Barroso EM, Carneseca EC, Paiva BSR. Understanding the differences between oncology patients and oncology health professionals concerning spirituality/religiosity: a cross-sectional study. Medicine [Internet]. 2015 [acesso 18 out 2023];94(47):2145. DOI: 10.1097/MD.0000000000002145 29. Cervellin AF, Kruse MHL. Espiritualidade e religiosidade nos cuidados paliativos: conhecer para governar. Esc Anna Nery Rev Enferm [Internet]. 2014 [acesso 18 out 2023];18(1):136-42. DOI: 10.5935/1414- 8145.20140020 30. World Health Organization. Op. cit. p. 15-6. Tradução livre. 30. World Health Organization. Op. cit. p. 15-6. Tradução livre. 31. Paiva FCL, Almeida JJ Jr, Damásio AC. Ética em cuidados paliativos: concepções sobre o fim da vida. Rev. bioét. (Impr.) [Internet]. 2014 [acesso 18 out 2023];22(3):550-60. DOI: 10.1590/1983-80422014223038 32. Byock I. Principles of palliative medicine. In: Walsh D, editor. Palliative medicine. Philadelphia: Saunders Elsevier; 2009. p. 33-41. 33. Costa RS, Santos AGB, Yarid SD, Sena ELS, Boery RNSO. Reflexões bioéticas acerca da promoção de cuidados paliativos a idosos. Saúde Debate [Internet]. 2016 [acesso 18 out 2023];40(108):170-7. DOI: 10.1590/0103- 1104-20161080014 34. Koerich MS, Machado RR, Costa E. Ética e bioética: para dar início à reflexão. Texto Contexto Enferm [Internet]. Referências 2005 [acesso 18 out 2023];14(22):106-10. DOI: 10.1590/S0104-07072005000100014 34. Koerich MS, Machado RR, Costa E. Ética e bioética: para dar início à reflexão. Texto Contexto Enferm [Internet]. 2005 [acesso 18 out 2023];14(22):106-10. DOI: 10.1590/S0104-07072005000100014 35. Nunes L. Ética em cuidados paliativos: limites ao investimento curativo. Rev. bioét. (Impr.) [Internet]. 2009 [acesso 18 out 2023];16(1):41-50. Disponível: https://bit.ly/47xVi3a 35. Nunes L. Ética em cuidados paliativos: limites ao investimento curativo. Rev. bioét. (Impr.) [Internet]. 2009 [acesso 18 out 2023];16(1):41-50. Disponível: https://bit.ly/47xVi3a Rev. bioét. 2023; 31: e3532PT 1-14 13 http://dx.doi.org/10.1590/1983-803420233532PT 13 Bioética aplicada aos cuidados paliativos: questão de saúde pública Bifulco VA, Caponero RA. Cuidados paliativos: conversas sobre a vida e a morte na saúde. Barueri: Manole; 2015. A comunicação no final da vida; p. 107-1. 49. Bifulco VA, Caponero RA. Cuidados paliativos: conversas sobre a vida e a morte na saúde. Barueri: Manole; 2015. A comunicação no final da vida; p. 107-1. 14 Rev. bioét. 2023; 31: e3532PT 1-14 http://dx.doi.org/10.1590/1983-803420233532PT Mario Angelo Cenedesi Júnior – Doutorando – mariocenedesi@outlook.com 0000-0001-8441-8321 Correspondência Rua do Professor, 600, ap. 64, Jardim Irajá CEP 14020-280. Ribeirão Preto/SP, Brasil. Recebido: 23.7.2023 Revisado: 18.10.2023 Aprovado: 19.10.2023 Pesquisa Pesquisa Mario Angelo Cenedesi Júnior – Doutorando – mariocenedesi@outlook.com 0000-0001-8441-8321 Correspondência Rua do Professor, 600, ap. 64, Jardim Irajá CEP 14020-280. Ribeirão Preto/SP, Brasil. Recebido: 23.7.2023 Revisado: 18.10.2023 Aprovado: 19.10.2023 http://dx.doi.org/10.1590/1983-803420233532PT Bioética aplicada aos cuidados paliativos: questão de saúde pública 36. Clotet J. A. Bioética: uma aproximação. 33ª ed. Porto Alegre: EdiPUCRS; 2003. 37. Beauchamp TL, Childress JF. Princípios de ética biomédica. 4ª ed. São Paulo: Loyola; 2002.ii 38. Crippa A, Lufiego CAF, Feijó AGS, De Carli GA, Gomes I. Aspectos bioéticos nas publicações sobre cuidados paliativos em idosos: análise crítica. Rev. bioét. (Impr.) [Internet]. 2015 [acesso 18 out 2023];23(1):149-60. p. 150. DOI: 10.1590/1983-80422015231055 39. Santos FS. Cuidados paliativos: diretrizes, humanização e alívio de sintomas. São Paulo: Athei 40. Vianna MLGS, Silva DSN. Existem cuidados paliativos sem interdisciplinaridade? In: Corradi-Perini C, Sperandio MRG, Souza W, organizadores. Bioética e cuidados paliativos. Curitiba: Prismas; 2016. p. 65-80. 41. Matsumoto DY. Cuidados paliativos: conceito, fundamentos e princípios. In: Carvalho RT, Parsons HA, organizadores. Manual de cuidados paliativos ANPC [Internet]. 2ª ed. São Paulo: Academia Nacional de Cuidados Paliativos; 2012 [acesso 18 out 2023]. p. 23-30. Disponível: https://bit.ly/3GqGhUb 42. Paiva CE, Paiva BSR, Yennurajalingam S, Hui D. The impact of religiosity and individual prayer activities on advanced cancer patients’ health: is there any difference in function of whether or not receiving palliative anti-neoplastic therapy? J Relig Health [Internet]. 2014 [acesso 18 out 2023];53(6):1717-27. DOI: 10.1007/ s10943-013-9770-6 43. Muldoon M, King N. Spirituality, health care, and bioethics. J Relig Health [Internet]. 1995 [acesso 18 out 2023];34(4):329-50. DOI: 10.1007/bf02248742 44. Brasil. Ministério da Saúde. Resolução nº 41, de 31 de outubro de 2018. Dispõe sobre as diretrizes para a organização dos cuidados paliativos, à luz dos cuidados continuados integrados, no âmbito Sistema Único de Saúde (SUS). Diário Oficial da União [Internet]. 23 nov 2018 [acesso 6 dez 2023]. Disponível: https://bit.ly/3uNfq2U 45. Conselho Federal de Medicina. Código de Ética Médica: Resolução CFM nº 2.217, de 27 de setembro de 2018, modificada pelas Resoluções CFM nº 2.222/2018 e 2.226/2019 [Internet]. Brasília: CFM; 2019 [acesso 29 nov 2023]. Disponível: https://bit.ly/40IcAqv 46. Pessini L, Bertachini L. Humanização e cuidados paliativos. São Paulo: Loyola; 2004. 47. Dadalto L. Testamento vital. 4ª ed. Indaiatuba: Foco; 2018. 47. Dadalto L. Testamento vital. 4ª ed. Indaiatuba: Foco; 2018. 48. Machado S, Reis-Pina P, Mota A, Marques R. Morrer num serviço de medicina interna: as últimas horas de vida. Medicina Interna (Lisboa) [Internet]. 2018 [acesso 18 out 2023];25(4):286-92. DOI: 10.24950/rspmi/ original/90/4/2018 48. Machado S, Reis-Pina P, Mota A, Marques R. Morrer num serviço de medicina interna: as últimas horas de vida. Medicina Interna (Lisboa) [Internet]. 2018 [acesso 18 out 2023];25(4):286-92. DOI: 10.24950/rspmi/ original/90/4/2018 49.
https://openalex.org/W4386347763	https://link.springer.com/content/pdf/10.1007/s10936-023-10008-x.pdf	English	null	The Access to Grammatical Number in Spanish Children and Adults	Journal of psycholinguistic research	2,023	cc-by	8,601	Journal of Psycholinguistic Research (2023) 52:2499–2515 https://doi.org/10.1007/s10936-023-10008-x Journal of Psycholinguistic Research (2023) 52:2499–2515 https://doi.org/10.1007/s10936-023-10008-x Abstract In Spanish, the plural form in plural dominant frequency pairs, like “diente/dientes” [tooth/ teeth], occurs more frequently than the corresponding singular form. On the other hand, for the singular dominant frequency pairs such as “cometa/cometas” [kite/kites], the singular form is more common than the plural. The recognition of singular forms by adult readers is dependent on the dominance factor, while the identification of plural forms relies on the frequency of the stem. Given that age and reading experience may influence morphological processing of words, we investigate the representation of singulars and plurals in Spanish primary school children in Third Grade (8/9) and Sixth Grade (11/12) and adults through a lexical decision task. Though children’s lexical decisions were twice as slow as adults, the pattern of morphological processing was consistent across ages: dominant plural forms resulted in decision times that were comparable to those of non-dominant singular forms, while recognition of singular-dominant forms was quicker than recognition of plural non- dominant forms. It appears that singulars are accessed and stored in the lexical memory as separate entities, while plurals depend on their morphological closer relatives, in this case, the singular forms. Keywords Grammatical number · Lexical decision · Plural processing · Number acquisition · Dominant frequency * Alberto Dominguez adomin@ull.es Alberto Dominguez1 · Anthea Santos1 · Yang Fu1 Accepted: 10 August 2023 / Published online: 1 September 2023 © The Author(s) 2023 Accepted: 10 August 2023 / Published online: 1 September 2023 © The Author(s) 2023 1 Instituto Universitario de Neurociencia (IUNE), Universidad de La Laguna, 38205 Tenerife, Spain The Access to Grammatical Number in Spanish Children and Adults Alberto Dominguez1 · Anthea Santos1 · Yang Fu1 Introduction After 30 years of research on the processing of morphological number, the recognition of plurals is a topic in word recognition research that yields seemingly inconsistent results. A usual methodology to investigate how morphologically complex words are accessed and represented is the lexical decision task on the dominant and non-dominant singular and plural nouns. Pairs of words such as “diente/dientes” [tooth/teeth], in which the plural form is much more frequent than the singular one, are called plural dominants. On the contrary, pairs of singular dominant frequency words are those, such as “cometa/cometas” [kite/kites], in which the singular form is higher in frequency than the corresponding plural 56789) 3 23451 Journal of Psycholinguistic Research (2023) 52:2499–2515 2500 form. A well-established interpretation suggests that the correlation between dominance and reaction times indicates independent access and representation of singular and plu- ral forms. On the contrary, if the plural access depends on the singular forms, no reac- tion times advantage of plural dominant forms over the corresponding singular forms is expected. The classical Dual Route (race) Model by Schreuder and Baayen (1995) accounts for two routes in the process of singular and plural forms: a direct route which allows to accede and represent the whole word, and a decomposition route through which stem and affixes are identified. The velocity of both routes depends on several factors, among them, the fre- quency of the specific form, plural or singular, and the possibility to segment the word, not always possible, because Spanish, English or French, marks plural with a –s but they do not mark the singular, reason why it is not necessary to segment the singular. This asymmetry could be the reason for different recognition times for singular and plural forms of sin- gular-dominant pairs “cometa/cometas” [kite/kites] versus plural-dominant pairs “diente/ dientes” [tooth/teeth]. The dual route model would predict that singular forms of a singular dominant pair depend on their own frequency whereas plural forms could depend on the frequency of their corresponding singular. On the other side, the recognition of singular and plural forms of plural dominant items will not differ because the singular, e.g. “diente” [tooth] is accessed directly, employing long reaction times due to their low frequency. Plural recognition “dientes” [teeth] do not benefit on their higher frequency because their entry is done through a singular low-frequency form. 1 3 Introduction On the other hand, these results seem congruent with the probabilistic point of view of Baayen et al., (2007, 2011), who consider two types of information to be taken into account 1 3 Journal of Psycholinguistic Research (2023) 52:2499–2515 2501 by the linguistic processor to recognize compound words. Paradigmatic information basi- cally refers to the number of words using a given constituent (e.g. a morpheme). It refers to the well-known effect of morphological family size. Syntagmatic information, on the other side, refers to the probability that a series of letters is followed by another series in different words; for example, the probability that a final -s determines the plural of a word. In both cases, the meaning attributed to the word can be extracted with some degree of ambiguity, e.g. the suffix -s may be in Spanish a plural (“las compras -the purchases-”) or it can be the second person of a verb (“tu compras -you buy-”). From a probabilistic perspec- tive, predictions regarding the effects of surface and root frequencies warrant reinterpreta- tion, given that both variables provide information about the syntagmatic combinatorial properties of morphological structures inside the words (roots and suffixes), and not about the access and representation of the whole word. Against the traditional perspective, the surface frequency could be interpreted (Baayen et al., 2007) as an indicator of morpho- logical structure and not as the information of the whole word familiarity. In particular, the surface frequency of inflectional plurals would indicate the frequency with which a root combines with the suffix –s to form a plural. This interpretation, therefore, predicts surface frequency effects for plural dominant over its corresponding non-dominant forms as a sign of morphological computation of number. However, the results of previous experiments in Spanish (Dominguez et al. 1999) do not align well with this prediction. The absence of a dominant frequency effect for plural dominant items supports better the prediction of the classical point of view in which singulars are not decomposed whereas plurals are accessed through their corresponding singular form. From our point of view, these apparently contradictory results and interpretations could be partially reconciled by adopting a particular paradigm, such as is the singular/plural of the same stem, to achieve a well-controlled situation, manipulating only the surface fre- quency of each word. Introduction However, beyond controlling for spurious variables that might influ- ence morphological processing experiment outcomes, investigating the age and experience of readers with simple and compound words might be crucial. Introduction These are the results obtained in visual word recognition tasks in Spanish (Dominguez et al., 1999). The superficial mark “-s” in “diente-s” [teeth]) seems to be the information used by readers to recognize the plural of a noun, but the recognition of the complete word demands the identification of a stem, which corresponds to the singular form. Similar results had been obtained also in German (Baayen et al., 1997a) English (Sereno & Jong- man, 1997) French (New et al., 2004), Italian (Baayen et al., 1997b) and Dutch (Reifegerste et al., 2017) and not only in comprehension but also in production tasks (Beyersmann et al., 2015). These consistent effects would be supporting an access to the entry of plurals across the singular corresponding form, because in Spanish, English, French, Italian, Dutch or German languages the plural marks are added to the singular form. This panorama becomes complicated by the results of Gimenes et al. (2016) in a mega- study making an analysis of regression on the lexical decision times of 2954 French words, 1475 English words and 544 German words. A prediction of the authors, later confirmed by data, was that “…for the three languages, as plurals should activate singulars when- ever they are processed, base frequency should be the only predictor for singulars (no effect of surface frequency)” (p. 318). The regression data obtained by the authors showed that times for singulars were better predicted by the frequency of the base than by its sur- face frequency, because singulars are, in fact, the base of the corresponding plural forms, “diente” [tooth] is a part of “dientes” [teeth]. Reaction times corresponding to the plural words, on the contrary, were well predicted by the base frequency as for the surface fre- quency. At first glance, these regression results appear incompatible with the dominance paradigm data, because Gimenes et al. would predict that singulars dominant and singulars non-dominant would obtain similar reaction times but, actually, singulars dominant spend less time than singulars non-dominant to be recognized. However, these results can hardly be compared with those derived from the classical model, since no direct comparison of singular and plural forms of the same stem was carried out. Age and Inflectional Morphology Although the more frequent inflections are “–a”, for feminine words, and “–o”, for masculine words, there are some exceptions to this rule and many other endings of words marking gender (see Teschner & Russel, 1984). This ambiguity is also present in Spanish plurals, as it was remarked before, given that the final -s could be indicating not a plural but also the second person of verbs (e.g. las compras vs. tú compras [the purchases vs. you buy).f Given the differences due to age in a morphologically complex paradigm, as is plu- ral in German, and also due to previous data obtained in Spanish gender processing, an additional goal of this study is to know the impact of age over the Spanish morphological processing of number. We intended to contrast Spanish children of Third grade (aged 8–9), Sixth grade (aged 11–12) and Adults, in a lexical decision task, testing the hypothesis that less experience in reading will promote the extended influence of base frequency even in a language with a unique suffix of plural. We might think that the simplicity of the plural suffix makes the question posed by our research irrelevant: since identifying the plural is as simple as recognizing a letter at the end of the word, why not simply do this? However, this process may be slower than direct access to a representation of the complete plural word in memory, along with its associated meaning. The morphological process requires segment- ing the suffix, identifying the stem of the word, and accessing lexical memory through the singular form. Therefore, direct access could be faster and more efficient, but it requires prior storage of the plural form, which can only be achieved through repeated exposure to that plural form, or in other words, experience. This is why it is so important to measure lexical access at different ages to observe the effect that experience can have on the use of the direct or indirect morphological route. By using pairs of singular/plural words with the same stem (e.g., diente/dientes) that differ only in surface frequency, we control certain intervening variables in visual word recognition, such as family size or orthographic similarity. This allows us to focus directly on the difference between singulars and plurals when these forms are frequent or infrequent compared to their corresponding counterpart form. It is expected that the results of Dominguez et al. Age and Inflectional Morphology Reifergerste et al. (2017) proposed two new conditions that determine the recognition of morphologically complex words: the experience of readers, and the morphological com- plexity of the language (see also Clashen & Reifergerste, 2017; Lorenz et al., 2019, and Reifegerste et al., 2019 for morphological differences due to age). Unlike adults, young German readers (i.e. pregraduate students) do not exhibit the typical disparity in lexical decision times between singular dominant-singulars (e.g. cometa) and plural dominant- singulars (e.g. diente). The lexical decision times of young readers for singulars, as for plu- rals, depend on the cumulative frequency of singular and plural forms but will be shorter than those of plurals, because, in addition to being affected by the base frequency, they have to be segmented in stem and affix. In view of these results, the authors defended an extended morphological analysis for all lexical items (Taft, 2004; Taft & Forster, 1975) in young readers. Similar to Reifergerste et al. (2017), age-related differences were observed in our study of gender processing in Spanish. Santos et al. (2022) compared 8 and 11-year-old chil- dren with adults. The data revealed a distinction between feminine and masculine forms of masculine-dominant pairs, whereas for feminine-dominant pairs, no differences between masculine and feminine forms emerged. On the contrary, the adult processing of gender 1 3 3 Journal of Psycholinguistic Research (2023) 52:2499–2515 2502 in Spanish seems to be carried out through a direct route of processing, given that the rec- ognition of feminine and masculine words was influenced only by its surface frequency and not by the base frequency (Butterworth, 1983; Mannelis & Tharp, 1977; Rueckl et al., 1997). In terms of probabilistic interpretations as Baayen et al. (2007) account, however, the extended effect of surface frequency indicates not only the familiarity of the whole word but also the familiarity of the combination of root and affixes in compound words. However, the mechanisms underpinning children’s access differ for masculine and feminine genders, pointing towards a classical dual-access explanation. That is, a route of segmenta- tion would be used for feminine words, better predicted by the base frequency, whereas the direct route would operate with masculine words, better predicted by surface frequency. Differences with adults could be determined by the paradigmatic ambiguity of gender in Spanish. Stimuli and Design The experiment includes two within-participant variables: number (singular and plural) and dominance (dominant and not dominant), as well as a between-participants vari- able: group, (third grade, aged 8–9; sixth grade, aged 11–12; and university adults). Sixty pairs of singular-plural words were selected. Thirty of these pairs were singu- lar dominant “cometa/cometas” [kite/kites], and the other thirty were plural dominant “diente/dientes” [tooth/teeth]. Each participant saw fifteen singular words and fifteen plural words from singular dominant pairs, and fifteen singular words and fifteen plural words from plural dominant pairs. All the nouns and adjectives were extracted from the “Diccionario de Frecuencias del Castellano Escrito en niños de 6 a 12 años –Frequency dictionary of written Castilian for 6 to12 year old children” (Martinez-Martin & Garcia Perez, 2004), a specific dictionary of lexical frequency for children. The construction of this dictionary started with the selection of a sample of children from first to sixth grade in elementary school. The frequency of the words was obtained from the books read by these children over the academic year, including the textbooks. Consequently, it is a measure of the written frequency of words similar to that of adult frequency dictionaries. The participants also encountered a total of sixty pseudowords, with half being plu- rals ending in –s, and the other half singulars, typically ending in –a or –o, reflecting the common gender endings in Spanish. The pseudowords were formed changing one letter from an existing word in the initial, middle or final part (except for the last letter, which was never changed). All categories of stimulus were matched in lexical frequency (see Table 1). By design, plural forms always had one more letter than singular forms, due to the addition of the suffix -s to denote plurality in Spanish. Therefore, the difference between singular and plurals is necessarily confounded with length, but it must not compromise the main interest of this study on the interaction of the dominance of frequency and the number (see stimuli sets in the Appendix). Since the same stimuli were presented also to the adult group, a valid question might be whether the dominance relationships in frequency, as established using the children’s dictionary, would still hold when measured using a general adult dictionary. It can be seen in the Appendix the frequencies from the children’s dictionary (Frequency column) and those from a general adult frequency dictionary (NIM column) of Guasch et al. Age and Inflectional Morphology (1999) will be replicated in the group of adult participants: the dominant singular will produce shorter times than its correspond- ing plural. However, the dominant plural will produce the same reaction times as its cor- responding singular. Any other result should be interpreted according to the parameters of probabilistic models such as that of Baayen et al., (2007, 2011), especially if a dominance effect is obtained in the plural form over the singular. With respect to the children’s groups, it would be possible to find no influence of the dominant frequency variable, like in Ger- man. However, unlike German, Spanish simplicity of plural, (i.e. -s) could facilitates the recognition of the singular and the plural forms, receiving, in this case, the influence of the frequency of each form. 1 3 Journal of Psycholinguistic Research (2023) 52:2499–2515 2503 Table 1 Average values of frequency and length in each experimental condition Participants The experiment involved two groups of children: fifty third-graders aged 8–9 years (29 boys, 21 girls), and fifty-two sixth-graders aged 11–12 years. In third grade, students are acquiring fundamental skills in phonological decoding and reading fluency. By sixth grade, it is expected that students have developed a broader vocabulary based on read- ing more sophisticated texts. Comparing these two grade levels allows for monitoring progress to observe any changes that may occur in the use of the direct and indirect routes of morphological processing. None of the students were enrolled in special education or reading recovery pro- grams. Their academic performance was around the average for their respective grades. All participants had normal or corrected-to-normal vision and were Spanish native speakers. They voluntarily took part in the experiment. Parents were informed about the aim and procedures of the experiment via a letter, and they signed the corresponding consent forms. The two participating schools are located in downtown Santa Cruz and serve neighbor- hoods with a diverse socioeconomic mix. Most families in these schools are of high to middle socioeconomic status, with parents having at least a high school or university level education. The group of adult readers was composed of 72 participants between 18 and 30 years old (63 women and 9 men), undergraduate students from the Speech and Therapy degree of the University of La Laguna. All of them were informed of the procedure and general goal of the experiment and participated voluntarily. They received academic credits for their participation. Stimuli and Design (2004). The dominance relationships established with the frequencies from the chil- dren’s dictionary were preserved for all singular-plural pairs in the adult dictionary. The correlation between both measures was significant (r2 = 0.744, p < 0.0001), thereby con- firming that the stimuli selected for children were also appropriate for the adult group. 1 3 Table 1 Average values of frequency and length in each experimental condition Singular Plural Frequency Length Frequency Length Singular Dominant 560.02 5.96 186.95 7.13 Plural Dominant 188.7 6.26 566.7 7.56 1 3 Table 1 Average values of frequency and length in each experimental condition Singular Plural Frequency Length Frequency Length Singular Dominant 560.02 5.96 186.95 7.13 Plural Dominant 188.7 6.26 566.7 7.56 Table 1 Average values of frequency and length in each experimental condition Journal of Psycholinguistic Research (2023) 52:2499–2515 2504 1 3 Results Table 2 shows mean reaction times (RTs) calculated once errors were removed as well as mean percentage of accuracy for each experimental condition in each group of age. As for the trimming RTs data, response errors were first removed from the analysis (n = 447, 4% of trials). Further, trials with RTs faster than 0.01-quantile (385 ms, n = 100) or slower than 0.98-quantile (2617 ms, n = 204) were considered as absolute outliers and excluded from the RTs analysis (0.1%). Trials with RTs that deviated more than 2.5 SDs from the partici- pant’s mean latency (relative outlier trials) were also discarded (n = 264, 3%), leaving 9605 data points for main data analysis.f p y The data were analyzed with generalized linear mixed-effects (gLME) models (Baayen et al., 2008) using function glmer of the lme4 package v. 1.1–27 (Bates et al., 2015) in R version 4.1.0 (R Core Team, 2018). Specifically, Gamma family and identity link were used in models fitted to continuous data, and the Binomial family and logit link to binomial variables. Reaction times and accuracy data were predicted by Dominance (singular vs. plural dominance), Number (singular vs. plural) and Groups (the third- and sixth-grade children and adults). All categorical variables are sum coded (2-level predictor Domi- nance and Number, − 0.5, 0.5; 3-level predictor Groups, − 0.5, 0.5, 1), such that the effect estimates would be evaluated at the grand average across all predictors and thus can be interpreted as simple main effects. In all models, we applied a model trimming approach starting with a maximal model including random intercepts for participants and items, by- participants random slopes for Dominance and Number and their interactions and by-item random slopes for Groups. Subsequently, if a model failed to converge, we followed Barr (2013) and Barr et al. (2013) suggestions simplifying the maximal model, by removing correlations between random factors until nonsingular convergence was achieved. Procedure The participants were individually tested in a soundproof and isolated room either at their respective schools, or in the Faculty Laboratory for undergraduates. All participants per- formed a lexical decision task, a common method for studying lexical access (Meyer & Schvaneveldt, 1971, Ehri & Wilce, 1983; Perfetti & Hogaboam, 1975). The participants were instructed to respond as quickly and accurately as possible, identifying whether or not the presented stimulus was a word or a pseudoword. They should press the “SI” button with the index or heart finger of their right hand (labeled on the “L” key of the keyboard) when a word was presented, and press “NO” with their left index or heart finger when a nonword was presented (labeled on the “S” key of the keyboard). Stimuli were adminis- tered with the E-prime 2.0 program (Schneider et al., 2002) which also recorded the reac- tion times and errors. The stimuli were preceded by an asterisk as a fixation point during 1000 ms and appeared in the center of the screen in white letters on a black background, remaining there until the response of the participant. Latencies from the stimulus appear- ance to the participant response, and errors, were recorded. The experimental stimuli started to be delivered after the children completed 10 items of training and once the researcher was sure the participant had understood the instruc- tions. The stimuli were randomized across participants and presented in the center of the screen with a 70-Hz refresh rate. The letters, presented in lower-case case Courier 16 font, appeared as white characters on a black background. Each character covered approximately 0.38° of visual angle from a distance of 60 cm. 1 3 Journal of Psycholinguistic Research (2023) 52:2499–2515 2505 Results Result- ing models were compared to the full model (with maximal random structure) using the Table 2 Mean reaction times, square error and percentage of accuracy obtained for the experimental condi- tions of the experiment Target word Singular Dominant Plural Dominant Singular Plural Singular Plural Groups Paseo (walk) Paseos (walks) Diente (tooth) Dientes (Teeth) Third Mean RT 1271 1346 1315 1332 SE 20.38 21.48 21.78 21.84 Accuracy (%) 96.5 94.6 92.3 95.2 Sixth Mean RT 1100 1185 1144 1172 SE 16.56 18.40 17.90 18.65 Accuracy (%) 98.1 96.3 98.3 95.6 Adults Mean RT 592 638 622 630 SE 4.58 5.92 5.07 5.56 Accuracy (%) 95.9 95.9 93.7 96.1 le 2 Mean reaction times, square error and percentage of accuracy obtained for the experimental condi- s of the experiment Journal of Psycholinguistic Research (2023) 52:2499–2515 2506 Analysis of RT data across groups Factors Dominance χ2(1) = 65.33, p < .0001* Number χ2(1) = 19.87, p < .0001* Group (Sixth grade) χ2(2) = 70,931.78, p < .0001* Dominance × Number χ2(1) = 121.42, p < .0001* Dominance × Group χ2(1) = 17.29, p < .0001* Number × Group χ2(2) = 11.54, p = .003 Number × Dominance × Group χ2(2) = 104.83, p < .0001*** Table 3 Chi-square values and probability for each of the factors of the experiment and interactions Table 3 Chi-square values and probability for each of the factors of the experiment and interactions Table 4 Pos-hoc comparisons to disentalgle the triple interaction between the three main factores Follow-up comparisons Plural and singular contrasts across dominance in each group Group Contrasts Plural dominant Singular dominant Estimate SE z p Estimate SE z p Third Plural versus Singular 13.17 6.26 2.104 0.106 83.48 7.05 11.845 < .0001 Sixth Plural versus Singular 12.29 6.88 1.786 0.148 73.48 8.89 8.262 < .0001 Adult Plural versus Singular − 3.49 6.86 -0.510 0.610 38.11 7.77 4.906 < .0001 Pos-hoc comparisons to disentalgle the triple interaction between the three main factores chi-square difference test and Bayesian’s information criterion (BIC), for which a lower value indicates better model fit (Kline, 2011). We called function car:Anova() from the car package to test the significance of effects and calculate p-value. Using chi-square instead of F value avoids issues in estimating the dominator degree of freedom in unbalanced design and is analogous to treating t-distribution as a z-distribution for the individual coefficients (Alday et al., 2017). 1 3 Journal of Psycholinguistic Research (2023) 52:2499–2515 2507 Table 5 Analysis of accuracy data across groups Factors Dominance χ2(1) = 1.05, p = .3 Number χ2(1) = 4.52, p = .03* Group (Sixth grade) χ2(2) = 6.89, p = .03* Dominance × Number χ2(1) = 1.91, p = .17 Dominance × Group χ2(1) = 0.86, p = .65 Number × Group χ2(2) = 6.16, p = .04* Number × Dominance × Group χ2(2) = 1.48, p = .48 Table 6 Follow-up comparisons for the differences between plural and singular in the three groups analysed Follow-up comparisons Groups × Number Group Contrasts Estimate SE z p Third Plural versus Singular − 0.14 0.36 − 0.38 .9 Sixth Plural versus Singular − 1.18 0.39 − 3.05 .007 Adult Plural versus Singular − 0.21 0.28 − 0.75 .9 75 ms difference between the plural and the singular forms in the group Third resulted in a statistically significant effect, as well as the 85 ms difference for group Sixth and the 46 ms difference for the group Adult.i 75 ms difference between the plural and the singular forms in the group Third resulted in a statistically significant effect, as well as the 85 ms difference for group Sixth and the 46 ms difference for the group Adult.i f Regarding the analysis of accuracy (see Table 5), the singular forms produced a signifi- cantly higher number of correct lexical decisions than the plural forms. The group of Sixth was more exact in his/her responses than the group of third and unexpectedly, also than the group of Adults. The interaction between number and group showed a higher number of right responses for singular than for the plural words in the group Third and also in the group of Adults, but on the contrary, in the group of Sixth, the highest correct responses corresponded to the plural words. This difference was confirmed in the post-hoc analyses as may be seen at Table 6. In summary, the groups of children of 8 and 11 years old, and adults show a similar pat- tern of results (see Fig. 1) with an interaction between dominance and number that support a different processing for singular and plural words. These experimental results replicate a previous study (Dominguez et al., 1999) carried out only with adults. Table 6 Follow-up comparisons for the differences between plural and singular in the three groups analysed Results Post-hoc comparisons were performed using the glht function in the multcomp package (Hothorn et al., 2015), with free method in order to take the correlation of the model parameters into account. The data and scripts to replicate the analyses are available at the Open Science Framework at https://osf.io/ka7g9/?view_only=787c16b7a0 5e4272a7ddf38828629b8d.if The analyses (Table 3) showed a significant effect of Number factor indicating that sin- gular words received shorter reaction times than plural words and the Dominance factor indicating that dominant gender produced shorter lexical decision. Also, the factor Group produced significant differences given the group Third produces longer reaction times the group Sixth and the last produces slower responses than the adult group. Both factors inter- acted significantly, indicating a greater difference between singular and plural when the singular is dominant (69 ms) than when the plural is dominant (18 ms). This influence of number dominance on the number variable was confirmed in the triple significant interac- tion across groups. The Table 4 show the follow up comparisons between plural and sin- gular for plural dominant pairs revealing a non-significant difference of 17 ms in the group Third, a non-significant difference of 28 ms in the group Sixth and also a non-significant difference of 8 ms in the adults group. On the contrary, for the plural dominant pairs the 1 3 2507 Journal of Psycholinguistic Research (2023) 52:2499–2515 Table 5 Analysis of accuracy data across groups Factors Dominance χ2(1) = 1.05, p = .3 Number χ2(1) = 4.52, p = .03* Group (Sixth grade) χ2(2) = 6.89, p = .03* Dominance × Number χ2(1) = 1.91, p = .17 Dominance × Group χ2(1) = 0.86, p = .65 Number × Group χ2(2) = 6.16, p = .04* Number × Dominance × Group χ2(2) = 1.48, p = .48 Table 6 Follow-up comparisons for the differences between plural and singular in the three groups analysed Follow-up comparisons Groups × Number Group Contrasts Estimate SE z p Third Plural versus Singular − 0.14 0.36 − 0.38 .9 Sixth Plural versus Singular − 1.18 0.39 − 3.05 .007 Adult Plural versus Singular − 0.21 0.28 − 0.75 .9 Journal of Psycholinguistic Research (2023) 52:2499–2515 Discussion Two groups of children from the Third and Sixth grades of Elementary School and a group of Adults were tested reading plural and singular words of the same stem. The participants were asked to make a lexical decision of singular and plural words from singular dominant pairs and plural dominant pairs. Dominance was defined by a higher frequency of the sin- gular or plural words of the same lexeme: “dientes-diente” [tooth-teeth] is a plural-domi- nant pair and “paseo-paseos” [walk-walks] is a singular-dominant pair. 3 3 Journal of Psycholinguistic Research (2023) 52:2499–2515 2508 500 600 700 800 900 1000 1100 1200 1300 Singular Dominant Plural Dominant Singular Dominant Plural Dominant Singular Dominant Plural Dominant Third Sixth Adults Singular Plural Fig. 1 Evolution of the lexical decision times (and standard error) on the words of singular dominant pairs and plural dominant pairs from the third grade (left) and the sixth grade (center) to the adult’s group (right). Note the superficial frequency determine the difference between singular and plural in the singular-domi- nant pairs across ages but not in the plural-dominant pairs Fig. 1 Evolution of the lexical decision times (and standard error) on the words of singular dominant pairs and plural dominant pairs from the third grade (left) and the sixth grade (center) to the adult’s group (right). Note the superficial frequency determine the difference between singular and plural in the singular-domi- nant pairs across ages but not in the plural-dominant pairs Lexical decisions of children were two times slower than those of adults (see Fig. 1), but the pattern of children’s reaction times behaved as in the adult group and also as in previ- ous research (Dominguez et al., 1999): plurals of plural-dominant pairs produced similar lexical decision times as singulars of plural-dominant pairs, providing evidence that plu- ral recognition could be dependent on the accumulative frequency of the lexeme. On the contrary, the singular of singular-dominant pairs speeds the responses due to its own high frequency, whereas plurals of these pairs are probably recognized through previous recog- nition of the stem (i.e. the singular form), slowing down its reaction time. 1 3 Discussion The rich morphology of the German language compels young readers to give priority to the stem, probably by identifying it as a significant segment which, in combination with a few suffixes allows to obtain the meaning of thousands of morpho- logically complex words found in the first time. However, the Spanish inflectional system has only one suffix for the plural: “–s”. When children and adults encountering a plural, they are likely to segment it to access the base form, the singular, regardless of the plu- ral’s frequency. In view of the results, the age of participants does not seem to be a suffi- cient condition to explain by itself the prevalence of cumulative frequency of singulars and plurals over the surface frequency found by Reifergerste et al. (2017), because even when the age of the participants of our study was notably lower, our results differ substantially of those. Consequently, the simplicity of Spanish, with a unique plural ending, facilitates the identification of the morphological structure of the word in the case of plurals but the frequency of the plural is not taken in account when recognizing the singular form. Addi- tionally, the ambiguity of the orthographic mark of the plural, which can also function as a verbal form, does not influence a different pattern of morphological structure learning. Note that this ambiguity does not exist when the words appear inside sentences. The word “compras –purchases-” can be a noun and then it will be preceded by an article, i.e. “las compras –the purchases-”, or it can be a verb, and then it will be preceded by a second person pronoun, i.e. “tú compras –you buy-”, or by the omission of the same, also valid in Spanish. In any case, when the word “compras” appears alone, it does not matter whether the final -s corresponds to a noun or a verb, as morphological segmentation is applicable in both cases. Therefore, it would produce stem-dependent reaction times as well.l An additional issue raised in light of these results is the flexibility and adaptability of the morphological system of processing. One interesting point arising from these results is the efficiency of a reading system that requires two different procedures—direct access and the use of segmentation rules—to reach the meaning of words. Discussion The results support a system of recognition based on the classical dual procedure: a direct route to the whole word representation in memory which is affected by the frequency of items and an indirect subsidiary route that separates the lexeme and the suffix to accede to the morphological information of number. This indirect route, in the case of number processing, is only possible for plurals, because there is no singular suffix in Spanish. The recognition times of singulars do not differ of plurals when they are more unfamiliar than their corresponding plural but differs when the singular is more frequent than plural. The race between the direct and the indirect routes starts at once but the final result is deter- mined, in the case of singulars, by their superficial frequency, and in the case of plurals, by the process of segmentation and the frequency of the base morpheme (Schreuder & Baayen, 1995). Any other alternative explanation like those based on the interpretation of surface frequency as a sign of morphological treatment of the word (Baayen et al., 2007, 2011) does not seem plausible, given the absence of an advantage of plural dominant items over the singular nondominant forms.f No differences were found between the results obtained by the children and those of adults in the experiment. The age does not determine a change in the procedure used to recognize plural of words by young beginner readers and young adult skilled Spanish read- ers. This is a different result than in German, language in which the adults recognize singu- lar and plural differently than do young people (Reifergerste et al., 2017). Young German 1 3 Journal of Psycholinguistic Research (2023) 52:2499–2515 2509 readers recognize singulars and plurals through the stem, which accumulates the frequency of both, the singular and the plural forms, and as a result, singulars and plurals of the same stem received similar lexical decision times irrespectively of the dominance. The decisive factor in explaining results in the young German readers, offered by the authors, was an interaction between the accumulated experience with the language and the high number of suffixes of plurals. Discussion The choice between these procedures depends on several factors, as suggested by the probabilistic account of morphological processing (Baayen et al., 2007, 2011). These factors include the frequency, age, number of suffixes in the language, productivity and type of suffixes (inflections or derivations), and regularity among others. Our point of view is that most of these determi- nant variables may be reduced to a common factor: the experience of readers with the roots and the suffixes forming different words. The familiarity of words to the beginner read- ers is reduced due to their lack of experience. Morphological segmentation allows them to grasp the meaning of unfamiliar complex words through the more frequent and likely known morphemes that make up these words, as opposed to recognizing the entire word. Age of acquisition of base words is, in fact, earlier than the age of complex words (Davies et al., 2016), perhaps because these base words take part in many other words, and read- ers are more familiar with these configurations. Adult readers, on the contrary, have many more representations of whole words by repeated exposure, increasing the capacity of the direct access route. Therefore, age and frequency could be consolidated into a single factor, namely, the number of exposures to the words in the language.fii Conversely, morphological productivity and the number of suffixes in a specific para- digm, such as gender or number, enable speakers and/or readers to understand the meaning of a word via the stem and affixes, even if they do not recognize the entire word. Thus, we 1 3 3 Journal of Psycholinguistic Research (2023) 52:2499–2515 2510 posit that the language system toggles between direct and indirect procedures through a statistical exploration of the language. This view is in consonance with the probabilistic model of Baayen (2007, 2011), which points to paradigmatic and syntagmatic factors influ- encing the access and representation of morphologically complex words. According to the results presented here, the plurals are formed in such a simple way in Spanish, and some other languages, that the use of both procedures, direct and indirect, does not change throughout the life. Another factor simplifying the learning of plurals in Spanish is that the ‘-s’ of plural nouns repeats in adjectives, determinants, pronouns, and other functional elements that agree with the noun to conform to the sentence’s syntac- tic structure, such as ‘niños guapos’, ‘los niños’, ‘estos niños’, etc. Discussion ([handsome boys], [the boys], [these boys]). Therefore, the application of rules is useful not only at the lexical level but also at the syntactic level. The Dual Route Race Model (Schreuder & Baayen, 1995) remains the best option to accommodate the access and representation of number along the life in Spanish. Some conclusions could be derived of our results: Some conclusions could be derived of our results: • The plural form of each noun or adjective is identified through the singular form, which is directly recognized in the lexical memory • These two forms of recognition do not vary from childhood to adulthoodf • The developmental differences observed in other languages, such as German, seem to be dictated by the simple structure of plurals in Spanishifi • The identification of a suffix -s at the end of the word and some syntactic agreement cues in the sentence allow the child to discriminate the plural from the beginning of their reading experience. A limitation of this study could be the exclusive use of isolated words. Future develop- mental research could investigate whether the advantage of dominant singular frequency words disappears when they are inserted into a sentence context with determiners that agree with them in number. Appendix This table shows the singular and plural words used in the experiments. This table shows the singular and plural words used in the experiments. NOL = Number of letters; NIM = Adult frequency from NIM dictionary (Guasch et al., 13). NOL = Number of letters; NIM = Adult frequency from NIM dictionary (Guasch et al., 2013). NOL = Number of letters; NIM = Adult frequency from NIM dictionary (Guasch et al., 2013). Appendix Fund- ing was provided by Ministerio de Ciencia e Innovación (Grant No. PID2020-114246 GB-100). Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Com- mons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Appendix Frequency NIM NOL Plural Frequency NIM NOL Singular dominant List 1 lugar 1568.38 367.72 5 lugares 267.84 63.774 7 nombre 1466.15 271.083 6 nombres 589.11 79.939 7 cuento 382.64 48.852 6 cuentos 187.28 23.271 7 regalo 251.68 29.133 6 regalos 132.61 14.211 7 domingo 228.65 62.886 7 domingos 86.31 22.383 8 banco 192.72 46.72 5 bancos 47.3 36.594 6 camión 165.12 21.317 6 camiones 35.56 15.455 8 Journal of Psycholinguistic Research (2023) 52:2499–2515 2511 Frequency NIM NOL Plural Frequency NIM NOL cero 131.34 19.363 4 ceros 65.68 1.954 5 metal 84.84 18.297 5 metales 30.47 9.415 7 retrato 68.9 29.489 7 retratos 11.55 8.527 8 cometa 65 7.461 6 cometas 30.66 3.02 7 cómic 57.31 0.711 5 cómics 12.89 0.711 6 concierto 53.24 12.435 9 conciertos 18.81 8.349 10 chándal 50.25 1.244 7 chándales 3.13 0.178 8 puzzle 33.52 2.842 6 puzzles 3.57 0.355 7 319.98 62.63 6 101.51 19.20 7.2 List 2 número 1477.86 237.686 6 números 775.82 25.936 7 texto 488.24 66.971 5 textos 53.86 30.377 6 sueño 358.95 131.456 5 sueños 141.81 50.273 6 círculo 248.7 26.291 7 círculos 62.98 14.034 8 tesoro 195.09 15.455 6 tesoros 37.54 7.461 7 paseo 181.52 25.758 5 paseos 28.25 11.547 6 tejado 146.36 11.014 6 tejados 44.64 11.192 7 premio 116.97 33.752 6 premios 30.75 14.389 7 milagro 71.26 35.884 7 milagros 30.34 14.034 8 truco 68.34 9.06 5 trucos 33.2 6.928 6 saludo 61.22 13.501 6 saludos 15.54 3.553 7 taxi 53.99 22.383 4 taxis 3.75 4.619 5 televisor 53.18 20.429 9 televisores 6.1 4.619 11 molino 48.16 8.527 6 molinos 11.94 2.487 7 mantel 30.7 5.862 6 manteles 4.93 1.954 8 240.03 44.26 5.93 85.43 13.56 7.07 Plural dominant List 1 ojos 1950.8 482.833 4 ojo 300.9 70.879 3 animales 1370.28 108.54 8 animal 583.3 72.656 6 instrumentos 337.15 20.074 12 instrumento 181.57 28.245 11 peces 263.51 16.698 5 pez 145.04 15.455 3 huesos 216.69 35.173 6 hueso 96.5 15.1 5 caramelos 190.51 4.974 9 caramelo 48.75 2.842 8 centímetros 158.42 30.91 11 centímetro 35.92 7.461 10 grados 123.69 53.65 6 grado 58.64 29.84 5 arbustos 86.08 4.086 8 arbusto 31.81 1.421 7 rotuladores 69.1 0.711 11 rotulador 31.12 0.533 9 cuernos 68.83 10.836 7 cuerno 38.74 6.395 6 aplausos 54.3 12.257 8 aplauso 13.96 6.04 7 bombones 51.03 3.731 8 bombón 16.17 1.776 6 patines 50.14 1.421 7 patín 6.95 0.711 5 pétalos 32.08 6.217 7 pétalo 5.38 1.421 6 334.84 52.80 7.8 106.31 17.38 6.46 Journal of Psycholinguistic Research (2023) 52:2499–2515 2512 Frequency NIM NOL Plural Frequency NIM NOL List 2 Plural años 1434.7 1168.18 4 año 704.45 342.14 3 dientes 475.63 67.504 7 diente 59.63 6.928 6 huevos 329.09 29.311 6 huevo 191.58 20.251 5 zapatos 236.76 44.233 7 zapato 66.46 12.968 6 labios 191.28 109.428 7 labio 28.9 7.994 6 apóstoles 174.81 3.908 9 apóstol 15.43 3.731 7 juguetes 140.71 7.106 8 juguete 58.92 5.329 7 insectos 115.24 14.389 8 insecto 27.08 5.507 7 sellos 72.03 6.217 6 sello 28.49 4.974 5 cromos 68.94 3.02 6 cromo 12.89 1.954 5 guantes 58.15 9.948 7 guante 14.55 8.349 6 cereales 52.53 5.685 8 cereal 3.76 1.421 6 calzoncillos 50.51 8.172 12 calzoncillo 4.77 0.888 11 reptiles 47.16 5.329 8 reptil 13.56 3.198 6 talones 30.33 7.816 7 talón 5.23 3.731 5 231.85 99.34 7.33 82.38 28.62 6.06 Pseudowords Singular Pseudowords Number of letters Plural Pseudowords Number of letters babo 4 descos 6 piobla 6 parros 6 despeto 7 medres 6 teso 4 sories 6 gablante 8 tenostos 8 repla 5 últamos 7 lamello 7 pideos 6 planja 6 delatos 7 guetero 7 serpiuntes 10 potanco 8 hulados 7 cutado 6 chalos 6 maire 5 chimoneas 9 cala 4 vajeos 6 ponchu 6 cervecarías 11 infoliz 7 celabros 8 zita 4 acoros 6 pelato 6 brochizos 9 cantesto 8 crominales 10 asmo 4 guscazos 8 blonca 6 posisos 7 bafetón 7 ratrasos 9 azécar 6 tateajes 8 1 3 Journal of Psycholinguistic Research (2023) 52:2499–2515 2513 Pseudowords Singular Pseudowords Number of letters Plural Pseudowords Number of letters corbuta 7 teclos 6 depirte 7 zarrones 8 ostrella 8 licros 6 lunu 4 colmallos 9 llavo 5 cenozas 7 refaño 6 ostrenos 8 sapa 5 sóptimos 8 lluvoa 6 infocciones 11 5.97 7.63 Funding Open Access funding provided thanks to the CRUE-CSIC agreement with Springer Nature. References https://doi.org/10.1023/A:1023216326448i p y g p g Ehri, L. C., & Wilce, L. S. (1983). Development of word identification speed in skilled and less skilled beginning readers. Journal of Educational Psychology, 75, 3–18. https://doi.org/10.1037/0022-0663. 75.1.3 Gimenes, M., Brysbaert, M., & New, B. (2016). The processing of singular and plural nouns in English, French, and Dutch: New insights from megastudies. Canadian Journal of Experimental Psychology / Revue Canadienne De Psychologie Expérimentale, 70(4), 316–324. https://doi.org/10.1037/cep00 00074 asch, M., Boada, R., Ferré, P., & Sánchez-Casas, R. (2013). NIM: A web-based Swiss army knife to select stimuli for psycholinguistic studies. Behavior Research Methods, 45, 765–771. Hothorn T., Bretz F. & Westfall P. (2015) Multcomp: Simultaneous Inference in General Parametric M els. R Package Ver. 1.4–0. http://CRAN.R-project.org/package=multcomp. Kline, R. B. (2011). Principles and practice of structural equation modeling (3rd ed.). Guilford Press.f Lorenz, A., Zwitserlood, P., Regel, S., & Abdel Rahman, R. (2019). Age-related effects in compound pro- duction: Evidence from a double-object picture naming task. Quarterly Journal of Experimental Psy- chology, 72(7), 1667–1681. https://doi.org/10.1177/1747021818806491fi Mannelis, L., & Tharp, D. (1977). The processing of affixed words. Memory and Cognition, 5(6), 690–695. https://doi.org/10.3758/BF03197417 Martínez Martín, J. A., & García Pérez, E. (2004). Diccionario de frecuencias del castellano escrito en niños de 6 a 12 años. Servicio de Publicaciones. Universidad Pontificia de Salamanca. i Meyer, D. E., & Schvaneveldt, R. W. (1971). Facilitation in recognizing pairs of words: Evidence of a dependence between retrieval operations. Journal of Experimental Psychology, 90(2), 227–234. https://doi.org/10.1037/h0031564 New, B., Brysbaert, M., Segui, J., Ferrand, L., & Rastle, K. (2004). The processing of singular and plu- ral nouns in French and English. Journal of Memory and Language, 51, 568–585. https://doi.org/10. 1016/j.jml.2004.06.010 j j Perfetti, C. A., & Hogaboam, T. (1975). Relationship between single word decoding and reading compre- hension skill. Journal of Educational Psychology, 67(4), 461–469. https://doi.org/10.1037/h0077013f Reifegerste, J., Elin, K., & Clahsen, H. (2019). Persistent differences between native speakers and late bilin- guals: Evidence from inflectional and derivational processing in older speakers. Bilingualism: Lan- guage and Cognition, 22(3), 425–440. https://doi.org/10.1017/S1366728918000615lf Reifegerste, J., Meyer, A. S., & Zwitserlood, P. (2017). Inflectional complexity and experience affect plural processing in younger and older readers of Dutch and German. Language, Cognition and Neurosci- ence, 32(4), 471–487. https://doi.org/10.1080/23273798.2016.1247213 Rueckl, J. G., Mikolinski, M., Raveh, M., Miner, C. S., & Mars, F. (1997). Morphological priming, frag- ment completion, and connectionist networks. Journal of Memory and Language, 36, 382–405. References Alday, P. M., Schlesewsky, M., & Bornkessel-Schlesewsky, I. (2017). Electrophysiology reveals the neu- ral dynamics of naturalistic auditory language processing: Event- related potentials reflect continuous model updates. ENeuro. https://doi.org/10.1523/ENEURO.0311-16.2017f Baayen, R. H., Burani, C., & Schreuder, R. (1997a). Effects of semantic markedness in the processing of regular nominal singulars and plurals in Italian. In G. E. B. J. V. Marle (Ed.), Yearbook of morphology (pp. 13–34). Kluwer. Baayen, R. H., Dijkstra, T., & Schreuder, R. (1997). Singulars and plurals in Dutch: Evidence for a paral- lel dual-route model. Journal of Memory and Language, 37(1), 94–117. https://doi.org/10.1006/jmla. 1997.2509 Baayen, R. H., Milin, P., Đurđević, D. F., Hendrix, P., & Marelli, M. (2011). An amorphous model for mor- phological processing in visual comprehension based on naive discriminative learning. Psychological Review, 118(3), 438–481. https://doi.org/10.1037/a0023851 p g Baayen, R. H., Wurm, L. H., & Aycock, J. (2007). Lexical dynamics for low-frequency complex words: A regression study across tasks and modalities. The Mental Lexicon, 2, 419–463. https://doi.org/10.1075/ ml.2.3.06baaff Barr, D. J. (2013). Random effects structure for testing interactions in linear mixed-effects models. Fronti in Psychology, 4, 328. https://doi.org/10.3389/FPSYG.2013.00328fi Barr, D. J., Levy, R., Scheepers, C., & Tily, H. J. (2013). Random effects structure for confirmatory hypoth- esis testing: Keep it maximal. Journal of Memory and Language. https://doi.org/10.1016/j.jml.2012. 11.001f Bates, D., Mächler, M., Bolker, B. M., & Walker, S. C. (2015). Fitting linear mixed-effects models using lme4. Journal of Statistical Software, 67(1), 1–48. https://doi.org/10.18637/JSS.V067.I01 Beyersmann, E., Dutton, E. M., Amer, S., Schiller, N. O., & Biedermann, B. (2015). The production of sin- gular- and plural- dominant nouns in Dutch. Language, Cognition and Neuroscience, 30(7), 867–876. https://doi.org/10.1080/23273798.2015.1027236 1 3 1 3 3 Journal of Psycholinguistic Research (2023) 52:2499–2515 2514 Butterworth, B. (1983). Lexical representation. In B. Butterworth (Ed.), Language production: Develop- ment, writing and other language processes (pp. 257–294). Academic. Clahsen, H., & Reifegerste, J. (2017). Morphological processing in old-age bilinguals. In M. Libben, M. Goral, & G. Libben (Eds.), Bilingualism: A framework for understanding the mental lexicon (pp. 217– 248). Benjamins. j Davies, S. K., Izura, C., Socas, R., & Domínguez, A. (2016). Age of Acquisition and Imageability norms for base and morphologically complex words in English and in Spanish. Behavior Research Methods, 48(1), 349–365. https://doi.org/10.3758/s13428-015-0579-y ( ) p g y Dominguez, A., Cuetos, F., & Segui, J. (1999). The Processing of Grammatical Gender and Number in Spanish. Journal of Psycholinguistic Research, 28, 485-498. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Teschner, R. V., & Russell, W. M. (1984). The gender patterns of Spanish nouns: An inverse dictionary- based analysis. Hispanic Linguistics, 1, 115–132. References https:// doi.org/10.1006/jmla.1996.2489 g j Santos, A., Bermúdez-Margaretto, B., Alvarez-González, C. J., & Dominguez, A. (2022). The frequency of word gender as a variable for lexical access in Spanish. Linguistics, 60(4), 1149–1167. https://doi.org/ 10.1515/ling-2020-0054 Schneider, W., Eschman, A., & Zuccolotto, A. (2002). E-prime user’s guide. Psychology Software Tools Inc. Schreuder, R., & Baayen, R. H. (1995). Modeling morphological processing. In L. B. Feldman (Ed.), M phological aspects of language processing (pp. 131–154). Lawrence Erlbaum.l Sereno, J., & Jongman, A. (1997). Processing of English inflectional morphology. Memory and Cognit 25(4), 425–437. https://doi.org/10.3758/BF03201119f Sereno, J., & Jongman, A. (1997). Processing of English in 25(4), 425–437. https://doi.org/10.3758/BF03201119 Taft, M. (2004). Morphological decomposition and the reverse base frequency effect. The Quarterly Journal of Experimental Psychology Section A, 57(4), 745–765. https://doi.org/10.1080/02724980343000477i Taft, M., & Forster, K. I. (1975). Lexical storage and retrieval of prefixed words. Journal of Verbal Learning and Verbal Behavior, 14, 638–647. https://doi.org/10.1016/S0022-5371(75)80051-X 1 1 3 Journal of Psycholinguistic Research (2023) 52:2499–2515 2515 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. 3 1
https://openalex.org/W4390620001	http://www.revistas.espol.edu.ec/index.php/compendium/article/download/1213/1026	es		PROCESOS DE SUCESIÓN Y PERMANENCIA DE LAS PYMES FAMILIARES BASADAS EN EL MODELO TRIDIMENSIONAL O DE GERSICK	Compendium/Compendium (En línea)	2,023	cc-by	7,330	Revista Compendium: Cuadernos de Economía y Administración 2023, Vol 10, No.3,176-189 https://doi.org/10.46677/compendium.v10i3.1213 Artículo Académico PROCESOS DE SUCESIÓN Y PERMANENCIA DE LAS PYMES FAMILIARES BASADAS EN EL MODELO TRIDIMENSIONAL O DE GERSICK. SUCCESSION AND PERMANENCE PROCESSES OF FAMILY SMEs BASED ON THE THREE-DIMENSIONAL OR GERSICK MODEL Tamara Cristina Mogrovejo Pintado1 Palabras clave: Empresas familiares, Metodología tridimensional, Dinámica empresarial, Sostenibilidad empresarial 1 Resumen Las pymes familiares son clave en la economía de muchos países, incluyendo el cantón Cuenca. No solo impulsan el crecimiento económico y el empleo, sino que también representan la continuidad de tradiciones y valores familiares. Sin embargo, enfrentan desafíos únicos debido a la intersección de intereses empresariales y familiares. La sucesión y permanencia exitosa son fundamentales para su sostenibilidad y éxito a largo plazo. El objetivo de este estudio fue examinar las empresas familiares en el cantón Cuenca utilizando la metodología tridimensional (empresa, familia y propiedad) para comprender su dinámica y desafíos. La metodología implica la recopilación de datos de múltiples fuentes, como entrevistas y análisis de documentos, para evaluar aspectos relacionados con la gestión empresarial, las dinámicas familiares y la estructura de propiedad a una muestra de 329 empresas familiares. Los resultados fueron tabulados en SPSS. El estudio reveló que estas entidades en el cantón Cuenca enfrentan desafíos significativos en áreas como la gestión administrativa, la falta de políticas de sucesión claras y la necesidad de equilibrar las dinámicas familiares y empresariales, específicamente se identificaron áreas críticas para mejorar la sostenibilidad y el 30% señala problemas importantes en la relación entre la familia y la empresa y posiblemente caracterizada por conflictos significativos. Códigos JEL: M10; O21 Universidad Politécnica Salesiana, Facultad de Administración y Economía (Ecuador). E-mail: tmogrovejop@est.ups.edu.ec ORCID # https://orcid.org/0009-0003-1223-5886 Recibido: octubre 22 de 2023 Aceptado: diciembre 20 de 2023 177 Revista Compendium: Cuadernos de Economía y Administración Keywords: Family businesses, Three Circles Methodology, business dynamics, Business sustainability Family SMEs are key to the economy of many countries, including the Cuenca canton. They not only drive economic growth and employment, but also represent the continuity of family traditions and values. However, they face unique challenges due to the intersection of business and family interests. Successful succession and retention are critical to its long-term sustainability and success. The objective of this study was to examine family businesses in the Cuenca canton using three-dimensional methodology (business, family and property) to understand their dynamics and challenges. The methodology involves collecting data from multiple sources, such as interviews and document analysis, to evaluate aspects related to business management, family dynamics and ownership structure in a sample of 329 family businesses. The results were tabulated in SPSS. The study revealed that these entities in the Cuenca canton face significant challenges in areas such as administrative management, the lack of clear succession policies and the need to balance family and business dynamics, specifically critical areas were identified to improve sustainability and 30 % indicates important problems in the relationship between family and business and possibly characterized by significant conflicts. INTRODUCCIÓN Las pequeñas y medianas empresas (Pymes) familiares representan una parte significativa del tejido empresarial en todo el mundo. Estas empresas, a menudo fundadas con pasión y dedicación, se enfrentan a desafíos únicos a lo largo de su ciclo de vida, particularmente en lo que respecta a la sucesión y la permanencia en el mercado (Rave y Moreno, 2023). La transición de una generación a otra y la adaptación a un entorno empresarial en constante cambio son elementos cruciales para el éxito y la supervivencia de estas organizaciones. En este contexto, el modelo Tridimensional de Gersick se ha destacado como una herramienta fundamental para comprender y gestionar los procesos de sucesión y permanencia en las Pymes familiares (Pimentel, 2021). Este modelo, desarrollado por Leslie E. Gersick en la década de 1990, se basa en una profunda investigación y ofrece una perspectiva holística que abarca no solo la sucesión generacional, sino también las dinámicas de la empresa y la familia (Barbosa, 2020). Proporciona una hoja de ruta sólida para las Pymes familiares que buscan perdurar en el tiempo y prosperar a medida que pasan de una generación a la siguiente. Uno de los enfoques teóricos relevantes que ha ganado atención es el Modelo Tridimensional, también conocido como el Modelo de Gersick, desarrollado por Leslie Gersick (Araoz y Bringas, 2023). Este modelo se ha convertido en un marco conceptual influyente para analizar la evolución de las empresas familiares a lo largo del tiempo (Cunha et al., 2020). El Modelo Tridimensional de Gersick se caracteriza por su enfoque integral que abarca tres dimensiones interrelacionadas: la dimensión de la familia, la propiedad y de la gestión (Filippim, 2019). Estas dimensiones son esenciales para comprender cómo se desarrollan y gestionan las PYMES familiares a medida que atraviesan diferentes etapas de crecimiento y sucesión (De León, 2019). En la dimensión de la familia, se presta especial atención a la dinámica y los roles familiares, que desempeñan un papel fundamental en la toma de decisiones relacionadas con la empresa (Andrews, 2010; Business y Finance, 2023). Además, la comunicación y la resolución de conflictos dentro de la familia son aspectos cruciales que afectan la estabilidad y la continuidad del negocio. En la dimensión de la propiedad, se exploran cuestiones como la estructura de propiedad y la 178 gobernanza familiar. La forma en que se distribuye y gestiona la propiedad entre los miembros de la familia tiene un impacto directo en la planificación de la sucesión y en la sostenibilidad de la empresa (Arista y De la Garza Ramos, 2022; Velarde, 2020). En la dimensión de la gestión, se examina el liderazgo y la transición de roles clave en la empresa familiar. La preparación de sucesores y la gestión de la cultura organizacional son factores críticos para asegurar una transición exitosa y la continuidad del negocio (Güldenkoh y Silberg, 2019). Además del enfoque tridimensional de Gersick, otros investigadores han contribuido a este campo mediante la identificación de factores contextuales que influyen en los procesos de sucesión y permanencia de las PYMES familiares. Estos factores incluyen el entorno empresarial, las leyes y regulaciones, así como las condiciones económicas y sociales que rodean a la empresa familiar. En este sentido Ward es conocido por su trabajo en el campo de la sucesión en empresas familiares. Sus investigaciones han destacado la importancia de la planificación de la sucesión y la gestión de conflictos familiares en el éxito continuado de las PYMES familiares. El autor, resalta la crítica importancia de la sucesión adecuada en el éxito continuado de las empresas familiares, especialmente en PYMES. Sus investigaciones han iluminado aspectos esenciales, como la necesidad de planificar la sucesión minuciosamente a largo plazo para evitar conflictos internos y desafíos operativos. Además, ha subrayado la gestión efectiva de conflictos familiares en empresas con dinámicas complejas y ha abogado por la preparación de sucesores y líderes, enfatizando la capacitación necesaria para una transición sin problemas. También ha promovido la idea de una gobernanza familiar eficaz y ha destacado la importancia de un enfoque a largo plazo en la gestión, donde las decisiones basadas en la continuidad y el legado suelen ser más exitosas que las de corto plazo (De Visscher et al., 2011; Schuman et al., 2010). Revista Compendium: Cuadernos de Economía y Administración De Ciantis y Lansberg( 2020) ha investigado ampliamente la dinámica de la sucesión en empresas familiares. Sus contribuciones se centran en la preparación de sucesores y la gobernanza familiar, así como en la gestión de conflictos y la comunicación efectiva en este contexto. Su trabajo ha subrayado la necesidad de identificar y desarrollar líderes potenciales dentro de la familia empresarial (Molina et al., 2016). Ha defendido la idea de que la preparación efectiva de sucesores implica proporcionar a estos individuos las habilidades, la experiencia y la orientación necesarias para asumir roles de liderazgo de manera efectiva. Lansberg ha destacado que la preparación de sucesores no se limita únicamente a la capacitación técnica, sino que también se centra en el desarrollo de habilidades de gestión, liderazgo y toma de decisiones (Feliu & Lansberg, 2022). Por otra parte, Visscher et al (2011) ha abordado temas relacionados con la gestión de la propiedad en empresas familiares, incluida la planificación patrimonial y la estructura de propiedad. Sus investigaciones han proporcionado ideas valiosas sobre cómo la propiedad puede afectar la continuidad de la empresa. Uno de los aspectos fundamentales que Visscher y su equipo han abordado es la planificación patrimonial en empresas familiares. Han examinado cómo las decisiones sobre la propiedad y la planificación de la sucesión pueden tener un impacto directo en la estabilidad y la continuidad de la empresa (Betancourt et al., 2015). Su trabajo ha destacado la importancia de considerar no solo los aspectos financieros de la propiedad, como la distribución de acciones o participaciones, sino también los aspectos emocionales y relacionados con la identidad de la familia empresaria. Esto ha llevado a un enfoque más holístico de la planificación patrimonial que reconoce la complejidad de las empresas familiares ( García y García, 2023). Carlock y Ward (2001) han trabajado en conjunto y han desarrollado el concepto de gobernanza familiar en el contexto de las empresas familiares. Han explorado cómo la gobernanza efectiva puede influir en la toma de 179 Revista Compendium: Cuadernos de Economía y Administración decisiones y la gestión en estas empresas (Samsudin y Fuza, 2021). imparcial y efectiva (Moores y Barrett, 2005; Thomas, 2002). En lo que respecta a los Roles y Responsabilidades en empresas familiares, Carlock et al (2010) han enfatizado la importancia de establecer claramente quién desempeña qué funciones dentro de la empresa, tanto para los miembros de la familia que trabajan en ella como para aquellos que no tienen un papel directo en la gestión. Esta claridad en los roles y las responsabilidades es esencial para prevenir malentendidos y posibles conflictos que puedan surgir debido a expectativas poco definidas. El modelo tridimensional, también conocido como el modelo de Gersick o el modelo de la sucesión en tres fases, es una metodología desarrollada por la investigadora Leslie W. Gersick para comprender y analizar el proceso de sucesión en las empresas familiares (García, 2005). Este modelo se utiliza para describir cómo las empresas familiares atraviesan cambios y transiciones en la dirección y gestión a lo largo del tiempo. El modelo se basa en estudios de casos de empresas familiares y se ha convertido en una herramienta valiosa para entender el ciclo de vida de las empresas familiares y cómo se manejan las sucesiones (Molina et al., 2016). Sin embargo, sobre la transparencia y comunicación, los investigadores han subrayado que una gobernanza familiar exitosa se fundamenta en la apertura y la comunicación efectiva. Esto implica la necesidad de compartir información de manera proactiva y transparente entre los miembros de la familia y otros interesados en la empresa. La comunicación abierta contribuye a evitar malentendidos y a fomentar un ambiente de confianza en el que las decisiones se toman de manera informada y colaborativa (Węcławski y Żukowska, 2019). La Planificación de la Sucesión, otro aspecto crucial de la gobernanza familiar, está estrechamente vinculada a la continuidad y la estabilidad de la empresa. Tucker y Lind (2022) que han investigado cómo las estructuras de gobernanza pueden facilitar la transición de liderazgo de una generación a la siguiente. Esto involucra la identificación y preparación de sucesores, así como la creación de procesos y políticas que garanticen una sucesión sin problemas y la preservación del legado familiar en la empresa. En lo que concierne a conflictos y resolución, han explorado cómo las estructuras de gobernanza pueden proporcionar mecanismos formales para abordar y resolver disputas que puedan surgir en el seno de la familia empresaria o en la gestión de la empresa. Estos mecanismos pueden incluir la implementación de consejos de familia o comités de gobierno que se encarguen de mediar y resolver conflictos de manera La metodología del modelo tridimensional o de Gersick se compone de tres fases distintas: en esta primera fase como se aprecia en la figura 1, la empresa familiar se encuentra en una etapa estable y funcional bajo la dirección del fundador o el líder principal. En esta etapa, el líder suele ser altamente autocrático y toma la mayoría de las decisiones clave. Sin embargo, la preparación para la sucesión comienza a gestarse, aunque de manera no formalizada. Los futuros sucesores pueden estar involucrados en la empresa, adquiriendo experiencia y habilidades (Andrade, 2019; Molina, 2012). La Fase de Transición se desencadena por un evento desencadenante, como la jubilación del fundador o un cambio significativo en la empresa o la familia. Durante esta fase, se produce una transición de poder y responsabilidad desde el líder fundador hacia el sucesor o sucesores. Esto suele ser un proceso complejo y puede estar acompañado de tensiones y conflictos familiares. Se implementan estructuras de gestión más formales y se redefinen los roles de liderazgo (Teston y Filippim, 2018). En la Fase de Consolidación, la empresa familiar ha logrado consolidar la transición de manera exitosa. El sucesor o sucesores asumen roles de liderazgo plenos y efectivos, y la empresa funciona bajo una estructura de liderazgo más colaborativa y profesionalizada. Se establecen reglas y sistemas para facilitar la comunicación y 180 Revista Compendium: Cuadernos de Economía y Administración la toma de decisiones en la familia y la empresa (Dantas et al., 2020). Figura 1 Modelo tridimensional o de Gersick. METODOLOGÍA La investigación se desarrolla en varias etapas con el objetivo de abordar las preguntas de investigación planteadas. En primer lugar, se describe el diseño de la investigación, que se basó en un enfoque cuantitativo con alcance descriptivo. El marco teórico utilizado fue el modelo tridimensional, que considera las dimensiones de Empresa, Familia y Propiedad en las empresas familiares. En cuanto a los procedimientos, se llevaron a cabo varias etapas. Inicialmente, se realizó una revisión del estado del arte en el campo de la sucesión en empresas familiares. Para analizar y sintetizar la información recopilada durante esta revisión, se utilizaron métodos analíticossintéticos y deductivo-inductivos. Nota. Obtenido de Rodríguez (2019) El modelo de los tres círculos se enfoca en abarcar situaciones problemáticas de las empresas familiares y describe las tres estructuras principales que son: 1. Familia, 2 Propiedad, 3. Empresa. La intersección 4 indica que la familia; trabaje o no en la empresa; es propietario. La intersección 5 indica a quienes trabajan en la empresa, pero no son parte de la familia. En la intersección 6 están quienes trabajan en la empresa, son parte de la familia, pero no poseen el capital en la misma. Y la intersección 7 están quienes forman parte de la familia, trabajan en la empresa y poseen capital. La metodología del modelo tridimensional de Gersick se enfoca en el proceso de sucesión en las empresas familiares y cómo evoluciona con el tiempo. Reconoce que la sucesión es un proceso complejo que involucra no solo la transición de roles de liderazgo sino también cambios en la dinámica familiar y organizativa. Este modelo proporciona una forma de comprender las diferentes etapas de la sucesión y puede ser útil para las empresas familiares que buscan planificar y gestionar con éxito la transición de liderazgo de una generación a otra. La recolección de datos se realizó a través de una encuesta compuesta por 21 preguntas cerradas. Esta encuesta se administró mediante la plataforma de Formularios de Google y se dirigió a empresas familiares registradas en la Superintendencia de Compañías, Valores y Seguros de Ecuador durante el año 2021. La unidad de análisis para este estudio fueron 2198 mediana y pequeñas empresas del cantón Cuenca (INEC, 2023). En lo que respecta al diseño muestral, debido al tamaño de la muestra, se calculó por muestreo probabilístico para poblaciones finitas según la formula siguiente: 𝑛= 𝑛= 𝑁 (𝑍)2 (𝑝)(1 − 𝑝) 𝑒 2 (𝑁 − 1) + 𝑍 2 (𝑝)(1 − 𝑝) 2198 (1.96)( 0.5 )(1 − 0.5) 0.052 (2198 − 1) + 1.962 (0.5)(1 − 0.5) = 329 Dónde: n: Tamaño de la muestra e: Error de muestreo (e= 0,05) p: Probabilidad a favor (p= 0,5) q: Probabilidad en contra (q=0,5) Z: nivel de confianza (Z=1.96) N: Población= 2198 181 Revista Compendium: Cuadernos de Economía y Administración Las fuentes de datos principales son las respuestas proporcionadas por las empresas familiares a la encuesta. Además, se recurrió a información secundaria obtenida de bases de datos públicas, como la Superintendencia de Compañías, el Banco Central y el Servicio de Rentas Internas. Para el análisis estadístico de los datos recopilados, se utiliza el software SPSS. Una vez obtenidos los resultados se procede a presentarlos en forma de gráficos, tablas y diagramas que facilitaron la exposición de los hallazgos. RESULTADOS Conforme a los propósitos de la investigación, se procede a delinearse el perfil de las 18 compañías que componen el análisis de caso. A continuación, se organiza de manera sistemática la exposición de los hallazgos. Figura 2 Dimensión empresa Políticas empresariales 7% Planificación de sucesión 6% 7% 31% 36% 26% 42% 45% 21% Políticas de Sucesión Relación Empresa / Familia Estructura Organizacional Gestión Administrativa Edad de la Empresa 3% 6% 10% 11% 9% 4% 16% 22% 16% 20% 4% 22% 21% 32% 25% 71% 70% Muy Bajo Bajo Medio 58% Bueno 61% 0% 20% 40% 60% 80% En el análisis de la dimensión Empresa en el contexto de examen, se revelan importantes hallazgos que arrojan luz sobre diversos aspectos de su funcionamiento y gestión. Utilizando la metodología de los tres círculos, podemos desglosar estos resultados de manera más detallada. En primer lugar, la Edad de la Empresa muestra una distribución diversa en las categorías de madurez. El hecho de que el 25% se ubique en la categoría Bueno indica que una parte sustancial de las empresas familiares tiene una antigüedad adecuada en el mercado, lo que podría implicar estabilidad y experiencia. Sin embargo, el 32% en Medio sugiere que existen empresas relativamente jóvenes en la muestra, lo que podría plantear desafíos en términos de su consolidación y desarrollo. Además, el 21% en Bajo y el 22% en Muy Bajo reflejan la presencia de empresas con una antigüedad limitada, en una etapa incipiente de su crecimiento, lo que puede requerir una atención especial para su fortalecimiento y estabilización. Por otro lado, la Gestión Administrativa es un aspecto crítico que necesita mejoras sustanciales en la mayoría de las empresas familiares. El hecho de que solo el 4% obtenga una clasificación Bueno sugiere que muy pocas empresas tienen prácticas administrativas sólidas y eficientes. El 20% en Medio indica que la mayoría de las empresas tienen margen para mejorar sus procesos administrativos. Sin embargo, es alarmante ver que el 61% se encuentre en Bajo y el 16% en Muy Bajo, lo que refleja una preocupante falta de eficacia en la gestión administrativa, lo que puede impactar negativamente en la operación de las empresas. La Estructura Organizacional presenta una situación más alentadora, con el 58% en la categoría Bueno. Esto indica que muchas empresas familiares tienen una estructura bien definida y eficiente, lo que es fundamental para un funcionamiento organizado. Sin embargo, el 22% en Medio y el 16% en Bajo señalan que algunas empresas pueden necesitar revisar y mejorar sus estructuras. Es importante destacar que solo un pequeño 4% se ubica en Muy Bajo, lo que sugiere que la situación crítica en este aspecto es menos común. La Relación Empresa / Familia es otro aspecto de gran relevancia. La presencia de un 9% en la categoría Bueno indica que algunas empresas han logrado establecer una relación armoniosa entre la empresa y la familia propietaria. Sin embargo, el 70% en Medio y el 11% en Bajo resaltan que en muchas empresas existen oportunidades significativas para mejorar esta relación crucial. El 10% en Muy Bajo refleja que en algunas empresas la relación es 182 Revista Compendium: Cuadernos de Economía y Administración problemática, lo que podría afectar tanto la dinámica familiar como el funcionamiento empresarial. Las Políticas de Sucesión son un componente esencial para la continuidad de las empresas familiares. La presencia de solo un 6% en Bueno indica que algunas empresas tienen políticas de sucesión sólidas y bien definidas, lo que es alentador. Sin embargo, el 71% en Medio y el 3% en Bajo ponen de manifiesto que la mayoría de las empresas familiares necesitan mejorar y desarrollar políticas de sucesión más claras y efectivas. El 21% en Muy Bajo refleja una preocupante falta de atención a este aspecto crucial. La Planificación de Sucesión también requiere una atención significativa, ya que solo el 7% se encuentra en Bueno. El 45% en Medio y el 6% en Bajo indican que la mayoría de las empresas familiares deben trabajar en una planificación más efectiva de la sucesión. El 42% en Muy Bajo refleja una falta crítica de atención a este aspecto, lo que puede tener implicaciones a largo plazo para la continuidad de las empresas. Figura 3 Dimensión familia 36% 31% 26% Formación de nuevos miembros 7% 36% 31% 26% Transmisión de liderazgo 7% Grado de satisfacción con los demás familiares empleados 31% 26% 36% 7% 31% 36% 26% Manejo de conflictos 7% 36% 31% 26% Relación familiar 7% Relación entre generaciones 36% 31% 26% 7% 7% 26% 25% # de familiares en altos cargos 0% Muy Bajo Bajo 10% 20% Medio 30% 41% 40% 50% Bueno Finalmente, las Políticas Empresariales son una dimensión que también demanda mejoras considerables. La presencia de un 7% en Bueno sugiere que algunas empresas tienen políticas empresariales sólidas. Sin embargo, el 26% en Medio, el 36% en Bajo y el 31% en Muy Bajo indican que la mayoría de las empresas familiares deben mejorar de manera significativa en este aspecto, ya que las políticas empresariales son fundamentales para una gestión efectiva y la toma de decisiones. Utilizando la metodología de los tres círculos, se obtienen resultados que ofrecen una visión detallada de esta dimensión clave. El número de familiares en altos cargos refleja una distribución diversa en las categorías de madurez. El 25% en Bueno indica que algunas empresas han logrado una presencia adecuada de familiares en roles de liderazgo, lo que puede ser beneficioso para la continuidad y la cohesión. Sin embargo, el 41% en Medio sugiere que la mayoría de las empresas familiares aún tienen margen para aumentar la representación de familiares en cargos clave. El 26% en Bajo y el 7% en Muy Bajo indican que existe un desafío significativo en este aspecto para muchas empresas. La Relación entre generaciones es un aspecto crítico de la dinámica familiar en empresas familiares. El 26% en Bueno indica que algunas empresas han logrado establecer una relación armoniosa y colaborativa entre diferentes generaciones familiares. Sin embargo, el 7% en Medio sugiere que existen tensiones o desafíos en la relación entre generaciones en algunas empresas. El 31% en Bajo y el 36% en Muy Bajo señalan que la mayoría de las empresas familiares enfrentan problemas significativos en este aspecto, lo que puede afectar la cohesión y la toma de decisiones. La Relación familiar es otro componente crucial para la estabilidad y el éxito de la empresa familiar. El 7% en Bueno indica que algunas empresas tienen relaciones familiares sólidas y saludables dentro del entorno laboral. Sin embargo, el 26% en Medio indica que en muchas empresas existe espacio para mejorar las relaciones familiares. El 31% en Bajo y el 36% en Muy Bajo reflejan que la relación familiar es un área de preocupación en la mayoría de las empresas familiares, lo que puede dar lugar a conflictos y tensiones. 183 El Manejo de conflictos es un aspecto directamente relacionado con la relación familiar. El 7% en Bueno sugiere que algunas empresas han implementado efectivamente estrategias para resolver conflictos de manera constructiva. Sin embargo, el 26% en Medio y el 36% en Bajo indican que la mayoría de las empresas familiares enfrentan desafíos en la gestión de conflictos. El 31% en Muy Bajo refleja una preocupante falta de habilidades para abordar y resolver conflictos, lo que puede tener un impacto negativo en el ambiente laboral y la productividad. El Grado de satisfacción con los demás familiares empleados es un indicador importante de la armonía en la empresa familiar. El 7% en Bueno indica que algunas empresas han logrado un alto nivel de satisfacción entre los familiares empleados. Sin embargo, el 36% en Medio y el 31% en Bajo sugieren que en la mayoría de las empresas familiares existen oportunidades para mejorar la satisfacción de los empleados familiares. El 36% en Muy Bajo refleja una insatisfacción significativa entre los familiares empleados, lo que puede generar un ambiente laboral tenso y poco productivo. La Transmisión de liderazgo es fundamental para la continuidad de la empresa familiar. El 7% en Bueno indica que algunas empresas tienen procesos efectivos de transmisión de liderazgo. Sin embargo, el 26% en Medio y el 31% en Bajo indican que la mayoría de las empresas necesitan mejorar sus prácticas en este sentido. El 36% en Muy Bajo refleja una falta crítica de atención a la transmisión de liderazgo, lo que puede poner en riesgo la continuidad de la empresa. La Formación de nuevos miembros es un aspecto complementario a la transmisión de liderazgo. El 7% en Bueno sugiere que algunas empresas familiares tienen programas efectivos de formación para preparar a las nuevas generaciones. Sin embargo, el 26% en Medio y el 31% en Bajo indican que la mayoría de las empresas necesitan fortalecer sus programas de formación. El 36% en Muy Bajo refleja una falta crítica de inversión en la capacitación de nuevos miembros, lo que puede comprometer la preparación de futuros líderes. Revista Compendium: Cuadernos de Economía y Administración Estos desafíos incluyen la representación de familiares en cargos clave, la relación entre generaciones, las relaciones familiares, la gestión de conflictos, la satisfacción de los empleados familiares, la transmisión de liderazgo y la formación de nuevos miembros. Abordar estos desafíos es esencial para fortalecer la cohesión familiar y garantizar la continuidad exitosa de la empresa a lo largo del tiempo. Figura 4. Dimensión propiedad 7% 28% 29% # Socios por generación 36% 8% 34% # Familiares en altos cargos 26% 32% 5% # Socios con cargos estratégicos 26% 38% 30% 7% # Socios que trabajan en la Empresa 26% 31% 36% 13% 20% # Socios Fundadores Muy Bajo 36% 31% 0% 10% Bajo Medio 20% 30% 40% Bueno En el contexto de la dimensión de la Propiedad en las empresas y cómo se distribuyen en términos de generaciones. Esta dimensión se considera esencial, ya que influye en la toma de decisiones, la continuidad y la estabilidad de la empresa. La metodología de los tres círculos se ha empleado de manera efectiva para explorar estos aspectos clave. En primer lugar, el número socios fundadores refleja la importancia de aquellos individuos que participaron en la creación de la empresa. Los resultados indican que aproximadamente un tercio de las empresas se encuentra en una posición sólida con respecto a la presencia de estos socios fundadores. Esto sugiere una base sólida de experiencia y tradición en el núcleo de la empresa. Sin embargo, también es relevante notar que un porcentaje significativo se encuentra en el rango Medio, lo que sugiere la presencia de oportunidades para aumentar la participación de los fundadores en algunas empresas. 184 En segundo lugar, los socios que trabajan en las empresas revelan la medida en que los socios están activamente involucrados en las operaciones diarias de la empresa. Los resultados indican que aproximadamente un tercio de las empresas tiene una participación sólida de socios que contribuyen al funcionamiento general. Sin embargo, un porcentaje similar se encuentra en la categoría Medio, lo que sugiere que aún hay margen para que algunas empresas fomenten una mayor implicación de sus socios en las operaciones cotidianas. Por otro lado, un número significativo de empresas se sitúa en los niveles Bajo y Muy Bajo, lo que indica la necesidad de superar desafíos para involucrar a los socios en el trabajo diario. En tercer lugar, los socios con cargos estratégicos evalúan la representación de socios en roles estratégicos dentro de la empresa. Aquí, se destaca que algunas empresas tienen una sólida representación de socios en estas posiciones, lo que puede contribuir a la toma de decisiones y la dirección de la empresa. Sin embargo, un gran porcentaje se encuentra en la categoría Medio, lo que sugiere que muchas empresas podrían mejorar su presencia en roles estratégicos. Además, la presencia de empresas en las categorías Bajo y Muy Bajo señala la necesidad de abordar la falta de representación de socios en posiciones clave. En cuarto lugar, el número de familiares en altos cargos pone de relieve la cantidad de familiares que ocupan puestos directivos de alto nivel en la empresa. Los resultados indican que algunas empresas tienen una sólida representación de familiares en cargos de liderazgo, lo que puede ser beneficioso para mantener la cohesión y una visión a largo plazo. No obstante, la presencia de empresas en las categorías Medio, Bajo y Muy Bajo indica que aún se debe mejorar la presencia de familiares en roles directivos en muchas organizaciones. Finalmente, el número de socios por generación evalúa cómo se distribuyen los socios en diferentes generaciones. Un resultado positivo es que un tercio de las empresas demuestra una distribución equitativa, lo que puede promover la continuidad y la integración de diversas perspectivas generacionales. Sin embargo, la Revista Compendium: Cuadernos de Economía y Administración presencia de empresas en las categorías Medio, Bajo y Muy Bajo sugiere que algunas organizaciones pueden enfrentar desafíos relacionados con una concentración significativa de socios en una sola generación, lo que podría plantear problemas en términos de sucesión y visión a largo plazo. Estos resultados son esenciales para la toma de decisiones estratégicas relacionadas con la estructura de propiedad y la participación de los socios en la empresa familiar. La diversidad de situaciones destaca la necesidad de enfoques adaptados y estrategias específicas para cada empresa en función de sus desafíos y fortalezas particulares. Figura 5 Puntuación por dimensión para determinar grado de madurez 20% 8% Muy Bajo 30% 22% 27% 27% 30% 24% Bajo 33% 32% Medio 27% 39% 20% 33% Bueno 13% 15% 0% 10% 20% 30% 40% 50% Puntuación Puntuación por dimensiones Propiedad Puntuación por dimensiones Familia Puntuación por dimensiones Empresa El análisis del grado de madurez de las empresas se llevó a cabo mediante la aplicación de una semaforización y la evaluación de las puntuaciones en tres dimensiones fundamentales: Empresa, Familia y Propiedad. Estas dimensiones se estratificaron en cuatro niveles: Bueno, Medio, Bajo y Muy Bajo. En lo que respecta a la dimensión de la Empresa, se observa que un 15% de la puntuación se ubica en la categoría Bueno, indicando la existencia de aspectos sólidos en el 185 funcionamiento empresarial. Sin embargo, el 39% se sitúa en Medio, lo que denota áreas donde la empresa podría mejorar y fortalecer sus prácticas y procesos. Además, un 24% se clasifica como Bajo, lo que señala la presencia de áreas sustanciales de mejora, mientras que un 22% se califica como Muy Bajo, evidenciando desafíos significativos en diversos aspectos empresariales. Por otro lado, en la dimensión de la Familia, se destaca que un 13% se considera Bueno, lo que sugiere que ciertos aspectos relacionados con la dinámica familiar se gestionan de manera armoniosa en la empresa. Sin embargo, un 27% se encuentra en la categoría Medio, lo que podría indicar la existencia de algunas tensiones o desafíos familiares, aunque en general, la dinámica es aceptable. Un 30% se clasifica como Bajo, lo que señala problemas importantes en la relación entre la familia y la empresa, y otro 30% se ubica en Muy Bajo, sugiriendo una situación problemática en la dimensión familiar, posiblemente caracterizada por conflictos significativos. En lo que concierne a la dimensión de la Propiedad, se destaca que un 33% se encuentra en la categoría Bueno, lo que implica que la propiedad de la empresa está bien estructurada y gestionada de manera efectiva. Sin embargo, el 32% se ubica en Medio, lo que podría indicar la presencia de algunas áreas de mejora o ambigüedad en la estructura y gestión de la propiedad. Además, un 27% se califica como Bajo, señalando problemas considerables en la gestión de la propiedad, mientras que un 8% se sitúa en Muy Bajo, reflejando una situación crítica en esta dimensión, posiblemente marcada por conflictos y falta de claridad. El análisis del grado de madurez de revela una variabilidad significativa en las tres dimensiones evaluadas (Empresa, Familia y Propiedad). Aunque existen fortalezas en ciertas áreas, también se identifican desafíos importantes en otras. Abordar los problemas identificados en las dimensiones de la Familia y la Propiedad se vuelve esencial, ya que estos pueden tener un impacto significativo en el funcionamiento global de la empresa. La semaforización y el análisis detallado por dimensiones proporcionan una visión completa de la salud y madurez de la Revista Compendium: Cuadernos de Economía y Administración empresa, lo que puede servir como base para la implementación de medidas específicas de mejora en áreas críticas. Al aplicar la metodología de los tres círculos, se observa que las empresas enfrentan desafíos importantes en las dimensiones de la Familia, la Propiedad y la Empresa. La interacción y alineación entre estas tres dimensiones parecen necesitar mejoras sustanciales. Esto subraya la importancia de abordar los problemas identificados en cada dimensión para lograr una mayor madurez y armonía en la empresa familiar. La metodología de los tres círculos proporciona una visión holística que puede servir como guía para la implementación de estrategias de mejora específicas en cada área. CONCLUSIONES Las empresas familiares del cantón Cuenca se encuentran ante un panorama de desafíos y oportunidades sustanciales que abarcan las esferas de Empresa, Familia y Propiedad. La identificación de áreas críticas, tales como una gestión administrativa deficiente y la carencia de políticas de sucesión claras, otorga a estas empresas la ocasión de centrarse en la mejora de dichos aspectos, asegurando así su perdurabilidad a largo plazo. Además, se ha constatado que la aplicación exitosa de la metodología de los tres círculos se revela como un recurso efectivo para evaluar y comprender la intrincada dinámica que involucra a la empresa, la familia y la propiedad en el contexto de las empresas familiares. Esto sienta una sólida base para la toma de decisiones estratégicas y la implementación de medidas específicas destinadas a la mejora. En la dimensión de Empresa, se observa que el 25% de las empresas se ubica en la categoría Bueno en cuanto a la Edad de la Empresa, indicando una antigüedad adecuada en el mercado. Sin embargo, el 53% se encuentra en las categorías Bajo y Muy Bajo, señalando empresas en etapas incipientes de su crecimiento que necesitan atención especial para fortalecerse y estabilizarse. En cuanto a la Gestión Administrativa, solo el 4% obtiene una 186 clasificación Bueno, revelando la necesidad urgente de mejoras en este aspecto crítico. Asimismo, el 77% se sitúa en las categorías Bajo y Muy Bajo, indicando una preocupante falta de eficacia en la gestión administrativa. Por otro lado, la Estructura Organizacional muestra que el 58% de las empresas tiene una estructura bien definida y eficiente, lo que es fundamental para un funcionamiento organizado. No obstante, aún hay un 38% que puede revisar y mejorar sus estructuras para un mejor desempeño. En la dimensión de Familia, se destaca que el 26% de las empresas tiene una Relación entre generaciones en la categoría Bueno, indicando que han logrado establecer una relación armoniosa entre diferentes generaciones familiares. Sin embargo, el 67% se encuentra en las categorías Bajo y Muy Bajo, evidenciando problemas significativos en esta área que pueden afectar la cohesión y la toma de decisiones. Asimismo, el Manejo de conflictos es un aspecto crítico, donde el 67% se clasifica en las categorías Bajo y Muy Bajo, mostrando una falta de habilidades para abordar y resolver conflictos de manera constructiva. En la dimensión de Propiedad, la representación de los socios en distintos aspectos es variada. El 33% se encuentra en la categoría Bueno en Número de socios fundadores, señalando una base sólida de experiencia en el núcleo de la empresa. Sin embargo, el 35% se sitúa en las categorías Medio y Bajo, mostrando oportunidades para aumentar la participación de los fundadores en algunas empresas. En el Número de familiares en altos cargos, el 42% está en las categorías Bajo y Muy Bajo, resaltando la necesidad de mejorar la presencia de familiares en cargos clave para mantener la cohesión y una visión a largo plazo en la empresa. Por último, el Número de socios por generación revela que el 32% de las empresas tiene una distribución equitativa de socios por generación, lo que es fundamental para una gestión eficiente y una visión a largo plazo. En este sentido, este análisis demuestra que, si bien algunas empresas muestran fortalezas en ciertas áreas, la mayoría enfrenta desafíos importantes en aspectos cruciales de Empresa, Familia y Propiedad. Es imperativo que estas Revista Compendium: Cuadernos de Economía y Administración empresas se enfoquen en la mejora de la gestión administrativa, la relación empresa/familia, las políticas de sucesión y la gestión de conflictos para garantizar su sostenibilidad y éxito a largo plazo. Estas conclusiones proporcionan una base sólida para la implementación de estrategias específicas destinadas a abordar los desafíos identificados y mejorar el desempeño global de las empresas familiares en el cantón Cuenca. Este estudio contribuye de manera sustancial al corpus de conocimientos académicos, puesto que ha aplicado eficazmente la metodología de los tres círculos en el ámbito de las empresas familiares del cantón Cuenca. Los académicos hallarán en este enfoque una referencia valiosa para futuras investigaciones relacionadas con la gestión de empresas familiares. Asimismo, esta investigación añade un valor significativo a la comunidad de investigadores al proporcionar datos específicos y análisis detallados sobre las empresas familiares en una región geográfica concreta. Esto posibilita una comparación precisa y una contextualización más adecuada de los desafíos y fortalezas a los que se enfrentan estas empresas. Los gerentes y líderes de empresas familiares pueden valerse de las conclusiones de este estudio como un recurso inestimable para la identificación de áreas de mejora dentro de sus organizaciones. Las recomendaciones específicas proporcionan una orientación clara sobre cómo abordar desafíos tales como la gestión administrativa y la planificación de sucesión. El éxito de las empresas familiares repercute directamente en la economía a nivel local y nacional. La atención a las áreas de mejora identificadas en este estudio puede contribuir al crecimiento económico y la estabilidad tanto en el cantón Cuenca como en el ámbito más amplio. En este sentido, se recomienda encarecidamente a las empresas familiares del cantón Cuenca que prioricen la mejora de su gestión administrativa, implementen políticas de sucesión sólidas y se dediquen a una planificación de sucesión efectiva. Adicionalmente, se sugiere la necesidad de llevar a cabo investigaciones adicionales que profundicen en áreas específicas identificadas en 187 Revista Compendium: Cuadernos de Economía y Administración este estudio, como la dinámica entre generaciones, las relaciones familiares y la gestión de conflictos en el contexto de las empresas familiares. REFERENCIAS Andrade, D. E. (2019). Aplicación del modelo Gersick en la empresa Andelas Cía. Ltda. Pontificia Universidad Católica del Ecuador. http://repositorio.pucesa.edu.ec/handle/12 3456789/2754 Andrews, J. M. (2010). Managing Growth: Best Practices of Family-Owned Businesses. https://digitalcommons.bryant.edu/cgi/vie wcontent.cgi?article=1005&context=hon ors_management#:~:text=The%20current %20three%2Ddimensional%20model,% E2%80%93%20ownership%2C%20famil y%20and%20business. Araoz, R. G., & Bringas, M. de los A. L. (2023). Modelos de gobernanza y continuidad en la empresa familiar. Sur Academia: Revista Académica-Investigativa de la Facultad Jurídica, Social y Administrativa, 10(19). https://revistas.unl.edu.ec/index.php/sura cademia/article/view/1605 Arista, R., & De la Garza Ramos, M. I. (2022). Influencia de la familiness en la relación de la heterogeneidad familiar y la postura innovadora de la empresa familiar. CIENCIA ergo-sum: revista científica multidisciplinaria de la Universidad Autónoma del Estado de México, 29(3). https://www.ssoar.info/ssoar/handle/docu ment/86685 Barbosa, R. A. (2020). Invernadero Inteligente, la contribución para el mejoramiento de la calidad de vida en la Guajira Colombiana Junio de 2020. Unimilitar. https://repository.unimilitar.edu.co/handl e/10654/36083 Betancourt, J. B., Barahona, H., Parra, L. F., & Lagos, D. (2015). Factores que influyen en la selección de mecanismos de planeación de sucesión patrimonial en empresas familiares de Cali. LIBRE EMPRESA, 12(1), 11–28. https://doi.org/10.18041/libemp.v23n1.23 101 Business & Finance. (2023). -Dimensional Development Model. Ebrary. https://ebrary.net/272305/business_financ e/-dimensional_development_model Carlock, R. S., & Ward, J. L. (2001). Strategic planning for the family business: Parallel planning to unify the family and business. Springer. https://link.springer.com/book/10.1057/9 780230508750 Carlock, R. S., Ward, J. L., Carlock, R. S., & Ward, J. L. (2010). Making the parallel family and business planning process work. When Family Businesses are Best: The Parallel Planning Process for Family Harmony and Business Success, 28–49. Cunha, V. M., Nadae, J., & da Rocha Grangeiro, R. (2020). Planejamento estratégico de sucessão em empresas familiares. Folha de Rosto, 6(3), 97–111. Dantas, V. M. da C., Nadae, J. de, & Grangeiro, R. da R. (2020). Strategic succession planning in family businesses. Folha de Rosto; v. 6 n. 3 (2020): set./dez.; 97-111, 24(2), 111–197. De Ciantis, D., & Lansberg, I. (2020). What family businesses need to adapt to a crisis. Harvard Business Review, Retrieved on, 17. https://netfamilybusiness.com/wpcontent/uploads/2020/05/What-FamilyBusinesses-Need-to-Adapt-to-aCrisis.pdf De León, E. R. R. P. (2019). Principales factores de la continuidad en las empresas familiares. Quipukamayoc, 27(53), 97– 103. De Visscher, F. M., Mendoza, D. S., & Ward, J. L. (2011). Financing transitions: Managing capital and liquidity in the family business. Macmillan. https://books.google.com/books?hl=es&l r=&id=sGTmJv_fPUMC&oi=fnd&pg=P 188 Revista Compendium: Cuadernos de Economía y Administración P2&dq=John+L.+Ward&ots=99R_tM48i D&sig=ObM41u31Xy8EVCkq78qkdOR Azs4 Feliu, N., & Lansberg, I. (2022). Women in family enterprise: Understanding the unique ownership challenges of daughters in business families. En The Power of Inclusion in Family Business (pp. 135– 151). Emerald Publishing Limited. https://www.emerald.com/insight/content /doi/10.1108/978-1-80117-578420221008 Filippim, E. S. (2019). O MODELO TRIDIMENSIONAL E O PROCESSO DE SUCESSÃƒO. RGC-Revista de Governança Corporativa, 5(2). https://rgc.org.br/ojs/index.php/rgc/article /view/46 García, I. G. (2005). Estudio de empresas familiares en escenarios partiendo del modelo evolutivo tridimensional. Omnia, 11(2), 29–52. García, K. M. V., & García, P. R. E. (2023). EL DESARROLLO LOCAL Y LA ACCIÓN DE LOS CLUBES ROTARY COMO VÍA PARA SALIR DE LA CRISIS ECONÓMICA. V CICAE, 336. Güldenkoh, M., & Silberg, U. (2019). The Characteristics of Family Businesses: A Comparative Analysis of East European Countries. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.3484792 INEC. (2023). Consultas personalizadas del Directorio de Empresas y Establecimientos 2021 SI. Tableau Public. https://public.tableau.com/app/profile/inst ituto.nacional.de.estad.stica.y.censos.inec ./viz/ConsultaspersonalizadasdelDirectori odeEmpresasyEstablecimientos2021SI_1 6722361486880/Dcruce Molina, A. I. P. (2012). El modelo de empresa familiar: Los cuatro pilares fundamentales. 3c Empresa: investigación y pensamiento crítico, 1(3), 1. Molina, P. A., Botero, S., & Montoya, J. N. (2016). Empresas de familia: Conceptos y modelos para su análisis. Pensamiento & Gestión, 41, 116–149. Moores, K., & Barrett, M. (2005). Understanding tensions and conflict: A phases of learning approach for leading family businesses. En Managing the family business (pp. 67– 85). Heidelberg Press. https://research.bond.edu.au/en/publicatio ns/understanding-tensions-and-conflict-aphases-of-learning-approach Pimentel, T. D. (2021). Modelos de sucesión en empresas familiares. Revista da Micro e Pequena Empresa, 15(3), 6–23. Rave, E. D., & Moreno, J. E. (2023). Proceso de sucesión en la empresa familiar. Estudio de caso, grandes empresas vs. Pymes. Revista Virtual Universidad Católica del Norte, 69, 157–189. https://doi.org/10.35575/rvucn.n69a7 Rodíguez, D. M., Ruiz, J. C., & Díaz, A. P. (2019). Cambios generacionales en las empresas de familia del municipio de Chiquinquirá (Boyacá-Colombia). Revista Espacios, 40(18). https://www.revistaespacios.com/a19v40 n18/19401801.html Samsudin, N. A., & Fuza, Z. I. M. (2021). Attribution of Family Business-Based in Cultural Heritage Management in Malaysia. 683–697. https://link.springer.com/chapter/10.1007 /978-3-031-08090-6_42 Schuman, A., Stutz, S., & Ward, J. L. (2010). Family business as paradox. Springer. https://link.springer.com/book/10.1057/9 780230291768 Teston, S., & Filippim, E. S. (2018). O modelo tridimensional e o processo de sucessão. Iberoamerican Journal of Corporate Governance, 5(2). https://www.rgc.org.br/Journals/article/vi ew/26 Thomas, J. (2002). Freeing the shackles of family business ownership. Family Business Review, 15(4), 321–336. 189 Tucker, D., & Lind, S. (2022). Family Firms and Mergers and Acquisitions: The Importance of Transfer of Trust. En Effective Implementation of Transformation Strategies: How to Navigate the Strategy and Change Interface Successfully (pp. 387–406). Springer. https://link.springer.com/chapter/10.1007 /978-981-19-2336-4_16 Velarde, J. T. (2020). El gobierno corporativo y la profesionalización de las empresas familiares. Universidad Nacional Revista Compendium: Cuadernos de Economía y Administración Federico Villarreal. http://repositorio.unfv.edu.pe/bitstream/h andle/20.500.13084/4564/VELARDE%2 0SUSSONI%20JULIA%20TERESA%20 %20DOCTORADO.pdf?sequence=1&is Allowed=y Węcławski, J., & Żukowska, B. (2019). Refreshing typologies based on family firms’ goals: The perspective of “firstgeneration enterprises”. Central European Management Journal, 27, 128– 146.
https://openalex.org/W4226138662	http://cds.cern.ch/record/2254048/files/CLICdp-Note-2017-001.pdf	English	null	CLICdet: The post-CDR CLIC detector model	CERN Document Server (European Organization for Nuclear Research)	2,021	cc-by	18,101	CLICdet: The post-CDR CLIC detector model N. Alipour Tehrani∗, J.-J. Blaising†, B. Cure∗, D. Dannheim∗, F. Duarte Ramos∗, K. Elsener∗, A. Gaddi∗, H. Gerwig∗, S. Green‡, C. Grefe§, D. Hynds∗, W. Klempt∗, L. Linssen∗, N. Nikiforou∗, A. Nurnberg∗, J.S. Marshall‡, M. Petric∗, S. Redford∗, P. Roloff∗, A. Sailer∗, F. Sefkow¶, E. Sicking∗, N. Siegrist∗, F. Simon∥, R. Simoniello∗, S. Spannagel∗, S. Sroka∗, L.R. Strom∗, M.A. Weber∗ ∗CERN, Geneva, Switzerland, † LAPP, Annecy, France, ‡ Cavendish Laboratory, University of Cambridge, Cambridge, United Kingdom, § University of Bonn, Germany, ¶ DESY, Hamburg, Germany, ∥Max-Planck-Institut fuer Physik, Munich, Germany, ∗∗now at PSI, Villigen, Switzerland CLICdp-Note-2017-001 28 February 2018 (revised 05 April 2019) NB: This is a modiﬁed version of the CLICdp-Note-2017-001 - the changes, introduced on 5 April 2019, are listed in Appendix III. 1 Introduction A state-of-the-art detector, built using cutting-edge technology and optimised through simulation, is crucial to exploit the physics potential of CLIC. Two detector models were previously deﬁned, based on concepts for the ILC detectors and adapted for the higher centre-of-mass energies at CLIC. The CLIC_ILD [1] and CLIC_SiD models [2] were used in the CDR [3] and physics studies in 2013 and 2014. Based on the lessons learnt during these years as well as the experience from several additional optimisation studies, a new model, dubbed CLICdet, has been designed for the forthcoming physics benchmark studies. This note presents the version of CLICdet as deﬁned in spring 2017. The physics requirements and the experimental conditions (beam-beam effects and backgrounds) are most challenging at the highest collision energy. Therefore, similarly to the CDR studies, the focus of this detector optimisation work is on 3 TeV CLIC. As a result, this note describes the new 3 TeV CLIC detector model. The crossing angle between the colliding beams is 20 mrad at 3 TeV. For the initial stage of CLIC (380 GeV in the present staging scenario [4]) the innermost and very forward region of the detector will have to be designed differently (due to a smaller crossing angle, and due to less incoherent electron-positron pair background allowing for a smaller radius of the ﬁrst vertex barrel layer). The changes in the detector layout are signiﬁcant, but not very complex – engineering details of the detector model for lower energy CLIC operation will be studied at a later stage. In chapter 2, this note provides details on the working hypotheses while chapter 3 gives an overview of the overall detector dimensions and most important parameters. Chapters 4 and 5 describe the vertex detector and the silicon tracker system. The electromagnetic and hadronic calorimeters are described in chapters 6 and 7, while chapter 8 gives details of the magnet system (solenoid and yoke) as well as the muon detectors. The very forward region is described in chapter 9, and the beam pipe and vacuum system is discussed in chapter 10. Chapter 11 recapitulates the detector opening and maintenance scenarios, which are very similar to those described in the CDR. Finally, Appendix I contains a table with the sensor areas of each sub-detector system, the pixel/pad sizes and the resulting total number of channels. 1 Introduction Throughout this note, the term "fast simulation" refers to the use of the LiC Detector Toy [5], while "full simulation" refers to detector simulations using GEANT4. Where "overlay of background" is mentioned, this refers to background events from γγ →hadrons interactions stemming from 60 bunch- crossings. 2 Working Hypotheses A number of detector design issues, and some matters concerning the layout of the CLIC experimental area, result from a complex interplay of CLIC accelerator optimisation and the envisaged CLIC physics reach. In order to make progress with the new CLIC detector model, it was inevitable to make some as- sumptions – these are here called "working hypotheses". Future detailed studies might lead to a revision of these assumptions. Abstract A new model for the CLIC detector has been deﬁned based on lessons learnt while working with the CDR detector models and after a series of simulation studies. The new model, dubbed "CLICdet", also incorporates the experience from various R&D activities linked to a future experiment at CLIC. This note describes the studies and thoughts leading to the new detector model, and gives details on all of its sub-detector systems. NB: This is a modiﬁed version of the CLICdp-Note-2017-001 - the changes, introduced on 5 April 2019, are listed in Appendix III. 2 Working Hypotheses 2.1 A single detector at CLIC Following the ILC model [6], the CLIC detector and physics studies were pursued for the CDR assuming the construction of two experiments, installed in a push-pull scheme. While the feasibility of such an arrangement was conceptually proven, it became clear that the two-detector scheme implies considerable extra cost, and beam time would inevitably be lost in the push-pull operations. Moreover, the results of benchmark analyses indicated that, with some exceptions in the forward direction, the physics reach was very similar for CLIC_ILD and CLIC_SiD. In parallel, it also became evident that the CLIC beam and background conditions are not favourable towards a TPC (as planned in ILD), at least not at 3 TeV CLIC (see [7] and [8]). This, in turn, implied 2 2 Working Hypotheses that the possible two CLIC detector concepts would look remarkably similar. In conclusion, and as a ﬁrst working hypothesis, it was agreed to consider one detector only for the present studies. In order to retain a very compact layout of the installation around the interaction point (IP), the opening of the detector for maintenance and repairs is supposed to happen in a dedicated area in close vicinity of the IP ("garage position"). 2.2 QD0 outside of the detector region In the two CDR detector concepts for CLIC, the last quadrupole magnet of the accelerator (the QD0) was embedded in the detector. This layout choice was enforced to allow for the highest possible luminosities, generally obtained with a short distance L∗between the IP and the near end of the QD0 magnet. In analogy to the ILC case at that time, the two detector concepts did not have the same L∗: for CLIC_ILD, this distance was 4.3 m, for CLIC_SiD it was 3.5 m, mainly arising from the different overall length of the tracking detectors. Integrating the QD0 within the complex forward region of the detectors was a challenge, which was further aggravated by the very stringent requirements to keep this magnet stable in position. In order to limit luminosity losses due to vibrations, the stability criterion was set to be less than 0.2 nm r.m.s. at frequencies above 4 Hz. This implied that additional space was required for a support tube and stabilisation systems. Together with the requirements for the opening of the detector for maintenance, the minimal support tube radius was found to be about 50 cm, resulting in a much reduced forward coverage of the hadron calorimeters in the CLIC CDR detector concepts. During discussions within the CLIC detector and physics collaboration (CLICdp) and with accelerator experts, it became clear that it was impossible to assess, a priori, if the possible luminosity loss due to a longer L∗can be off-set by the potential gain in physics reach due to a better forward coverage. In both cases, rather complex detailed studies would be needed. Therefore, after studies of the magnet and yoke system with a possible reduction in overall detector length, it was decided – as a second working hypothesis – to move the QD0 to an L∗of 6 m, i.e. outside of the detector region. With such a layout, a signiﬁcantly better forward HCAL coverage is possible, which is desirable given several important physics scenarios which require reconstruction of physics objects at low polar angles. 3 3 3 Overall Dimensions and Parameters 3 Overall Dimensions and Parameters This chapter provides information about the general considerations leading to the choice of the main detector parameters such as tracker radius and magnetic ﬁeld strength. A comparison to the CDR detector models is presented in Table 1. An illustration of the CLICdet model compared to CLIC_SiD is given in Figures 1 and 2. Table 1: Comparison of key parameters of the different CLIC detector concepts. CLIC_ILD and CLIC_SiD values are taken from the CDR [3]. The inner radius of the electromagnetic calori- meter (ECAL) is given by the smallest distance of the calorimeter to the main detector axis. For the hadronic calorimeter (HCAL), materials are given separately for the barrel and the endcap. Concept CLICdet CLIC_ILD CLIC_SiD Vertex inner radius [mm] 31 31 27 Tracker technology Silicon TPC/Silicon Silicon Tracker half length [m] 2.2 2.3 1.5 Tracker outer radius [m] 1.5 1.8 1.3 ECAL absorber W W W ECAL X0 22 23 23 ECAL barrel rmin [m] 1.5 1.8 1.3 ECAL barrel ∆r [mm] 202 172 139 ECAL endcap zmin [m] 2.31 2.45 1.66 ECAL endcap ∆z [mm] 202 172 139 HCAL absorber barrel / endcap Fe / Fe W / Fe W / Fe HCAL λI 7.5 7.5 7.5 HCAL barrel rmin [m] 1.74 2.06 1.45 HCAL barrel ∆r [mm] 1590 1238 1177 HCAL endcap zmin [m] 2.45 2.65 1.80 HCAL endcap ∆z [mm] 1590 1590 1595 Solenoid ﬁeld [T] 4 4 5 Solenoid bore radius [m] 3.5 3.4 2.7 Solenoid length [m] 8.3 8.3 6.5 Overall height [m] 12.9 14.0 14.0 Overall length [m] 11.4 12.8 12.8 Overall weight [t] 8100 10800 12500 1The multiple scattering contribution to the track momentum resolution is proportional to 1/(BL). For variations of B and L considered in the options for the CLIC detector, the changes in the multiple scattering contribution are minor. 3.1 Tracker volume radius vs. magnetic ﬁeld Transverse momentum resolution, angular track resolution and jet energy resolution using particle ﬂow [9] beneﬁt from a larger tracker radius. The dependence on tracker radius is stronger than that for a change in the magnetic ﬁeld strength, as anticipated by the Gluckstern formula1 [10]. Results from fast simu- lation are shown in Figure 3 for 500 GeV muons at θ = 90◦. A tracker radius of 1.5 m was chosen as a compromise between CLIC_SiD (1.3 m) and CLIC_ILD (1.8 m). (In fact, 1.5 m was later deﬁned as the inner radius of the inscribed circle of the ECAL barrel, see Chapter 6). In a ﬁrst stage, a magnetic ﬁeld between 3.5 and 4.5 T was being discussed. 4 3 Overall Dimensions and Parameters Fe - Yoke Steel - HCAL Steel - HCAL ECAL Coil - 4T Si - Tracker Figure 1: Longitudinal cross section showing the top right quadrant of CLICdet (left) and CLIC_SiD (right). In CLICdet, the structures shown on the right of the image (i.e. outside of the yoke) represent the end coils, while in CLIC_SiD these are a schematic illustration of shielding rings. Figure 1: Longitudinal cross section showing the top right quadrant of CLICdet (left) and CLIC_SiD (right). In CLICdet, the structures shown on the right of the image (i.e. outside of the yoke) represent the end coils, while in CLIC_SiD these are a schematic illustration of shielding rings. Figure 2: Transverse (XY) cross section of CLICdet (left) and CLIC_SiD (right). Figure 2: Transverse (XY) cross section of CLICdet (left) and CLIC_SiD (right). 3.2 Magnetic ﬁeld vs. occupancies and ﬂavour tagging Occupancies in the vertex detector due to incoherent electron-positron pairs from beamstrahlung are directly affected by the magnetic ﬁeld strength [11]. With an outer tracker radius of 1.5 m a magnetic ﬁeld strength of 3.5 T would be sufﬁcient to retain the transverse momentum resolution achieved with the CLIC_SiD concept. The bore of the solenoid would thus have a radius of around 3 m in case of a tungsten based HCAL, 3.4 m for an HCAL with steel absorbers. Such a free bore is similar to the CMS coil (but with a shorter coil, e.g. 8 m in CLIC_ILD vs. 13 m in CMS). Given these dimensions, a magnetic ﬁeld strength of 4.5 T should be technically feasible. After investigating the impact of the magnetic ﬁeld 5 3 Overall Dimensions and Parameters [mm] max R 1200 1300 1400 1500 ] -1 ) [GeV 2 T /p T p ∆ RMS( 10 15 20 25 30 35 40 -6 10 × -µ Single ° = 90 θ p = 500 GeV, B = 3.5 T B = 4.0 T B = 4.5 T B = 5.0 T B = 5.5 T Figure 3: Momentum resolution as a function of tracker radius for different strengths of the B-ﬁeld. Results obtained in fast simulation, for 500 GeV muons at θ=90◦and for a silicon strip tracker with a point resolution of 7 µm. Note that at such high momenta, the contribution from multiple scattering is negligible. Figure 3: Momentum resolution as a function of tracker radius for different strengths of the B-ﬁeld. Results obtained in fast simulation, for 500 GeV muons at θ=90◦and for a silicon strip tracker with a point resolution of 7 µm. Note that at such high momenta, the contribution from multiple scattering is negligible. strength on the pair background in the vertex detector, studies on the ﬂavour tagging performance as a function of vertex innermost radius were performed - these showed a weak dependence [12]. Finally, following discussions with CMS engineers on magnet design feasibility and risks, the magnetic ﬁeld strength for the new CLIC detector model was chosen to be 4 T. 3.4 Calorimeters The work on ECAL for the new CLIC detector went through two distinct phases: during the earlier studies, emphasis was on jet energy resolution (with less attention paid to energy resolution for high energy photons). This led to an ECAL similar to CLIC_ILD, but with 25 instead of 30 layers (see Section 6.1). Later, however, the importance of medium to high energy photons for CLIC physics was stressed, with a request for a sampling term of the energy resolution function smaller than 20%, and with a constant term as small as possible (the latter is crucially important for the very high energy photons). Optimisation studies found that a ﬁner longitudinal sampling of the ECAL is needed to meet these requirements. An ECAL with 30 equally spaced layers marginally fulﬁlls the requirements, but further improvements could be obtained with an ECAL of 40 layers. In conclusion, for the new detector model a silicon-tungsten (SiW) ECAL with 40 layers is chosen, with 1.9 mm thick W absorber plates, and with silicon sensor cells of 5 × 5 mm2. The HCAL depth remains at 7.5 λI, as required by physics studies for the CDR. Reducing the tracker radius (with respect to CLIC_ILD) and keeping the magnetic ﬁeld at 4 T allows more radial space for the HCAL opening up for the possibility of using steel absorber plates instead of the more expensive tungsten. Further detailed studies, described in Chapter 7, have shown that the performances of a W- HCAL are not superior to those of a Fe-HCAL of the same depth in λI. As a result, steel is foreseen as absorber material for CLICdet in both the endcap and barrel region. The required inner bore radius for the solenoid is 3.5 m. However, the decision to opt for a Fe-HCAL also has implications on the size, and thus the cost of the magnet system. The required bore radius increases by 0.37 metres when compared to the option of a W-HCAL. According to the parametric model developed in [13], this corresponds to an additional cost for the total magnet system (incl. the yoke) of about 25 MCHF. However, this must be compared to the cost of tungsten plates for a barrel HCAL (these were estimated to amount to 112 MCHF for CLIC_ILD and 60 MCHF for CLIC_SiD). 3.3 Tracker volume length and tracker layout A good acceptance in the forward region is particularly important for the higher energy stages at CLIC. A long main tracker is crucial for the forward tracking performance since the momentum resolution depends strongly on the lever arm at lower angles (1/R2 for pT and 1/L for the polar angle, see [10]). Here, R is the radius at which the particle leaves the tracker, R < Rmax. By extending the tracker length, R and R BdR increases for the particles emitted in the forward direction. Moreover, a longer distance from the IP to the calorimeter endcaps provides a better angular coverage of the hadronic calorimeter endcap. In the CDR detector models the angular coverage of the HCAL endcaps had been limited at their inner radius by the large support tube, which is not present in the new CLIC detector model. A drastic reduction of the endcap yoke thickness, compensated by end coils, would a priori allow for an even longer tracking volume – however, the importance of an L∗of not more than 6 m and the on-axis B-ﬁeld requirements in the region of the QD0 ﬁnally led to the value of 2.2 m (similar to the length chosen in CLIC_ILD) as the best compromise for the half-length of the tracker volume. The layout of the silicon tracking layers was revisited. A CLIC_SiD style design would lead to outer tracker barrel layers of 4.4 m, was was found unrealistically long. A layout with a large support tube was chosen, separating the tracker volume into an inner and outer tracker region. Both regions consist of three barrel layers, complemented by seven and four forward tracker disks for the inner and outer tracker, respectively. The new layout is discussed in detail in Chapter 5. 6 3 Overall Dimensions and Parameters 3.4 Calorimeters Finally, the choice of a Fe-HCAL and thus the larger radius changes the aspect ratio and thus slightly impacts the inhomogeneity of the magnetic ﬁeld in the tracker region. The change from aspect ratio 1.07 (for CLICdet with its Fe-HCAL) to 1.20 (had a W-HCAL been chosen) is deemed to be of minor importance. Engineering considerations lead, at this stage of the conceptual detector design, to 12-fold calorimeter cross sections throughout. Discussions on cable routing from the inner detectors and calorimeters out through the yoke region lead to the layout shown in Figure 1 (left): The ECAL barrel is aligned with the HCAL barrel, and the ECAL endcap is inserted into the HCAL endcap to allow for a free vertical passage for cables and other services. Moreover, the inner top corner of the HCAL endcap is modiﬁed for a smooth passage of the services (90◦angles are not feasible for services routing when little space is available). Details of the optimisation studies for ECAL and HCAL are described in Chapters 6 and 7. 2In the CDR detector models, such an anti-solenoid needed to be introduced to shield (partially) the beam and the QD0 against the ﬁeld from the detector solenoid. 3.5 Magnet system The magnet system of CLICdet consists of a superconducting solenoid producing a 4 T ﬁeld in the centre of the detector, and a steel return yoke. Given that there is only one detector, the requirements on stray ﬁelds and radiation shielding [14] can be relaxed, leading to a reduced thickness of the yoke. The overall dimensions of the magnet system are shown in Figure 4. Also shown in the ﬁgure are the end coils, which are necessary to compensate for the reduced thickness of the yoke endcap w.r.t. the ILD magnet system [15]. Details of the magnet system are described in 8. Contrary to the CDR, there appears to be no need for an anti-solenoid [16]2 as a consequence of placing the QD0 outside of the detector region. 7 3 Overall Dimensions and Parameters Figure 4: Schematic RZ view of the CLICdet magnet system. All dimensions are given in mm. To simplify the drawing, only one half of the magnet section is shown and calorimeters as well as other detector details are omitted. Figure 4: Schematic RZ view of the CLICdet magnet system. All dimensions are given in mm. To simplify the drawing, only one half of the magnet section is shown and calorimeters as well as other detector details are omitted. 3.6 Muon detectors For engineering reasons, a minimal thickness of the iron plates in the return yoke of about 10 cm is required. Reducing the yoke endcap thickness therefore implies that the number of muon detection layers has to be reduced. In CLICdet, there are 6 muon detection layers (as compared to 9 in the CDR models). These 6 layers are distributed with equal spacing in the yoke. In the barrel region, and in analogy to the CDR detector models, an additional muon detection layer as close as possible to the solenoid is foreseen. 4 Vertex Detector The vertex detector in CLICdet, similarly to the CDR detector models, consists of a cylindrical barrel detector closed off in the forward directions by "disks". The layout is based on double layers, i.e. two sensitive layers ﬁxed on one support structure, in both barrel and forward region. The barrel consists of three double layers. In the forward region, the three "disks" are split up in 8 segments which are arranged to create a "spiral". This spiral geometry allows efﬁcient air-ﬂow cooling of the vertex detector [17]. The air-ﬂow imposes that both spirals have the same sense of rotation - this leads to an asymmetric layout of the vertex "disks". An illustration of the vertex detector surrounding the vaccum pipe is shown in Figure 6. Figure 6: Schematic view of the CLIC vertex detector consisting of three double layers in the barrel and forward region spirals. Figure 6: Schematic view of the CLIC vertex detector consisting of three double layers in the barrel and forward region spirals. Figure 6: Schematic view of the CLIC vertex detector consisting of three double layers in the barrel and forward region spirals. The material budget envisaged is 0.2% X0 per detection layer (0.4% X0 per double layer). The CLIC- det vertex detector is built from 25 × 25 µm2 pixels. Using pulse height information on charge sharing, a single point resolution of 3 µm is aimed for. The inner radius of the innermost vertex barrel layer is limited by the occupancy from incoherent pairs. Along with the choice of the magnetic ﬁeld of 4 T the inner radius has been increased from 27 mm (in CLIC_SiD) to 31 mm. The impact on the ﬂavour tag performance due to this change is less than 10%, as shown in Figure 7, and is considered acceptable. The overall length of the barrel vertex detector, built from staves, is 260 mm. The total area of the sensors in the three barrel double layers is 0.487 m2. The double layer structure is shown in Figure 8, in the two options presently available in the simulations. Further details on the dimensions of the vertex detector barrel layers used in the simulation model are given in Table 2. The eight petals per "disk" of the spirals (24 petals in total per side) are ﬂat trapezoids installed at 90◦ to the detector axis. 3.7 Very forward region The main elements of the very forward region of CLICdet, containing the LumiCal and BeamCal de- tectors, remain as in the CDR models. The noteable exception is the absence of the QD0 in the detector volume - this allows to simplify the vacuum system. Contrary to the CDR detector models, no vacuum valve is needed inside the detector volume (a large valve near LumiCal added considerable complexity in the CDR layout). The layout of the forward region is shown in Figure 5. Further details on the forward region, the vacuum system and the detector opening scenario are given in Chapters 9 to 11. Figure 5: Layout of the forward region in CLICdet, seen from the top. The LumiCal is shown in dark blue, the BeamCal in light blue colour. Figure 5: Layout of the forward region in CLICdet, seen from the top. The LumiCal is shown in dark blue, the BeamCal in light blue colour. 8 4 Vertex Detector 4 Vertex Detector The ﬁrst petal is located at Z = 160.0 mm from the interaction point, the last petal at Z = 298.8 mm. The longitudinal distance between the individual petals is 5.5 mm. In radial direction, the petals cover the range R = 33 mm to R = 102 mm (centre of the trapezoid base). The inner edge of the petals is 28 mm long, the outer edge measures 85 mm. These dimensions lead to an overlap of 2 mm from petal to petal. An overview of the vertex petal arrangement as implemented in the simulation is shown in Figure 9. The total area of the silicon sensors of all petals is 0.351 m2. In order to reach the very low material budget required, each sensor layer is built from 50 µm thick 9 4 Vertex Detector 0.5 0.6 0.7 0.8 0.9 1 Misidentification eff. -3 10 -2 10 -1 10 1 Charm Background r=31mm r=27mm LF Background r=31mm r=27mm ° =90 θ =500 GeV, s dijet events at Beauty eff. 0.5 0.6 0.7 0.8 0.9 1 r=31mm/r=27mm 0.8 0.9 1 1.1 Charm Background LF Background 0.5 0.6 0.7 0.8 0.9 1 Misidentification eff. -3 10 -2 10 -1 10 1 Charm Background r=31mm r=27mm LF Background r=31mm r=27mm ° =90 θ =500 GeV, s dijet events at Beauty eff. 0.5 0.6 0.7 0.8 0.9 1 r=31mm/r=27mm 0.8 0.9 1 1.1 Charm Background LF Background Figure 7: Comparison of ﬂavour tag performance for two different radii of the innermost layer of the vertex detector (from [18]). Note that in this study, the magnetic ﬁeld remained unchanged at 5 T. ° =90 θ =500 GeV, s dijet events at Figure 7: Comparison of ﬂavour tag performance for two different radii of the innermost layer of the vertex detector (from [18]). Note that in this study, the magnetic ﬁeld remained unchanged at 5 T. Figure 7: Comparison of ﬂavour tag performance for two different radii of the innermost layer of the vertex detector (from [18]). Note that in this study, the magnetic ﬁeld remained unchanged at 5 T. Table 2: Vertex barrel layout as implemented in the simulations. Barrel layers Inner radius [mm] No. of staves Stave width [mm] 1 - 2 31 - 33 16 13 3 - 4 44 - 46 12 26 5 - 6 58 - 60 16 26 sensors attached to 50 µm thick ASICs. 4 Vertex Detector A central structure, consisting of a 1800 µm thick Rohacell glued inside two 30 µm thick CFRP skins, supports two such sensor+ASIC assemblies, thus forming a double-layer. On the outer side of each ASIC, 300 µm are foreseen for a combination of connectivity and power pulsing. The total thickness of the assembly is 2.8 mm. The support structure for the vertex petals needs to be stronger and is thus re-enforced with two additional CFRP skins, and a slightly thinner Rohacell core, resulting in the same assembly thickness as for the barrel layers. Details of the conceptual engineering design for the vertex layers can be found in [19]. sensors attached to 50 µm thick ASICs. A central structure, consisting of a 1800 µm thick Rohacell glued inside two 30 µm thick CFRP skins, supports two such sensor+ASIC assemblies, thus forming a double-layer. On the outer side of each ASIC, 300 µm are foreseen for a combination of connectivity and power pulsing. The total thickness of the assembly is 2.8 mm. The support structure for the vertex petals needs to be stronger and is thus re-enforced with two additional CFRP skins, and a slightly thinner Rohacell core, resulting in the same assembly thickness as for the barrel layers. Details of the conceptual engineering design for the vertex layers can be found in [19]. A simpliﬁed layout is implemented in the simulation model. The 50 µm silicon sensors are separated by a 2 mm air-gap. On the outside of each sensor, an additional 140 µm of silicon represents the com- bined material of ASIC, support structure, connectivity and power pulsing. The resulting material budget corresponds to the thickness in X0 in the engineering design, and is summarized in Table 3. Air-cooling of the vertex detector, which is possible thanks to the reduced power dissipation reached using power-pulsing, has been extensively studied in simulations as well as in wind-channel and mock- up tests [17, 20]. The air needs to be guided from the outside of the CLICdet through piping to the central area of the detector, and out again on the opposite side. Moreover, since a sufﬁcient cooling efﬁciency is 10 4 Vertex Detector Table 3: Vertex detector double layer material budget as implemented in the simulations. Function Material Thickness Material budget Thickness Material budget Barrel layers [µm ] Barrel layers [%X0 ] Petals [µm ] Petals [%X0 ] ASIC, support etc. 4 Vertex Detector Silicon 140 0.149 170 0.181 Sensor Silicon 50 0.053 50 0.053 Gap Air 2000 0.001 2000 0.001 Sensor Silicon 50 0.053 50 0.053 ASIC, support etc. Silicon 140 0.149 170 0.181 [cm] [cm] 0 2 4 6 0 2 4 6 (a) Option 1 - considered feasible from an engin- eering point of view [cm] [cm] 0 2 4 6 0 2 4 6 (b) Option 2 - presently used in the simulation model, but disfavoured from an engineering point of view Figure 8: The double layer arrangement in the vertex barrel detector (XY view). Table 3: Vertex detector double layer material budget as implemented in the simulations [cm] [cm] 0 2 4 6 0 2 4 6 [cm] [cm] 0 2 4 6 0 2 4 6 (b) Option 2 - presently used in the simulation model, but disfavoured from an engineering point of view (a) Option 1 - considered feasible from an engin- eering point of view Figure 8: The double layer arrangement in the vertex barrel detector (XY view). [cm] [cm] 0 2 4 6 8 10 0 2 4 6 8 10 Figure 9: Schematic view of the vertex forward petals arrangement, showing the 2 mm overlap between subsequent petals (XY view). Figure 9: Schematic view of the vertex forward petals arrangement, showing the 2 mm overlap between subsequent petals (XY view). 11 11 4 Vertex Detector only achieved with a spiralling air-ﬂow around the vertex detector barrel layers, the air must arrive at the vertex forward petals already with a rotational velocity component. Guiding the air to the vertex detector is achieved by a double-walled structure. The inner wall is the conical beam pipe made of 4.8 mm thick stainless steel, the outer wall consists of a < 1 mm CFRP sheet. A spiralling structure made of CFRP, inserted in the 5 mm space between the two walls, steers the air into a rotational ﬂow. The layout as presently envisaged is schematically shown in Figure 10. In order not to limit the forward acceptance of the detector, a pointing geometry with the CFRP sheet limiting the aperture is chosen, with an angle of 6.6◦. As a result, the half-length of the cylindrical (beryllium) part of the vacuum pipe is 308 mm (as compared to 260 mm in CLIC_ILD). 5 Silicon Tracker The tracking systems considered for the two CDR detector models were a TPC for CLIC_ILD and an all- silicon tracker for CLIC_SiD. Occupancy studies using the CLIC 3 TeV beam conditions [7, 8] found an occupancy of about 30% in the TPC pads (without safety factors), caused mainly by the long readout time and the fact that background hits are integrated over the full CLIC bunch train. Moreover, it turned out that the contribution of the TPC to an accurate momentum measurement is limited [3]. It was therefore concluded that the next CLIC detector model should feature an all-silicon tracker. As a result, the vertex and tracker are regarded as one uniﬁed tracking system in the reconstruction process. (The dimensions of the cells in vertex and tracker, as presently envisaged, are given in Appendix I 13.) A conceptual design for supporting the vertex detector and beam vacuum tube inside a support cylinder had been proposed for the CDR detector models [21]. This support cylinder had a diameter just large enough to ﬁt around the conical vacuum pipe. However, such a layout prevents the forward tracker disks to cover the smallest possible angles, thus penalising tracking in a region which is already suffering from the small R B dl. A support tube at a larger radius, as chosen for CLICdet, helps avoiding this issue. The overall layout of the silicon tracker in CLICdet is shown in Figure 11. The tracking volume has a radius of 1.5 m and a half-length of 2.2 m. The main support tube has an inner and outer radius of 0.575 and 0.600 m, respectively, and a half-length of 2.25 m. This support tube effectively divides the tracker volume into two regions: the "Inner Tracker" and "Outer Tracker". The Inner Tracker contains three tracker barrel layers (ITB1-3) and, on each side of the barrel, seven inner tracker disks (ITD1-7). The Outer Tracker is built from three large barrel layers (OTB1-3) complemented on either side by four outer tracker disks (OTD1-4). When compared to CLIC_SiD, CLICdet has a much larger tracking system, in particular extending the forward region acceptance. The number of expected hits in CLICdet as a function of polar angle θ is shown in Figure 12. Figure 11: Overall layout of the tracking system: the area in darker red illustrates the main support tube for the inner tracking region, the vertex detector and the vacuum system. 4 Vertex Detector only achieved with a spiralling air-ﬂow around the vertex detector barrel layers, the air must arrive at the vertex forward petals already with a rotational velocity component. Guiding the air to the vertex detector is achieved by a double-walled structure. The inner wall is the conical beam pipe made of 4.8 mm thick stainless steel, the outer wall consists of a < 1 mm CFRP sheet. A spiralling structure made of CFRP, inserted in the 5 mm space between the two walls, steers the air into a rotational ﬂow. The layout as presently envisaged is schematically shown in Figure 10. In order not to limit the forward acceptance of the detector, a pointing geometry with the CFRP sheet limiting the aperture is chosen, with an angle of 6.6◦. As a result, the half-length of the cylindrical (beryllium) part of the vacuum pipe is 308 mm (as compared to 260 mm in CLIC_ILD). Figure 10: Engineering layout for the innermost region of the detector, indicating the conical beam pipe (stainless steel – SSt), and the outer shell (carbon-ﬁber-reinforced polymer – CFRP) used to create an air-channel for the vertex detector cooling. The positions of the vertex barrel and petal layers as well as the innermost tracker barrel layers and disks are indicated. A cylinder keeping the air ﬂow in the close vicinity of the vertex detector is also shown. Figure 10: Engineering layout for the innermost region of the detector, indicating the conical beam pipe (stainless steel – SSt), and the outer shell (carbon-ﬁber-reinforced polymer – CFRP) used to create an air-channel for the vertex detector cooling. The positions of the vertex barrel and petal layers as well as the innermost tracker barrel layers and disks are indicated. A cylinder keeping the air ﬂow in the close vicinity of the vertex detector is also shown. 12 5 Silicon Tracker 5 Silicon Tracker The central grey area is the envelope for the air cooling of the vertex detector. Figure 11: Overall layout of the tracking system: the area in darker red illustrates the main support tube for the inner tracking region, the vertex detector and the vacuum system. The central grey area is the envelope for the air cooling of the vertex detector. 13 5 Silicon Tracker ]° [ θ 0 50 100 150 hits n 0 5 10 15 20 25 Vertex Barrel Vertex Disks Inner Tracker Barrel Inner Tracker Disks Outer Tracker Barrel Outer Tracker Disks Total Figure 12: The coverage of the tracking systems with respect to the polar angle θ. Shown is the mean number of hits created by 500 GeV muons in full simulation. Only primary muon hits are taken into consideration (hits from secondary particles are ignored). At least eight hits are measured for all tracks with a polar angle down to about 8◦. ]° [ θ 0 Figure 12: The coverage of the tracking systems with respect to the polar angle θ. Shown is the mean number of hits created by 500 GeV muons in full simulation. Only primary muon hits are taken into consideration (hits from secondary particles are ignored). At least eight hits are measured for all tracks with a polar angle down to about 8◦. Engineering studies are under way to deﬁne, in a preliminary manner, the support structures, cooling systems etc. needed for the tracker barrel layers and disks. The material budget needed in addition to the 200 µm thick layer of silicon (sensors plus ASICS or monolithic structure) can be estimated. In addition to the cylindrical main support tube, two carbon ﬁbre structures ("interlink structures") are needed, to mount the inner and outer tracker barrel layers and to route connections. Only very preliminary sketches of these two interlink structures exist. The building blocks from which the tracker detection layers are constructed, are modules of sensor plus ASIC. They are glued on one side to multi-layer carbon ﬁber structures acting as supports and containing the cooling. On the other side, these modules are glued to the elements needed for connectivity. In the tracker disks, modules are arranged into petals, which in turn are assembled into the full disks. In the inner and outer barrel, the silicon sensor size for all modules is 30 × 30 mm2. 5 Silicon Tracker Presently, an overlap between modules of approximately 2 mm is foreseen in radial direction - no overlap along the detector axis. The outer tracker disks are assembled from the same type of modules, 30 × 30 mm2, while the inner tracker disks are made of modules with 15 × 15 mm2 sensors. In all tracker disks, a considerable overlap between petals is foreseen while modules inside the petals have no overlap. Details of the present ideas on module support, overlaps etc. can be found in [19]. This preliminary engineering model is implemented in the simulation model of the tracker, with em- phasis on the correct total material budget per layer (in X0). The current implementation is shown in Figures 13 and 14. Simpliﬁcations with respect to the engineering model include the use of larger rect- angular surfaces instead of the small modules in the tracker disk petals, as shown in Figure 15. The overall geometrical parameters of the tracker are given in Table 4 and 5 for the barrel and disks, respect- ively. The material budget for the different tracker layers is listed in Tables 6 and 7. Details of the presently envisaged design can be found in [19]. The material budget for the modules (sensor and electronics) plus cooling and connectivity, is estimated to be 0.89% X0 for ITB’s and 1.02% X0 for OTB’s and tracker disks. 14 5 Silicon Tracker This material budget does not include the tracker support structures, made of carbon ﬁbre components, which differs from layer to layer and amounts to 0.13% X0 – 0.37% X0. The CLICdet simulation model includes this additional material. This material budget does not include the tracker support structures, made of carbon ﬁbre components, which differs from layer to layer and amounts to 0.13% X0 – 0.37% X0. The CLICdet simulation model includes this additional material. The main support tube, in its preliminary design, amounts to 1.25% X0. The interlink structure for the outer barrel layers is estimated to contribute 0.3% X0, while the inner interlink amounts to 0.5% X0. Cables from the vertex detector, which are to be routed outwards along ITB1 and further out along the conical vacuum tube, are represented by an additional 0.47% X0 (deemed to be a conservative estimate). In the CLICdet simulation model, both interlink structures are inserted with 0.6% X0, approximately accounting for the actual graphite structure plus some of the cables. 5 Silicon Tracker The total material budget for the vertex plus tracker region as a function of polar angle is shown in Figure 16. Preliminary results of a ﬁrst validation of the tracking in CLICdet are shown in Figure 17. 0 0.5 1 1.5 2 0 0.5 1 1.5 z[m] x[m] Figure 13: XZ-view of the tracker as implemented in the simulation model. The black lines indicate the tracker support structures including cooling and cables, the green lines represent the tracker sensor layers. The blue lines show the main support tube and the interlink structures. The orange line indicates the vacuum tube. The vertex detector is shown in the centre (in red). Cables going outwards from the vertex detector are represented in magenta. 0 0.5 1 1.5 2 0 0.5 1 1.5 z[m] x[m] Figure 13: XZ-view of the tracker as implemented in the simulation model. The black lines indicate the tracker support structures including cooling and cables, the green lines represent the tracker sensor layers. The blue lines show the main support tube and the interlink structures. The orange line indicates the vacuum tube. The vertex detector is shown in the centre (in red). Cables going outwards from the vertex detector are represented in magenta. 15 5 Silicon Tracker 0.5 1.0 1.5 0.5 1.0 1.5 y[m] x[m] Figure 14: XY-view of the tracker barrel layers as implemented in the simulation model. The black lines indicate the support structures including cooling and cables, the green lines represent the tracker sensor layers. The blue line shows the main support tube. Vertex detector (in purple) and vacuum tube (in orange) are shown in the centre. Figure 14: XY-view of the tracker barrel layers as implemented in the simulation model. The black lines indicate the support structures including cooling and cables, the green lines represent the tracker sensor layers. The blue line shows the main support tube. Vertex detector (in purple) and vacuum tube (in orange) are shown in the centre. Figure 15: XY-illustration of the tracker disk implementation in the simulation model: the overlay between petals is visible as darker, smaller wedges. The petals are constructed from three (or four) rectangular volumes (example highlighted in yellow/brown), a simpliﬁcation with respect to the engineering layout in [19]. N.B. This graph is a view from z = 1800 mm in downstream direction, thus showing the ITD6 and OTD3 inner and outer tracker disks. 5 Silicon Tracker Figure 15: XY-illustration of the tracker disk implementation in the simulation model: the overlay between petals is visible as darker, smaller wedges. The petals are constructed from three (or four) rectangular volumes (example highlighted in yellow/brown), a simpliﬁcation with respect to the engineering layout in [19]. N.B. This graph is a view from z = 1800 mm in downstream direction, thus showing the ITD6 and OTD3 inner and outer tracker disks. 16 5 Silicon Tracker 5 Silicon Tracker Table 4: Main parameters of the tracker barrel layout, radius R and half-length L/2. Layer No. Name R [mm] L/2 [mm] 1 ITB1 127 482 2 ITB2 340 482 3 ITB3 554 692 4 OTB1 819 1264 5 OTB2 1153 1264 6 OTB3 1486 1264 Table 5: Main parameters of the tracker disks. Disk No. Name Z [mm] Rin [mm] Rout [mm 1 ITD1 524 72 404 2 ITD2 808 99 552 3 ITD3 1093 131 555 4 ITD4 1377 164 542 5 ITD5 1661 197 544 6 ITD6 1946 231 548 7 ITD7 2190 250 537 8 OTD1 1310 618 1430 9 OTD2 1617 618 1430 10 OTD3 1883 618 1430 11 OTD4 2190 618 1430 Table 6: Material budget of the tracker barrel layers - total per barrel layer, as im- plemented in the simulation. Layer Name X0 [%] ITB1 - 3 0.89 OTB1 - 3 1.02 Table 7: Material budget of the tracker disks - total per disk, as implemented in the simulation. Disk Name X0 [%] ITD1 - 7 1.02 OTD1 - 4 1.02 ]° [ θ 0 20 40 60 80 ] 0 Material Budget [%X 0 10 20 30 Figure 16: Total material budget of the vertex plus tracker system, including beam pipe, supports and cables, as a function of the polar angle. 17 Table 5: Main parameters of the tracker disks. Disk No. Name Z [mm] Rin [mm] Rout [mm] 1 ITD1 524 72 404 2 ITD2 808 99 552 3 ITD3 1093 131 555 4 ITD4 1377 164 542 5 ITD5 1661 197 544 6 ITD6 1946 231 548 7 ITD7 2190 250 537 8 OTD1 1310 618 1430 9 OTD2 1617 618 1430 10 OTD3 1883 618 1430 11 OTD4 2190 618 1430 Table 7: Material budget of the tracker disks - total per disk, as implemented in the simulation. 5 Silicon Tracker Disk Name X0 [%] ITD1 - 7 1.02 OTD1 - 4 1.02 Table 5: Main parameters of the tracker disks. Disk No. Name Z [mm] Rin [mm] Rout [mm] 1 ITD1 524 72 404 2 ITD2 808 99 552 3 ITD3 1093 131 555 4 ITD4 1377 164 542 5 ITD5 1661 197 544 6 ITD6 1946 231 548 7 ITD7 2190 250 537 8 OTD1 1310 618 1430 9 OTD2 1617 618 1430 10 OTD3 1883 618 1430 11 OTD4 2190 618 1430 Table 7: Material budget of the tracker disks - total per disk, as implemented in the simulation. Disk Name X0 [%] ITD1 - 7 1.02 OTD1 - 4 1.02 Table 5: Main parameters of the tracker disks. Disk No. Name Z [mm] Rin [mm] Rout [mm] 1 ITD1 524 72 404 2 ITD2 808 99 552 3 ITD3 1093 131 555 4 ITD4 1377 164 542 5 ITD5 1661 197 544 6 ITD6 1946 231 548 7 ITD7 2190 250 537 8 OTD1 1310 618 1430 9 OTD2 1617 618 1430 10 OTD3 1883 618 1430 11 OTD4 2190 618 1430 Table 5: Main parameters of the tracker disks. Table 4: Main parameters of the tracker barrel layout, radius R and half-length L/2. Layer No. Name R [mm] L/2 [mm] 1 ITB1 127 482 2 ITB2 340 482 3 ITB3 554 692 4 OTB1 819 1264 5 OTB2 1153 1264 6 OTB3 1486 1264 Table 4: Main parameters of the tracker barrel layout, radius R and half-length L/2. Table 6: Material budget of the tracker barrel layers - total per barrel layer, as im- plemented in the simulation. ]° [ θ 0 20 40 60 80 ] 0 Material Budget [%X 0 10 20 30 Figure 16: Total material budget of the vertex plus tracker system, including beam pipe, supports and cables, as a function of the polar angle. ]° [ θ 0 20 40 60 80 ] 0 Material Budget [%X 0 10 20 30 Figure 16: Total material budget of the vertex plus tracker system, including beam pipe, supports and cables, as a function of the polar angle. 6.1 ECAL optimisation with emphasis on jets The initial round of ECAL simulation studies towards a new CLIC detector model was concluded as follows: • It had been agreed previously that the inner radius of the ECAL should be 1500 mm. • The ILD detector concept assumes 30 ECAL layers, with the silicon-tungsten3 option considered the baseline. The optimal cell size was found to be 5 × 5 mm2. Reducing the number of ECAL layers (while keeping a ﬁxed ECAL thickness in X0) results in a moderate increase of the sampling term in the photon energy resolution at lower energies. For example, the energy resolution for 10 GeV and 100 GeV photons as a function of the number of ECAL layers is shown in Figure 18. Furthermore, the electron energy resolution is mostly based on the track momentum resolution for all but the highest energy electrons. On this basis, it was felt that the penalty for reducing the baseline ECAL design from 30 to 25 layers would be acceptable. Accordingly, the number of layers for the two ECAL sections was set to 17 for the ﬁrst part (thinner plates) and 8 for the second part (thicker absorber plates). • The ILD detector concept assumes 30 ECAL layers, with the silicon-tungsten3 option considered the baseline. The optimal cell size was found to be 5 × 5 mm2. Reducing the number of ECAL layers (while keeping a ﬁxed ECAL thickness in X0) results in a moderate increase of the sampling term in the photon energy resolution at lower energies. For example, the energy resolution for 10 GeV and 100 GeV photons as a function of the number of ECAL layers is shown in Figure 18. Furthermore, the electron energy resolution is mostly based on the track momentum resolution for all but the highest energy electrons. On this basis, it was felt that the penalty for reducing the baseline ECAL design from 30 to 25 layers would be acceptable. Accordingly, the number of layers for the two ECAL sections was set to 17 for the ﬁrst part (thinner plates) and 8 for the second part (thicker absorber plates). • The jet energy resolution is weakly affected by the lateral segmentation of the ECAL, as shown in Figure 19(a). The impact of larger cell sizes would need to be studied in more detail, e.g. for hadronic tau decays. 6.1 ECAL optimisation with emphasis on jets Pending such additional studies, the baseline cell size of 5 × 5 mm2 is kept as baseline. Similarly, the jet energy resolution only weakly depends on the number of ECAL layers (see Figure 19(b)). • Using thin scintillators as active layers remains an option for a future CLIC detector, but the feasibility of a ﬁne segmentation, the cost implications and calibration issues need to be understood in more detail. • Using thin scintillators as active layers remains an option for a future CLIC detector, but the feasibility of a ﬁne segmentation, the cost implications and calibration issues need to be understood in more detail. 6 Electromagnetic Calorimeter Following the completion of the CDR, the detector optimisation efforts for the ECAL were pursued. The PANDORA particle ﬂow algorithm has been used throughout. The work can be separated in two parts: (1) ECAL optimisation using the ILD detector model for ILC, with emphasis on jets, and with the paradigm that photon energy resolution is not expected to be a major physics driver, and (2) ECAL optimisation studies with a DD4hep based new CLIC detector model and the push for excellent energy resolution for high energy photons. The results of both approaches are summarized in this chapter – the second design philosophy was ﬁnally chosen for the new CLIC detector model. 5 Silicon Tracker 17 17 5 Silicon Tracker p [GeV] 1 10 2 10 3 10 ] -1 ) [GeV 2 T,true /p T p ∆ ( σ 5 − 10 4 − 10 3 − 10 2 − 10 ° =90 Θ ° =40 Θ ° =20 Θ Figure 17: Preliminary results of the momentum resolution as a function of particle momentum obtained with the CLICdet model (CLIC_o03_v08, head release of 2 Feb 2017). Single muons are passed through the full simulation and reconstruction chain of the CLIC software (simulation and reconstruction head release of 2 Feb 2017). p [GeV] 1 10 2 10 3 10 ] -1 ) [GeV 2 T,true /p T p ∆ ( σ 5 − 10 4 − 10 3 − 10 2 − 10 ° =90 Θ ° =40 Θ ° =20 Θ 10 Figure 17: Preliminary results of the momentum resolution as a function of particle momentum obtained with the CLICdet model (CLIC_o03_v08, head release of 2 Feb 2017). Single muons are passed through the full simulation and reconstruction chain of the CLIC software (simulation and reconstruction head release of 2 Feb 2017). 18 6 Electromagnetic Calorimeter 3In the simulations, the "tungsten" is a mixture of 93% tungsten, 6.1% nickel and 0.9% iron. 6.2 ECAL optimisation with emphasis on high energy photons Following the ﬁndings in the earlier round of studies, as presented in Section 6.1, the ECAL performance in terms of energy resolution for high energy photons was questioned. At 3 TeV CLIC, such photons are very important in a number of physics scenarios, e.g. for the measurement of Higgs decays, possible new heavy resonances, mass of dark matter candidates. For a recent overview, see [23]. Moreover, good ECAL performance is also crucial for the reconstruction of the luminosity spectrum using Bhabha events, since at the highest energies the ECAL resolution is better than the momentum resolution from the tracker [24]. In order to investigate the ECAL performance for different sampling options, a series of full simulation studies was performed [25]. Different ECAL models were simulated, as summarized in Table 8. Since 19 6 Electromagnetic Calorimeter nLayers 15 20 25 30 RMS(E) / Mean(E) 0.04 0.05 0.06 0.07 0.08 0.09 2 SiW 5x5mm (a) 10 GeV photons nLayers 15 20 25 30 RMS(E) / Mean(E) 0.024 0.026 0.028 0.03 0.032 0.034 2 SiW 5x5mm (b) 100 GeV photons Figure 18: Silicon-tungsten (SiW) ECAL energy resolution in the ILD detector for 10 GeV and 100 GeV photons in the barrel, as a function of the number of ECAL layers. Error bars indicate the statistical accuracy of the simulation results. The total thickness of the ECAL is about 23 X0, for all the cases shown (from [22], see Appendix II 14 for software details). nLayers 15 20 25 30 RMS(E) / Mean(E) 0.04 0.05 0.06 0.07 0.08 0.09 2 SiW 5x5mm nLayers 15 20 25 30 RMS(E) / Mean(E) 0.024 0.026 0.028 0.03 0.032 0.034 2 SiW 5x5mm (a) 10 GeV photons (b) 100 GeV photons Figure 18: Silicon-tungsten (SiW) ECAL energy resolution in the ILD detector for 10 GeV and 100 GeV photons in the barrel, as a function of the number of ECAL layers. Error bars indicate the statistical accuracy of the simulation results. The total thickness of the ECAL is about 23 X0, for all the cases shown (from [22], see Appendix II 14 for software details). 6.2 ECAL optimisation with emphasis on high energy photons ECAL Cell Size [mm] 0 5 10 15 20 25 ) [%] j (E 90 ) / Mean j (E 90 RMS 0 1 2 3 4 5 45 GeV Jets 100 GeV Jets 180 GeV Jets 250 GeV Jets (a) as function of ECAL cell size nLayers 15 20 25 30 ) [%] j (E 90 ) / Mean j (E 90 RMS 0 1 2 3 4 5 45 GeV Jets 100 GeV Jets 180 GeV Jets 250 GeV Jets (b) as function of number of ECAL layers Figure 19: Jet energy resolution in the ILD detector for jets of different energies, as a function of the SiW ECAL cell size and number of ECAL layers (keeping a constant depth of about 23 X0). Error bars indicate the statistical accuracy of the simulation results (from [22], see Appendix II 14 for software details). ECAL Cell Size [mm] 0 5 10 15 20 25 ) [%] j (E 90 ) / Mean j (E 90 RMS 0 1 2 3 4 5 45 GeV Jets 100 GeV Jets 180 GeV Jets 250 GeV Jets nLayers 15 20 25 30 ) [%] j (E 90 ) / Mean j (E 90 RMS 0 1 2 3 4 5 45 GeV Jets 100 GeV Jets 180 GeV Jets 250 GeV Jets (a) as function of ECAL cell size (b) as function of number of ECAL layers Figure 19: Jet energy resolution in the ILD detector for jets of different energies, as a function of the SiW ECAL cell size and number of ECAL layers (keeping a constant depth of about 23 X0). Error bars indicate the statistical accuracy of the simulation results (from [22], see Appendix II 14 for software details). leakage into the HCAL was observed for high energy photons, a simultaneous ECAL and HCAL calib- ration using high energy photons was performed. A selection of simulation results, shown in Table 9, illustrates the main ﬁndings. The total energy resolution (from ECAL plus HCAL) as a function of photon energy is shown in Figure 20, for photons hitting the ECAL in the centre of the barrel (θ=90◦, φ=0◦). leakage into the HCAL was observed for high energy photons, a simultaneous ECAL and HCAL calib- ration using high energy photons was performed. A selection of simulation results, shown in Table 9, illustrates the main ﬁndings. 6.2 ECAL optimisation with emphasis on high energy photons The total energy resolution (from ECAL plus HCAL) as a function of photon energy is shown in Figure 20, for photons hitting the ECAL in the centre of the barrel (θ=90◦, φ=0◦). It is noted that CLICdet_17_8 indeed performs worse than an ECAL with 30 layers. Moreover, a 30-layer ECAL with uniform layer thickness (CLICdet_30) was found to give better resolution than the previous 30-layer ECAL assembled in two groups (CLICdet_20_10), with thinner layers ﬁrst and thicker layers later. This is readily understood from the evolution of shower depth as a function of energy: At 20 6 Electromagnetic Calorimeter Table 8: List of the different ECAL models studied. Label of the model Layer structure Thickness Space for Total thickness tungsten alloy sensor and electronics per layer [mm] [mm] [mm] CLICdet_17_8 17 thin + 8 thick 2.4 + 4.8 2 4.4 + 6.8 CLICdet_20_10 20 thin + 10 thick 2 + 4 2 4 + 6 CLICdet_30 30 uniform 2.65 2 4.65 CLICdet_40_a 40 uniform 1.9 2 3.9 CLICdet_40_b 40 uniform 1.9 3.15 5.05 Table 8: List of the different ECAL models studied. Table 9: Photon energy resolution for different ECAL conﬁgurations as deﬁned in Table 8, for photons hitting the central region of the ECAL barrel (θ=90◦, φ=0◦) . The sigma (from a ﬁt to a Gaussian distribution) of the distribution (EECAL+EHCAL)/E is given in [%]. Energy [GeV] 10 50 200 500 1000 1500 CLICdet_17_8 5.90±0.05 2.89±0.02 1.64±0.01 1.14±0.01 0.816±0.006 0.675±0.005 CLICdet_20_10 5.36±0.04 2.66±0.02 1.46±0.01 0.981±0.007 0.711±0.005 0.600±0.004 CLICdet_30 5.43±0.04 2.44±0.02 1.230±0.009 0.790±0.006 0.572±0.004 0.485±0.004 CLICdet_40_a 4.58±0.03 2.07±0.02 1.050±0.008 0.673±0.005 0.491±0.004 0.403±0.003 CLICdet_40_b 4.72±0.03 2.11±0.02 1.077±0.008 0.707±0.005 0.537±0.004 0.438±0.003 higher energies the peak of the distribution is located deeper inside the ECAL, i.e. in the region with thicker tungsten plates in the model CLICdet_20_10. The detector models with uniform ﬁne sampling and 40 layers with 1.9 mm tungsten plates were found to perform best. A slightly larger (but more realistic) space reserved for sensor and electronics slightly reduces the performance, as shown by the comparison of CLICdet_40_a and _b. [GeV] γ true E 0 500 1000 1500 )/E [%] HCal +E ECal (E σ 0 1 2 3 4 5 6 CLICdet_17_8 CLICdet_20_10 CLICdet_30 CLICdet_40_b Figure 20: Resolution for high energy photons of the total ECAL plus HCAL energy, for different ECAL models as deﬁned in Table 8. Photons are incident at θ=90◦and φ=0◦. 6.2 ECAL optimisation with emphasis on high energy photons Figure 20: Resolution for high energy photons of the total ECAL plus HCAL energy, for different ECAL models as deﬁned in Table 8. Photons are incident at θ=90◦and φ=0◦. 21 6 Electromagnetic Calorimeter The energy resolution for the most promising model CLICdet_40_b was further investigated for photons uniformly distributed in the entire barrel region. Results are shown in Table 10. As expected, the resulting energy resolution is slightly worse than the one obtained in the (θ=90◦, φ=0◦) case. The energy resolution for the most promising model CLICdet_40_b was further investigated for photons uniformly distributed in the entire barrel region. Results are shown in Table 10. As expected, the resulting energy resolution is slightly worse than the one obtained in the (θ=90◦, φ=0◦) case. Finally, using the software chain as given in Appendix II 14, the energy resolution studies for photons were also extended to the ECAL endcaps, again using CLICdet_40_b. The initial results showed that the reconstructed energy in the endcap was signiﬁcantly lower than the one of the barrel. A scaling factor of 1.07 on the endcap response had to be introduced to match the response of the barrel. Detailed investigations into the causes of this difference revealed that the effect is due to the magnetic ﬁeld, as described below. Table 10: Photon energy resolution for the ECAL conﬁguration CLICdet_40, with particles uniformly distributed in the entire barrel, compared to photons hitting the centre of the barrel. The sigma of the distribution (EECAL+EHCAL)/E is given. Energy [GeV] 50 500 1500 CLICdet_40_b entire barrel 2.28±0.02 0.955±0.007 0.775±0.006 CLICdet_40_b θ=90◦, φ=0◦ 2.11±0.02 0.707±0.007 0.438±0.003 : Photon energy resolution for the ECAL conﬁguration CLICdet_40, with particles uniformly distributed in the entire barrel, compared to photons hitting the centre of the barrel. The sigma 6.3 Impact of the B-ﬁeld on the ECAL response Investigating the detailed ECAL response for different polar angle regions, it was found that the barrel region gave a 7% higher raw hit energy sum than the endcap region, for all photon energies studied. Detailed studies revealed that the difference was not due to detector geometry, but is instead caused by the magnetic ﬁeld: switching off the detector magnetic ﬁeld moved the barrel response to the lower values found for the endcap. This observation was conﬁrmed with photons using the full PFA reconstruction chain, as shown in Figure 21 for the case of 500 GeV. Aspects of the B-ﬁeld effect on the e.m. shower development in ILD have been investigated earlier, see [26]. h_E_phRECO_over_true Entries 946 Mean 1.001 Std Dev 0.02268 / ndf 2 χ 517.8 / 27 Constant 1.2 ± 22.3 Mean 0.002 ± 1.004 Sigma 0.00113 ± 0.02797 /E_true PFA γ E 0.9 0.95 1 1.05 1.1 0 20 40 60 80 100 h_E_phRECO_over_true Entries 946 Mean 1.001 Std Dev 0.02268 / ndf 2 χ 517.8 / 27 Constant 1.2 ± 22.3 Mean 0.002 ± 1.004 Sigma 0.00113 ± 0.02797 h_E_phRECO_over_true Entries 946 Mean 1.001 Std Dev 0.02268 / ndf 2 χ 517.8 / 27 Constant 1.2 ± 22.3 Mean 0.002 ± 1.004 Sigma 0.00113 ± 0.02797 (a) Before correction - the left peak is endcap domin- ated, the right peak is barrel dominated. h_E_phRECO_over_true Entries 951 Mean 1.029 Std Dev 0.01023 / ndf 2 χ 27.9 / 13 Constant 8.4 ± 204.5 Mean 0.000 ± 1.029 Sigma 0.000229 ± 0.009023 /E_true PFA γ E 0.9 0.95 1 1.05 1.1 0 20 40 60 80 100 120 140 160 180 200 h_E_phRECO_over_true Entries 951 Mean 1.029 Std Dev 0.01023 / ndf 2 χ 27.9 / 13 Constant 8.4 ± 204.5 Mean 0.000 ± 1.029 Sigma 0.000229 ± 0.009023 h_E_phRECO_over_true Entries 951 Mean 1.029 Std Dev 0.01023 / ndf 2 χ 27.9 / 13 Constant 8.4 ± 204.5 Mean 0.000 ± 1.029 Sigma 0.000229 ± 0.009023 (b) After correction - the response frome endcap and barrel overlap. Figure 21: The ECAL response in the endcap-barrel transition region for 500 GeV photons, before and after the correction needed to compensate for the B-ﬁeld effect. 6.3 Impact of the B-ﬁeld on the ECAL response h_E_phRECO_over_true Entries 951 Mean 1.029 Std Dev 0.01023 / ndf 2 χ 27.9 / 13 Constant 8.4 ± 204.5 Mean 0.000 ± 1.029 Sigma 0.000229 ± 0.009023 /E_true PFA γ E 0.9 0.95 1 1.05 1.1 0 20 40 60 80 100 120 140 160 180 200 h_E_phRECO_over_true Entries 951 Mean 1.029 Std Dev 0.01023 / ndf 2 χ 27.9 / 13 Constant 8.4 ± 204.5 Mean 0.000 ± 1.029 Sigma 0.000229 ± 0.009023 h_E_phRECO_over_true Entries 951 Mean 1.029 Std Dev 0.01023 / ndf 2 χ 27.9 / 13 Constant 8.4 ± 204.5 Mean 0.000 ± 1.029 Sigma 0.000229 ± 0.009023 (b) After correction - the response frome endcap and barrel overlap. (a) Before correction - the left peak is endcap domin- ated, the right peak is barrel dominated. (a) Before correction - the left peak is endcap domin- ated, the right peak is barrel dominated. (b) After correction - the response frome endcap and barrel overlap. Figure 21: The ECAL response in the endcap-barrel transition region for 500 GeV photons, before and after the correction needed to compensate for the B-ﬁeld effect. Figure 21: The ECAL response in the endcap-barrel transition region for 500 GeV photons, before and after the correction needed to compensate for the B-ﬁeld effect. 22 6 Electromagnetic Calorimeter 6.4 Optimized ECAL layout as implemented in CLICdet The basic ECAL structure of the CLIC detector is a silicon-tungsten sampling calorimeter with 5 × 5 mm2 silicon detector cells. For improved photon energy resolution, in particular at high energy, the ECAL con- sists of 40 layers, with tungsten plates of 1.9 mm thickness and a distance of 3.15 mm between plates (i.e. the model CLICdet_40_b). The total thickness of the ECAL corresponds to about 22 X0. The overall dimensions of the ECAL as implemented in the simulations are provided in Table 11. Opening the detector for maintenance or repairs, as described in Chapter 11, implies that the ECAL endcap has to be built in two parts, with a small so-called ECAL plug remaining in place attached to the support structures, while the bulk of the endcap is retracted. The necessary empty space between ECAL endcap and ECAL plug is presently assumed to be shaped as a ring, with a width of 30 mm. Table 11: ECAL layout as implemented in the simulation model. ECAL barrel rmin [mm] 1500 ECAL barrel rmax [mm] 1702 ECAL barrel zmax [mm] 2210 ECAL endcap/plug zmin [mm] 2307 ECAL endcap/plug zmax [mm] 2509 ECAL endcap rmin [mm] 410 ECAL endcap rmax [mm] 1700 ECAL plug rmin [mm] 260 ECAL plug rmax [mm] 380 Table 11: ECAL layout as implemented in the simulation model. The ECAL segmentation as implemented in the simulation model is given in Table 12, and shown in Figure 22. Note that the ECAL starts with an absorber layer, followed by a sensor/electronics layer, and so on. The last element in the ECAL stack is a sensor/electronics layer. meters for the ECAL segmentation as implemented in the simulation model, with a total SiW layers Parameters for the ECAL segmentation as implemented in the simulation model, with a total of 40 SiW layers. ble 12: Parameters for the ECAL segmentation as implemented in the simulation mode of 40 SiW layers. Function Material Layer thickness [mm] absorber tungsten alloy 1.90 Insulator G10 0.15 Connectivity mixed (86% Cu) 0.10 Sensor silicon 0.50 Space air 0.10 PCB mixed (82% Cu) 1.30 Space air 0.25 Insulator G10 0.75 Total between W plates 3.15 Total SiW layer 5.05 23 6 Electromagnetic Calorimeter gure 22: Implementation of ECAL and HCAL in the simulation model (the reader may need to zo in to see all the details). 7 Hadronic Calorimeter Similarly to the case of the ECAL, detailed optimisation studies have been performed for the HCAL. Details are described in [27]. As an example result, Figure 23 shows the dependence of the jet energy resolution as a function of the number of layers in the HCAL (keeping the number of λI constant). The dependence of the jet energy resolution on the cell size in the HCAL is shown in Figure 24. Other parameters, such as the scintillator tile thickness, were varied but only a moderate impact on the jet energy resolution was found. Number Of Layers In The HCal 20 30 40 50 60 ) [%] j (E 90 ) / Mean j (E 90 RMS 3 3.5 4 4.5 45 GeV Jets 100 GeV Jets 180 GeV Jets 250 GeV Jets Figure 23: Jet energy resolution in the ILD detector for jets of different energies, as a function of the number of HCAL layers, keeping the number of λI constant. Figure 23: Jet energy resolution in the ILD detector for jets of different energies, as a function of the number of HCAL layers, keeping the number of λI constant. As a result of these studies, the proposed hadronic calorimeter of CLICdet consists of 60 steel absorber plates, each of them 19 mm thick, interleaved with scintillator tiles, similar to the CALICE calorimeter design for the ILD detector. The gap for the sensitive layers and their cassette is 7.5 mm. The overall dimensions of the HCAL are summarized in Table 13. In the simulations, the part of the HCAL endcap which surrounds the ECAL endcap (see Figure 2) is treated as a separate entity called the "HCAL ring". The space needed to place the LumiCal inside the HCAL endcap is referred as "LumiCal cutout". The detailed segmentation parameters as implemented in the simulation model are given in Table 14. The polystyrene scintillator in the cassette is 3 mm thick with a tile size of 30×30 mm2. Analog readout of the tiles with SiPMs is envisaged. Further details are presented in [27]. A section of the HCAL barrel as implemented in the simulations is shown in Figure 22. Both, the endcap and the barrel HCAL, are around 7.5 λI deep, which brings the combined thickness of ECAL and HCAL to 8.5 λI (see Figure 25). 6.4 Optimized ECAL layout as implemented in CLICdet The region of a junction between two sectors (of the dodecagon) the barrel region is shown. In the ECAL, the olive-green regions indicate the tungsten laye while the red regions symbolize the silicon sensors. Purple layers are G10, green is PCB a connectivity, and white is air. In the HCAL, blue regions indicate the steel layers (with th steel sheets for the cassette), red stands for the scintillator, while green and white are PCB a air as in the ECAL. gure 22: Implementation of ECAL and HCAL in the simulation model (the reader may need to zoo in to see all the details). The region of a junction between two sectors (of the dodecagon) the barrel region is shown. In the ECAL, the olive-green regions indicate the tungsten laye while the red regions symbolize the silicon sensors. Purple layers are G10, green is PCB a connectivity, and white is air. In the HCAL, blue regions indicate the steel layers (with th steel sheets for the cassette), red stands for the scintillator, while green and white are PCB a air as in the ECAL. Figure 22: Implementation of ECAL and HCAL in the simulation model (the reader may need to zoom in to see all the details). The region of a junction between two sectors (of the dodecagon) in the barrel region is shown. In the ECAL, the olive-green regions indicate the tungsten layers, while the red regions symbolize the silicon sensors. Purple layers are G10, green is PCB and connectivity, and white is air. In the HCAL, blue regions indicate the steel layers (with thin steel sheets for the cassette), red stands for the scintillator, while green and white are PCB and air as in the ECAL. Figure 22: Implementation of ECAL and HCAL in the simulation model (the reader may need to zoom in to see all the details). The region of a junction between two sectors (of the dodecagon) in the barrel region is shown. In the ECAL, the olive-green regions indicate the tungsten layers, while the red regions symbolize the silicon sensors. Purple layers are G10, green is PCB and connectivity, and white is air. In the HCAL, blue regions indicate the steel layers (with thin steel sheets for the cassette), red stands for the scintillator, while green and white are PCB and air as in the ECAL. 24 7 Hadronic Calorimeter 7 Hadronic Calorimeter In the studies performed in preparation of the CDR, this depth of the calorimetry for hadrons was found to be sufﬁcient [28]. The overlap of mW and mZ measurements from the invariant mass of the two jets in WW →νℓj j and ZZ →νν j j events, generated at various c.m. energies, is similar for a steel- and a tungsten-based calor- imeter, as illustrated in Figure 26. The solid lines in the ﬁgure show the overlap for two different HCAL models, one using steel (blue) and the other using tungsten (red) as absorber material for identical λI, in the absence of beam-induced background. The dashed lines show the same measurements performed after the overlay of 60 bunch crossings of γγ →hadrons background events generated at √s = 3 TeV. The overlap is used as an indication of the jet energy resolution (JER) obtained from each HCAL model. 25 7 Hadronic Calorimeter HCal Cell Lateral Size [mm] 0 20 40 60 80 100 ) [%] j (E 90 ) / Mean j (E 90 RMS 2.5 3 3.5 4 45 GeV Jets 100 GeV Jets 180 GeV Jets 250 GeV Jets Figure 24: Jet energy resolution in the ILD detector for jets of different energies, as a function of the HCAL cell size. The cells are squared. Figure 24: Jet energy resolution in the ILD detector for jets of different energies, as a function of the HCAL cell size. The cells are squared. Table 13: HCAL overall layout as implemented in the simulation model. HCAL barrel rmin [mm] 1740 HCAL barrel rmax [mm] 3330 HCAL barrel zmax [mm] 2210 HCAL endcap zmin [mm] 2539 HCAL endcap zmax [mm] 4129 HCAL endcap rmin [mm] 250 HCAL endcap rmax [mm] 3246 HCAL ring zmin [mm] 2360 HCAL ring zmax [mm] 2539 HCAL ring rmin [mm] 1730 HCAL ring rmax [mm] 3246 LumiCal cutout in HCAL rmax [mm] 180 LumiCal cutout in HCAL ztot [mm] 200 Table 13: HCAL overall layout as implemented in the simulation model. A larger overlap corresponds to a poorer separation between the W and Z invariant mass peaks which suggests a reduced JER. Since the performance of the two calorimeter options is very similar, steel was chosen as absorber material, for reasons of cost and complexity of machining and assembling of the tungsten plates. 7 Hadronic Calorimeter Moving the QD0 outside of the detector region reduces the diameter of the support tube w.r.t. the CDR detector models. This in turn allows one to improve the forward coverage of the HCAL, which in CLICdet has an endcap inner radius of 250 mm. As an illustration of the advantage of better forward coverage, a di-jet invariant mass measurement for jets in the forward region in ZZ →νν j j events for three HCAL models featuring different HCAL endcap inner radii has been performed in full simulations for several jet energies. For example, Figure 27 shows the results for physics events overlaid with 60 bunch crossings of γγ →hadrons beam induced background. For reasons of computational efﬁciency, the ZZ were generated at √s = 1 TeV while 26 7 Hadronic Calorimeter Table 14: Parameters for the HCAL segmentation as implemented in the simulation model, with a total of 60 Fe-Scintillator layers. Function Material Layer thickness [mm] Absorber steel 19 Space air 2.7 Cassette Steel 0.5 PCB mixed 0.7 Conductor Cu 0.1 Scintillator Polystyrene 3 Cassette Steel 0.5 Total between steel plates 7.5 Total Fe-scint. layer 26.5 [deg] θ Iλ # 0 2 4 6 8 10 12 0 10 20 30 40 50 60 70 80 90 ECal HCal Full Calorimetry Coil Figure 25: Nuclear interaction lengths λI in the calorimeters with respect to the polar angle θ. The in- teraction length corresponding to the material of the superconducting coil is shown for com- pleteness. : Parameters for the HCAL segmentation as implemented in the simulation model, with a total of 60 Fe-Scintillator layers. [deg] θ Iλ # 0 2 4 6 8 10 12 0 10 20 30 40 50 60 70 80 90 ECal HCal Full Calorimetry Coil [deg] θ Iλ # 0 2 4 6 8 10 12 0 10 20 30 40 50 60 70 80 90 ECal HCal Full Calorimetry Coil Figure 25: Nuclear interaction lengths λI in the calorimeters with respect to the polar angle θ. The in- teraction length corresponding to the material of the superconducting coil is shown for com- pleteness. the background events were taken from the worst case scenario, i.e. generated for CLIC operating at √s = 3 TeV. 7 Hadronic Calorimeter The kt algorithm of FastJet was used to reconstruct exclusively two jets on the same side of the detector (after optimisation, the FastJet parameter R=0.5 and the "default" criterion for PFOs were used throughout to estimate the relative performance of the models) . The Rin = 240 mm model provides an improved di-jet invariant mass, without signiﬁcantly increasing the acceptance to background compared to the Rin = 360 mm case. The Rin = 120 mm case, considered not feasible from an engineering perspective, is included for comparison. the background events were taken from the worst case scenario, i.e. generated for CLIC operating at √s = 3 TeV. The kt algorithm of FastJet was used to reconstruct exclusively two jets on the same side of the detector (after optimisation, the FastJet parameter R=0.5 and the "default" criterion for PFOs were used throughout to estimate the relative performance of the models) . The Rin = 240 mm model provides an improved di-jet invariant mass, without signiﬁcantly increasing the acceptance to background compared to the Rin = 360 mm case. The Rin = 120 mm case, considered not feasible from an engineering perspective, is included for comparison. While the advantage of a better HCAL forward coverage is evident even including the overlay of γγ →hadrons background, detailed studies for CLIC_ILD revealed a too high occupancy in the inner regions of the HCAL endcap. As a remedy, a thick shielding (polyethylene and tungsten) inside the support tube was proposed [29]. In CLICdet, a shielding of this type would have to be placed inside the HCAL endcap and would therefore reduce its forward acceptance. For ﬁrst physics studies with the new 27 7 Hadronic Calorimeter Typical Jet Energies [GeV] 100 200 300 400 500 Overlap [%] Z /m W m 0 5 10 15 20 25 Fe W Fe, 60 BX overlay W, 60 BX overlay Typical Jet Energies [GeV] 100 200 300 400 500 Overlap [%] Z /m W m 0 5 10 15 20 25 Fe W Fe, 60 BX overlay W, 60 BX overlay Figure 26: Overlap, in percent, of mW and mZ measurements from the invariant mass of the two jets in WW →νℓj j and ZZ →νν j j events, respectively, as a function of the typical energy of the jets in the events. 8.1 Superconducting Solenoid The solenoid magnetic ﬁeld for the new CLIC detector model is chosen to be 4 T. This is understood to be a reasonable compromise between what is strictly necessary from the minimal requirements for tracking (3.5 T would be sufﬁcient) and what is desirable, i.e. to have some margin in the detector performance while maintaining realistic solenoid parameters. In the simulation model, the magnetic ﬁeld in CLICdet is 4 T throughout the volume inside the super- conducting coil. The ﬁeld in the yoke is 1.5 T pointing in the opposite direction with respect to the inner ﬁeld. Beyond the end of the coil the ﬁeld decreases rapidly in Z-direction. Therefore, the simulation model currently assumes no ﬁeld in the yoke endcap. In the CDR, the design of the more challenging solenoid of CLIC_SiD, with smaller dimensions but a 5 T ﬁeld, was described. The parameters of the solenoid for CLICdet are similar to the one of the ILD detector at the ILC, described in detail in [15]. The overall dimensions of the solenoid and yoke of CLICdet are shown in Figure 4. At CLIC an anti-DiD dipole such as described for ILD is excluded: the luminosity loss due to synchro- tron radiation by the beam in such a dipole ﬁeld would be too large [30]. The main design parameters of the superconducting solenoid are given in Table 15. The solenoid of CLICdet is implemented in the simulation model with parameters as shown in Table 16. The material budget of the solenoid, in terms of λI, is indicated in Figure 25. The solenoid for CLICdet is more challenging than the one of CMS, but is considered feasible using the technology available today. Due to the larger coil radius, the total magnetic ﬂux in CLICdet is about 45% higher than the one in CMS. It was also noted during the discussions that the aspect ratio of the solenoid in CLICdet is not favourable to achieve a very homogenous ﬁeld – however, external constraints (position of QD0) make it difﬁcult to make the solenoid longer for the same diameter. 7 Hadronic Calorimeter Solid lines show results without background events, dashed lines results with 60 bunch-crossings of γγ →hadrons background overlaid. Figure 26: Overlap, in percent, of mW and mZ measurements from the invariant mass of the two jets in WW →νℓj j and ZZ →νν j j events, respectively, as a function of the typical energy of the jets in the events. Solid lines show results without background events, dashed lines results with 60 bunch-crossings of γγ →hadrons background overlaid. detector model, it was decided not to include such a shielding. At a later stage, smaller sensor cell sizes in the forward HCAL, possibly combined with shielding, may have to be introduced. [GeV] JJ m 0 50 100 150 200 Entries/2 GeV 0 50 100 150 200 250 300 =120 mm in R =240 mm in R =360 mm in R =3 TeV s had @ → γ γ 60 BX =1 TeV s jj @ ν ν → ZZ Jets with R=0.5 Default Selected PFOs Figure 27: Di-jet invariant mass measurement for jets in the forward region in ZZ →νν j j events for three HCAL models featuring different HCAL endcap radii. Figure 27: Di-jet invariant mass measurement for jets in the forward region in ZZ →νν j j events for three HCAL models featuring different HCAL endcap radii. 28 8 Magnet System 8.1 Superconducting Solenoid Table 15: CLICdet solenoid main parameters Solenoid central ﬁeld [T] 4.0 Nominal current [kA] 20 Maximum ﬁeld on conductor [T] 4.6 Total ampere-turns solenoid [MAt] 29.8 Inductance [H] 11.6 Coil inner radius [mm] 3649 Stored energy [GJ] 2.32 Coil outer radius [mm] 3993 Stored energy per unit of cold mass [kJ/kg] 12.9 Coil length [mm] 7800 Table 15: CLICdet solenoid main parameters Table 16: Description of the coil elements as implemented in the simulation model. For all elements the material, the longitudinal extent in one half of the detector zmin/max and the radial extent rmin/max are given. Table 16: Description of the coil elements as implemented in the simulation model. For all elements the material, the longitudinal extent in one half of the detector zmin/max and the radial extent rmin/max are given. Material zmin [mm] zmax [mm] rmin [mm] rmax [mm] Steel 0 4129 3483 3523 Vacuum 0 4129 3523 3649 Aluminium 0 3900 3649 3993 Vacuum 0 4129 3993 4250 Steel 0 4129 4250 4290 Vacuum 3900 4089 3649 3993 Steel 4089 4129 3483 4290 Material zmin [mm] zmax [mm] rmin [mm] rmax [mm] Steel 0 4129 3483 3523 Vacuum 0 4129 3523 3649 Aluminium 0 3900 3649 3993 Vacuum 0 4129 3993 4250 Steel 0 4129 4250 4290 Vacuum 3900 4089 3649 3993 Steel 4089 4129 3483 4290 29 8 Magnet System 4Note that these end coils are not part of the detector simulation model. 4Note that these end coils are not part of the detector simulation model. ese end coils are not part of the detector simulation model. 8.2 Yoke and Muon Detectors The iron return yoke is structured into three rings in the barrel region and the two endcaps, as shown in Figure 28. The outer radius of the yoke is reduced w.r.t. the CDR detector models, since the external constraints (stray magnetic ﬁeld, shielding) are relaxed in a scenario with only one detector. The yoke endcaps are much thinner than in the CDR to allow for an L∗of 6 m, with the QD0 outside of the detector region. Figure 28: The iron return yoke of CLICdet. The end coils are not shown in this illustration. Figure 28: The iron return yoke of CLICdet. The end coils are not shown in this illustration. In order to close the magnetic ﬂux, a set of 4 concentric end coils is foreseen4, as shown schematically in Figure 4. An earlier version of end coils is presented in [31], where the feasibility is demonstrated by a comparison with the coils used at the LHCb experiment. The main design parameters of the new end coils are given in Table 17. The effect of these end coils on the magnetic ﬁeld outside of the detector region is shown in Figure 29. The peak value of the ﬁeld decreases outside of the detector region (z > 6 m) when end coils are included, e.g. from about 0.5 T to 0.08 T at z = 6.5 m. Table 17: Main design parameters of the four end coils per endcap. The dimensions of the coils are shown in Figure 4. Coil # turns Copper mass [ton] Resistance [mΩ] Voltage drop [V] Power [kW] RC1 4 × 12 5.6 2.7 16.5 101 RC2 3 × 20 13.3 6.4 39.1 240 RC3 4 × 24 46.7 22.4 137.6 844 RC4 4 × 18 45.8 22.0 135.1 829 : Main design parameters of the four end coils per endcap. The dimensions of the coils are shown in Figure 4. In the design of the yoke, mechanical constraints impose steel plate thicknesses of at least 100 mm. This implies that, in a thinner yoke, the number of muon detection layers needs to be reduced with respect to the CDR. Fewer muon layers are not expected to impair the performance of the muon identiﬁcation at CLIC. This had already been anticipated in the note describing the design studies [32], which states that the "muon identiﬁcation algorithm within PANDORAPFANEW leaves room for improvement". 9 LumiCal, BeamCal and Intratrain Feedback Two smaller electromagnetic calorimeters, LumiCal and BeamCal, are installed in the very forward region, on either side of the IP. Their speciﬁc purposes (luminosity measurement and monitoring of collisions) require that they are aligned with the outgoing beam axis. Moreover, these two calorimeters provide coverage for reconstructing electrons and photons down to very small angles. The positioning and overall dimensions of LumiCal and BeamCal are given in Table 18. Behind BeamCal, which acts as shielding for this more sensitive equipment, a beam position monitor (BPM) and kicker system for the intra-train feedback is installed. The basic design of the calorimeters and the feedback components remains as in CLIC_ILD. The positioning and radii of these elements are sligthly changed in the new detector model CLICdet. 8.2 Yoke and Muon Detectors A 30 8 Magnet System Figure 29: The longitudinal magnetic ﬁeld on axis, with and without end coils (RCs). Note that the yoke endcap extends to z = 5.7 m. Figure 29: The longitudinal magnetic ﬁeld on axis, with and without end coils (RCs). Note that the yoke endcap extends to z = 5.7 m. muon system with 6 instead of 9 layers (as in the CDR) should therefore be sufﬁcient, with an additional 7th layer in the barrel as close as possible to the coil. Similarly, [32] states that the performance for equidistant muon detection layers is very similar to the model chosen for the CDR, where three groups of three layers were proposed. The muon system layout in CLICdet is shown in Figure 30. 2 As in the CDR, the muon detection layers are proposed to be built as RPCs with cells of 30×30 mm2 (alternatively, crossed scintillator bars could be envisaged). The free space between yoke steel layers is 40 mm, which is considered generous given present-day technologies for building RPCs. In analogy to CMS, the yoke layers and thus the muon detectors are staggered to avoid gaps (see Figure 2). As in the CDR, the muon detection layers are proposed to be built as RPCs with cells of 30×30 mm2 (alternatively, crossed scintillator bars could be envisaged). The free space between yoke steel layers is 40 mm, which is considered generous given present-day technologies for building RPCs. In analogy to CMS, the yoke layers and thus the muon detectors are staggered to avoid gaps (see Figure 2). 31 8 Magnet System Figure 30: Schematic cross section of the muon system layout in the yoke of CLICdet. The staggering of the the layers is not visible in this cross section. Figure 30: Schematic cross section of the muon system layout in the yoke of CLICdet. The staggering of the the layers is not visible in this cross section. 32 9 LumiCal, BeamCal and Intratrain Feedback 9.2 Beam Calorimeter (BeamCal) The BeamCal completes the coverage of the electromagnetic calorimeter down to 10 mrad (see Fig- ure 31). Like the LumiCal, the BeamCal consists of 40 layers of 3.5 mm tungsten plates. The sensor material has to be radiation hard (up to 1 MGy per year), because the BeamCal has to absorb a large fraction of the incoherent electron-positron pairs from the beam-beam interaction. In the simulation 0.3 mm thick diamond layers are used as active elements. The BeamCal segmentation implemented in the software model of CLICdet consists of a regular pattern of 8 × 8 mm2 cells. To reduce the number of particles scattering back into the detector, the side of the BeamCal facing the IP is covered with a 100 mm thick graphite layer. The layout as implemented in the simulation model is shown in Figure 31. Table 18: Positioning and dimensions of LumiCal and BeamCal – the longitudinal dimension indicated for BeamCal includes the 100 mm of graphite absorber. The geometrical acceptance angles are given. In practice, the physical acceptance depends on shower containment. Note that LumiCal is partially in the shadow of the ECAL plug (see Figure 19). Zstart [mm] Zend [mm] Rin [mm] Rout [mm] θin [mrad] θout [mrad] LumiCal 2539 2710 100 340 39 134 BeamCal 3181 3441 32 150 10 46 Table 18: Positioning and dimensions of LumiCal and BeamCal – the longitudinal dimension indicated for BeamCal includes the 100 mm of graphite absorber. The geometrical acceptance angles are given. In practice, the physical acceptance depends on shower containment. Note that LumiCal is partially in the shadow of the ECAL plug (see Figure 19). Table 18: Positioning and dimensions of LumiCal and BeamCal – the longitudinal dimension indicated for BeamCal includes the 100 mm of graphite absorber. The geometrical acceptance angles are given. In practice, the physical acceptance depends on shower containment. Note that LumiCal is partially in the shadow of the ECAL plug (see Figure 19). Zstart [mm] Zend [mm] Rin [mm] Rout [mm] θin [mrad] θout [mrad] LumiCal 2539 2710 100 340 39 134 BeamCal 3181 3441 32 150 10 46 9.1 Luminosity Calorimeter (LumiCal) The main purpose of the LumiCal is the precise measurement of electrons and positrons from Bhabha scattering events. It consists of 40 layers of 3.5 mm tungsten and 0.32 mm silicon sensors. The seg- mentation of the sensor layers into sectors, and of these sectors into pads, is described in [33], however, due to its different longitudinal position in CLICdet, the LumiCal now has 64 radial and 48 azimuthal segments per layer. Each layer also contains 0.2 mm support (consisting of epoxy, kapton and copper) and 0.25 mm air gaps (see Table 19). The position and size of the LumiCal provide an overlap of the coverage with the ECAL endcap (see Figure 31). 9.3 Intra-train Feedback A system consisting of BPMs and fast kickers on either side of the interaction point is foreseen [34], allowing to partially correct luminosity loss due to vibrations and other imperfections. As schematically shown in Figure 31, these elements are placed as close as possible to the IP to reduce latency in the feedback. On the other hand, it is mandatory to install the BPM and kicker beyond BeamCal, which is acting as shielding for these elements as well as for the QD0. In the layout of CLICdet, the centre of the 33 9 LumiCal, BeamCal and Intratrain Feedback 2539 3181 3455 3730 z[mm] Figure 31: Layout of the forward region in CLICdet, seen from the top, as implemented in the simulation model. LumiCal and BeamCal are shown in black, a graphite absorber upstream of BeamCal (to reduce back-scattering) is shown in white. Downstream of BeamCal the kicker (on the incoming beam) and the beam position monitor (on the outgoing beam) are indicated. Both are components of the intra-train feedback system. The arrows indicate the direction of the beams. 2539 3181 3455 3730 z[mm] Figure 31: Layout of the forward region in CLICdet, seen from the top, as implemented in the simulation model. LumiCal and BeamCal are shown in black, a graphite absorber upstream of BeamCal (to reduce back-scattering) is shown in white. Downstream of BeamCal the kicker (on the incoming beam) and the beam position monitor (on the outgoing beam) are indicated. Both are components of the intra-train feedback system. The arrows indicate the direction of the beams. Table 19: Parameters of the forward calorimeter layer structure. The number of layers NLayers, their thickness dLayer, the absorber material, its thickness dAbs, the active material and its thickness dAct are given. NLayers dLayer [mm] Absorber dAbs [mm] Active dAct [mm] LumiCal 40 4.27 Tungsten 3.5 Silicon 0.32 BeamCal 40 4.00 Tungsten 3.5 Diamond 0.30 Table 19: Parameters of the forward calorimeter layer structure. The number of layers NLayers, their thickness dLayer, the absorber material, its thickness dAbs, the active material and its thickness dAct are given. NLayers dLayer [mm] Absorber dAbs [mm] Active dAct [mm] LumiCal 40 4.27 Tungsten 3.5 Silicon 0.32 BeamCal 40 4.00 Tungsten 3.5 Diamond 0.30 BPM (installed on the outgoing beam, assumed to be 90 mm long) is located at 3775 mm from the IP. 10 Beam Pipe and Vacuum System The vacuum system in the detector is separated by valves from the accelerator vacuum (QD0 and bey- ond). A schematic view of the vacuum system is given in Figure 32, showing the thinner incoming and the larger outgoing beam pipes at the crossing angle of 20 mrad. The separating valves and additional pumping ports are considered part of the accelerator system and are not shown in the ﬁgure. The main difference between this vacuum system layout and the one proposed in the CDR is the ab- sence of a large vacuum valve just downstream of the LumiCal detector. The CDR layout was mainly motivated by the need for a rapid "push-pull" exchange of the detector on the IP, with little time for pumping the central region. However, without "push-pull" the rather modest requirements for the beam- line vacuum of about 10−6 mbar in the detector region imply that the full detector can be vented with dry nitrogen when an intervention (detector opening for maintenance) is needed. Another difference concerns the beam pipe shapes and diameters: for the outgoing beam, the conical beam pipe in the region of BeamCal and BPM is replaced by a cylindrical pipe, and for the incoming beam, the cylindrical beam pipe has a larger diameter than in the CDR, making it less fragile. This is possible since, in the absence of QD0, more space is available inside the support tube. Figure 32: Schematic view of the components of the vacuum system in the detector region (the IP is located at the right hand edge of the ﬁgure). Figure 32: Schematic view of the components of the vacuum system in the detector region (the IP is located at the right hand edge of the ﬁgure). Figure 32: Schematic view of the components of the vacuum system in the detector region (the IP is located at the right hand edge of the ﬁgure). The geometrical parameters of the different parts of the beam pipe as implemented in the simulation model are described in Table 20. Around the interaction point the beam pipe consists of a beryllium cylinder. The next part of the beam pipe is a steel cone with a half-opening of 6.6◦. 9.3 Intra-train Feedback The centre of the kicker (on the incoming beam, assumed to be 260 mm long) is located at 3585 mm. BPM (installed on the outgoing beam, assumed to be 90 mm long) is located at 3775 mm from the IP. The centre of the kicker (on the incoming beam, assumed to be 260 mm long) is located at 3585 mm. 34 10 Beam Pipe and Vacuum System D Conical beam pipe with an half-opening angle of 10 mrad. A Alignment of beam pipe part: 0 (aligned on detector axis), 1 (aligned on incoming beam axis), 2 (aligned on outgoing beam axis). B O 10 Beam Pipe and Vacuum System The wall thickness of this cone increases from 4 mm in the most central region to 4.8 mm in the region of largest diameter of the cone (engineering rule of thumb: the wall thickness should not be less than 1% of the diameter). The steel pipe acts as absorber for particles scattering back from the forward region [11]. The maximal cone radius is given by the inner radius of the ECAL endcap "plug". At a radius of 240 mm the beam pipe becomes cylindrical until it reaches the front face of the LumiCal. Just upstream of the LumiCal, the beam pipe diameter is reduced from its full opening to ﬁt the 100 mm inner radius of LumiCal, which is centred on the outgoing beam axis. The corresponding vertical end- ﬂange is assumed to be 3 mm thick. The beam pipe changes again before the BeamCal, when it splits into two separate pipes: A thin cylindrical beam pipe for the incoming beam, and a larger cylindrical beam pipe for the outgoing beam. 35 10 Beam Pipe and Vacuum System Table 20: Parameters for the beam pipe parts as implemented in the simulation model. Each par cylinder barrel or cone positioned between Z1 to Z2, the radii RIn/Out 1/2 are the inner and radii at position Z1/2 respectively. &A Z1 [mm] Z2 [mm] RIn 1 [mm] RIn 2 [mm] ROut 1 [mm] ROut 2 [mm] Material 0 0 308 29.4 29.4 30.0 30.0 Beryllium 0 308 337 29.4 29.4 30.0 33.4 Iron 0 337 2080 29.4 235.2 33.4 240.0 Iron 0 2080 2528 235.2 235.2 240.0 240.0 Iron 0B 2528 2531 0.0 98.0 240.0 240.0 Iron 2 2531 3170 98.0 98.0 99.0 99.0 Iron 2C 3170 3173 2.7 31.0 99.0 99.0 Iron 2 3173 3500 31.0 31.0 32.0 32.0 Iron 2D 3500 12500 31.0 125.0 32.0 127.0 Iron 1 3173 3281 2.7 2.7 3.7 3.7 Iron 1 3281 3835 2.7 2.7 3.7 3.7 Iron 1 3835 3845 2.7 12.0 3.7 13.5 Iron 1 3845 12500 12.0 12.0 13.5 13.5 Iron A Table 20: Parameters for the beam pipe parts as implemented in the simulation model. Each part is a cylinder barrel or cone positioned between Z1 to Z2, the radii RIn/Out 1/2 are the inner and outer radii at position Z1/2 respectively. D Conical beam pipe with an half-opening angle of 10 mrad. 10 Beam Pipe and Vacuum System A Alignment of beam pipe part: 0 (aligned on detector axis), 1 (aligned on incoming beam axis), 2 (aligned on outgoing beam axis). B O B Beam pipe end in front of LumiCal: ROut 1 is the size of the hole where the beam pipe inside LumiCal is connected. The hole is centred on the outgoing beam axis. C I C Beam pipe end in front of BeamCal: RIn 1 is the size of the hole for the incoming beam pipe, ROut 1 is the size of the hole for the outgoing beam pipe. D 36 11 Detector Opening and Maintenance 11 Detector Opening and Maintenance The preliminary procedure for detector opening and maintenance, as outlined in the CDR, can be applied to CLICdet with minor changes. The main steps are: The preliminary procedure for detector opening and maintenance, as outlined in the CDR, can be applied to CLICdet with minor changes. The main steps are: • close all beam pipe vacuum valves, vent the beam pipe in the detector region with a dry gas (e.g. nitrogen); • open the vacuum on the detector-side of QD0, close the detector vacuum system by ﬂanges; • move the detector from the IP to the garage position in the cavern; • move the detector from the IP to the garage position in the cavern; • the support tube installation/extraction tool is installed behind the endcap; e support tube installation/extraction tool is installed behind the endcap; • the jacks holding the support tube in position are retracted; • the jacks holding the support tube in position are retracted; • the endcap is slid back, giving access to the vacuum connection between forward region and central detector; • open vacuum connection near LumiCal, remove bellows as shown in Figure 33; • LumiCal is opened sideways, and the support tube can be retracted some 50 cm by the extraction tool; • close the vacuum system on all sides by ﬂanges; • remove the support tube as a whole by a crane; In case the inner detector system needs to be extracted, both endcaps need to be opened as described above: On one side, the endcap needs to be retracted by 4–5 m to make space for a support platform. On the opposite side, the opening needs to be sufﬁcient to disconnect the vacuum system at the level of the bellows near the LumiCal. Tooling and procedures needed to extract the inner detector system are currently under study. Figure 33: Removal of the bellows downstream of LumiCal. Figure 33: Removal of the bellows downstream of LumiCal. 37 37 12 Summary and Outlook 12 Summary and Outlook This note describes the new CLIC detector model, CLICdet, and all of its sub-systems in its version of spring 2017. This detector model has been implemented in the new, DD4hep-based CLIC detector software chain. The simulation and reconstruction processes are currently being ﬁnalized. Validation studies are ongoing, in view of physics benchmark studies with full detector simulations. 38 13 Appendix I 13 Appendix I 13 Appendix I Table 21: Overview of CLICdet sub-detector parameters. Note that numbers for barrel detectors or disks contain the sum of both sides of the experiment. Subdetector Sensor area [m2] Cell size [mm2] Number of channels [106] VTX barrel 0.487 0.025×0.025 780 VTX spirals 0.351 0.025×0.025 560 Inner Tracker Disk ITD1 1.26 0.025×0.025 2000 Inner Tracker Disk ITD2 2.26 0.05×1 46 Inner Tracker Disk ITD3 2.20 0.05×1 44 Inner Tracker Disk ITD4 2.06 0.05×1 42 Inner Tracker Disk ITD5 1.96 0.05×1 40 Inner Tracker Disk ITD6 1.88 0.05×1 38 Inner Tracker Disk ITD7 1.82 0.05×1 36 Outer Tracker Disk OTD1 13.92 0.05×10 27.8 Outer Tracker Disk OTD2 13.92 0.05×10 27.8 Outer Tracker Disk OTD3 13.92 0.05×10 27.8 Outer Tracker Disk OTD4 13.92 0.05×10 27.8 Inner Tracker Barrel ITB1 0.79 0.05×1 16 Inner Tracker Barrel ITB2 2.20 0.05×1 44 Inner Tracker Barrel ITB3 5.22 0.05×5 21 Outer Tracker Barrel OTB1 14.30 0.05×10 29 Outer Tracker Barrel OTB2 20.32 0.05×10 41 Outer Tracker Barrel OTB3 26.04 0.05×10 52 ECAL barrel 1808 5×5 72 ECAL endcaps (incl. ECAL plugs) 726 5×5 29 HCAL barrel 4346 30×30 4.8 HCAL endcaps 4041 30×30 4.5 HCAL rings 340 30×30 0.4 MUON barrel 1942 30×30 2.2 MUON endcaps 1547 30×30 1.7 LumiCal 26.5 3.75×(13 – 44) 0.25 BeamCal 5.4 8×8 0.093 39 14 Appendix II 14 Appendix II Table 22: Pandora parameters and software versions used to produce the results shown in Figures 18 and 19 [22]. HCAL Timing Cuts 100 ns ECAL Timing Cuts 100 ns HCAL Hadronic Cell Truncation 1 GeV (Optimal for Default HCAL) Software ilcsoft_v01-17-07, including PandoraPFA v02-00-00 Digitiser ILDCaloDigi, realistic ECAL and HCAL digitisation options enabled Calibration PandoraAnalysis toolkit v01-00-00 Table 23: Pandora parameters and software versions used to produce the results shown in Figures 20 and 21 - from [35]. Detector model CLIC_o3_v06 HCAL Timing Cuts 10 ns ECAL Timing Cuts 10 ns HCALHadronic Cell Truncation 1 GeV (Optimal for Default HCAL) Software simulation DD4hep 0.18, iLCSoft package from CLICdp builds 2016-11-09 Reconstruction DDMarlinPandora, using PandoraPFA v03-00-00 Digitiser DDMarlinCaloDigi, realistic ECAL and HCALl digitisation options NOT enabled Calibration PandoraAnalysis toolkit v01-00-00 40 15 Appendix III 15 Appendix III 15 Appendix III List of changes introduced on 5 April 2019: List of changes introduced on 5 April 2019: • numbers in Appendix I were corrected and LumiCal / BeamCal numbers added; • the present layout of LumiCal, i.e. 64 radial and 48 azimuthal segments per layer, was added in section 9.1. • the present layout of LumiCal, i.e. 64 radial and 48 azimuthal segments per layer, was added in section 9.1. 41 41 References References [1] A. Munnich A., A. Sailer, The CLIC-ILD-CDR Geometry for the CDR Monte Carlo Mass Production (2011), LCD-Note-2011-002. [2] C. Grefe, A. Munnich, The CLIC-SiD-CDR Detector Model for the CLIC CDR Monte Carlo Mass Production (2011), LCD-Note-2011-009. [3] L. Linssen et al., eds., Physics and Detectors at CLIC: CLIC Conceptual Design Report, vol. CERN-2012-003, 2012. [4] CLIC, CLICdp, Updated baseline for a staged Compact Linear Collider, CERN-2016-004, CERN, 2016. [5] M. Regler, et al., LiC Detector Toy, Vienna Fast Simulation Tool for Charged Tracks, http://www.hephy.at/project/ilc/lictoy/. [6] T. Behnke, et al., The International Linear Collider Technical Design Report - Volume 4: Detectors, CERN-ATS-2013-037, 2013. [7] M. Killenberg, Occupancy in the CLIC ILD Time Projection Chamber (2011), LCD-Note-2011-029. [8] M. Killenberg, Occupancy in the CLIC ILD Time Projection Chamber using Pixelised Readout (2013), LCD-Note-2013-005. [9] M. Thomson, Particle Flow Calorimetry and the PandoraPFA Algorithm, Nucl. Instrum. Meth. A 611 (2009) 25. [10] R. Gluckstern, Uncertainties in track momentum and direction, due to multiple scattering and measurement errors, Nucl. Instrum. Meth. 24 (1963) 381. [11] D. Dannheim, A. Sailer, Beam-Induced Backgrounds in the CLIC Detectors (2011), LCD-Note-2011-021. [12] N. Alipour Tehrani, Optimisation studies for the CLIC vertex-detector geometry (2015), CLICdp-Conf-2015-002. [13] A. Herve, Different presentations on design and costing of superconducting magnet systems for experiments, https://edms.cern.ch/document/1740563/, 2008 - 2010. [14] B. Parker et al., Functional Requirements on the Design of the Detectors and the Interaction Region of an e+e- Linear Collider with a Push-Pull Arrangement of Detectors (2009), ILD-Note-2009-050. [15] F. Kircher et al., Conceptual Design of the ILD Detector Magnet System (2012), LC-DET-2012-081. [16] A. Bartalesi, M. Modena, 3D FEA Computation of the CLIC Machine Detector Interface Magnets (2012). [17] F. Duarte Ramos, H. Gerwig, M. Villarejo Bermudez, CLIC inner detectors cooling simulations (2013), LCD-Note-2013-007. [18] N. Alipour Tehrani, P. Roloff, Optimisation Studies for the CLIC Vertex-Detector Geometry (2014), CLICdp-Note-2014-002. [19] S. Sroka et al., New CLIC detector model - engineering input to the vertex and tracker layout, EDMS document No. 1706383, 2016. 42 References [20] F. Duarte Ramos, W. Klempt, F.-X. Nuiry, Experimental tests on the air cooling of the CLIC vertex detector (2015), CLICdp-Note-2016-002. [20] F. Duarte Ramos, W. Klempt, F.-X. Nuiry, Experimental tests on the air cooling of the CLIC vertex detector (2015), CLICdp-Note-2016-002. [20] F. Duarte Ramos, W. Klempt, F.-X. Nuiry, Experimental tests on the air cooling of the CLIC vertex detector (2015), CLICdp-Note-2016-002. [21] M. References Villarejo Bermudez, F. Duarte Ramos, H. Gerwig, Mechanical integration studies for the CLIC vertex and inner tracking detectors (2015), CLICdp-Note-2015-002. [22] S. Green, J. S. Marshall, B. Xu, private communication, 2016. [23] P. Roloff, Photon reconstruction requirements, presentation at the ECFA Linear Collider workshop 2016, Santander, Spain, https://agenda.linearcollider.org/event/7014/contributions/34682/. [24] S. Poss, A. Sailer, Luminosity Spectrum Reconstruction at Linear Colliders, EPJC 74 (2014) 2833. [25] N. Nikiforou, M. Weber, Update on ECAL photon performance / optimisation, presentation at the ECFA Linear Collider Workshop 2016, Santander, Spain, https://agenda.linearcollider.org/event/7014/contributions/34659/. [26] M. Reinhard, CP violation in the Higgs sector with a next-generation detector at the ILC, PhD thesis, École polytechnique Laboratoire Leprince-Ringuet, 2009. [27] N. Nikiforou, Hadronic Calorimeter Optimization for a New Detector Model at CLIC (2016), Draft Note, https://edms.cern.ch/document/1756529/. [28] L. Linssen et al., eds., Physics and Detectors at CLIC: CLIC Conceptual Design Report, Fig. 6.2, vol. CERN-2012-003, 2012. [29] S. van Dam, A. Sailer, The occupancy in the Hadronic Calorimeter endcap of the CLIC detector (2014), CLICdp-Note-2014-004. [30] B. Dalena, D. Schulte, R. Thomas, Impact of the Experiment Solenoid on the CLIC Luminosity (2010), CERN-ATS-2010-081. [31] H. Gerwig, A. Herve, Ring Coils on the Endcap Yoke of a CLIC Detector (2011), LCD-Note-2011-017. [32] E. van der Kraaij, B. Schmidt, Muon System Design Studies for Detectors at CLIC (2011), LCD-Note-2011-008. [33] H. Abramowicz et al., A Luminosity Calorimeter for CLIC (2009), LCD-Note-2009-002. [34] J. Resta-Lopez, P. N. Burrows, G. Christian, Luminosity Performance Studies of the Compact Linear Collider with Intra-train Feedback System at the Interaction Point, JINST 5 (2010) 09007. [35] M. Weber, Update on ECAL photon optimisation, presentation at the CLICdp working group on detector optimisation, CERN, Switzerland, https://indico.cern.ch/event/581371/. 43
https://openalex.org/W4304091780	https://discovery.ucl.ac.uk/id/eprint/10157417/1/journal.pcbi.1010548.pdf	English	null	Chaos in synthetic microbial communities	PLOS computational biology/PLoS computational biology	2,022	cc-by	17,450	OPEN ACCESS Citation: Karkaria BD, Manhart A, Fedorec AJH, Barnes CP (2022) Chaos in synthetic microbial communities. PLoS Comput Biol 18(10): e1010548. https://doi.org/10.1371/journal. pcbi.1010548 Citation: Karkaria BD, Manhart A, Fedorec AJH, Barnes CP (2022) Chaos in synthetic microbial communities. PLoS Comput Biol 18(10): e1010548. https://doi.org/10.1371/journal. pcbi.1010548 Editor: Christopher Rao, University of Illinois at Urbana-Champaign, UNITED STATES Received: December 8, 2021 Accepted: September 7, 2022 Published: October 10, 2022 PLOS COMPUTATIONAL BIOLOGY PLOS COMPUTATIONAL BIOLOGY a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 * christopher.barnes@ucl.ac.uk * christopher.barnes@ucl.ac.uk Abstract Predictability is a fundamental requirement in biological engineering. As we move to building coordinated multicellular systems, the potential for such systems to display chaotic behav- iour becomes a concern. Therefore understanding which systems show chaos is an impor- tant design consideration. We developed a methodology to explore the potential for chaotic dynamics in small microbial communities governed by resource competition, intercellular communication and competitive bacteriocin interactions. Our model selection pipeline uses Approximate Bayesian Computation to first identify oscillatory behaviours as a route to find- ing chaotic behaviour. We have shown that we can expect to find chaotic states in relatively small synthetic microbial systems, understand the governing dynamics and provide insights into how to control such systems. This work is the first to query the existence of chaotic behaviour in synthetic microbial communities and has important ramifications for the fields of biotechnology, bioprocessing and synthetic biology. RESEARCH ARTICLE Chaos in synthetic microbial communities Behzad D. KarkariaID1, Angelika ManhartID2, Alex J. H. FedorecID1, Chris P. BarnesID1* 1 Department of Cell & Developmental Biology, University College London, London, United Kingdom, 2 Department of Mathematics, University College London, London, United Kingdom ¤ Current address: Department of Cell & Developmental Biology, University College London, London, United Kingdom * christopher.barnes@ucl.ac.uk Abstract Predictability is a fundamental requirement in biological engineering. As we move to building ONAL BIOLOGY RESEARCH ARTICLE Chaos in synthetic microbial communities Behzad D. KarkariaID1, Angelika ManhartID2, Alex J. H. FedorecID1, Chris P. BarnesID1* 1 Department of Cell & Developmental Biology, University College London, London, United Kingdom, 2 Department of Mathematics, University College London, London, United Kingdom ¤ Current address: Department of Cell & Developmental Biology, University College London, London, United Kingdom * christopher.barnes@ucl.ac.uk Abstract Predictability is a fundamental requirement in biological engineering. As we move to building ONAL BIOLOGY Introduction Funding: B.D.K was funded through the BBSRC LIDo Doctoral Training Partnership, (Grant No 1758911). A.J.H.F and C.P.B received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 Research and Innovation Programme (Grant No. 770835). C.P.B. received salary funding from the Wellcome Trust (209409/Z/17/Z). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Chaos can be defined as deterministic behaviour that displays aperiodic orbits and sensitivity to initial conditions [1]. Infinitesimally small differences in the initial conditions of a chaotic system will become amplified over time, making forecasting and prediction of behaviour impossible [2]. Despite being deterministic, chaotic systems possess an inherent uncertainty due to the fact that we can never describe the initial conditions of a system in sufficient detail. Building systems which behave in a predictable and repeatable manner is essential across fields invested in engineering biology and its applications. Evidence from studies of neural networks suggests the increasing probability of chaotic behaviour as the number of dimensions in the network grows [3–5]. Therefore we might expect opportunities for unpredictable behaviour to become more probable as we try and implement larger synthetic communities, or edit existing networks such as the human gut microbiome. Steps to date have not been taken to investigate the existence of chaos in small synthetic microbial networks. A long-term goal of engineering biology is to create truly scalable and robust synthetic microbial communities [6, 7]. Therefore understanding and evaluating the possibility of chaotic behaviour in a system becomes an important consideration. Competing interests: The authors have declared that no competing interests exist. Observations of chaotic behaviour in biological systems have been reported. A three species system containing one predator and two prey species has been demonstrated to produce cha- otic behaviour, with dilution rate a key parameter in enabling aperiodic behaviours [8]. An eight year study of a planktonic food web measured chaotic behaviours, resulting in subpopu- lation abundance predictability being limited to 15–30 days, despite constant external condi- tions [9]. These experimental examples demonstrate that a low number of species are capable of producing chaotic behaviour and are therefore unpredictable. In order to predict chaotic behaviour in synthetic microbial communities, we need to develop models that capture interactions between different community species. Generalised Lotka-Volterra equations (gLV) have previously been used to model pair-wise interactions and infer inter-species relationships [10]. PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities Author summary In chaotic systems, infinitesimally small differences in the initial conditions will become amplified over time, making forecasting and prediction of behaviour impossible. Although we know that chaos can be observed in the complex networks of natural ecosys- tems, the field of biotechnology is interested in designing and building new microbial communities and the presence of chaotic behaviour is unexplored. In this paper, we pres- ent a statistical pipeline that can tell us how, when and why chaos arises in small microbial communities. We apply this approach to study a set of communities involving quorum sensing systems and amensal interactions through antimicrobial peptides. Out of 4182 interaction networks in these three strain communities, we identify the networks that have the highest propensity to produce chaos. We then explore the levers we can pull to bring these networks in and out of chaotic regimes. Our work is the first to look at chaos in synthetic microbial communities and indicates that chaos is an important design consideration. Peer Review History: PLOS recognizes the benefits of transparency in the peer review process; therefore, we enable the publication of all of the content of peer review and author responses alongside final, published articles. The editorial history of this article is available here: https://doi.org/10.1371/journal.pcbi.1010548 C i h 2022 K k i l Thi i Copyright: © 2022 Karkaria et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: Data and scripts required to reproduce the figures in this manuscript can be accessed from the following Zenondo repository: https://doi.org/10.5281/ d 5764686 Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 1 / 24 PLOS COMPUTATIONAL BIOLOGY Introduction However, gLV models provide an incomplete description of interactions we expect to find in microbial communities. They are unable to capture the existence of chaos in three species networks [11]. Furthermore, gLV models have failed to predict community formation from pairwise interactions in microbial communities [12]. gLV models lack dynamics that occur with the accumulation and depletion of extracellu- lar species, which can be important for predicting the true dynamics of a community [13]. Modified Lotka-Volterra equations produce chaotic behaviour in predator-prey systems by including time-delayed feedback [13, 14], or in one predator two prey systems, by adding dampening effects [15]. While these abstractions are suitable in some circumstances, by modelling the intermediates involved in competitive interactions we can include experimen- tally measurable mechanisms and parameters. In previous work, we have modelled quorum sensing (QS) to regulate bacteriocin expression and engineer inter-population interactions. These methods allowed us to tune experimental parameters of an engineered two strain system [16], and predict the most promising topologies for producing stability in two and three strain systems [17]. The existence of chaos in dynamical models and the distribution of chaotic parameter space can be identified using various optimisation techniques. The unscented Kalman filter has previously been used to investigate chaos in electrical circuits and biological systems, obtaining parameters yielding chaos [18]. Simulated annealing has been applied to finding chaotic parameters in four species standard Lotka Volterra models [19]. Evidence also suggests that perturbation of system parameters can be used to drive systems towards or away from Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 PLOS Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 2 / 24 PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities chaotic attractors [20]. The possibility of chaos in synthetic microbial communities, to our knowledge, has not been previously considered. Standard competitive gLV models can produce chaotic behaviour in four species networks [11]. We use these previous findings to demonstrate and validate our methodology, before applying it to models that describe interpopulation interactions that are more specific to mech- anisms found in synthetic microbial communities. Development of a novel statistical approach to identifying chaotic regions in a multidimensional parameter space Approximate Bayesian Computation Sequential Monte Carlo (ABC SMC) is a method that can be used for model selection and parameter inference in dynamical systems [21] (Algo- rithm 1). This flexible algorithm can also be used to tackle the design question, namely what model topologies and parameters are able to give rise to some specified target qualitative behaviour [22]. ABC SMC requires a distance function, describing how far away a simulation is from the objective behaviour. When searching for chaotic beahviour, we use the maximal Lyapunov exponent (λ1) to create a distance function. We calculate λ1 by initialising two nearby orbits and measuring their divergence or convergence over the course of a simulation (see Methods and Algorithm 1). λ1 < 0 corresponds to linear stability, λ1 = 0 corresponds to periodic oscillations, and λ1 > 0 corresponds to chaos. While these rules hold true for infinite time, our simulations run for a finite time, meaning these boundary rules can be noisy. To identify chaos, we therefore define a threshold above 0 where we can be sure simulations have chaotic behaviour. First, we demonstrate the use of ABC SMC in resolving a chaotic parameter distribution in a competitive gLV system. Competitive gLV equations are commonly used in ecological popu- lation modelling, and have similarly been used to model microbial communities [23]. They describe generic negative interactions between species that could represent competition for nutrients or amensal interactions. Competitive gLV systems take the form nutrients or amensal interactions. Competitive gLV systems take the form dNi dt ¼ riNið1 X n j¼1 aijNjÞ where Ni is the size of a species population, ri is the growth rate, n is the number of species in the population and α the interaction matrix, describes the amensal interactions between pairs of species in the system. To simulate the chemostat environment, we set the diagonal as a dilu- tion rate, D, which is the same for all species. The diagonal of α can also be thought of as defin- ing the carrying capacity of each species. a ¼ D a12 a13 a14 a21 D a23 a24 a31 a32 D a34 a41 a42 a43 D 2 66666664 3 77777775 Vano et al. previously identified a chaotic attractor in this system using a brute-force parameter search [11]. Fig 1A shows the parameters identified, Fig 1B shows the resulting cha- otic timeseries of the four species. Development of a novel statistical approach to identifying chaotic regions in a multidimensional parameter space A Shows the d h h d ll f h l f h h Fig 1. Demonstration of chaotic attractor identified by Vano et al. in a four species competitive Lotka-Volterra model [11]. A Shows the parameters used in the chaotic attractor and an illustration of the interaction topology. B Time series of the chaotic attractor. C Posterior parameter distribution for chaotic objective, identified using ABC SMC (red) and the individual chaotic particle identified by Vano et al. (black). Center grid shows 2D parameter distributions, left and top rows 1D parameter distribtuions. https://doi org/10 1371/journal pcbi 1010548 g001 identified a threshold of λ > 0.015 was sufficient for classifying chaotic behaviour and ran ABC SMC for this chaotic objective. We show the posterior of several parameters in Fig 1C. The black point corresponds to the parameters found by Vano et al, while the red scatter points correspond to chaotic behaviour we identified using ABC SMC. We can see that the interspe- cies interaction parameters, a42 and a43, are constrained, indicating their importance for pro- ducing chaotic behaviour, given the prior parameter distributions. Conversely, the initial population of N1 is not constrained, indicating the chaotic behaviours are robust to changing initial conditions. Similarly, r3, defining the growth rate of N3 is not constrained. The full pos- terior parameter distribution is shown in S1 Fig. Mechanisms of interaction in microbial communities such as crossfeeding and toxin inter- actions would be subjected to time delays, accumulation of intermediate species and dynamic genetic regulation, contributing to non-linearity of these systems. gLV equations simplify these mechanisms and as such, are unable to capture chaotic behaviour with three species. In the next sections we move to studying more biochemically realistic systems. Development of a novel statistical approach to identifying chaotic regions in a multidimensional parameter space We wanted to see if our ABC SMC methods could provide a posterior distribution for chaotic behaviour, capturing the findings of Vano et al. We PLOS Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 3 / 24 dNi dt ¼ riNið1 X n j¼1 aijNjÞ l ti i th th dNi dt ¼ riNið1 X n j¼1 aijNjÞ h N i th i f i l ti i th th t where Ni is the size of a species population, ri is the growth rate, n is the number of species in the population and α the interaction matrix, describes the amensal interactions between pairs of species in the system. To simulate the chemostat environment, we set the diagonal as a dilu- tion rate, D, which is the same for all species. The diagonal of α can also be thought of as defin- ing the carrying capacity of each species. g the carrying capacity of each species. a ¼ D a12 a13 a14 a21 D a23 a24 a31 a32 D a34 a41 a42 a43 D 2 66666664 3 77777775 g y g p y p a ¼ D a12 a13 a14 a21 D a23 a24 a31 a32 D a34 a41 a42 a43 D 2 66666664 3 77777775 Vano et al. previously identified a chaotic attractor in this system using a brute-force parameter search [11]. Fig 1A shows the parameters identified, Fig 1B shows the resulting cha- otic timeseries of the four species. We wanted to see if our ABC SMC methods could provide a posterior distribution for chaotic behaviour, capturing the findings of Vano et al. We Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 PLOS Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 3 / 24 PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities Fig 1. Demonstration of chaotic attractor identified by Vano et al. in a four species competitive Lotka-Volterra model [11]. A Shows the parameters used in the chaotic attractor and an illustration of the interaction topology. B Time series of the chaotic attractor. C Posterior parameter distribution for chaotic objective, identified using ABC SMC (red) and the individual chaotic particle identified by Vano et al. (black). Center grid shows 2D parameter distributions, left and top rows 1D parameter distribtuions. https://doi.org/10.1371/journal.pcbi.1010548.g001 Fig 1. Demonstration of chaotic attractor identified by Vano et al. in a four species competitive Lotka-Volterra model [11]. Searching for chaos across synthetic microbial community models 4182 models are generated forming our initial model space. A(ii) We then perform ABC SMC for an oscillatory objective, which yielded 117 models that were capable of producing oscillations. A(iii) These form the prior model space for the chaos objective, using a threshold of λ1 > 0.003, we identify models capable of producing chaotic behaviour B The barchart shows the probability of models for the chaotic objective. The error bars represent the standard deviation. C An example time series representative of the chaos objective posterior distribution. Population densities as optical density (OD) show sustained, nonrepetitive oscillatory behaviour for the three species community. Fig 2A shows the pipeline we developed to search for chaos in synthetic three strain sys- tems. Three strains, N1, N2, N3, optionally express bacteriocins, B1, B2, under the control of optionally expressed QS molecules, A1, A2. The QS molecules regulate expression of bacterio- cins positively or negatively. Each strain can be optionally sensitive to a bacteriocin. The initial model space describes an enumeration of possible combinations of bacteriocin and QS sys- tems, and forms the first uniform prior model space of 4182 models (Fig 2A(i)). Prior parame- ter distributions describe the range of characteristics for the different parts (Table 1). We expected the existence of chaos to be sparse in this three strain model space, and therefore computationally expensive to explore. Oscillations are a known route to chaos [1], therefore, in order to narrow down the search, we defined a novel set of three distances that are used to classify oscillatory behaviours. These were the period of the signal (defined through the Fourier transform), the number of expected peaks, and the amplitude of the signal (see Methods). We also define an extinction threshold of 10−5; if a strain population falls below this it is classi- fied as extinct. Using these distances, we performed ABC SMC for an oscillations objective Fig 2A shows the pipeline we developed to search for chaos in synthetic three strain sys- tems. Three strains, N1, N2, N3, optionally express bacteriocins, B1, B2, under the control of optionally expressed QS molecules, A1, A2. The QS molecules regulate expression of bacterio- cins positively or negatively. Each strain can be optionally sensitive to a bacteriocin. The initial model space describes an enumeration of possible combinations of bacteriocin and QS sys- tems, and forms the first uniform prior model space of 4182 models (Fig 2A(i)). Searching for chaos across synthetic microbial community models These were the period of the signal (defined through the Fourier transform), the number of expected peaks, and the amplitude of the signal (see Methods). We also define an extinction threshold of 10−5; if a strain population falls below this it is classi- fied as extinct. Using these distances, we performed ABC SMC for an oscillations objective Fig 2. Overview of the pipeline for identifying chaotic topologies A(i) The initial model space is built from different combinations of engineering options. N1, N2, N3 are the three strains being engineered, and can optionally express QS molecules A1, A2 and bacteriocins B1, B2, B3. 4182 models are generated forming our initial model space. A(ii) We then perform ABC SMC for an oscillatory objective, which yielded 117 models that were capable of producing oscillations. A(iii) These form the prior model space for the chaos objective, using a threshold of λ1 > 0.003, we identify models capable of producing chaotic behaviour B The barchart shows the probability of models for the chaotic objective. The error bars represent the standard deviation. C An example time series representative of the chaos objective posterior distribution. Population densities as optical density (OD) show sustained, nonrepetitive oscillatory behaviour for the three species community. https://doi.org/10.1371/journal.pcbi.1010548.g002 Fig 2. Overview of the pipeline for identifying chaotic topologies A(i) The initial model space is built from different combinations of engineering options. N1, N2, N3 are the three strains being engineered, and can optionally express QS molecules A1, A2 and bacteriocins B1, B2, B3. 4182 models are generated forming our initial model space. A(ii) We then perform ABC SMC for an oscillatory objective, which yielded 117 models that were capable of producing oscillations. A(iii) These form the prior model space for the chaos objective, using a threshold of λ1 > 0.003, we identify models capable of producing chaotic behaviour B The barchart shows the probability of models for the chaotic objective. The error bars represent the standard deviation. C An example time series representative of the chaos objective posterior distribution. Population densities as optical density (OD) show sustained, nonrepetitive oscillatory behaviour for the three species community. Fig 2. Overview of the pipeline for identifying chaotic topologies A(i) The initial model space is built from different combinations of engineering options. N1, N2, N3 are the three strains being engineered, and can optionally express QS molecules A1, A2 and bacteriocins B1, B2, B3. Searching for chaos across synthetic microbial community models In previous work we developed a model framework to describe QS regulated bacteriocin inter- actions in a three strain model space, and predicted topologies that form stable communities [17]. Here we use this same model space to investigate the existence of chaos in three strain synthetic microbial communities. Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 PLOS Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 4 / 24 PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities Fig 2. Overview of the pipeline for identifying chaotic topologies A(i) The initial model space is built from different combinations of engineering options. N1, N2, N3 are the three strains being engineered, and can optionally express QS molecules A1, A2 and bacteriocins B1, B2, B3. 4182 models are generated forming our initial model space. A(ii) We then perform ABC SMC for an oscillatory objective, which yielded 117 models that were capable of producing oscillations. A(iii) These form the prior model space for the chaos objective, using a threshold of λ1 > 0.003, we identify models capable of producing chaotic behaviour B The barchart shows the probability of models for the chaotic objective. The error bars represent the standard deviation. C An example time series representative of the chaos objective posterior distribution. Population densities as optical density (OD) show sustained, nonrepetitive oscillatory behaviour for the three species community. https://doi.org/10.1371/journal.pcbi.1010548.g002 i O f h l f d f h Fig 2A shows the pipeline we developed to search for chaos in synthetic three strain sys- tems. Three strains, N1, N2, N3, optionally express bacteriocins, B1, B2, under the control of optionally expressed QS molecules, A1, A2. The QS molecules regulate expression of bacterio- cins positively or negatively. Each strain can be optionally sensitive to a bacteriocin. The initial model space describes an enumeration of possible combinations of bacteriocin and QS sys- tems, and forms the first uniform prior model space of 4182 models (Fig 2A(i)). Prior parame- ter distributions describe the range of characteristics for the different parts (Table 1). We expected the existence of chaos to be sparse in this three strain model space, and therefore computationally expensive to explore. Oscillations are a known route to chaos [1], therefore, in order to narrow down the search, we defined a novel set of three distances that are used to classify oscillatory behaviours. Searching for chaos across synthetic microbial community models Prior parame- ter distributions describe the range of characteristics for the different parts (Table 1). We expected the existence of chaos to be sparse in this three strain model space, and therefore computationally expensive to explore. Oscillations are a known route to chaos [1], therefore, in order to narrow down the search, we defined a novel set of three distances that are used to classify oscillatory behaviours. These were the period of the signal (defined through the Fourier transform), the number of expected peaks, and the amplitude of the signal (see Methods). We also define an extinction threshold of 10−5; if a strain population falls below this it is classi- fied as extinct. Using these distances, we performed ABC SMC for an oscillations objective Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 5 / 24 PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities Chaos in synthetic microbial communities (Fig 2A(ii)). We identified 117 models capable of producing oscillations. These models become the new uniform prior model distribution for the next stage, where we perform ABC SMC for the previously described chaotic objective (Fig 2A(iii)). In this model framework we identified λ1 > 0.003 as sufficient for classifying chaos. The posterior probabilities of the models for the chaotic objective given the prior distribu- tions used are shown in Fig 2B. Fig 2C shows a representative chaotic trajectory, demonstrat- ing aperiodic non-repeating behaviour, satisfying the qualitative features of chaos. Properties of chaotic models We next explored some of the properties of chaotic topologies identified using ABC SMC. Fig 3A shows the top performing models when subsetting for complexity, based on the number of parts expressed. mk refers to the k-th model from the initial model space. m850 contains four expressed parts and possesses the highest posterior probability for chaotic behaviour. Systems containing fewer parts all had a posterior probability of zero. As complexity increases to five and six parts (m3177 and m2547), the posterior probability decreases. Our previous work dem- onstrated that system stability increased with the number of parts [17]. The peak in the poste- rior probability for chaos at four parts reflects a balance that includes enough mechanisms to enable co-existence, without the tighter network of negative interactions that are associated with linear stability [17, 24]. We highlight that these observations are limited to the small com- munities defined in our prior. Searching for chaos across synthetic microbial community models These properties may not be reflective of larger communities, however, we hypothesise that a trade-off between stabilizing interactions that enable co-exis- tence, and destabilising interactions to prevent linear stability, will remain important for pro- ducing chaotic population dynamics. Table 1. Prior distributions for both two and three strain systems are sampled uniformly between the min and max values listed below. Constant parameters have the same min and max value. Parameter / State variable Description Prior (min) Prior (max) Units Citation Parameters CN OD to cell number scaling factor 1e9 1e9 None N/A CB Microcin scaling factor 1e−9 1e−9 None N/A CA QS scaling factor 1e−9 1e−9 None N/A D Dilution rate 0.01 0.5 h−1 N/A KAyBz Half maximal QS promoter activation/repression from Ay to Bz 1e−9 1e−6 M [43] K Monod’s half saturation constant 3.9e−5 3.9e−5 M [44] Kω Half saturation killing constant 1e−7 1e−6 M [45, 46] S0 Substrate concentration of input media (0.4% glucose) 0.02 0.02 M M9 media γ E. coli substrate yield 1e11 1e11 cell M−1 [47] kAy Production rate of AHL per cell 1e−22 1e−15 M h−1 [48] KBmax z Maximal expression rate of microcin 1e−22 1e−15 M h−1 [49] mxmax Maximum growth rate 0.4 3 h−1 [50, 51] nz Hill coefficient AHL induced expression 1 2 M [43] nω Hill coefficient for killing 1 2 M [43] ωmax Maximum rate of bacteriocin killing 0.5 2.0 M−1 h−1 [45, 46, 52] Initial state variable N OD of strain 0.01 0.5 OD N/A S 0.4% glucose concentration 0.02 0.02 M N/A B Microcin concentration 1e−81 1e−81 M CB N/A A QS concentration 1e−10 1e−10 M CA N/A https://doi.org/10.1371/journal.pcbi.1010548.t001 PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities Table 1. Prior distributions for both two and three strain systems are sampled uniformly between the min and max values listed below. Constant parameters have the same min and max value. Table 1. Prior distributions for both two and three strain systems are sampled uniformly between the min and max values listed below. Constant parameters have the same min and max value. Searching for chaos across synthetic microbial community models Parameter / State variable Description Prior (min) Prior (max) Units Citation Parameters CN OD to cell number scaling factor 1e9 1e9 None N/A CB Microcin scaling factor 1e−9 1e−9 None N/A CA QS scaling factor 1e−9 1e−9 None N/A D Dilution rate 0.01 0.5 h−1 N/A KAyBz Half maximal QS promoter activation/repression from Ay to Bz 1e−9 1e−6 M [43] K Monod’s half saturation constant 3.9e−5 3.9e−5 M [44] Kω Half saturation killing constant 1e−7 1e−6 M [45, 46] S0 Substrate concentration of input media (0.4% glucose) 0.02 0.02 M M9 media γ E. coli substrate yield 1e11 1e11 cell M−1 [47] kAy Production rate of AHL per cell 1e−22 1e−15 M h−1 [48] KBmax z Maximal expression rate of microcin 1e−22 1e−15 M h−1 [49] mxmax Maximum growth rate 0.4 3 h−1 [50, 51] nz Hill coefficient AHL induced expression 1 2 M [43] nω Hill coefficient for killing 1 2 M [43] ωmax Maximum rate of bacteriocin killing 0.5 2.0 M−1 h−1 [45, 46, 52] Initial state variable N OD of strain 0.01 0.5 OD N/A S 0.4% glucose concentration 0.02 0.02 M N/A B Microcin concentration 1e−81 1e−81 M CB N/A A QS concentration 1e−10 1e−10 M CA N/A https://doi.org/10.1371/journal.pcbi.1010548.t001 ree strain systems are sampled uniformly between the min and max values listed below. Constant parameters have the (Fig 2A(ii)). We identified 117 models capable of producing oscillations. These models become the new uniform prior model distribution for the next stage, where we perform ABC SMC for the previously described chaotic objective (Fig 2A(iii)). In this model framework we identified λ1 > 0.003 as sufficient for classifying chaos. The posterior probabilities of the models for the chaotic objective given the prior distribu- tions used are shown in Fig 2B. Fig 2C shows a representative chaotic trajectory, demonstrat- ing aperiodic non-repeating behaviour, satisfying the qualitative features of chaos. The posterior probabilities of the models for the chaotic objective given the prior distribu- tions used are shown in Fig 2B. Fig 2C shows a representative chaotic trajectory, demonstrat- ing aperiodic non-repeating behaviour, satisfying the qualitative features of chaos. Properties of chaotic models P W 3 p e c a o r e w m h t d We next explored some of the properties of chaotic topologies identified using ABC SMC. Fig 3A shows the top performing models when subsetting for complexity, based on the number of parts expressed. mk refers to the k-th model from the initial model space. m850 contains four expressed parts and possesses the highest posterior probability for chaotic behaviour. Systems containing fewer parts all had a posterior probability of zero. As complexity increases to five and six parts (m3177 and m2547), the posterior probability decreases. Our previous work dem- onstrated that system stability increased with the number of parts [17]. The peak in the poste- rior probability for chaos at four parts reflects a balance that includes enough mechanisms to enable co-existence, without the tighter network of negative interactions that are associated with linear stability [17, 24]. We highlight that these observations are limited to the small com- munities defined in our prior. These properties may not be reflective of larger communities, however, we hypothesise that a trade-off between stabilizing interactions that enable co-exis- tence, and destabilising interactions to prevent linear stability, will remain important for pro- ducing chaotic population dynamics. Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 6 / 24 PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities Fig 3. Topologies and properties associated with chaotic behaviour A Shows the models with highest posterior probability when subsetted for number of parts expressed, in order of increasing complexity (4, 5 and 6 expressed parts). The bar chart shows the mean model posterior probability across three experiments, the error bars indicate the standard deviation. B Comparison between average posterior probabilities with different properties. In order from left to right, the barcharts compare: The number of QS systems used, the modes by which QS regulates bacteriocin expression (positive, negative or both), the number of bacteriocins used, and systems containing self-limiting (SL), other-limiting (OL) or SL and OL interactions. https://doi.org/10.1371/journal.pcbi.1010548.g003 Fig 3B provides summaries of how different parts contribute to chaotic behaviour in the three strain models. We can see that one QS system and positive regulation of bacteriocin is strongly favoured for producing chaos. This ensures all system bacteriocins are regulated in tandem. Expression rates are all dependent upon the same QS, resulting in stronger negative or positive correlations defined by the mode of regulation. Properties of chaotic models Two bacteriocin systems also domi- nate the model posterior. Bacteriocin interactions can be categorised as either self-limiting (SL), whereby the strain is inhibited by the bacteriocin it produces, or other-limiting (OL) where a strain is inhibited by a bacteriocin produced by a different strain. Both SL only and a combination of SL and OL interactions are associated with producing chaotic behaviour. These observations are interesting in comparison to other work on ecological systems. Coop- erative interactions were previously found to give rise to unstable systems, whereas competi- tion was more indicative of stability [24]. The same effect might occur here in systems with Fig 3. Topologies and properties associated with chaotic behaviour A Shows the models with highest posterior probability when subsetted for number of parts expressed, in order of increasing complexity (4, 5 and 6 expressed parts). The bar chart shows the mean model posterior probability across three experiments, the error bars indicate the standard deviation. B Comparison between average posterior probabilities with different properties. In order from left to right, the barcharts compare: The number of QS systems used, the modes by which QS regulates bacteriocin expression (positive, negative or both), the number of bacteriocins used, and systems containing self-limiting (SL), other-limiting (OL) or SL and OL interactions. https://doi.org/10.1371/journal.pcbi.1010548.g003 Fig 3. Topologies and properties associated with chaotic behaviour A Shows the models with highest posterior probability when subsetted for number of parts expressed, in order of increasing complexity (4, 5 and 6 expressed parts). The bar chart shows the mean model posterior probability across three experiments, the error bars indicate the standard deviation. B Comparison between average posterior probabilities with different properties. In order from left to right, the barcharts compare: The number of QS systems used, the modes by which QS regulates bacteriocin expression (positive, negative or both), the number of bacteriocins used, and systems containing self-limiting (SL), other-limiting (OL) or SL and OL interactions. https://doi.org/10.1371/journal.pcbi.1010548.g003 Fig 3. Topologies and properties associated with chaotic behaviour A Shows the models with highest posterior probability when subsetted for number of parts expressed, in order of increasing complexity (4, 5 and 6 expressed parts). The bar chart shows the mean model posterior probability across three experiments, the error bars indicate the standard deviation. B Comparison between average posterior probabilities with different properties. Properties of chaotic models The same effect might occur here in systems with Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 7 / 24 PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities one QS, rather than two, as the system would be expected to have increased correlation. While chaotic behaviour may seem to be very different from linear stability, both behaviours share the necessity for coexistence. Our previous work showed that SL interactions were important for producing linear stability, while OL interactions more frequently associated with extinction events and non-linear stability [17]. This may explain why we see tendencies for topologies to share a mixture of stability associated SL interactions, and instability associated OL interac- tions. We also find models with three bacteriocins, and hence higher suppression of growth, have a low posterior probability for chaos, given the prior distributions used. Properties of chaotic models In order from left to right, the barcharts compare: The number of QS systems used, the modes by which QS regulates bacteriocin expression (positive, negative or both), the number of bacteriocins used, and systems containing self-limiting (SL), other-limiting (OL) or SL and OL interactions. https://doi.org/10.1371/journal.pcbi.1010548.g003 Fig 3B provides summaries of how different parts contribute to chaotic behaviour in the three strain models. We can see that one QS system and positive regulation of bacteriocin is strongly favoured for producing chaos. This ensures all system bacteriocins are regulated in tandem. Expression rates are all dependent upon the same QS, resulting in stronger negative or positive correlations defined by the mode of regulation. Two bacteriocin systems also domi- nate the model posterior. Bacteriocin interactions can be categorised as either self-limiting (SL), whereby the strain is inhibited by the bacteriocin it produces, or other-limiting (OL) where a strain is inhibited by a bacteriocin produced by a different strain. Both SL only and a combination of SL and OL interactions are associated with producing chaotic behaviour. These observations are interesting in comparison to other work on ecological systems. Coop- erative interactions were previously found to give rise to unstable systems, whereas competi- tion was more indicative of stability [24]. The same effect might occur here in systems with Fig 3B provides summaries of how different parts contribute to chaotic behaviour in the three strain models. We can see that one QS system and positive regulation of bacteriocin is strongly favoured for producing chaos. This ensures all system bacteriocins are regulated in tandem. Expression rates are all dependent upon the same QS, resulting in stronger negative or positive correlations defined by the mode of regulation. Two bacteriocin systems also domi- nate the model posterior. Bacteriocin interactions can be categorised as either self-limiting (SL), whereby the strain is inhibited by the bacteriocin it produces, or other-limiting (OL) where a strain is inhibited by a bacteriocin produced by a different strain. Both SL only and a combination of SL and OL interactions are associated with producing chaotic behaviour. These observations are interesting in comparison to other work on ecological systems. Coop- erative interactions were previously found to give rise to unstable systems, whereas competi- tion was more indicative of stability [24]. Parameter importance for chaos in m850 The model with the highest posterior probability for chaotic behaviour was m850; the topology is shown in Fig 4A. It consists of a single QS system, produced by N1, that positively regulates two bacteriocins. B1 is produced by N1 and N2 but it inhibits the growth of N1 only. B2 is pro- duced by N3 and inhibits the growth of N3 only. The system in total consists of four expressed parts. m850 also ranked highly for the oscillatory objective, ranking 3rd out of the initial 4182 Fig 4. Examining chaos in m850. A Topology of m850 with key parameters labelled. kA1 is the rate of QS molecule production, KBmax1 and KBmax2 are the maximal expression rates of bacteriocins B1 and B2 respectively. B Posterior parameter distributions of m850 for chaos (red) and oscillatory (blue) objectives for key parameters in system design. The borders show 1D posterior distributions for each parameter and the off-diagonal element the 2D posterior marginals. C Feature importance calculated using random forest regression. The information gain (bits) is calculated as an average of the reduction in entropy across all trees in the forest (2000 trees). The error bars indicate the standard deviation of the entropy for each feature across all trees. D Sensitivity analysis of a chaotic input vector with chaotic region in red. Black stars refer to the identified stable steady state. The fixed parameter values are shown in Table 2 https://doi.org/10.1371/journal.pcbi.1010548.g004 Fig 4. Examining chaos in m850. A Topology of m850 with key parameters labelled. kA1 is the rate of QS molecule production, KBmax1 and KBmax2 are the maximal expression rates of bacteriocins B1 and B2 respectively. B Posterior parameter distributions of m850 for chaos (red) and oscillatory (blue) objectives for key parameters in system design The borders show 1D posterior distributions for each parameter and the off-diagonal element the 2D posterior marginals. C Feature importance calculated using random forest regression. The information gain (bits) is calculate as an average of the reduction in entropy across all trees in the forest (2000 trees). The error bars indicate the standa deviation of the entropy for each feature across all trees. D Sensitivity analysis of a chaotic input vector with chaotic region in red. Black stars refer to the identified stable steady state. The fixed parameter values are shown in Table 2 https://doi.org/10.1371/journal.pcbi.1010548.g004 PLOS Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 8 / Fig 4. Parameter importance for chaos in m850 Examining chaos in m850. A Topology of m850 with key parameters labelled. kA1 is the rate of QS molecule production, KBmax1 and KBmax2 are the maximal expression rates of bacteriocins B1 and B2 respectively. B Posteri di ib i f f h ( d) d ill (bl ) bj i f k i d i Fig 4. Examining chaos in m850. A Topology of m850 with key parameters labelled. kA1 is the rate of QS molecule production, KBmax1 and KBmax2 are the maximal expression rates of bacteriocins B1 and B2 respectively. B Posterior parameter distributions of m850 for chaos (red) and oscillatory (blue) objectives for key parameters in system design. The borders show 1D posterior distributions for each parameter and the off-diagonal element the 2D posterior marginals. C Feature importance calculated using random forest regression. The information gain (bits) is calculated as an average of the reduction in entropy across all trees in the forest (2000 trees). The error bars indicate the standard deviation of the entropy for each feature across all trees. D Sensitivity analysis of a chaotic input vector with chaotic region in red. Black stars refer to the identified stable steady state. The fixed parameter values are shown in Table 2 https://doi.org/10.1371/journal.pcbi.1010548.g004 Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 PLOS Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 8 / 24 PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities d l Thi i i bl h b d l h h i i Table 2. Fixed parameters used in Figs 4D and 5. Parameter/State variable value Parameters CN 1e9 CB 1e−9 CA 1e−9 D 0.167 KA1B1 3.37e−9 KA1B1 4.26e−8 K 3.9e−5 Kω 1.6e−7 S0 0.02 γ 1e11 kAy 3.5e−17 KBmax1 3.58e−17 KBmax2 8.89e−16 m1max 2.61 m2max 1.17 m3max 1.48 n1 1.2 n2 1.43 nω 1.87 ωmax 0.79 Initial state variable N1 0.24 N2 0.25 N3 0.27 S 0.02 B1 1e−71 B2 1e−71 A1 1e−10 https://doi.org/10.1371/journal.pcbi.1010548.t002 models This presents an interesting problem whereby a model that has promising use as an Table 2. Fixed parameters used in Figs 4D and 5. Parameter importance for chaos in m850 Parameter/State variable value Parameters CN 1e9 CB 1e−9 CA 1e−9 D 0.167 KA1B1 3.37e−9 KA1B1 4.26e−8 K 3.9e−5 Kω 1.6e−7 S0 0.02 γ 1e11 kAy 3.5e−17 KBmax1 3.58e−17 KBmax2 8.89e−16 m1max 2.61 m2max 1.17 m3max 1.48 n1 1.2 n2 1.43 nω 1.87 ωmax 0.79 Initial state variable N1 0.24 N2 0.25 N3 0.27 S 0.02 B1 1e−71 B2 1e−71 A1 1e−10 https://doi.org/10.1371/journal.pcbi.1010548.t002 models. This presents an interesting problem whereby a model that has promising use as an oscillator also has a high potential to produce chaos, relative to other candidate models. Identi- fying the parameters and initial conditions important for differentiating between chaotic and oscillatory behaviour gives us insight into how to control this behaviour when constructing genetic circuits or selecting chemostat settings. As a first step, we analyzed the model to quantify the possible steady states and basins of attraction. Our analysis gave analytical conditions for the existence and stability for complete extinction and for single strain survival (See Methods). For three-strain co-existence, we find the following necessary conditions: maxfD K þ S0 S0 ; m1max D D þ omax ; m3max D D þ omax g < m2max < minfm1max; m3maxg This shows that for three-strain co-existence, the maximal growth rate of N2 has to lie between certain upper and lower bounds. In particular, it has to be smaller than the maximal growth rate of N1 or N3. We can see from the topology of m850 (Fig 4A) that the growth of N2 is not limited by any bacteriocin, therefore the only limitation on growth comes through Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 9 / 24 PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities resource competition. If N2 had a higher growth rate than N1 or N3, it would out compete these strains and cause an extinction event. We then wanted to explore the most important parameters that separate oscillatory and chaotic behaviours in m850 only. We refer to a set of parameters and initial conditions as an input vector. Using ABC SMC, we performed parameter inference on m850 for the chaotic and oscillatory objectives, generating 3750 input vectors for each objective. We can use this dataset of labelled input vectors to understand the importance of individual parameters, initial condi- tions and nearby steady states. Fig 4B shows multivariate parameter distributions for the oscillator and chaotic objectives for the experimentally accessible parameters. Parameter importance for chaos in m850 The dilution rate (D) is a directly controllable parameter of the chemostat. The production rate of A1 (kA1) can be tuned by using an induc- ible promoter to control expression of the AHL synthase species. Strain maximal growth rates (μmax1, μmax2, μmax3) can be controlled by using different base strains or through the combined use of auxotrophic strains and defined media. Finally, the initial population densities (N1, N2, N3) can easily be set when inoculating the initial culture. Divergence between two parameter distributions indicates its importance in differentiating between the two objectives. We can see that the oscillatory objective distributions for D, N1 and μmax2 are all constrained towards lower values relative to the prior. However, for all these distributions we can see that the cha- otic and oscillatory regions overlap. This again implies that chaotic and oscillatory behaviour exist close to one another in parameter space, and highlights the multidimensional nature that determines the behaviour. To further investigate the importance of parameters and initial conditions we trained a ran- dom forest classifier model using the input vectors as features [25]. We curated a label-bal- anced dataset of oscillating input vectors and chaotic input vectors. Using a train test ratio of 0.5, the trained classifier model was able to classify the test set with a 90% accuracy (Meth- ods). Fig 4C shows the average information gain across all decision tree classifiers in the forest for all free parameters. This can be used as an indicator of feature importance in correctly clas- sifying an input vector. KA1B1 and KA1B2 describe the concentration of A1 required to produce half-maximal repression of bacteriocins B1 and B2 respectively. While the feature importance indicates these parameters are the most important, they are more difficult to tune compared with other parameters in this system. The error bars indicate the variability in the importance of a feature across all trees in the forest. Large error bars suggest single features are not essen- tial for classification, and that redundancy exists between the features used [26]. describe the conc s B1 and B2 respec important, they a he error bars indi Large error bars s ncy exists between d i From the set of chaotic input vectors, we used numerical methods to identify nearby steady states that could be reached by changing the initial state of the system only. Parameter importance for chaos in m850 Fig 4D shows the sensitivity analysis of a chaotic input vector. The black stars indicate stable steady states identi- fied by numerical analysis. We perturbed the initial species values of either N1, B1 or B2 indi- vidually. The plots show how changing these initial states yields different Lyapounv exponents, highlighting the chaotic region in red. The range of Lyapunov exponents shown in Fig 4D sug- gest that by changing the initial conditions only we are able to produce a range of different behaviours. Perturbing N2, N3 or A1 did not produce chaotic behaviour. It is interesting that the initial state of N1 as the A1 producing strain appears to be more important whereas the ini- tial concentration of A1 itself is not. Exploring the parameters in the transition to chaos E Real-time ramp up tuning of D, moving the system from a chaotic state to a stable steady state. https://doi.org/10.1371/journal.pcbi.1010548.g005 between different population dynamics [27–29]. The posterior parameter distribution shown in Fig 4B strongly indicated the dilution rate, D, to be important for defining chaotic behaviour We previously identified the QS production rate k and D as important param Fig 5. Parameters kA1 and D can be tuned to control transitions between chaotic, oscillatory and stable states. The fixed parameter values are shown in Table 2. A Map showing how different combinations of kA1 and D change population behaviour. The grid fill colour corresponds to the maximum Lyapunov exponent measured, the grid outlines indicate the approximate classification where green is stable, yellow is oscillatory, red is chaotic and white is extinction. B Bifurcation diagram showing the community states visited for different values of kA1. C Bifurcation diagram showing the community states visited for different values of D. D Real-time ramp down tuning of kA1, moving the system from a chaotic state to a stable steady state. E Real-time ramp up tuning of D, moving the system from a chaotic state to a stable steady state. https://doi.org/10.1371/journal.pcbi.1010548.g005 Fig 5. Parameters kA1 and D can be tuned to control transitions between chaotic, oscillatory and stable states. The fixed parameter values are shown in Table 2. A Map showing how different combinations of kA1 and D change population behaviour. The grid fill colour corresponds to the maximum Lyapunov exponent measured, the grid outlines indicate the approximate classification where green is stable, yellow is oscillatory, red is chaotic and white is extinction. B Bifurcation diagram showing the community states visited for different values of kA1. C Bifurcation diagram showing the community states visited for different values of D. D Real-time ramp down tuning of kA1, moving the system from a chaotic state to a stable steady state. E Real-time ramp up tuning of D, moving the system from a chaotic state to a stable steady state. between different population dynamics [27–29]. The posterior parameter distribution shown in Fig 4B strongly indicated the dilution rate, D, to be important for defining chaotic behaviour. We previously identified the QS production rate, kA1 and D as important param- eters for transitioning between co-existence and extinction states [16]. Exploring the parameters in the transition to chaos Being able to move a system from a chaotic state to a fixed point could be important in a bioprocess control scenario so we explored this in more detail. Previous studies have fre- quently identified the bioreactor dilution rate as an important parameter for transitioning Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 10 / 24 PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities between different population dynamics [27–29]. The posterior parameter distribution shown in Fig 4B strongly indicated the dilution rate, D, to be important for defining chaotic behaviour. We previously identified the QS production rate, kA1 and D as important param- eters for transitioning between co-existence and extinction states [16]. We hypothesised that the antagonistic effect of kA1 to D would make it a useful parameter for controlling pop- ulation behaviour. First, we took an input vector known to produce chaotic behaviour and randomly sampled new values for kA1 and D from the prior and calculated the Lyapunov exponent of the new input vector. Fig 5A shows the results where filled colour indicates the maximal Lyapunov exponent calculated at each grid reference. The grid outline indicates the classification, the red grid region of Fig 5A shows the chaotic region. Fig 5A illustrates that, changing D and kA1 affects the Lyapunov exponent. The bifurcation diagrams in Fig 5B and 5C for kA1 and D respectively, illustrate the antagonistic transitions in behaviour that occur when changing the two parameters. Fig 5A and 5B show transitions through one strain extinctions (Nx < 10−5), stable steady state, oscillations and chaotic behaviour. Fig 5A and 5C both show that increasing kA1 results in transitions from stable co-existence, through oscillations and then to chaos, Fig 5. Parameters kA1 and D can be tuned to control transitions between chaotic, oscillatory and stable states. The fixed parameter values are shown in Table 2. A Map showing how different combinations of kA1 and D change population behaviour. The grid fill colour corresponds to the maximum Lyapunov exponent measured, the grid outlines indicate the approximate classification where green is stable, yellow is oscillatory, red is chaotic and white is extinction. B Bifurcation diagram showing the community states visited for different values of kA1. C Bifurcation diagram showing the community states visited for different values of D. D Real-time ramp down tuning of kA1, moving the system from a chaotic state to a stable steady state. Exploring the parameters in the transition to chaos We hypothesised that the antagonistic effect of kA1 to D would make it a useful parameter for controlling pop- ulation behaviour. between different population dynamics [27–29]. The posterior parameter distribution shown in Fig 4B strongly indicated the dilution rate, D, to be important for defining chaotic behaviour. We previously identified the QS production rate, kA1 and D as important param- eters for transitioning between co-existence and extinction states [16]. We hypothesised that the antagonistic effect of kA1 to D would make it a useful parameter for controlling pop- ulation behaviour. First, we took an input vector known to produce chaotic behaviour and randomly sampled new values for kA1 and D from the prior and calculated the Lyapunov exponent of the new input vector. Fig 5A shows the results where filled colour indicates the maximal Lyapunov exponent calculated at each grid reference. The grid outline indicates the classification, the red grid region of Fig 5A shows the chaotic region. Fig 5A illustrates that, changing D and kA1 affects the Lyapunov exponent. The bifurcation diagrams in Fig 5B and 5C for kA1 and D respectively, illustrate the antagonistic transitions in behaviour that occur when changing the two parameters. Fig 5A and 5B show transitions through one strain extinctions (Nx < 10−5), stable steady state, oscillations and chaotic behaviour. Fig 5A and 5C both show that increasing kA1 results in transitions from stable co-existence, through oscillations and then to chaos, Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 11 / 24 PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities followed abruptly by an extinction event. Fig 5B and c both show that a lower dilution rate is associated with chaos; increasing the dilution rate reduces instability to produce oscillations, which abruptly transitions to a stable extinction state. In a bioreactor control scenario it is interesting to understand if a community could be switched between states in real time. Fig 5D and 5E show how this is possible by modifying kA1 and D respectively. The red arrows on Fig 5A indicate the position of the single start point and two end points in these real-time transitions. It is important to note that when ramping up the dilution rate in real-time, we reach stable steady state in a region that would not be obtainable with a fixed dilution rate. Discussion It would be interest- ing to compare the existence of chaotic attractors in systems that use toxin-antitoxin systems [32], combinations of cooperative and competitive interactions [33], or mutualistic only interac- tions [34]. Genome scale metabolic models contain a large number of linear reactions [35];, they can be combined to describe microbial communities and used to model industrial biopro- cesses [36, 37]. Given the high dimensional nature of metabolic networks, it would be interest- ing to investigate whether these models yield chaotic behaviour in small community networks. Conclusion To conclude, we have developed methods for identifying chaotic parameter regions using ABC SMC. We have demonstrated the application of this method to resolve a previously iden- tified chaotic attractor in a gLV model, and identified models susceptible to chaos in three strain synthetic microbial communities. Although chaotic attractors are generally thought to be sparse in low dimensional systems, we have shown their existence in realistic synthetic microbial systems. They may also exist in close proximity to stable steady state regions. This work demonstrates that deterministic chaos will be an important factor in microbial commu- nity design and should be studied in much more detail. T A t s b m w n M T M t o Discussion We have developed a novel methodology to explore parameter regions that give rise to chaotic dynamics. We have applied it to the exploration of chaotic dynamics in synthetic microbial communities and found a high prevalence of such dynamics in these systems. This work is the first to query the existence of chaotic behaviour in synthetic microbial communities. We show that we can expect to find chaotic states in relatively small synthetic microbial systems, which has important ramifications for the field. has important ramifications for the field. By first running ABC SMC for the oscillatory objective we were able to drastically reduce the model space for the search for chaos. However, the timecourse simulation and parameter sam- pling makes this pipeline computationally costly. In the future we can consider using eigenvalue stability methods to reject particles without simulation, improving the efficiency of our approach and therefore improving the number of samples available for posterior estimation [30, 31]. We expect it will become increasingly important to consider the location of chaotic attractors in parameter space as the microbial communities we build or interact with become more com- plex. These methods can easily be applied to parametrise different models. It would be interest- ing to compare the existence of chaotic attractors in systems that use toxin-antitoxin systems [32], combinations of cooperative and competitive interactions [33], or mutualistic only interac- tions [34]. Genome scale metabolic models contain a large number of linear reactions [35];, they can be combined to describe microbial communities and used to model industrial biopro- cesses [36, 37]. Given the high dimensional nature of metabolic networks, it would be interest- ing to investigate whether these models yield chaotic behaviour in small community networks. Conclusion By first running ABC SMC for the oscillatory objective we were able to drastically reduce the model space for the search for chaos. However, the timecourse simulation and parameter sam- pling makes this pipeline computationally costly. In the future we can consider using eigenvalue stability methods to reject particles without simulation, improving the efficiency of our approach and therefore improving the number of samples available for posterior estimation [30, 31]. We expect it will become increasingly important to consider the location of chaotic attractors in parameter space as the microbial communities we build or interact with become more com- plex. These methods can easily be applied to parametrise different models. Three strain synthetic communities model space definition Models are generated from a set of parts, that are expressed by different strains in the sys- tem. We represent an expression configuration through a set of options. We define the options for expression of A in each strain, where the options are: not expressed, expression Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 12 / 24 PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities of A1, or expression of A2 (0, 1 or 2 respectively). We define the options for expression of bacteriocin as: no expression, expression of B1, expression of B2 or expression of B3 (0, 1, 2 and 3 respectively). Lastly we define the mode of regulation, R, for each bacteriocin, which can be either induced or repressed (0 and 1). This is redundant if a bacteriocin is not expressed. of A1, or expression of A2 (0, 1 or 2 respectively). We define the options for expression of bacteriocin as: no expression, expression of B1, expression of B2 or expression of B3 (0, 1, 2 and 3 respectively). Lastly we define the mode of regulation, R, for each bacteriocin, which can be either induced or repressed (0 and 1). This is redundant if a bacteriocin is not expressed. (0 and 1). This is redundant if a bacteriocin is not A ¼ f0; 1; 2g B ¼ f0; 1; 2; 3g R ¼ f0; 1g This enables us to build possible part combinations that can be expressed by a population. Let Pc be a family of sets, where each set is a unique combination of parts. PC ¼ A B R Each strain in a system can be sensitive to up to one bacteriocin. Let I represent the options for strain sensitivity. The options are: insensitive, sensitive to B1, sensitive to B2 or sensitive to B3 (0, 1, 2 and 3 respectively). Each strain is defined by its sensitivities, and expression of parts. Let PE be all unique engi- neered strains: PE ¼ I PC Which can be combined to form a model, yielding unique combinations: PM ¼ PE PE PE Finally we use a series of rules to remove redundant models A system is removed if: Each strain is defined by its sensitivities, and expression of parts. Let PE be all unique engi- neered strains: Each strain is defined by its sensitivities, and expression of parts. Three strain synthetic communities model space definition Let PE be all unique engi- neered strains: PE ¼ I PC Which can be combined to form a model, yielding unique combinations: PM ¼ PE PE PE Each strain is defined by its sensitivities, and expression of parts. Let PE be all unique engi neered strains: PE ¼ I PC Which can be combined to form a model, yielding unique combinations: PM ¼ PE PE PE Finally, we use a series of rules to remove redundant models. A system is removed if: 1. Two or more strains are identical, concerning bacteriocin sensitivity and combination of expressed parts. 2. The QS regulating a bacteriocin is not present in the system. 3. A strain is sensitive to a bacteriocin that does not exist in the system. 4. A bacteriocin exists that no strain is sensitive to. This cleanup yields the options which are used to generate ODE equations for a system. System equations State variables in each system are rescaled to improve speed of obtaining numerical approxi- mations. NX describes the concentration of a strain, Bz describes the concentration of a bacteri- ocin and Ay describes the concentration of a quorum molecule. CN, CB and CA are scaling factors: N0 x ¼ NxCN B0 z ¼ BzCB A0 y ¼ AyCA Each model is represented as sets where N defines the number of strains, B defines the set of bacteriocins and A defines the set of QS systems. The following differential equations are PLOS Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 13 / 24 Finally, we use a series of rules to remove redundant models. A system is removed if: 1. Two or more strains are identical, concerning bacteriocin sensitivity and combination of expressed parts. 2. The QS regulating a bacteriocin is not present in the system. 3. A strain is sensitive to a bacteriocin that does not exist in the system. 4. A bacteriocin exists that no strain is sensitive to. This cleanup yields the options which are used to generate ODE equations for a system. System equations State variables in each system are rescaled to improve speed of obtaining numerical approxi- mations. NX describes the concentration of a strain, Bz describes the concentration of a bacter ocin and Ay describes the concentration of a quorum molecule. Three strain synthetic communities model space definition oðB0 zÞ ¼ omax B0no z K no o þ B 0no z Induction or repression of bacteriocin expression by QS, is defined by kB(z, y), where z defines the bacteriocin being expressed and y defines the quorum molecule regulating its expression. KBmax z is the maximal expression rate of the bacteriocin and KBz is the concentra- tion of quorum molecule at which bacteriocin is half-maximal. nz defines the cooperativity of the AHL binding. kBðz; yÞ ¼ KBmaxz A0nz y K nz Bz þ A 0nz y kBðz; yÞ ¼ KBmaxz K nz Bz K nz Bz þ A 0nz y Software packages and simulation settings The ABC SMC model selection algorithm was written in python using Numpy [38], Pandas and Scipy [39]. ODE simulations were conducted in C++ with a Rosenbrock 4 stepper from the Boost library [40]. All simulations use an absolute error tolerance of 1e−9, and relative error tolerance of 1e−4. Simulations were conducted for 5000hrs, and were stopped early if the population of any strain fell below 1e−5 (extinction event). Simulations with an extinction event have distances set to maximum in order to prevent excessive time spent simulating col- lapsed populations. PLOS Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 14 / 24 Growth is modelled by Monod’s equation for growth limiting nutrient, S. mxmax defines the maximal growth rate of the strain and KX defines the concentration of substrate required for half-maximal growth. mxðSÞ ¼ mxmaxS KX þ S d fi d b ðB0Þ h d fi th i l k Killing by bacteriocin is defined by oðB0 zÞ, where ωmax defines the maximal killing rate which is set to 0 if the strain is insensitive. Kω defines the concentration at which half-maximal killing occurs. oðB0 zÞ ¼ omax B0no z K no o þ B 0no z I d i i f b i i i b QS i d fi d b k ( ) h defines the concentration omax B0no z K no o þ B 0no z b d f d b Induction or repression of bacteriocin expression by QS, is defined by kB(z, y), where z defines the bacteriocin being expressed and y defines the quorum molecule regulating its expression. KBmax z is the maximal expression rate of the bacteriocin and KBz is the concentra- tion of quorum molecule at which bacteriocin is half-maximal. Three strain synthetic communities model space definition CN, CB and CA are scaling factors: N0 x ¼ NxCN B0 z ¼ BzCB A0 y ¼ AyCA Each model is represented as sets where N defines the number of strains, B defines the set of bacteriocins and A defines the set of QS systems. The following differential equations are 2. The QS regulating a bacteriocin is not present in the system. 3. A strain is sensitive to a bacteriocin that does not exist in the system. 4. A bacteriocin exists that no strain is sensitive to. This cleanup yields the options which are used to generate ODE equations for a system. System equations State variables in each system are rescaled to improve speed of obtaining numerical approxi- mations. NX describes the concentration of a strain, Bz describes the concentration of a bacteri- ocin and Ay describes the concentration of a quorum molecule. CN, CB and CA are scaling factors: N0 x ¼ NxCN B0 z ¼ BzCB A0 y ¼ AyCA Each model is represented as sets where N defines the number of strains, B defines the set of bacteriocins and A defines the set of QS systems. The following differential equations are Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 PLOS Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 13 / 24 PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities each model. dNx dt ¼ NxmxðSÞ Nx X B z¼1 oðB0 zÞ NxD dS dt ¼ DðS0 SÞ X N x¼1 mxN0 x g dBz dt ¼ X N x¼1 ðkBx;zN0 xÞ CB DBz dAy dt ¼ X N x¼1 kAx;yN0 x CA DAy used to represent each model. used to represent each model. dNx dt ¼ NxmxðSÞ Nx X B z¼1 oðB0 zÞ NxD dS dt ¼ DðS0 SÞ X N x¼1 mxN0 x g dBz dt ¼ X N x¼1 ðkBx;zN0 xÞ CB DBz dAy dt ¼ X N x¼1 kAx;yN0 x CA DAy Growth is modelled by Monod’s equation for growth limiting nutrient, S. mxmax defines the maximal growth rate of the strain and KX defines the concentration of substrate required for half-maximal growth. mxðSÞ ¼ mxmaxS KX þ S Killing by bacteriocin is defined by oðB0 zÞ, where ωmax defines the maximal killing rate which is set to 0 if the strain is insensitive. Kω defines the concentration at which half-maximal killing occurs. Three strain synthetic communities model space definition nz defines the cooperativity of the AHL binding. Software packages and simulation settings Software packages and simulation settings The ABC SMC model selection algorithm was written in python using Numpy [38], Pandas and Scipy [39]. ODE simulations were conducted in C++ with a Rosenbrock 4 stepper from the Boost library [40]. All simulations use an absolute error tolerance of 1e−9, and relative error tolerance of 1e−4. Simulations were conducted for 5000hrs, and were stopped early if the population of any strain fell below 1e−5 (extinction event). Simulations with an extinction event have distances set to maximum in order to prevent excessive time spent simulating col- lapsed populations. Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 14 / 24 PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities Approximate Bayesian computation with sequential monte-carlo (ABC SMC) Particles are first sampled from the prior distribution and simulated. A set of distances, d, are cal- culated from the simulation. If all distances are less than the intermediate threshold, ϵt, the parti- cle is accepted (d < ϵt). Accepted particles are weighted using importance sampling. The next population is sampled from the previous, and a new threshold is generated that is closer to the final threshold, ϵF. This process is repeated until we reach a distance threshold of ϵF. ABC SMC is highly parallel, allowing us to take advantage of high performance computing resources [21]. population is sampled from the previous, and a new threshold is generated that is closer to the final threshold, ϵF. This process is repeated until we reach a distance threshold of ϵF. ABC SMC is highly parallel, allowing us to take advantage of high performance computing resources [21]. Updating ϵt. At the end of each population of ABC SMC, the distance threshold ϵt is updated to approach the final population, ϵF. The quantile parameter, α, is defined. The dis- tances of the population are sorted in ascending order and the distance at quantile α is used as the threshold for the next population. If ϵt < ϵF, we set ϵt = ϵF, marking the next population as the final population. Algorithm 1: Algorithm for model selection with ABC SMC 1: Initialisation Set population indicator, t = 0 Set ϵt Set final epsilon, ϵF Set population size, N Set population particle count, i Set distance threshold quantile, α 2: Sample particle, consisting of a model (m) and parameters (θ): If t = 0, sample θ(m) from prior distribution, π(m, θ) If t > 0, sample θ(m) from previous distribution fyðmÞ i t1g with weights w(m)t−1 3: Perturb particle If t > 0, perturb particle using perturbation kernel Kt, yielding perturbed particle θ(m) 4: Simulate particle x f(x\|θ, m) 5: Calculate distance from objective d = ρ(x, x0) 6: Accept or reject particle If d > ϵt, reject particle and go to 2 If d < ϵt, accept particle, add θ, m to population fyðmÞ i tg 7: Set accepted particle weight Particle weight, w, is set to 1 for the initial population. Approximate Bayesian computation with sequential monte-carlo (ABC SMC) For sub- sequent populations, the weight of a particle is equal to the proba- bility of observing the particle given the prior, divided by the probability of observing the particle given the previous population. wi t ¼ pðyÞ PN j¼1 wj t1Ktðyi t1jyÞ i = i + 1 If i < N go ot 2 8: Population full Normalise population particle weights If ϵt == ϵF, return fyðmÞ i tg and wt, the approximation of the posterior distribution Prepare next population Set i = 0 Set t = t + 1 Update the distance threshold as a function of the distances in the population, d and the threshold quantile, α, ϵt = fe(α, d) go to 2 Updating ϵt. At the end of each population of ABC SMC, the distance threshold ϵt is updated to approach the final population, ϵF. The quantile parameter, α, is defined. The dis- tances of the population are sorted in ascending order and the distance at quantile α is used as the threshold for the next population. If ϵt < ϵF, we set ϵt = ϵF, marking the next population as the final population. Algorithm 1: Algorithm for model selection with ABC SMC g g 1: Initialisation Set population indicator, t = 0 Set ϵt Set final epsilon, ϵF Set population size, N Set population particle count, i Set distance threshold quantile, α Set population size, N 2: Sample particle, consisting of a model (m) and parameters (θ): If t = 0, sample θ(m) from prior distribution, π(m, θ) If t > 0, sample θ(m) from previous distribution fyðmÞ i t1g with weights w(m)t−1 3: Perturb particle If t > 0, perturb particle using perturbation kernel Kt, yielding perturbed particle θ(m) 4: Simulate particle x * f(x\|θ, m) 5: Calculate distance from objective d = ρ(x, x0) 6: Accept or reject particle If d > ϵt, reject particle and go to 2 If d < ϵt, accept particle, add θ, m to population fyðmÞ i tg 7: Set accepted particle weight 15 / 24 Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities Oscillatory population dynamic objective We define the oscillatory population dynamic using three summary statistics for each strain. First, we use Fourier transform of the population signal to find the maximum frequency, f, and convert this to the period, T. pulation signal to find the maximum frequency, f, and T ¼ 1=f i h i l i i i i d d A i We set a minimum period of t/2 where t is the simulation time, giving us do1. do1. Any simu- lations in which T < t/2, do1 is set to 0, this distance ensures that all we have frequencies of oscillations that are on a scale relevant to the time period being measured. It was found that using the signal frequency alone resulted in acceptance of many simulations with very small oscillations, or simulations that rapidly dampen. We therefore generated two additional dis- tances that account for oscillation amplitudes to select for sustained oscillations only. We can define the number of expected peaks in the simulation, p. p ¼ t T e identified by changes from a positive gradient to Peaks in the trajectory are identified by changes from a positive gradient to a negative gradi- ent, and troughs via changes from negative gradient to positive gradient. The peak-to-peak amplitudes are calculated by differences between consecutive peaks and troughs. AK is the number of amplitudes above the threshold, K = 0.05. do2 is the difference between the number of expected oscillations in the simulation, and the count of above threshold oscillations. Because incomplete oscillations at the time the simulation ends can impact the distance mea- surement, we set a lenient final distance threshold for do2:do3 compares the final amplitude AF in the simulation to the threshold. We set do3 ¼ 0 if do3 > K. do3 Maximal Lyapunov exponent calculation Maximal Lyapunov exponent calculation ABC SMC was conducted with population sizes of 10, repeated 255 times yielding a combined final population of 2550 particles. Random forest classifier model Using the sci-kit learn (sklearn) python package [42], a random forest classifier was trained using 2000 estimators. The data used consisted of 3750 oscillatory input vectors, and 3750 cha- otic input vectors. Training and test datasets were generated with a ratio of 0.5 by random sampling. Fig 7 shows the performance of the classifier model on the test data. Algorithm 2: Description of dual-orbit method, demonstrated with two-dimensional system 1 Set S = 0.0 2 Set parameters and initial state θi = (xi, yi) for orbit, f(θi) 3 Simulate f(θi) for transient time, tt, yielding state, θa0 4 Set initial state of nearby orbit, f(θb0), where, θb0 = θa0 + 40 Fig 6. Illustration of dual-orbit algorithm used to calculate the λ1. Two orbits with an initial state separation of 40 are followed. After each time step measure the separation, 41, is measured. The perturbed orbit (red) is readjusted to prevent excess separation. The average rate of separation between the two orbits corresponds with the λ1. https://doi.org/10.1371/journal.pcbi.1010548.g006 COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities Fig 6. Illustration of dual-orbit algorithm used to calculate the λ1. Two orbits with an initial state separation of 40 are followed. After each time step measure the separation, 41, is measured. The perturbed orbit (red) is readjusted to prevent excess separation. The average rate of of dual-orbit algorithm used to calculate the λ1. Two orbits with an initial state separation of 40 are followed. After each he separation, 41, is measured. The perturbed orbit (red) is readjusted to prevent excess separation. The average rate of the two orbits corresponds with the λ1. 71/journal.pcbi.1010548.g006 Fig 6. Illustration of dual-orbit algorithm used to calculate the λ1. Two orbits with an initial state sepa time step measure the separation, 41, is measured. The perturbed orbit (red) is readjusted to prevent exc separation between the two orbits corresponds with the λ1. https://doi.org/10.1371/journal.pcbi.1010548.g006 For each sampled particle a prescreening process was performed to minimise time spent conducting the more computationally time consuming dual-orbit method. Simulations in which a strain fell below 10−5 were rejected. The number of oscillations with an amplitude greater than 0.05 was counted for each strain signal. If any strain showed less than 2 oscilla- tions the particle was rejected. Maximal Lyapunov exponent calculation ABC SMC was conducted with population sizes of 10, repeated 255 times yielding a combined final population of 2550 particles. Maximal Lyapunov exponent calculation Lyapunov exponents can be used to measure chaotic behaviour; they describe the average exponential rate of divergence between two near trajectories of a dynamical system. The max- imal Lyapunov exponent, λ1, can be used as determinant of chaotic behaviour. Using a method described by Sprott et al. [41], we evolve two nearby orbits and measure their average rate of separation. This directly investigates whether small changes to an initial state will pro- duce a disproportionate separation. By periodically readjusting the distance of divergence after each time step we measure separation across a period of time, preventing a single event dominating subsequent states (Fig 6). The method is described in full by Algorithm 2. For all simulations we generate nearby orbits by perturbing one of the strain initial strain densities by 40 = 10−10. All simulations use a transient time equivalent to the first 10% of the time series. Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 16 / 24 PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities Chaos population dynamic objective dC1 is the only distance for the chaotic objective. If dC1 < 0, the particle is rejected. The final distance threshold, ϵC, is equivalent to all λ1 > 0.003. dC1 ¼ 1=ð1 þ l1Þ d ¼ ðdC1Þ ϵC ¼ f0:997g For each sampled particle a prescreening process was performed to minimise time spent conducting the more computationally time consuming dual-orbit method. Simulations in Fig 6. Illustration of dual-orbit algorithm used to calculate the λ1. Two orbits with an initial state separation of 40 are followed. After each time step measure the separation, 41, is measured. The perturbed orbit (red) is readjusted to prevent excess separation. The average rate of separation between the two orbits corresponds with the λ1. https://doi.org/10.1371/journal.pcbi.1010548.g006 UTATIONAL BIOLOGY Chaos in synthetic microbial communities Chaos population dynamic objective dC1 is the only distance for the chaotic objective. If dC1 < 0, the particle is rejected. The final distance threshold, ϵC, is equivalent to all λ1 > 0.003. dC1 ¼ 1=ð1 þ l1Þ d ¼ ðdC1Þ ϵC ¼ f0:997g For each sampled particle a prescreening process was performed to minimise time spent conducting the more computationally time consuming dual-orbit method. Simulations in which a strain fell below 10−5 were rejected. The number of oscillations with an amplitude greater than 0.05 was counted for each strain signal. If any strain showed less than 2 oscilla- tions the particle was rejected. Random forest classifier model Using the sci-kit learn (sklearn) python package [42], a random forest classifier was trained using 2000 estimators. The data used consisted of 3750 oscillatory input vectors, and 3750 cha- otic input vectors. Training and test datasets were generated with a ratio of 0.5 by random sampling. Fig 7 shows the performance of the classifier model on the test data. Using the sci-kit learn (sklearn) python package [42], a random forest classifier was trained using 2000 estimators. The data used consisted of 3750 oscillatory input vectors, and 3750 cha- otic input vectors. Training and test datasets were generated with a ratio of 0.5 by random sampling. Fig 7 shows the performance of the classifier model on the test data. Random forest classifier model Algorithm 2: Description of dual-orbit method, demonstrated with two-dimensional system 1 Set S = 0.0 2 Set parameters and initial state θi = (xi, yi) for orbit, f(θi) 3 Simulate f(θi) for transient time, tt, yielding state, θa0 4 Set initial state of nearby orbit, f(θb0), where, θb0 = θa0 + 40 5 Set t = 0 6 Advance f ðya0Þ and f ðyb0Þ by one step, dt, yielding states ya1 and yb1 respectively 7 Set t = t + dt 8 Calculate separation between the state variables of the two orbits, 41 ¼ ½ðxa1 xb1Þ 2 þ ðya1 yb1Þ 2 1=2 Algorithm 2: Description of dual-orbit method, demonstrated with two-dimensional system 1 Set S = 0.0 4 Set initial state of nearby orbit, f(θb0), where, θb0 = θa0 + 40 5 Set t = 0 6 Advance f ðya0Þ and f ðyb0Þ by one step, dt, yielding states ya1 and yb1 respectively 7 Set t = t + dt 8 Calculate separation between the state variables of the two orbits, 41 ¼ ½ðxa1 xb1Þ 2 þ ðya1 yb1Þ 2 1=2 Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 17 / 24 9 S = S + log2(\|41/40\|) 10 Readjust yb1 to align directionally with ya1, xb0 ¼ xa1 þ 40ðxb1 xa1Þ=41 and yb0 ¼ ya1 þ 40ðyb1 ya1Þ=41 11 Set xa0 ¼ xa1 and xb0 ¼ xb1 12 Repeat lines 6 to 11 for n iterations 13 Calculate maximal Lyapunov exponent as an average of the separation values, λ1 = S/n Analysis of m850 m850 is described by the following equations dN1 dt ¼ N1 m1maxS K þ S omax N1B 0no 1 K no o þ B 0no 1 N1D ð1aÞ dN2 dt ¼ N2 m2maxS K þ S N2D ð1bÞ dN3 dt ¼ N3 m3maxS K þ S omax N3B 0no 2 K no o þ B 0no 2 N3D ð1cÞ dS DðS SÞ m1N0 1 m2N0 2 m3N0 3 ð1dÞ Fig 7. Confusion matrix showing accuracy of random forest classifier on test data. org/10.1371/journal.pcbi.1010548 October 10, 2022 18 / 24 Random forest classifier model https://doi.org/10.1371/journal.pcbi.1010548.g007 LOGY Chaos in synthetic microbial communities PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities 9 S = S + log2(\|41/40\|) 10 Readjust yb1 to align directionally with ya1, xb0 ¼ xa1 þ 40ðxb1 xa1Þ=41 and yb0 ¼ ya1 þ 40ðyb1 ya1Þ=41 11 Set xa0 ¼ xa1 and xb0 ¼ xb1 12 Repeat lines 6 to 11 for n iterations 13 Calculate maximal Lyapunov exponent as an average of the separation values, λ1 = S/n Analysis of m850 m850 is described by the following equations dN1 dt ¼ N1 m1maxS K þ S omax N1B 0no 1 K no o þ B 0no 1 N1D ð1aÞ dN2 dt ¼ N2 m2maxS K þ S N2D ð1bÞ dN3 dt ¼ N3 m3maxS K þ S omax N3B 0no 2 K no o þ B 0no 2 N3D ð1cÞ dS dt ¼ DðS0 SÞ m1N0 1 g m2N0 2 g m3N0 3 g ð1dÞ dB1 dt ¼ kB1;1N0 1 CB DB1 ð1eÞ dB2 dt ¼ kB2;1N0 3 CB DB2 ð1fÞ dA1 dt ¼ kA1;1N0 1 CA DA1 ð1gÞ Fig 7. Confusion matrix showing accuracy of random forest classifier on test data. https://doi.org/10.1371/journal.pcbi.1010548.g007 PLOS Computational Biology \| https://doi org/10 1371/journal pcbi 1010548 October 10 2022 18 / 24 9 S = S + log2(\|41/40\|) 10 Readjust yb1 to align directionally with ya1, xb0 ¼ xa1 þ 40ðxb1 xa1Þ=41 and yb0 ¼ ya1 þ 40ðyb1 ya1Þ=41 11 Set xa0 ¼ xa1 and xb0 ¼ xb1 12 Repeat lines 6 to 11 for n iterations 13 Calculate maximal Lyapunov exponent as an average of the separation values, λ1 = S/n Analysis of m850 m850 is described by the following equations dN S N B 0n Fig 7. Confusion matrix showing accuracy of random forest classifier on test data. https://doi.org/10.1371/journal.pcbi.1010548.g007 ð1aÞ ð1bÞ ð1fÞ PLOS Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 18 / 24 PLOS Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 18 / 24 PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities N0 x ¼ NxCN B0 z ¼ BzCB A0 y ¼ AyCA kBz;y ¼ KBmaxz A0nz y K nz Bz þ A 0nz y By setting the left hand side of (1) to 0 we find a number of steady states P = (N1, N2, N3, S, B1, B2, A1). The trivial steady state. Random forest classifier model We inter- pret the right and left-hand-sides of (2) as a function of S, denoting them by R(S) and L(S) respectively. It is easy to see that A1(S) is a decreasing function of S, B1(S) increases as a function of A1 and R(S) is an increasing function of B1. Consequently the R(S) is a decreasing function of S. We also see that R(0) = ωm ax > 0 and R(S0) = 0. Further L(0) = −D and LðS0Þ ¼ m1max S0 KþS0 D and L(S) increases as a function of S. This proves the statement. To summarise, the single-strain survival steady state requires the corresponding maximal growth rate to be large compared to other parameters. PLOS Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 20 / 24 The steady state P1 = (N1, 0, 0, S, B1, 0, A1) is more complicated and can not be expressed explicitly. Instead it is given as follows: Assume there exists a solution S to the following equa- tion m1max S K þ S D ¼ omax ðB1ðSÞCBÞ no K no o1 þ ðB1ðSÞCBÞ no ; ð2Þ where B1ðSÞ ¼ KBmax1 kA1 ðA1ðSÞCBÞ n1þ1 K n1 AB1 þ ðA1ðSÞCBÞ n1 ; and A1ðSÞ ¼ kA1g CBm1max ðS0 SÞðK þ SÞ S : If such a solution S exists then B1 = B1(S), A1 = A1(S) and N1 ¼ DCB kA1 CN A1. Lemma 1 There exists a unique steady state P1 if and only if D < m1max S0 K þ S0 : The steady state P1 = (N1, 0, 0, S, B1, 0, A1) is more complicated and can not be expressed explicitly. Instead it is given as follows: Assume there exists a solution S to the following equa- tion m1max S K þ S D ¼ omax ðB1ðSÞCBÞ no K no o1 þ ðB1ðSÞCBÞ no ; ð2Þ ð2Þ If such a solution S exists then B1 = B1(S), A1 = A1(S) and N1 ¼ DCB kA1 CN A1. Lemma 1 There exists a unique steady state P1 if and only if D < m1 S0 K S : mma 1 There exists oof: We need the so Proof: We need the solution to (2) to fulfil S < S0 in order for A1 to be positive. We inter- pret the right and left-hand-sides of (2) as a function of S, denoting them by R(S) and L(S) respectively. Random forest classifier model P0 = (0, 0, 0, S0, 0, 0, 0). The Jacobian of the linearisation has eigenvalues D; D m1max S0 S0 þ K ; D m2max S0 S0 þ K ; D m3max S0 S0 þ K : Consequently the trivial steady state always exists and is linearly stable for Consequently the trivial steady state always exists and is linearly stable for D > S0 S0 þ K maxfm1max; m2max; m3maxg: D > S0 S0 þ K maxfm1max; m2max; m3maxg: This shows that if the dilution rate is high enough, no strain can survive. g g , One strain only steady states. There are three steady states where only one strain sur- vives, P1, P2, P3. While P2, and P3 can be calculated explicitly, P1 is given implicitly (see below). We start with P2: P2 ¼ ð0; N2; 0; S; 0; 0; 0Þ; where N2 ¼ g CN S0m2max DðS0 þ KÞ m2max D ; S ¼ DK m2max D : We see that P2 exists provided D < m2maxS0 S0 þ K : The linearisation at P2 has eigenvalues D; D 1 m1max m2max ! ; D 1 m3max m2max ! ; 1 Km2max m2max D m2maxS0 DðS0 þ KÞ : This shows that P2 exists and is linearly stable if One strain only steady states. There are three steady states where only one strain sur- vives, P1, P2, P3. While P2, and P3 can be calculated explicitly, P1 is given implicitly (see below). Random forest classifier model W t t ith P We start with P2: P2 We see that P2 exists provided This shows that P2 exists and is linearly stable if D < m2maxS0 S0 þ K ; and m2max > maxfm1max; m3maxg: PLOS Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 19 / 24 PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities The situation for P3 is very similar: P3 ¼ ð0; 0; N3; S; 0; 0; 0Þ; where N3 ¼ g CN S0m3max DðS0 þ KÞ m3max D ; S ¼ DK m3max D : We see that P3 exists provided D < m3maxS0 S0 þ K : The linearisation at P3 has eigenvalues D; Dð1 m2max m3max Þ; Dð1 m1max m3max Þ; 1 Km3max ðm3max DÞðm3maxS0 DðS0 þ KÞÞ: This shows that P3 exists and is linearly stable if D < m3maxS0 S0 þ K ; and m3max > maxfm1max; m2maxg: The steady state P = (N 0 0 S B 0 A ) is more complicated and can not be expressed GY Chaos in synthetic microbial communities The situation for P3 is very similar: where We see that P3 exists provided We see that P3 exists provided The linearisation at P3 has eigenvalues The linearisation at P3 has eigenvalues Dð1 m1max m3max Þ; 1 Km3max ðm3max DÞðm3maxS0 DðS0 þ KÞÞ: This shows that P3 exists and is linearly stable if D < m3maxS0 S0 þ K ; and m3max > maxfm1max; m2maxg: The steady state P1 = (N1, 0, 0, S, B1, 0, A1) is more complicated and can not be expressed explicitly. Instead it is given as follows: Assume there exists a solution S to the following equa- tion m1max S K þ S D ¼ omax ðB1ðSÞCBÞ no K no o1 þ ðB1ðSÞCBÞ no ; ð2Þ where B1ðSÞ ¼ KBmax1 kA1 ðA1ðSÞCBÞ n1þ1 K n1 AB1 þ ðA1ðSÞCBÞ n1 ; and A1ðSÞ ¼ kA1g CBm1max ðS0 SÞðK þ SÞ S : If such a solution S exists then B1 = B1(S), A1 = A1(S) and N1 ¼ DCB kA1 CN A1. Lemma 1 There exists a unique steady state P1 if and only if D < m1max S0 K þ S0 : Proof: We need the solution to (2) to fulfil S < S0 in order for A1 to be positive. Random forest classifier model It is easy to see that A1(S) is a decreasing function of S, B1(S) increases as a function of A1 and R(S) is an increasing function of B1. Consequently the R(S) is a decreasing function of S. We also see that R(0) = ωm ax > 0 and R(S0) = 0. Further L(0) = −D and LðS0Þ ¼ m1max S0 KþS0 D and L(S) increases as a function of S. This proves the statement. To summarise, the single-strain survival steady state requires the corresponding maximal growth rate to be large compared to other parameters. Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 PLOS Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 20 / 24 PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities The three-strain co-existence steady state. P123 = (N1, N2, N3, S, B1, B2, A1). From the equation for N2 we obtain that S ¼ DK m2max D m2max < minfm1max; m3maxg; D < minfm2max S0 K þ S0 ; omax m2max m1max m2max ; omax m2max m3max m2max g The three-strain co-existence steady state. P123 = (N1, N2, N3, S, B1, B2, A1). From the equation for N2 we obtain that S ¼ DK m2max D m2max < minfm1max; m3maxg; D < minfm2max S0 K þ S0 ; omax m2max m1max m2max ; omax m2max m3max m2max g Stability. Solved numerically using MATLAB. For each of the 3750 chaotic input vectors we used numerical root finding to calculate P123, and determined its stability by numerically determining the eigenvalues of the Jacobian. We found P123 existed for all 3750 input vectors and was stable for 7.8% of them. Supporting information S1 Fig. Posterior distribution of chaotic objective for gLV model. Posterior distribution of of all parameters. (TIF) Author Contributions Conceptualization: Behzad D. Karkaria, Alex J. H. Fedorec, Chris P. Barnes. Formal analysis: Behzad D. Karkaria. Methodology: Behzad D. Karkaria, Angelika Manhart, Alex J. H. Fedorec. Project administration: Chris P. Barnes. Resources: Chris P. Barnes. Software: Behzad D. Karkaria. Supervision: Chris P. Barnes. Visualization: Behzad D. Karkaria. Writing – original draft: Behzad D. Karkaria, Angelika Manhart. Author Contributions Conceptualization: Behzad D. Karkaria, Alex J. H. Fedorec, Chris P. Barnes. Formal analysis: Behzad D. Karkaria. Methodology: Behzad D. Karkaria, Angelika Manhart, Alex J. H. Fedorec. Project administration: Chris P. Barnes. Resources: Chris P. Barnes. Software: Behzad D. Karkaria. Supervision: Chris P. Barnes. Visualization: Behzad D. Karkaria. Writing – original draft: Behzad D. Karkaria, Angelika Manhart. References 1. Strogatz SH. Nonlinear Dynamics and Chaos. In: Nonlinear Dynamics and Chaos; 2018. 2. Lorenz EN. Deterministic Nonperiodic Flow. Journal of the Atmospheric Sciences. 1963; 20(2). https:// doi.org/10.1175/1520-0469(1963)020%3C0130:DNF%3E2.0.CO;2 3. Ispolatov I, Madhok V, Allende S, Doebeli M. Chaos in High-Dimensional Dissipative Dynamical Sys- tems. Scientific Reports. 2015; 5. https://doi.org/10.1038/srep12506 PMID: 26224119 4. Albers DJ, Sprott JC, Crutchfield JP. Persistent Chaos in High Dimensions. Physical Review E—Statis- tical, Nonlinear, and Soft Matter Physics. 2006; 74(5). PMID: 17280024 5. Dechert WD, Sprott JC, Albers DJ. On the Probability of Chaos in Large Dynamical Systems: A Monte Carlo Study. Journal of Economic Dynamics and Control. 1999; 23(8). https://doi.org/10.1016/S0165- 1889(98)00053-0 5. Dechert WD, Sprott JC, Albers DJ. On the Probability of Chaos in Large Dynamical Systems: A Monte Carlo Study. Journal of Economic Dynamics and Control. 1999; 23(8). https://doi.org/10.1016/S0165- 1889(98)00053-0 21 / 24 Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities 6. Jawed K, Yazdani SS, Koffas MA. Advances in the Development and Application of Microbial Consortia for Metabolic Engineering. Metabolic engineering communications. 2019; 9:e00095. https://doi.org/10. 1016/j.mec.2019.e00095 PMID: 31720211 7. Tsoi R, Dai Z, You L. Emerging Strategies for Engineering Microbial Communities. Biotechnology advances. 2019; 37(6):107372. https://doi.org/10.1016/j.biotechadv.2019.03.011 PMID: 30880142 8. Becks L, Hilker FM, Malchow H, Ju¨rgens K, Arndt H. Experimental Demonstration of Chaos in a Micro- bial Food Web. Nature. 2005; 435(7046):1226–1229. https://doi.org/10.1038/nature03627 PMID: 15988524 9. Benincà E, Huisman J, Heerkloss R, Jo¨hnk KD, Branco P, Van Nes EH, et al. Chaos in a Long-Term Experiment with a Plankton Community. Nature. 2008; 451(7180):822–825. https://doi.org/10.1038/ nature06512 PMID: 18273017 10. Hsu RH, Clark RL, Tan JW, Ahn JC, Gupta S, Romero PA, et al. Microbial Interaction Network Inference in Microfluidic Droplets. Cell Systems. 2019; 9(3):229–242.e4. https://doi.org/10.1016/j.cels.2019.06. 008 PMID: 31494089 11. Vano JA, Wildenberg JC, Anderson MB, Noel JK, Sprott JC. Chaos in Low-Dimensional Lotka-Volterra Models of Competition. Nonlinearity. 2006; 19(10):2391–2404. https://doi.org/10.1088/0951-7715/19/ 10/006 12. Momeni B, Xie L, Shou W. Lotka-Volterra Pairwise Modeling Fails to Capture Diverse Pairwise Micro- bial Interactions. eLife. 2017; 6. https://doi.org/10.7554/eLife.25051 PMID: 28350295 13. Fan G, Wolkowicz GSK. Chaotic Dynamics in a Simple Predator-Prey Model with Discrete Delay. Dis- crete and Continuous Dynamical Systems—Series B. 2021; 26(1). 14. Zhao H, Sun Y, Wang Z. Control of Hopf Bifurcation and Chaos in a Delayed Lotka-Volterra Predator- Prey System with Time-Delayed Feedbacks. Abstract and Applied Analysis. 2014; 2014. https://doi.org/ 10.1155/2014/104156 15. Gilpin ME. References Weiße AY, Oyarzu´n DA, Danos V, Swain PS. Mechanistic Links between Cellular Trade-Offs, Gene Expression, and Growth. Proceedings of the National Academy of Sciences. 2015; 112(9):E1038– E1047. https://doi.org/10.1073/pnas.1416533112 PMID: 25695966 30. Kirk P, Rolando DMY, MacLean AL, Stumpf MPH. Conditional Random Matrix Ensembles and the Sta- bility of Dynamical Systems. New Journal of Physics. 2015; 17(8):083025. https://doi.org/10.1088/ 1367-2630/17/8/083025 31. Allesina S, Tang S. Stability Criteria for Complex Ecosystems. Nature. 2012; 483(7388):205–208. https://doi.org/10.1038/nature10832 32. Liao MJ, Din MO, Tsimring L, Hasty J. Rock-Paper-Scissors: Engineered Population Dynamics Increase Genetic Stability. Science (New York, NY). 2019; 365(6457):1045–1049. https://doi.org/10. 1126/science.aaw0542 33. Kong W, Meldgin DR, Collins JJ, Lu T. Designing Microbial Consortia with Defined Social Interactions. Nature Chemical Biology. 2018; 14(8):821–829. https://doi.org/10.1038/s41589-018-0091-7 PMID: 29942078 34. Ziesack M, Gibson T, Oliver JKW, Shumaker AM, Hsu BB, Riglar DT, et al. Engineered Interspecies Amino Acid Cross-Feeding Increases Population Evenness in a Synthetic Bacterial Consortium. mSys- tems. 2019; 4(4):e00352–19. https://doi.org/10.1128/mSystems.00352-19 PMID: 31409662 35. Ha¨dicke O, Klamt S. EColiCore2: A Reference Network Model of the Central Metabolism of Escherichia Coli and Relationships to Its Genome-Scale Parent Model. Scientific Reports. 2017; 7(1):1–15. 36. Ye C, Zou W, Xu N, Liu L. Metabolic Model Reconstruction and Analysis of an Artificial Microbial Eco- system for Vitamin C Production. Journal of Biotechnology. 2014;182–183(1). 37. Ang KS, Lakshmanan M, Lee NR, Lee DY. Metabolic Modeling of Microbial Community Interactions for Health, Environmental and Biotechnological Applications. Current Genomics. 2018; 19(8). https://doi. org/10.2174/1389202919666180911144055 PMID: 30532650 38. Harris CR, Millman KJ, van der Walt SJ, Gommers R, Virtanen P, Cournapeau D, et al. Array Program- ming with NumPy. Nature. 2020; 585 (7825). https://doi.org/10.1038/s41586-020-2649-2 s://doi.org/10.1038/s41586 020 2649 2 M, Reddy T, Cournapeau D, et al. SciPy 1.0: Fu hon. Nature Methods. 2020; 17(3). h Library. In: Proceedings of The. vol. 243; 2002 3. 39. Virtanen P, Gommers R, Oliphant TE, Haberland M, Reddy T, Cournapeau D, et al. SciPy 1.0: Funda- mental Algorithms for Scientific Computing in Python. Nature Methods. 2020; 17(3). 40. Siek JG, Lee LQ, Lumsdaine A. The Boost Graph Library. In: Proceedings of The. vol. 243; 2002. 0. Siek JG, Lee LQ, Lumsdaine A. The Boost Graph Library. In: Proceedings of The. vol. 243; 2002. 1. Sprott JC. Chaos and Time-Series Analysis; 2003. 41. Sprott JC. Chaos and Time-Series Analysis; 2003. 41. Sprott JC. Chaos and Time-Series Analysis; 2003. 42. Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, et al. References Spiral Chaos in a Predator-Prey Model. The American Naturalist. 1979; 113(2):306–308. https://doi.org/10.1086/283389 https://doi.org/10.1086/283389 16. Fedorec AJH, Karkaria BD, Sulu M, Barnes CP. Single Strain Control of Microbial Consortia. Nature Communications 2021; 12(1):1–12 https://doi org/10 1038/s41467-021-22240-x PMID: 33785746 16. Fedorec AJH, Karkaria BD, Sulu M, Barnes CP. Single Strain Control of Microbial Consortia. Nature Communications. 2021; 12(1):1–12. https://doi.org/10.1038/s41467-021-22240-x PMID: 33785746 17. Karkaria BD, Fedorec AJH, Barnes CP. Automated Design of Synthetic Microbial Communities. Nature Communications. 2021; 12(1):1–12. https://doi.org/10.1038/s41467-020-20756-2 PMID: 33510148 ated Des ps://doi.o Moon S, et 18. Silk D, Kirk PDW, Barnes CP, Toni T, Rose A, Moon S, et al. Designing Attractive Models via Automated Identification of Chaotic and Oscillatory Dynamical Regimes. Nature Communications. 2011; 2(1). https://doi.org/10.1038/ncomms1496 PMID: 21971504 19. Roques L, Chekroun MD. Probing Chaos and Biodiversity in a Simple Competition Model. Ecological Complexity. 2011; 8(1). https://doi.org/10.1016/j.ecocom.2010.08.004 20. Liu YY, Baraba´si AL. Control Principles of Complex Systems. Reviews of Modern Physics. 2016; 88(3). https://doi.org/10.1103/RevModPhys.88.035006 21. Toni T, Welch D, Strelkowa N, Ipsen A, Stumpf MP. Approximate Bayesian Computation Scheme for Parameter Inference and Model Selection in Dynamical Systems. Journal of the Royal Society Inter- face. 2009; 6(31):187–202. https://doi.org/10.1098/rsif.2008.0172 PMID: 19205079 2 2 2 2 2 2 22. Barnes CP, Silk D, Sheng X, Stumpf MPH. Bayesian Design of Synthetic Biological Systems. Proceed- ings of the National Academy of Sciences. 2011; 108(37):15190–15195. https://doi.org/10.1073/pnas. 1017972108 PMID: 21876136 23. The´bault E, Fontaine C. Stability of Ecological Communities and the Architecture of Mutualistic and Tro- phic Networks. Science (New York, NY). 2010; 329(5993):853–856. https://doi.org/10.1126/science. 1188321 PMID: 20705861 24. Coyte KZ, Schluter J, Foster KR. The Ecology of the Microbiome: Networks, Competition, and Stabil- ity. Science (New York, NY). 2015; 350 (6261). https://doi.org/10.1126/science.aad2602 PMID: 26542567 25. Breiman L. Random Forests. Machine Learning. 2001; 45(1):5–32. https://doi.org/10.1023/ A:1010933404324 26. Genuer R, Poggi JM, Tuleau-Malot C. Variable Selection Using Random Forests. Pattern Recognition Letters. 2010; 31(14):2225–2236. https://doi.org/10.1016/j.patrec.2010.03.014 27. Balagadde´ FK, Song H, Ozaki J, Collins CH, Barnet M, Arnold FH, et al. A Synthetic Escherichia Coli Predator-Prey Ecosystem. Molecular systems biology. 2008; 4(187):187. https://doi.org/10.1038/msb. 2008.24 PMID: 18414488 22 / 24 PLOS Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities 28. Yurtsev EA, Conwill A, Gore J. Oscillatory Dynamics in a Bacterial Cross-Protection Mutualism. Pro- ceedings of the National Academy of Sciences of the United States of America. 2016; 113(22):6236– 41. https://doi.org/10.1073/pnas.1523317113 PMID: 27194723 29. References Scikit-Learn: Machine Learning in Python. Journal of Machine Learning Research. 2011; 12. 43. Kylilis N, Tuza ZA, Stan GB, Polizzi KM. Tools for Engineering Coordinated System Behaviour in Syn- thetic Microbial Consortia. Nature Communications. 2018; 9(1). https://doi.org/10.1038/s41467-018- 05046-2 PMID: 29992956 44. Senn H, Lendenmann U, Snozzi M, Hamer G, Egli T. The Growth of Escherichia Coli in Glucose-Limited Chemostat Cultures: A Re-Examination of the Kinetics. BBA—General Subjects. 1994. https://doi.org/ 10.1016/0304-4165(94)90072-8 PMID: 7803473 4 4 4 4 4 45. Destoumieux-Garzo´n D, Duquesne S, Peduzzi J, Goulard C, Desmadril M, Letellier L, et al. The Iron- Siderophore Transporter FhuA Is the Receptor for the Antimicrobial Peptide Microcin J25: Role of the Microcin Val11-Pro16 β-Hairpin Region in the Recognition Mechanism. Biochemical Journal. 2005. https://doi.org/10.1042/BJ20042107 PMID: 15862112 46. Kaur K, Tarassova O, Dangeti RV, Azmi S, Wishart D, McMullen L, et al. Characterization of a Highly Potent Antimicrobial Peptide Microcin N from Uropathogenic Escherichia Coli. FEMS Microbiology Let- ters. 2016; 363(11):1–6. https://doi.org/10.1093/femsle/fnw095 PMID: 27190283 47. Andersen KB, Von Meyenburg K. Are Growth Rates of Escherichia Coli in Batch Cultures Limited by Respiration? Journal of Bacteriology. 1980. https://doi.org/10.1128/jb.144.1.114-123.1980 PMID: 6998942 48. Marenda M, Zanardo M, Trovato A, Seno F, Squartini A. Modeling Quorum Sensing Trade-Offs between Bacterial Cell Density and System Extension from Open Boundaries. Scientific Reports. 2016. https://doi.org/10.1038/srep39142 PMID: 27966657 49. Weiße AY, Oyarzu´n DA, Danos V, Swain PS. Mechanistic Links between Cellular Trade-Offs, Gene Expression, and Growth. Proceedings of the National Academy of Sciences of the United States of America. 2015; 112(9):E1038–1047. https://doi.org/10.1073/pnas.1416533112 PMID: 25695966 23 / 24 PLOS Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 50. Marisch K, Bayer K, Scharl T, Mairhofer J, Krempl PM, Hummel K, et al. A Comparative Analysis of Industrial Escherichia Coli K-12 and b Strains in High-Glucose Batch Cultivations on Process-, Tran- scriptome- and Proteome Level. PLoS ONE. 2013. https://doi.org/10.1371/journal.pone.0070516 PMID: 23950949 51. Treloar NJ, Fedorec AJH, Ingalls B, Barnes CP. Deep Reinforcement Learning for the Control of Micro- bial Co-Cultures in Bioreactors. PLOS Computational Biology. 2020; 16(4):e1007783. https://doi.org/ 10.1371/journal.pcbi.1007783 PMID: 32275710 52. Destoumieux-Garzo´n D, Thomas X, Santamaria M, Goulard C, Barthe´le´my M, Boscher B, et al. Micro- cin E492 Antibacterial Activity: Evidence for a TonB-dependent Inner Membrane Permeabilization on Escherichia Coli. Molecular Microbiology. 2003. https://doi.org/10.1046/j.1365-2958.2003.03610.x PMID: 12890026 S COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities PLOS COMPUTATIONAL BIOLOGY Chaos in synthetic microbial communities 50. 50. Marisch K, Bayer K, Scharl T, Mairhofer J, Krempl PM, Hummel K, et al. A Comparative Analysis of Industrial Escherichia Coli K-12 and b Strains in High-Glucose Batch Cultivations on Process-, Tran- scriptome- and Proteome Level. PLoS ONE. 2013. https://doi.org/10.1371/journal.pone.0070516 PMID: 23950949 References Marisch K, Bayer K, Scharl T, Mairhofer J, Krempl PM, Hummel K, et al. A Comparative Analysis of Industrial Escherichia Coli K-12 and b Strains in High-Glucose Batch Cultivations on Process-, Tran- scriptome- and Proteome Level. PLoS ONE. 2013. https://doi.org/10.1371/journal.pone.0070516 PMID: 23950949 51. Treloar NJ, Fedorec AJH, Ingalls B, Barnes CP. Deep Reinforcement Learning for the Control of Micro- bial Co-Cultures in Bioreactors. PLOS Computational Biology. 2020; 16(4):e1007783. https://doi.org/ 10.1371/journal.pcbi.1007783 PMID: 32275710 51. Treloar NJ, Fedorec AJH, Ingalls B, Barnes CP. Deep Reinforcement Learning for the Control of Micro- bial Co-Cultures in Bioreactors. PLOS Computational Biology. 2020; 16(4):e1007783. https://doi.org/ 10.1371/journal.pcbi.1007783 PMID: 32275710 52. Destoumieux-Garzo´n D, Thomas X, Santamaria M, Goulard C, Barthe´le´my M, Boscher B, et al. Micro- cin E492 Antibacterial Activity: Evidence for a TonB-dependent Inner Membrane Permeabilization on Escherichia Coli. Molecular Microbiology. 2003. https://doi.org/10.1046/j.1365-2958.2003.03610.x PMID: 12890026 52. Destoumieux-Garzo´n D, Thomas X, Santamaria M, Goulard C, Barthe´le´my M, Boscher B, et al. Micro- cin E492 Antibacterial Activity: Evidence for a TonB-dependent Inner Membrane Permeabilization on Escherichia Coli. Molecular Microbiology. 2003. https://doi.org/10.1046/j.1365-2958.2003.03610.x PMID: 12890026 52. Destoumieux-Garzo´n D, Thomas X, Santamaria M, Goulard C, Barthe´le´my M, Boscher B, et al. Micro- cin E492 Antibacterial Activity: Evidence for a TonB-dependent Inner Membrane Permeabilization on Escherichia Coli. Molecular Microbiology. 2003. https://doi.org/10.1046/j.1365-2958.2003.03610.x PMID: 12890026 PLOS Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 24 / 24 Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 PLOS Computational Biology \| https://doi.org/10.1371/journal.pcbi.1010548 October 10, 2022 24 / 24
https://openalex.org/W3020177819	https://orbit.dtu.dk/files/211228524/brb3.1630.pdf	English	null	Changes in the left temporal microstate are a sign of cognitive decline in patients with Alzheimer’s disease	Brain and behavior	2,020	cc-by	12,287	Citation (APA): Musaeus, C. S., Engedal, K., Høgh, P., Jelic, V., Khanna, A. R., Kjaer, T. W., Mørup, M., Naik, M., Oeksengaard, A.-R., Santarnecchi, E., Snaedal, J., Wahlund, L.-O., Waldemar, G., & Andersen, B. B. (2020). Changes in the left temporal microstate are a sign of cognitive decline in patients with Alzheimer's disease. Brain and Behavior, 10(6), Article e01630. https://doi.org/10.1002/brb3.1630 Downloaded from orbit.dtu.dk on: Oct 24, 2024 Downloaded from orbit.dtu.dk on: Oct 24, 2024 Downloaded from orbit.dtu.dk on: Oct 24, 2024 General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.  Users may download and print one copy of any publication from the public portal for the purpose of private study or research.  You may not further distribute the material or use it for any profit-making activity or commercial gain  You may freely distribute the URL identifying the publication in the public portal If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. Changes in the left temporal microstate are a sign of cognitive decline in patients with Alzheimer’s disease \| 1 of 15 https://doi.org/10.1002/brb3.1630 wileyonlinelibrary.com/journal/brb3 This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2020 The Authors. Brain and Behavior published by Wiley Periodicals LLC. The peer review history for this article is available at https://publons.com/publon/10.1002/brb3.1630 Correspondence Christian S. Musaeus, Rigshospitalet, University of Copenhagen Blegdamsvej 9–section 6911, 2100 Copenhagen, Denmark. Email: christian.sandoee.musaeus@regionh.dk Funding information Kavli Charitable Trust Abstract Introduction: Large-scale brain networks are disrupted in the early stages of Alzheimer's disease (AD). Electroencephalography microstate analysis, a promising method for studying brain networks, parses EEG signals into topographies represent- ing discrete, sequential network activations. Prior studies indicate that patients with AD show a pattern of global microstate disorganization. We investigated whether any specific microstate changes could be found in patients with AD and mild cogni- tive impairment (MCI) compared to healthy controls (HC). Materials and methods: Standard EEGs were obtained from 135 HC, 117 patients with MCI, and 117 patients with AD from six Nordic memory clinics. We parsed the data into four archetypal microstates. Results: There was significantly increased duration, occurrence, and coverage of microstate A in patients with AD and MCI compared to HC. When looking at micro- states in specific frequency bands, we found that microstate A was affected in delta (1–4 Hz), theta (4–8 Hz), and beta (13–30 Hz), while microstate D was affected only Brain and Behavior. 2020;00:e01630. \| 1 of 15 https://doi.org/10.1002/brb3.1630 wileyonlinelibrary.com/journal/brb3 This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2020 The Authors. Brain and Behavior published by Wiley Periodicals LLC. The peer review history for this article is available at https://publons.com/publon/10.1002/brb3.1630 Correspondence Christian S. Musaeus, Rigshospitalet, University of Copenhagen Blegdamsvej 9–section 6911, 2100 Copenhagen, Denmark. Email: christian.sandoee.musaeus@regionh.dk Funding information Kavli Charitable Trust Abstract Introduction: Large-scale brain networks are disrupted in the early stages of Alzheimer's disease (AD). Electroencephalography microstate analysis, a promising method for studying brain networks, parses EEG signals into topographies represent- ing discrete, sequential network activations. Prior studies indicate that patients with AD show a pattern of global microstate disorganization. Changes in the left temporal microstate are a sign of cognitive decline in patients with Alzheimer’s disease We investigated whether any specific microstate changes could be found in patients with AD and mild cogni- tive impairment (MCI) compared to healthy controls (HC). Materials and methods: Standard EEGs were obtained from 135 HC, 117 patients with MCI, and 117 patients with AD from six Nordic memory clinics. We parsed the data into four archetypal microstates. Results: There was significantly increased duration, occurrence, and coverage of microstate A in patients with AD and MCI compared to HC. When looking at micro- states in specific frequency bands, we found that microstate A was affected in delta (1–4 Hz), theta (4–8 Hz), and beta (13–30 Hz), while microstate D was affected only Brain and Behavior. 2020;00:e01630. \| 1 of 15 https://doi.org/10.1002/brb3.1630 wileyonlinelibrary.com/journal/brb3 This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2020 The Authors. Brain and Behavior published by Wiley Periodicals LLC. The peer review history for this article is available at https://publons.com/publon/10.1002/brb3.1630 Correspondence Christian S. Musaeus, Rigshospitalet, University of Copenhagen Blegdamsvej 9–section 6911, 2100 Copenhagen, Denmark. Email: christian.sandoee.musaeus@regionh.dk Funding information Kavli Charitable Trust Abstract Introduction: Large-scale brain networks are disrupted in the early stages of Alzheimer's disease (AD). Electroencephalography microstate analysis, a promising method for studying brain networks, parses EEG signals into topographies represent- ing discrete, sequential network activations. Prior studies indicate that patients with AD show a pattern of global microstate disorganization. We investigated whether any specific microstate changes could be found in patients with AD and mild cogni- tive impairment (MCI) compared to healthy controls (HC). Materials and methods: Standard EEGs were obtained from 135 HC, 117 patients with MCI, and 117 patients with AD from six Nordic memory clinics. We parsed the data into four archetypal microstates. Results: There was significantly increased duration, occurrence, and coverage of microstate A in patients with AD and MCI compared to HC. When looking at micro- states in specific frequency bands, we found that microstate A was affected in delta (1–4 Hz), theta (4–8 Hz), and beta (13–30 Hz), while microstate D was affected only Changes in the left temporal microstate are a sign of cognitive decline in patients with Alzheimer's disease Musaeus, Christian S; Engedal, Knut; Høgh, Peter; Jelic, Vesna; Khanna, Arjun R; Kjaer, Troels Wesenberg; Mørup, Morten; Naik, Mala; Oeksengaard, Anne-Rita; Santarnecchi, Emiliano Total number of authors: 14 Citation (APA): Musaeus, C. S., Engedal, K., Høgh, P., Jelic, V., Khanna, A. R., Kjaer, T. W., Mørup, M., Naik, M., Oeksengaard, A.-R., Santarnecchi, E., Snaedal, J., Wahlund, L.-O., Waldemar, G., & Andersen, B. B. (2020). Changes in the left temporal microstate are a sign of cognitive decline in patients with Alzheimer's disease. Brain and Behavior, 10(6), Article e01630. https://doi.org/10.1002/brb3.1630 this document breaches copyright please contact us providing details, and we will remove access to the work immediate ur claim. Received: 25 October 2019 \| Revised: 5 March 2020 \| Accepted: 20 March 2020 Received: 25 October 2019 \| Revised: 5 March 2020 \| Accepted: 20 March 2020 DOI: 10.1002/brb3.1630 O R I G I N A L R E S E A R C H Changes in the left temporal microstate are a sign of cognitive decline in patients with Alzheimer’s disease Christian S. Musaeus1 \| Knut Engedal2 \| Peter Høgh3,4 \| Vesna Jelic5,6 \| Arjun R. Khanna7 \| Troels Wesenberg Kjær4,8 \| Morten Mørup9 \| Mala Naik10 \| Anne-Rita Oeksengaard5 \| Emiliano Santarnecchi11 \| Jon Snaedal12 \| Lars-Olof Wahlund5 \| Gunhild Waldemar1 \| Birgitte B. Andersen1 1Department of Neurology, Danish Dementia Research Centre (DDRC), Rigshospitalet, University of Copenhagen, Copenhagen, Denmark 2Norwegian National Advisory Unit on Ageing and Health (Ageing and Health), Vestfold Hospital Trust and Oslo University Hospital, Ullevaal, Oslo, Norway 3Regional Dementia Research Center, Department of Neurology, Zealand University Hospital, Roskilde, Denmark 4Department of Clinical Medicine, University of Copenhagen, Copenhagen, Denmark 5Division of Clinical Geriatrics, Department of Neurobiology, Care Sciences and Society, Karolinska Institutet, Stockholm, Sweden 6Department of Geriatric Medicine, Memory Clinic, Karolinska University Hospital, Huddinge, Sweden 7Department of Neurosurgery, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA 8Neurophysiology Center, Zealand University Hospital, Roskilde, Denmark 9Section for Cognitive Systems, DTU Compute, Technical University of Denmark, Lyngby, Denmark 10Department of Geriatric Medicine, Haraldsplass Deaconess Hospital, Bergen, Norway 11Berenson-Allen Center for Non-invasive Brain Stimulation, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA 12Department of Geriatric Medicine, Landspítali University Hospital, Reykjavik, Iceland 1Department of Neurology, Danish Dementia Research Centre (DDRC), Rigshospitalet, University of Copenhagen, Copenhagen, Denmark 2Norwegian National Advisory Unit on Ageing and Health (Ageing and Health), Vestfold Hospital Trust and Oslo University Hospital, Ullevaal, Oslo, Norway 3Regional Dementia Research Center, Department of Neurology, Zealand University Hospital, Roskilde, Denmark 4Department of Clinical Medicine, University of Copenhagen, Copenhagen, Denmark 5Division of Clinical Geriatrics, Department of Neurobiology, Care Sciences and Society, Karolinska Institutet, Stockholm, Sweden 6Department of Geriatric Medicine, Memory Clinic, Karolinska University Hospital, Huddinge, Sweden 7Department of Neurosurgery, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA 8Neurophysiology Center, Zealand University Hospital, Roskilde, Denmark 9Section for Cognitive Systems, DTU Compute, Technical University of Denmark, Lyngby, Denmark 10Department of Geriatric Medicine, Haraldsplass Deaconess Hospital, Bergen, Norway 11Berenson-Allen Center for Non-invasive Brain Stimulation, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA 12Department of Geriatric Medicine, Landspítali University Hospital, Reykjavik, Iceland Correspondence Christian S. Musaeus, Rigshospitalet, University of Copenhagen Blegdamsvej 9–section 6911, 2100 Copenhagen, Denmark. Email: christian.sandoee.musaeus@regionh.dk Funding information Kavli Charitable Trust Brain and Behavior. 2020;00:e01630. 1 \| INTRODUCTION Kircher, 1998; Strik et al., 1997), although some studies report a lon- ger duration (Ihl et al., 1993; Musaeus, Nielsen, et al., 2019) and one study did not find any significant differences (Nishida et al., 2013). Although examined inconsistently, the microstate syntax may be temporally disorganized in patients with AD (Koenig, Studer, Studer, Hubl, Melie, & Strik, 2005; Nishida et al., 2013). The various and con- flicting findings in previous studies are likely due to methodological differences (Dierks et al., 1997; Ihl et al., 1993; Stevens & Kircher, 1998; Strik et al., 1997) and a low number of study participants. Furthermore, in patients with AD, the changes in spectral power have previously been shown to be most prominent in the lower fre- quencies (Musaeus, Engedal, et al., 2018). However, studies have to our knowledge not investigated the frequency-specific microstates in patients with AD. 1 Large-scale functional brain networks are altered in patients with Alzheimer's disease (AD) (Dickerson & Sperling, 2009), even in a very early stage of the disease (Cummings, 2004; Selkoe, 2002). Such alterations are considered crucial elements of the neuropatho- logical cascade characterizing AD (Palop & Mucke, 2016). Multiple methods to investigate such brain networks have been proposed, the most common being functional magnetic resonance imaging (fMRI) (Buckner et al., 2005; Greicius, Srivastava, Srivastava, Reiss, & Menon, 2004). However, cortical network dynamics occur at a timescale order of magnitude faster than the blood oxygen level-de- pendent signal used in fMRI. The fine temporal resolution of elec- troencephalography (EEG) indicates that this method may be better suited for studying the fine temporal dynamics of distributed corti- cal network activity. In the current study, we employed a standardized, validated ap- proach to EEG microstate analysis in a large prospectively recruited cohort to investigate changes in microstates in patients with AD and mild cognitive impairment (MCI) compared to healthy aging. Furthermore, we investigated the frequency-specific microstate changes in patients with AD. In addition, we sought to determine whether microstates could be used as a diagnostic tool for AD and MCI using the same type of classification method as previously de- scribed (Musaeus, Engedal, et al., 2018; Musaeus, Engedal, Hogh, et al., 2019). Lastly, we wanted to examine whether microstates cor- related with clinical scores or cerebrospinal fluid biomarkers. 1 \| INTRODUCTION One technique for studying distributed brain networks using EEG is microstate analysis, which involves dividing the EEG signal into a number of distinct states (Lehmann, Ozaki, Ozaki, & Pal, 1987) defined by spatial topographies of electric potentials recorded at scalp level. Such parcellation in functional states has been shown to be reliable over multiple recordings (Khanna, Pascual-Leone, Pascual-Leone, & Farzan, 2014) and correlated with cognitive abili- ties (Santarnecchi et al., 2017). It is suggested that spontaneous EEG microstates reflect the activity of resting-state networks (Van de Ville, Britz, & Michel, 2010; Yuan, Zotev, Zotev, Phillips, Drevets, & Bodurka, 2012), and alterations in the structure and temporal repre- sentation of microstates have been documented in other diseases, such as frontotemporal dementia (Nishida et al., 2013) and schizo- phrenia (Andreou et al., 2014; Lehmann et al., 2005), underpinning it as a potential novel classifier of disease in neurological and psychi- atric disorders. K E Y W O R D S Alzheimer's disease, EEG, microstate, mild cognitive impairment, network Abstract Abstract Introduction: Large-scale brain networks are disrupted in the early stages of Alzheimer's disease (AD). Electroencephalography microstate analysis, a promising method for studying brain networks, parses EEG signals into topographies represent- ing discrete, sequential network activations. Prior studies indicate that patients with AD show a pattern of global microstate disorganization. We investigated whether any specific microstate changes could be found in patients with AD and mild cogni- tive impairment (MCI) compared to healthy controls (HC). Materials and methods: Standard EEGs were obtained from 135 HC, 117 patients with MCI, and 117 patients with AD from six Nordic memory clinics. We parsed the data into four archetypal microstates. Results: There was significantly increased duration, occurrence, and coverage of microstate A in patients with AD and MCI compared to HC. When looking at micro- states in specific frequency bands, we found that microstate A was affected in delta (1–4 Hz), theta (4–8 Hz), and beta (13–30 Hz), while microstate D was affected only Brain and Behavior. 2020;00:e01630. \| 1 of 15 https://doi.org/10.1002/brb3.1630 wileyonlinelibrary.com/journal/brb3 This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2020 The Authors. Brain and Behavior published by Wiley Periodicals LLC. The peer review history for this article is available at https://publons.com/publon/10.1002/brb3.1630 \| 1 o wileyonlinelibrary.com/journal/brb3 2 of 15 MUSAEUS et al. in the delta and theta bands. Microstate features were able to separate HC from AD with an accuracy of 69.8% and HC from MCI with an accuracy of 58.7%. Conclusions: Further studies are needed to evaluate whether microstates represent a valuable disease classifier. Overall, patients with AD and MCI, as compared to HC, show specific microstate alterations, which are limited to specific frequency bands. These alterations suggest disruption of large-scale cortical networks in AD and MCI, which may be limited to specific frequency bands. neurological disorders with dementia other than AD, Parkinson's dis- ease dementia, and Lewy body dementia; (b) major psychiatric disor- ders; and (c) alcohol or drug abuse. The HC (n = 146) were recruited from among family members of the patients, through advertising, or were employees at the recruiting hospitals. All participants gave written informed consent to participate in the study. The study was approved by the ethic committees from Norway, Sweden, Iceland, and Denmark. addition, three HC were excluded due to use of either antidepres- sants or antipsychotics, since the underlying condition and/or sever- ity was unknown (see Figure 1). Table 1 presents a full description of the final sample. 2.1 \| Participants The cohort used in the current study was recruited as part of a vali- dation study (Engedal et al., 2015). Spectral power and connectivity analyses are presented elsewhere (Musaeus, Engedal, et al., 2018; Musaeus, Engedal, Hogh, et al., 2019). Participants were recruited from six Nordic memory clinics, and each clinic had to include a mini- mum of 60 patients and 20 HC who were elderly. In most clinics, all the included patients (n = 365) were recruited during their first as- sessment using the following three predefined exclusion criteria: (a) The studies investigating the alterations in microstates in pa- tients with AD (Dierks et al., 1997; Ihl, Dierks, Dierks, Froelich, Martin, & Maurer, 1993; Musaeus, Nielsen, Nielsen, & Hogh, 2019; Nishida et al., 2013; Stevens & Kircher, 1998; Strik et al., 1997) have generated varied and, to some extent, conflicting results. Most stud- ies of AD and healthy controls (HC) find a shorter average duration of microstates in patients with AD (Dierks et al., 1997; Stevens & MUSAEUS et al. 3 of 15 2.2 \| Clinical diagnostic assessment All patients underwent a clinical diagnostic assessment comprised of (a) history from the patient and an informant; (b) physical examina- tion focusing on neurological and cardiology status; (c) blood tests to screen for disorders that could be associated with cognitive impair- ment, such as B12 vitamin deficiency or low thyroid hormone levels; and (d) CT or MRI of the brain to evaluate white matter changes, general atrophy, and atrophy of the medial temporal lobes. Some of the patients underwent neuropsychological tests covering various cognitive domains, and some underwent a lumbar puncture to ex- amine amyloid beta-42, total tau, and phosphorylated tau protein in the cerebrospinal fluid, when indicated (see Table 1). Some patients The cohort was collected as part of a validation study where pa- tients with other types of dementias were recruited (Engedal et al., 2015). In the current study, the following groups were excluded from analysis due to the low number of patients: vascular dementia (n = 15), Lewy body dementia (n = 10), Parkinson's disease dementia (n = 5), frontotemporal dementia (n = 4), mixed dementias (n = 8), and mixed AD and vascular dementia (n = 5). In addition, patients with subjective cognitive decline (n = 64) were also excluded since no follow-up data on progression were available. Since some of the EEGs were of poor quality or lost, we had to exclude eight EEGs from HC, eight from patients with MCI, and 15 from patients with AD. 2.3 2.3 2.2 \| Clinical diagnostic assessment In FI G U R E 1 Low diagram of the number of included participants in the current study and of the excluded participants after preprocessing 4 of 15 \| HC MCI AD p-value (n = 135) (n = 117) (n = 117) AD versus MCI versus HC Mean age (SD), years 66.44 (7.64) 70.15 (8.13) 75.49 (7.65) <.001 Sex (female), % 60.7 53 60.7 .377 Education, years (SD) 13.89 (3.61) 11.57 (3.92) 10.07 (3.39) <.001 Antipsychotics 0 1 4 Antidepressants 0 25 18 Tranquilizers/ hypnotics 3 12 6 Antidementia drugs 0 2 18 Painkillers 4 6 4 Lumbar punctures performed 38 32 Amyloid beta 42, mean (SD) 876.16 (354.92) 516.34 (114.52) <.001 Total tau, mean (SD) 420.98 (230.51) 532.31 (273.85) .069 Phosphorylated tau, mean (SD) 63.16 (26.72) 106.33 (114.52) .027 MMSE, mean (SD) 28.91 (1.34) 27.11 (2.16) 23.52 (3.79) <.001 Word list memory, mean (SD) 20.49 (3.99) 15.25 (4.29) 11.06 (3.83) <.001 Word list recall, mean(SD) 7.39 (1.69) 3.66 (2.34) 1.52 (1.64) <.001 Word list recognition, mean (SD) 19.47 (1.44) 17.92 (2.01) 15.84 (2.62) <.001 Number of one- second epochs 113.93 (55.49) 119.64 (48.23) 130.82 (53.14) .038 Abbreviations: AD, Alzheimer's disease; HC, healthy controls; MCI, mild cognitive impairment; MMSE, mini-mental state examination; SD, standard deviation. p-values show the differences when comparing patients with AD, patients with MCI, and HC. 4 of 15 5 \| MUSAEUS et al. TA B LE 1 Characteristics of study participants Abbreviations: AD, Alzheimer's disease; HC, healthy controls; MCI, mild cognitive impairment; MMSE, mini-mental state examination; SD, standard deviation. p-values show the differences when comparing patients with AD, patients with MCI, and HC. Abbreviations: AD, Alzheimer's disease; HC, healthy controls; MCI, mild cognitive impairment; MMSE, mini-mental state examination; SD, standard deviation. p-values show the differences when comparing patients with AD, patients with MCI, and HC. 2.4 \| EEG recording The EEGs were recorded using a 19-channel NicoletOne EEG System (Natus®). The sampling rate was different between the six sites rang- ing from 250 to 1,000 Hz. Electrodes were placed according to the international 10–20 system of electrode placement (Fp1, Fp2, F7, F3, Fz, F4, F8, T3, T5, T4, T6, C3, Cz, C4, P3, Pz, P4, O1, O2). Data were recorded using the average reference. Two bipolar electro-oculog- raphy channels and one electrocardiogram channel were recorded to monitor artifacts. The EEGs were recorded, alternating between 3-min periods each of eyes closed and eyes open, except for one center, where only continuous eyes-closed segments were recorded. The participants were alerted if they became visibly drowsy. \| Cognitive tests (A) to (D) assigned according to previous literature 1979). Each of the ten items was assessed as present or not, yielding a minimum score of 0 and a maximum score of 10. noise or artifacts. Channels with excessive noise, drift, or reduced connection were rejected and interpolated using spherical interpo- lation. Recordings with four or more electrodes with excessive ar- tifacts were excluded from the analysis. Afterward, the data were rereferenced to the average reference and independent component analysis was performed using the extended infomax algorithm (Lee, Girolami, Girolami, & Sejnowski, 1999), extracting up to nineteen components based on the data rank. This was done for each file, and components containing eye blinks, eye movement, ECG artifacts, or specific line-noise artifacts were removed. Finally, the EEGs were visually inspected again, and epochs with excessive noise or artifacts were removed. The additional step of epoch removal was performed to assure that any artifacts, which could not be removed using inde- pendent component analysis, were removed. The investigator was blinded to diagnosis. The controls were tested with the MMSE, clock-drawing test, and the CERAD ten-word tests (all three parts), while depression was rated with the modified Montgomery Åsberg Depression Rating Scale in the same way as the patients. \| Cognitive tests were assessed with fluorodeoxyglucose PET or Tc-HMPAO SPECT. A previously published paper contains the details of the clinical as- sessments (Braekhus, Ulstein, Ulstein, Wyller, & Engedal, 2011) out- lined above. For each patient, we conducted the MMSE (Engedal, Haugen, Haugen, Gilje, & Laake, 1988; Folstein, Folstein, Folstein, & McHugh, 1975), which is a measure of overall cognitive function consisting of 20 items with a minimum score of 0 and a maximum score of 30, with lower scores indicating more severe cognitive impairment; the clock-drawing test (Shulman, 2000), which is a short screening test of cognition and visuospatial function with a minimum score of 0 and a maximum score of 5, with lower scores indicating poorer func- tion, and the CERAD ten-word list) (Morris, Mohs, Mohs, Rogers, Fillenbaum, & Heyman, 1988), which tests the ability to learn ten words with three repetitions (CERAD learning max. score = 30), to re- call the ten words after ten minutes (CERAD recall max. score = 10), and to recognize the ten words among another ten different words (CERAD recognition max. score = 20). We used a modified version of the Montgomery Åsberg Depression Rating Scale to assess whether the patients had any depressive symptoms (Montgomery & Asberg, The clinical diagnoses, which applied the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision, and the McKhann criteria, were made at consensus confer- ences at each memory clinic, or by at least two experienced physi- cians (McKhann et al., 1984). Winblad criteria were used to diagnose MCI (Winblad et al., 2004). All diagnoses were made blinded to the EEG results. All HC were interviewed, and previous and present disorders and drug use were recorded. Any individual with a cognitive test score one standard deviation below the mean for age on the Mini-Mental State Examination (MMSE), the clock-drawing test (CDT), or the Consortium to Establish a Registry for Alzheimer's Disease (CERAD) score was excluded (see below). \| 5 of 15 MUSAEUS et al. FI G U R E 2 Global maps calculated based on the aggregated dataset from all participants and back-fitted to each EEG recording. (A) to (D) assigned according to previous literature \| 5 of 15 MUSAEUS et al. \| 5 of 15 FI G U R E 2 Global maps calculated based on the aggregated dataset from all participants and back-fitted to each EEG recording. 2.6 \| Microstate analysis Before performing the microstate analysis using the Microstate EEGlab Toolbox, we lowpass-filtered the data at 40 Hz with the same settings as mentioned above (Poulsen, Pedroni, Langer, & Hansen, 2018). For each subject, we began by extracting the first 1,000 electric field montages at global field power (GFP) peaks with a minimum peak distance of 10 ms. GFP peaks that exceeded two times the standard deviation of the GFPs of all maps were excluded. To identify topographic clusters within these data, we submitted all n × 1,000 electric potential topographies to a modified K-means clustering algorithm (Pascual-Marqui, Michel, Michel, & Lehmann, 1995). Polarity of the EEG topography was ignored (Lehmann, 1971; Pascual-Marqui et al., 1995; Wackermann, Lehmann, Lehmann, Michel, & Strik, 1993). We chose to predefine the sought number of microstates as four to remain consistent with the majority of prior studies of EEG microstates (Khanna et al., 2014) and because four have been confirmed to generate reproducible maps (Khanna, Pascual-Leone, Pascual-Leone, Michel, & Farzan, 2015). The number of repetitions was set at 50, while the maximum number of itera- tions was set at 1,000. To test whether the algorithm was affected by local minima issues, we repeated the entire analysis three times and found that the main findings can be robustly recovered (see Supplementary material, pp. 3–7; Figure S2; and Tables S1–S3). A set the delta band, we low-pass filtered the data below 4 Hz, w dpass filtering was performed for the other frequency b ng the same filter settings as mentioned above. For the su ent microstate analysis, we performed the same analysis as m ned above. Finally, we noted that the average age of subjects in the HC t was less than that of the AD and MCI groups. To explore w FI G U R E 3 General schema for EEG microstate analysis. (a) The global field power (GFP) of the multichannel EEG signal is calculated at each time point. Local maxima of the GFP curve represent points of greatest topographic signal-to-noise ratio. These points are sampled and re-expressed as a topography of electric potentials at each of the n electrodes in the electrode array. (b) These topographies are subjected to a modified k-means clustering analysis. Each topography is plotted in n-dimensional space, and the modified k-means algorithm is applied to partition the resulting n-dimensional space into optimal clusters. An arbitrary label, A, B, C, and D, is assigned to each cluster. (c) Each topography at local maxima of the GFP curve is assigned a label, A, B, C, or D, according to the cluster it belongs to. The multichannel EEG signal is re-expressed as a sequence of alternating labels. From this data re-representation, values such as the average microstate class duration, frequency of unique occurrences of microstates, and microstate transition probabilities can be calculated of four global maps was generated (see Figure 2) and back-fitted to the whole EEG. To reduce noise, we rejected microstate segments shorter than 30 ms, which is the default in the toolbox (Poulsen et al., 2018). This was done to assure that the short segments, which may have been due to noise, did not affect the results. After back-fitting the global maps, we calculated global explained variance (GEV), du- ration, occurrence, coverage, and the syntax for EEG files. Here, the duration is defined as the average time for each map to be present before transitioning to another map. Occurrence is defined as the average number of times a microstate occurred during the entire EEG. Coverage is defined as the total percent of the EEG for which a microstate was accounted for. GEV is defined as the variance of EEG activity explained by all four microstates. 2.5 \| Preprocessing EEG The data were imported to MATLAB (Mathworks, v2016a) using the EEGLAB toolbox (Delorme & Makeig, 2004). The eyes-closed segments were selected from the three-minute, eyes-open, and eyes-closed periods. If the EEG only contained eyes closed, then the segments were selected from the first 10 min of the record- ing to prevent inclusion of segments with drowsiness or sleep. The electrodes were computationally located on the scalp by means of the DIPFIT toolbox (Oostenveld, Fries, Fries, Maris, & Schoffelen, 2011). The auxiliary channels (electrocardiogram and reference electrodes) were removed, and the data were bandpass-filtered from 1–70 Hz and subsequently band-stop filtered from 45–55 Hz using the pop_firws function in EEGLab with a filter order of two. The Kaiser window parameter beta was estimated using a maxi- mum passband ripple of .001. The data were then down-sampled to 200 Hz. Next, the data were divided into one-second epochs that were visually inspected to remove epochs containing excessive MUSAEUS et al. 6 of 15 2.8 \| Prediction The dataset, which consisted of the z-scores of duration, occurrence, and coverage for each microstate, as well as analysis of the micro- state syntax (i.e., frequency of transitions among microstates), was compressed using principal component analysis such that 99% of the data variance was retained for the subsequent classification analysis. For the classification, we used multinomial regression for the three- class classification of AD, MCI, and HC, implemented using the min- Func optimization procedure (Schmidt, 2005). Afterward, we used logistic regression for the two-class classification, also implemented using the MATLAB code for logistic regression, which included HC and AD. We quantified the model using leave-one-out cross-valida- tion. The classification accuracies are reported averaged over the number of observations (i.e., subjects) left out one at a time in the leave-one-out cross-validation procedure. This procedure was also performed using the results from the main analysis (1–40 Hz) and the results from the frequency-specific microstates (24 values × 5 frequency bands). Furthermore, we used the same method using multinomial regression for the three-class classification and logis- tic regression for the two-class classification when removing the youngest 18 HC. FI G U R E 4 Global maps from the frequency-specific microstates for delta (1–4 Hz), theta (4–8 Hz), alpha (8–13 Hz), and beta (13–30 Hz). The maps were calculated based on the aggregated dataset from all participants and back-fitted to each EEG recording. (A) to (D) assigned according to previous literature effect this age difference may have had on microstate results, we conducted an additional analysis removing the 18 youngest HC sub- jects (thereby including 117 participants in each group) and repeated the microstate analysis as described above. MUSAEUS et al. FI G U R E 4 Global maps from the frequency-specific microstates for delta (1–4 Hz), theta (4–8 Hz), alpha (8–13 Hz), and beta (13–30 Hz). The maps were calculated based on the aggregated dataset from all participants and back-fitted to each EEG recording. (A) to (D) assigned according to previous literature Current medications were included as binary values for whether the person received antipsychotics, antidepressants, hypnotics, antide- mentia drugs, or painkillers. To correct for multiple comparisons, we performed false discovery rate (FDR) correction for 24 comparisons. Afterward, we computed post hoc tests using the same covariates as described above on the log-transformed data if the group com- parison was significant after FDR correction using a general linear model. The post hoc tests were considered significant at a p < .05. The same analyses were performed for the frequency-specific mi- crostates, and the correction for multiple comparisons was done for each frequency band separately. For correlations, we chose to use the recall score and the learn- ing score from the CERAD ten-word list since neuropsychological studies have demonstrated that learning and recall best discrimi- nate between HC and AD (Collie & Maruff, 2000; Twamley, Ropacki, Ropacki, & Bondi, 2006), with the recall score being the most sen- sitive in the early phases of AD (Welsh, Butters, Butters, Hughes, Mohs, & Heyman, 1991). We performed partial correlation using the same covariates as described above between duration, coverage, and occurrence (see definition under Microstate analysis) of micro- state A and the recall and learning scores from the CERAD ten-word list, including MMSE, amyloid, total tau, and phosphorylated tau. Afterward, we performed FDR correction for all p-values. If the ad- justed p < .05, it was considered significant. FI G U R E 4 Global maps from the frequency-specific microstates for delta (1–4 Hz), theta (4–8 Hz), alpha (8–13 Hz), and beta (13–30 Hz). The maps were calculated based on the aggregated dataset from all participants and back-fitted to each EEG recording. (A) to (D) assigned according to previous literature See Figure 3 for an overview of the microstate analysis pipeline. However, if transitions from one state to the next occurred ran- domly, observed transition values would be proportional to the rela- tive occurrence of the microstate classes. To test this, we performed syntax analyses based on the same analysis as previously described in detail (Lehmann et al., 2005; Nishida et al., 2013). We calculated the observed transitions based on all transitions, and then, the ex- pected transitions based on the occurrence of the microstates for each subject separately. Afterward, these values were averaged across subjects for each group, and the difference was assessed using the chi-square distance. To statistically test the difference, we performed a permutation test with 5,000 repetitions and randomly assigned the labels “expected” and “observed” to the subjects’ sets of twelve transition probabilities, after which the chi-square dis- tance was computed. The majority of studies have used four microstates (Michel & Koenig, 2018). This type of parcellation of the dataset leaves a large part of the GEV unexplained, which is why we repeated the analyses described above for three, five, and six microstates (see Supplementary material, pp. 8–14). For the delta band, we low-pass filtered the data below 4 Hz, while bandpass filtering was performed for the other frequency bands using the same filter settings as mentioned above. For the subse- quent microstate analysis, we performed the same analysis as men- tioned above. To investigate whether the microstate changes were located to specific frequency bands, we performed microstate calculations for the following frequency bands: delta (1–4 Hz), theta (4–8 Hz), alpha (8–13 Hz), and beta (13–30 Hz), see Figure 4 for global maps. Finally, we noted that the average age of subjects in the HC co- hort was less than that of the AD and MCI groups. To explore what MUSAEUS et al. 7 of 15 3.5 \| Prediction For prediction analysis, we first used multinomial regression be- tween AD, MCI, and HC and found an accuracy of 40.4% (sensitivi- tyMCI = 22.2%, sensitivityAD = 38.5%, specificity = 57.8%). Since it was not possible for the classifier to distinguish well between MCI and the two other groups, we then compared HC and AD, where we found an accuracy of 62.7% (sensitivity = 57.3%, specificity = 67.4%) and HC and MCI, where we found an accuracy of 55.2% (sensitiv- ity = 44.4%, specificity = 64.4%). For the syntax analysis, we found a significant difference be- tween the transition rate between AD, MCI, and HC from microstate For the syntax analysis, we found a significant difference be- tween the transition rate between AD, MCI, and HC from microstate D to C (p-valueuncorr = .006, p-valuecorr = .024, F = 5.244, df = 357), D to A (p-valueuncorr = .020, p-valuecorr = .002, F = 6.117, df = 357), C to A (p-valueuncorr = .005, p-valuecorr = .024, F = 5.419, df = 357), and B to A (p-valueuncorr = .012, p-valuecorr = .040, F = 4.500, df = 357), see Figure 5. However, no significant differences in the expected and observed transition probabilities were found for any of the groups (p > .05). When applying linear discriminant analysis, we found almost the same accuracy between the three groups with an accuracy of 43.6% (sensitivityMCI = 32.5%, sensitivityAD = 35.9%, specificity = 60.0%). The same pattern was found using quadratic discriminant analysis for the three-class classification with an accuracy of 39.8% (sensi- tivityMCI = 12.8%, sensitivityAD = 49.6%, specificity = 54.8%) and for support vector machine we found an accuracy of 41.2% (sensitivi- tyMCI = 18.8%, sensitivityAD = 29.1%, specificity = 71.1%). The microstate results for the comparisons between AD, MCI, and HC after removing the youngest 18 HC can be found in Supplementary material, pp. 15–16 (Figure S6 for global maps). We also performed the multinomial regression using the results from the main analysis (1–40 Hz) and the results from the frequen- cy-specific microstates together. Here, we found an accuracy of 42.0% (sensitivityMCI = 27.4%, sensitivityAD = 37.6%, specificity = 58.5%) between all three groups. coverage of microstate A and decreased duration and coverage of microstate D in the delta band (p < .05). In the theta band, we found significantly increased duration of microstate A/B in patients and significantly decreased occurrence and coverage in microstate D (p < .05). Lastly, we found significantly decreased duration of micro- state B and decreased occurrence and coverage of microstate A in the beta band. No significant differences were found in the alpha band. See Figure 4 and Table 3. When examining transitions, we found that B is significantly more likely to transition to A in AD as compared with both MCI and HC (p = .043, F = 6.233). No significant differences in the expected and observed transition probabilities were found for any of the groups (p > .05). was implemented using the default settings in MATLAB specified to use hyperparameter optimization of all parameters as defined in the code. 3.1 \| Demographics, cognitive tests, and AD biomarkers Table 1 provides a full overview of demographics, cognitive test scores, AD biomarkers, and number of 1-s epochs for the three groups: AD, MCI, and HC. We found significant difference in both age and education for patients with AD and MCI as compared with HC (p < .05). 3.4 \| Correlation When performing partial correlation analysis adjusting for covari- ates, we did not find any significant correlation between the features of microstate A and the recall and learning scores from the CERAD ten-word list, MMSE, amyloid, total tau, or phosphorylated tau when correcting for multiple comparisons using FDR. However, the largest rho obtained was between coverage of microstate A and recall from the CERAD (p = .018, ρ = −0.129). See Figure S1 for scatterplot of the correlation between coverage of microstate A and the recall score from the CERAD. 3.2 \| Microstate results There was no significant difference in the total GEV between the three diagnostic groups (p = .970, F = 0.031, df = 366), with an aver- age GEV across groups of 49.51%. We found a significant difference between AD, MCI, and HC for the duration of microstate A, with post hoc test using a gen- eral linear model showing a significant difference between AD and HC (p = .002, t-value = 9.748, df = 250), and between MCI and HC (p = .004, t-value = 8.633, df = 253). Furthermore, sig- nificant differences were found between AD, MCI, and HC for coverage of microstate A, with a significant difference between AD and HC (p < .001, t-value = 15.111, df = 250), and between MCI and HC (p = .003, t-value = 8.985, df = 250). We also found a significant difference in occurrence for microstate A, and the post hoc t tests showed significant difference between AD and HC (p < .001, t = 12.364, df = 250), and between MCI and HC (p = .012, t = 6.466, df = 250), see Table 2. 3.5 \| Prediction When using logistic regression, we found an accuracy of 69.8% (sensitivity = 68.4%, specificity = 71.1%) be- tween HC and AD, and between HC and MCI, we found an accuracy of 58.7% (sensitivity = 52.1%, specificity = 64.4%). 2.7 \| Statistics All statistics were performed in MATLAB (vR2016a). To compare sex, we performed a chi-square test. For number of one-second ep- ochs, age, years of education, the MMSE, and the ten-word list of the CERAD score and subscores (learning, recognition and recall), we performed a one-way ANOVA for the three groups. In addition, we performed linear discriminant analysis, quadratic discriminant analysis, and support vector machine for the three- class classifications. Here, we did not perform principal component analysis when using linear discriminant analysis or support vector machine. The support vector machine for the three-class problem When comparing the microstate values, we log-transformed the data first due to the non-normal distribution. For comparing AD, MCI, and HC, we performed an ANCOVA (Gruner, 2020) using age, sex, years of education, and current medications as covariates. MUSAEUS et al. 8 of 15 3.3 \| Frequency-specific microstates When investigating the frequency-specific microstates, we found that both AD, and MCI showed significant increased duration and \| 9 of 15 USAEUS et al. For the analysis of the groups when removing the 18 youngest C, we found an accuracy of 36.5% (sensitivityMCI = 22.2%, sen- tivityAD = 36.8%, specificity = 50.4%). When investigating two asses, we found an accuracy of 62.8% (sensitivity = 66.7%, speci- (4–8 Hz), and beta (13–30 Hz), while microstate D was affected only in the delta and theta bands. Specific syntax alterations (the fre- quency of transitions from microstate D, C, and B to microstate A) in both patients with AD and MCI compared to HC were found, but G U R E 5 Significant results for the yntax analysis between healthy controls C), mild cognitive impairment (MCI), nd Alzheimer's disease (AD). The first olumn is for HC, the second for MCI, and e third for AD. The values represent the ercentage of times when for example icrostates D transitioned to the other icrostates. The figure shows that icrostate B, C, and D were more likely to ansition to microstate A in patients with D and in patients with MCI MUSAEUS et al. \| 9 of 15 FI G U R E 5 Significant results for the syntax analysis between healthy controls (HC), mild cognitive impairment (MCI), and Alzheimer's disease (AD). The first column is for HC, the second for MCI, and the third for AD. The values represent the percentage of times when for example microstates D transitioned to the other microstates. The figure shows that microstate B, C, and D were more likely to transition to microstate A in patients with AD and in patients with MCI For the analysis of the groups when removing the 18 youngest HC, we found an accuracy of 36.5% (sensitivityMCI = 22.2%, sen- sitivityAD = 36.8%, specificity = 50.4%). When investigating two classes, we found an accuracy of 62.8% (sensitivity = 66.7%, speci- ficity = 59.0%) between AD and HC and an accuracy of 56.0% (sensi- tivity = 53.8%, specificity = 58.1%) between MCI and HC. (4–8 Hz), and beta (13–30 Hz), while microstate D was affected only in the delta and theta bands. 3.3 \| Frequency-specific microstates Specific syntax alterations (the fre- quency of transitions from microstate D, C, and B to microstate A) in both patients with AD and MCI compared to HC were found, but no significant differences in expected or observed transition prob- abilities were found. Furthermore, we found a diagnostic accuracy between HC and AD of 69.8%. Together, we find that microstate features show poor diagnostic accuracy in patients with AD, but we find significant changes in microstate A features, which previously has been associated with temporal lobe function. FI G U R E 5 Significant results for the syntax analysis between healthy controls (HC), mild cognitive impairment (MCI), and Alzheimer's disease (AD). The first column is for HC, the second for MCI, and the third for AD. The values represent the percentage of times when for example microstates D transitioned to the other microstates. The figure shows that microstate B, C, and D were more likely to transition to microstate A in patients with AD and in patients with MCI 4 \| DISCUSSION In the current study, we found significantly increased duration, oc- currence, and coverage of microstate A in patients with AD and MCI compared to HC. When examining frequency-specific microstates, we found that microstate A was affected in delta (1–4 Hz), theta The majority of prior studies found shorter durations of all mi- crostates in patients suffering from AD (Dierks et al., 1997; Stevens & Kircher, 1998; Strik et al., 1997), which has been suggested to in- dicate a temporal disorganization of global cortical networks in AD 10 of 15 MUSAEUS et al. \| E 2 Microstate features, including duration, occurrence, and coverage for microstates A-D, and the FDR-adjusted p-value and F-value for the comparison between AD, MCI, and HC. es of freedom for all comparisons were 357 Duration Occurrence Coverage HC MCI AD p-value F-value HC MCI AD p-value F-value HC MCI AD p-value F-value state A, (SD) 75.32 (9.59) 80.30 (11.31) 81.47 (11.14) .024 5.19 2.49 (0.69) 2.77 (0.70) 2.87 (0.61) .020 6.26 19.01 (6.48) 22.64 (7.65) 23.62 (6.68) .017 7.37 state B, (SD) 79.15 (10.81) 82.02 (12.31) 82.53 (12.23) .329 1.33 2.74 (0.63) 2.84 (0.61) 2.95 (0.55) .307 1.48 21.88 (6.37) 23.53 (6.73) 24.61 (6.98) .307 1.49 state C, (SD) 85.67 (17.39) 83.58 (23.32) 81.68 (17.95) .329 1.30 3.00 (0.67) 2.79 (0.67) 2.85 (0.68) .163 2.59 26.17 (8.99) 23.81 (10.10) 23.83 (9.19) .255 1.99 state D, (SD) 101.88 (32.55) 95.93 (26.18) 90.83 (23.92) .084 3.50 3.22 (0.51) 3.10 (0.49) 3.02 (0.55) .513 0.71 32.94 (11.29) 30.02 (9.84) 27.95 (10.10) .145 2.83 iations: AD, Alzheimer's disease; FDR, false discovery rage; HC, healthy controls; MCI, mild cognitive impairment; SD, standard deviation. p-values show the differences when comparing patients D ti t ith MCI d HC (Koenig et al., 2005; Nishida et al., 2013). However, these early stud- ies used adaptive segmentation and did not group the microstates into specific classes (Dierks et al., 1997; Ihl et al., 1993; Stevens & Kircher, 1998; Strik et al., 1997), which may account for some of the differences between the present and earlier findings. A more recent study using cluster analysis (Nishida et al., 2013) separating the mi- crostates into four clusters did not find any significant differences between patients with AD and HC but may have been underpow- ered. 4 \| DISCUSSION Another study using part of the same sample as described here found that microstate A was most affected in patients with AD and MCI compared to HC (Musaeus, Nielsen, et al., 2019). In the present study, overall microstate duration, occurrence, and coverage were either increased (microstates A and B) or decreased (microstates C and D) in AD and MCI compared to HC, suggesting aberrancy in the temporal dynamics of large-scale cortical networks in patients with AD and MCI. When comparisons were made by individual mi- crostate classes, only microstate A was significantly different (see Table 2). Furthermore, we also examined the microstates in specific frequency bands, which has to our knowledge not been done before in patients with AD. Here, we found that microstate A was specifi- cally altered in delta (1–4 Hz), theta (4–8 Hz), and beta (13–30 Hz), while microstate D was affected only in the delta and theta bands and microstate B showed decreased duration in the beta band. To understand the connection between EEG microstates and the spa- tial changes, studies have used source localization and found that the main sources of microstate A were localized in the left temporal lobe (Brechet et al., 2019; Custo et al., 2017). Furthermore, studies have also explored the association between microstates and both the blood oxygen level-dependent signal and resting-state networks measured with fMRI (Britz, Van De Ville, & Michel, 2010; Musso, Brinkmeyer, Brinkmeyer, Mobascher, Warbrick, & Winterer, 2010; Yuan et al., 2012). One study associated microstate A with activa- tions in the superior and middle temporal gyri (Britz et al., 2010). In patients with AD, other studies have found that the temporal areas, and especially the hippocampus, showed decreased activity during encoding of new information (Golby et al., 2005; Kato, Knopman, Knopman, & Liu, 2001; Machulda et al., 2003). This supports that microstate A is associated with areas of the brain that has been shown to be affected in patients with AD. This difference in affected side has previously been reported in MR studies, which showed that atrophy was more pronounced on the left side (Baron et al., 2001; Killiany et al., 2000) in patients with AD. Furthermore, the studies using source localization showed that microstate A corresponds to the left temporal lobe (Brechet et al., 2019; Custo et al., 2017). 4 \| DISCUSSION The reason for only finding significant differences in the left side may be due unintentional selection bias toward patients referred with lan- guage affection or evidence that early changes in perfusion as mea- sured with SPECT in AD are more prominent on the left side (Hogh, Madsen Sjo Madsen Sjo Gade & Waldemar 2004) Alternatively TA B LE 2 Microstate features, including duration, occurrence, and coverage for microstates A-D, and the FDR-adjusted p-value and F-value for the comparison between AD, MCI, and HC Degrees of freedom for all comparisons were 357 Duration Occurrence Coverage HC MCI AD p-value F-value HC MCI AD p-value F-value HC MCI AD p-value F-value Microstate A, (SD) 75.32 (9.59) 80.30 (11.31) 81.47 (11.14) .024 5.19 2.49 (0.69) 2.77 (0.70) 2.87 (0.61) .020 6.26 19.01 (6.48) 22.64 (7.65) 23.62 (6.68) .017 7.37 Microstate B, (SD) 79.15 (10.81) 82.02 (12.31) 82.53 (12.23) .329 1.33 2.74 (0.63) 2.84 (0.61) 2.95 (0.55) .307 1.48 21.88 (6.37) 23.53 (6.73) 24.61 (6.98) .307 1.49 Microstate C, (SD) 85.67 (17.39) 83.58 (23.32) 81.68 (17.95) .329 1.30 3.00 (0.67) 2.79 (0.67) 2.85 (0.68) .163 2.59 26.17 (8.99) 23.81 (10.10) 23.83 (9.19) .255 1.99 Microstate D, (SD) 101.88 (32.55) 95.93 (26.18) 90.83 (23.92) .084 3.50 3.22 (0.51) 3.10 (0.49) 3.02 (0.55) .513 0.71 32.94 (11.29) 30.02 (9.84) 27.95 (10.10) .145 2.83 Abbreviations: AD, Alzheimer's disease; FDR, false discovery rage; HC, healthy controls; MCI, mild cognitive impairment; SD, standard deviation. p-values show the differences when comparing patients with AD patients with MCI and HC MUSAEUS et al. 4 \| DISCUSSION 11 of 15 E 3 Microstate features, including duration, occurrence, and coverage for microstates A-D, and the FDR-adjusted p-value and F-value for the comparison between AD, MCI, and HC investigating frequency-specific microstates Duration Occurrence Coverage HC MCI AD p-value F-value HC MCI AD p-value F-value HC MCI AD p-value F-value (1–4 Hz) crostate , (SD) 93.55 (13.05) 102.43 (15.72) 103.03 (16.31) .041* 5.71 2.19 (0.43) 2.43 (0.43) 2.44 (0.38) .072 4.25 20.99 (6.51) 25.45 (7.47) 25.61 (7.46) .041* 5.35 crostate (SD) 100.02 (10.18) 98.79 (9.82) 98.12 (10.41) .916 0.20 2.44 (0.27) 2.41 (0.28) 2.39 (0.31) .857 0.45 24.57 (4.61) 24.01 (4.50) 23.62 (4.94) .857 0.39 crostate (SD) 107.54 (11.91) 105.31 (11.54) 105.54 (12.99) .857 0.34 2.59 (0.33) 2.53 (0.35) 2.50 (0.35) .857 0.41 28.13 (6.15) 26.95 (6.29) 26.68 (6.40) .919 0.13 crostate , (SD) 101.94 (9.79) 96.19 (10.26) 97.70 (12.49) .030* 6.80 2.56 (0.30) 2.43 (0.35) 2.43 (0.39) .134 2.91 26.30 (5.00) 23.59 (5.37) 24.09 (6.49) .046* 4.95 a (4–8 Hz) crostate , (SD) 111.92 (17.37) 120.41 (21.49) 120.25 (24.57) .054 4.30 2.16 (0.34) 2.11 (0.37) 2.11 (0.38) .611 0.61 24.31 (5.46) 25.43 (6.33) 25.56 (7.12) .588 0.94 crostate /B, (SD) 102.53 (18.25) 115.18 (22.04) 120.67 (32.29) .005* 8.65 1.92 (0.48) 1.95 (0.46) 2.04 (0.48) .397 1.40 20.00 (7.14) 22.77 (7.95) 25.08 (9.34) .054 4.21 crostate (SD) 102.74 (13.93) 104.97 (20.34) 103.50 (21.24) .611 0.58 2.30 (0.50) 2.09 (0.55) 2.07 (0.51) .203 2.22 23.84 (6.82) 22.41 (8.29) 21.73 (7.60) .611 0.59 crostate , (SD) 132.68 (33.64) 135.05 (49.43) 127.28 (30.99) .099 3.24 2.39 (0.32) 2.18 (0.36) 2.14 (0.37) .006* 7.75 31.86 (9.15) 29.39 (9.67) 27.64 (9.40) .028* 5.76 a (8–13 Hz) crostate , (SD) 157.38 (30.13) 169.29 (34.85) 169.74 (35.01) .255 2.78 1.23 (0.42) 1.39 (0.39) 1.44 (0.42) .181 3.60 20.00 (9.03) 24.14 (9.74) 25.24 (10.17) .181 3.94 crostate (SD) 176.28 (36.90) 170.85 (35.34) 169.36 (44.32) .793 0.46 1.39 (0.33) 1.44 (0.31) 1.45 (0.31) .793 0.37 24.66 (7.74) 24.92 (7.94) 24.89 (8.96) .996 0.03 crostate (SD) 196.86 (95.11) 184.15 (88.59) 177.39 (91.27) .475 1.53 1.29 (0.31) 1.26 (0.30) 1.33 (0.33) .340 2.07 25.57 (12.74) 23.67 (12.69) 24.18 (12.98) .571 0.97 crostate , (SD) 210.02 (85.30) 190.88 (69.84) 184.09 (92.05) .504 1.32 1.42 (0.28) 1.42 (0.26) 1.38 (0.31) .996 0.00 29.77 (12.24) 27.28 (10.68) 25.69 (12.31) .637 0.80 (13–30 Hz) Hz) 91.25 (10.51) 89.19 (10.80) 86.77 (9.55) .089 3.55 3.20 (0.32) 3.19 (0.34) 3.06 (0.41) .043* 5.02 29.35 (5.35) 28.62 (5.74) 26.83 (6.00) .043* 5.08 91.42 (9.49) 87.21 (8.57) 87.38 (9.53) .043* 5.28 3.18 (0.38) 3.13 (0.34) 3.11 (0.37) .584 0.63 29.31 (5.75) 27.50 (4.95) 27.40 (5.85) .154 2.62 78.27 (8.74) 79.10 (12.73) 80.17 (11.17) .348 1.24 2.44 (0.53) 2.49 (0.59) 2.58 (0.59) .174 2.37 19.42 (5.82) 20.25 (8.16) 21.22 (7.63) .189 2.15 81.23 (10.26) 82.38 (9.67) 83.75 (10.37) .556 0.72 2.65 (0.50) 2.82 (0.53) 2.88 (0.53) .125 3.09 21.93 (6.66) 23.63 (6.72) 24.55 (7.05) .183 2.25 AD, Alzheimer's disease; FDR, false discovery rage; HC, healthy controls; MCI, mild cognitive impairment; SD, standard deviation. Beta (13–30 Hz) 4.1 \| Limitations of the study and future directions The current study has some limitations. Cerebrospinal fluid markers were available for only 30% of the included patients, which limited the statistical power of the correlations between microstate fea- tures and cerebrospinal fluid markers. The lack of follow-up data in the MCI group prevented us from investigating which patients with MCI converted to AD, but we hypothesize that most of them would ultimately develop AD, as supported the substudy mentioned above where 50% of the MCI progressed to AD (Musaeus, Nielsen, et al., 2018). Demographically, we found that the HC were younger than the AD and MCI patients, which impacts microstate features (Koenig et al., 2002). When removing the 18 youngest participants in the HC and thereby having a total of 117 participants in all three groups, we found similar results. A large proportion of AD and MCI patients were treated with medications (34.62%) that may have impacted the EEG, as seen in patients with schizophrenia (Merrin, Meek, Meek, Floyd, & Callaway, 1990). Furthermore, the HC group had a higher level of education than both clinical groups, which may also have played in assessing memory function. However, we tried to correct for these confounders by including age, use of drugs, and educational level as covariates when performing ANCOVA. Future studies investigating the changes in EEG microstates in patients with AD should strive to do longer EEG recordings in an effort to make it possible to extract individual maps for each participant. We hypothesize that this will increase GEV for each of the participants and may make EEG micro- states an applicable diagnostic tool since differences between the diagnostic groups become clearer. Moreover, there was a significant difference in the number of one-second epochs per subject, which may have affected our findings. In addition, future studies should investigate the relationship between EEG microstates and changes in metabolism as measured with FDG-PET in patients with AD. We also examined whether microstates could be used as a po- tential classifier of disease and found that microstate features are not satisfactory available to distinguish between three groups (AD, MCI, and HC), which probably is due to nonsignificant differences between AD and MCI. microstates, and this difference may be due to the fact that previous studies recorded EEGs on younger participants. In the current analy- sis, we only included the first 534 GFP peaks in the segmentation to equalize contributions from longer EEG files and this may also lead to a lower GEV. In addition, the low number of GFP peaks also limits our ability to investigate topographical differences in maps between groups. Furthermore, in the current study we had only 19 channels, which is below the number of channels used in recent microstate studies (Michel & Koenig, 2018). As a result, future studies should include longer EEG recordings to better determine whether any topographical changes exist. in spectral power. Here, we found that especially the theta band is affected in the early stage of the disease (Musaeus, Engedal, et al., 2018). Due to short EEG recordings, we did not investigate any po- tential topographical changes. Future studies should include longer EEG recordings to investigate whether any topographical changes exist. When testing whether there were any differences in expected and observed transition probabilities, no significant changes were found. Since microstate A is associated with the temporal lobe, the changes may be related to the neuropathological findings in AD as described by the Braak stages (Braak & Braak, 1991; Thal, Rub, Rub, Orantes, & Braak, 2002), which are especially pronounced in the temporal lobes in early AD. Furthermore, follow-up studies using PiB-PET, which quantifies the beta-amyloid deposition, have found that the temporal lobes are one of the first parts of the brain with beta-amyloid deposition (Okello et al., 2009; Villemagne et al., 2011). Here, we hypothesize that the changes in microstate A may reflect the underlying pathological changes in the left temporal lobe. 4.1 \| Limitations of the study and future directions In fact, many patients with MCI may have AD at a subclinical dementia stage, which is supported by the notion that over 50% of the included patients with MCI progressed (“con- verted”) to AD within two years follow-up in the Danish substudy (Musaeus, Nielsen, Osterbye, & Hogh, 2018). When comparing HC with AD, we found a poor classification rate of 69.8%. This rate is lower than between AD and HC using EEG spectral power (Musaeus, Engedal, et al., 2018) and lower than the discriminatory power of EEG connectivity (Musaeus, Engedal, Hogh, et al., 2019). The un- derlying reason could be that the changes in EEG microstates may not be present before later in the disease stages. The EEG segments were also too short to obtain four optimal maps if each participant's EEG was segmented as previously described (Koenig et al., 1999). We suggest that by recording longer EEGs and using segmentation for each person individually, it may be possible to increase the diag- nostic accuracy. We chose to extract four microstates since they are the most commonly reported ones and have been shown to be reliable (Khanna et al., 2014). We found that the algorithm was robust, and the main findings could be replicated (see Supplementary material, pp. 3–7). When looking at three, five, and six microstates, we found that patients with AD and MCI had a global affection of microstates, but there was a significant affection of microstate A when extract- ing both three, five, and six microstates. The results suggest that changes in microstate A are the hallmark of EEG microstate changes in AD. In addition, the accuracy did not differ markedly from the main analysis, which may be due to microstate A being the most import- ant microstate in patients with AD. Furthermore, even though GEV was not significantly different between the three groups, it was low (average GEV = 49.51%) compared to what other studies have re- ported (normally reporting a GEV of >70% (Michel & Koenig, 2018)). The increase in GEV was minimal when increasing the number of 4 \| DISCUSSION p-values show the differences when comparing patients 12 of 15 MUSAEUS et al. AUTHORS' CONTRIBUTIONS Dickerson, B. C., & Sperling, R. A. (2009). Large-scale functional brain network abnormalities in Alzheimer's disease: Insights from func- tional neuroimaging. Behavioural Neurology, 21(1), 63–75. https://doi. org/10.1155/2009/610392 K.E., P.H., V.J., M.N., A.O., J.S., L.W., and B.A. initiated the study, recruited patients, and gathered patient data. C.M., M.M., and A.K. analyzed the data. C.M. wrote the first draft of the article. C.M., K.E., P.H., V.J., A.K., T.K., M.M., M.N., A.O., E.S., J.S., L.W., G.W., and B.A. have edited and revised the manuscript, and all authors have ap- proved the final version of the manuscript. Dierks, T., Jelic, V., Julin, P., Maurer, K., Wahlund, L. O., Almkvist, O., … Winblad, B. (1997). EEG-microstates in mild memory impairment and Alzheimer's disease: Possible association with disturbed information processing. J Neural Transm (Vienna), 104(4–5), 483–495. Engedal, K., Haugen, P., Gilje, K., & Laake, P. (1988). Efficacy of short mental tests in the detection of mental impairment in old age. Comprehensive Gerontology. Section A, 2(2), 87–93. Christian S. Musaeus https://orcid.org/0000-0002-6442-2910 Emiliano Santarnecchi https://orcid.org/0000-0002-6533-7427 Golby, A., Silverberg, G., Race, E., Gabrieli, S., O'Shea, J., Knierim, K., … Gabrieli, J. (2005). Memory encoding in Alzheimer's disease: An fMRI study of explicit and implicit memory. Brain, 128(Pt 4), 773–787. CONFLICTS OF INTEREST Delorme, A., & Makeig, S. (2004). EEGLAB: An open source toolbox for analysis of single-trial EEG dynamics including independent compo- nent analysis. Journal of Neuroscience Methods, 134(1), 9–21. https:// doi.org/10.1016/j.jneumeth.2003.10.009 None. found evidence that the microstates were due to microstate changes in specific frequency bands. This microstate has previously been as- sociated with activity in the left temporal lobe using source localiza- tion and changes in the blood oxygen level-dependent signal in the left temporal region, which is strongly affected by amyloid and tau pathology in patients with AD. Together, our results show that EEG microstate analysis may be a useful tool in examining dynamic net- work activity and disruption in AD. Future studies should examine the relationship between hypometabolism with FDG-PET and the microstate features to understand the applicability of EEG micro- states as a diagnostic tool. Britz, J., Van De Ville, D., & Michel, C. M. (2010). BOLD correlates of EEG topography reveal rapid resting-state network dynamics. NeuroImage, 52(4), 1162–1170. https://doi.org/10.1016/j.neuro image.2010.02.052 Buckner, R. L., Snyder, A. Z., Shannon, B. J., LaRossa, G., Sachs, R., Fotenos, A. F., … Mintun, M. A. (2005). Molecular, structural, and functional characterization of Alzheimer's disease: Evidence for a relationship between default activity, amyloid, and memory. Journal of Neuroscience, 25(34), 7709–7717. https://doi.org/10.1523/JNEUR OSCI.2177-05.2005 Collie, A., & Maruff, P. (2000). The neuropsychology of preclinical Alzheimer's disease and mild cognitive impairment. Neuroscience and Biobehavioral Reviews, 24(3), 365–374. https://doi.org/10.1016/ S0149-7634(00)00012-9 Cummings, J. L. (2004). Alzheimer's disease. New England Journal of Medicine, 351(1), 56–67. https://doi.org/10.1056/NEJMra040223 DATA AVAILABILITY STATEMENT The data that support the findings of this study are available from the corresponding author upon reasonable request. However, due to regulations, we are not able to share the EEG files. Engedal, K., Snaedal, J., Hoegh, P., Jelic, V., Bo Andersen, B., Naik, M., … Oeksengaard, A. R. (2015). Quantitative EEG applying the statisti- cal recognition pattern method: A useful tool in dementia diagnostic workup. Dementia and Geriatric Cognitive Disorders, 40(1–2), 1–12. Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). "Mini-mental state". A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12(3), 189–198. 5 \| CONCLUSION In the present study, we found evidence of EEG microstate changes in patients with AD. Specifically, there was a longer duration, larger coverage, and higher frequency of occurrence of microstate A among AD and MCI patients compared to HC. Furthermore, we MUSAEUS et al. 13 of 15 ACKNOWLEDGMENTS Custo, A., Van De Ville, D., Wells, W. M., Tomescu, M. I., Brunet, D., & Michel, C. M. (2017). Electroencephalographic resting-state net- works: source localization of microstates. Brain Connect, 7(10), 671– 682. https://doi.org/10.1089/brain.2016.0476 We would like to thank the Kavli Charitable Trust in Bergen, Norway, which funded this EEG study. Conversion of amyloid posi- tive and negative MCI to AD over 3 years: An 11C-PIB PET study. Neurology, 73(10), 754–760. https://doi.org/10.1212/WNL.0b013 e3181b23564 Lee, T. W., Girolami, M., & Sejnowski, T. J. (1999). Independent compo- nent analysis using an extended infomax algorithm for mixed sub- gaussian and supergaussian sources. Neural Computation, 11(2), 417– 441. https://doi.org/10.1162/089976699300016719 Lehmann, D. (1971). Multichannel topography of human alpha EEG fields. Electroencephalography and Clinical Neurophysiology, 31(5), 439–449. https://doi.org/10.1016/0013-4694(71)90165-9 Oostenveld, R., Fries, P., Maris, E., & Schoffelen, J. M. (2011). FieldTrip: Open source software for advanced analysis of MEG, EEG, and in- vasive electrophysiological data. Computational Intelligence and Neuroscience, 2011, 156869. Lehmann, D., Faber, P. L., Galderisi, S., Herrmann, W. M., Kinoshita, T., Koukkou, M., … Koenig, T. (2005). EEG microstate duration and syntax in acute, medication-naive, first-episode schizophrenia: A multi-center study. Psychiatry Research, 138(2), 141–156. Palop, J. J., & Mucke, L. (2016). Network abnormalities and interneu- ron dysfunction in Alzheimer disease. Nature Reviews Neuroscience, 17(12), 777–792. https://doi.org/10.1038/nrn.2016.141 Lehmann, D., Ozaki, H., & Pal, I. (1987). EEG alpha map series: Brain micro-states by space-oriented adaptive segmentation. Electroencephalography and Clinical Neurophysiology, 67(3), 271–288. https://doi.org/10.1016/0013-4694(87)90025-3 Pascual-Marqui, R. D., Michel, C. M., & Lehmann, D. (1995). Segmentation of brain electrical activity into microstates: Model estimation and validation. IEEE Transactions on Biomedical Engineering, 42(7), 658– 665. https://doi.org/10.1109/10.391164 Poulsen, A. T., Pedroni, A., Langer, N., & Hansen, L. K. (2018). Microstate EEGlab toolbox: An introductory guide. bioRxiv: 289850. Machulda, M. M., Ward, H. A., Borowski, B., Gunter, J. L., Cha, R. H., O'Brien, P. C., … Jack, C. R. Jr (2003). Comparison of memory fMRI response among normal, MCI, and Alzheimer's patients. Neurology, 61(4), 500–506. https://doi.org/10.1212/01.WNL.00000 79052.01016.78 Santarnecchi, E., Khanna, A. R., Musaeus, C. S., Benwell, C. S. Y., Davila, P., Farzan, F., … Shafi, M. M. (2017). EEG microstate correlates of fluid intelligence and response to cognitive training. Brain Topography, 30(4), 502–520. https://doi.org/10.1007/s10548-017-0565-z McKhann, G., Drachman, D., Folstein, M., Katzman, R., Price, D., & Stadlan, E. M. (1984). Clinical diagnosis of Alzheimer's disease: Report of the NINCDS-ADRDA Work Group under the auspices of depart- ment of health and human services task force on Alzheimer's disease. Neurology, 34(7), 939–944. https://doi.org/10.1212/WNL.34.7.939 Schmidt, M. (2005). minFunc: Unconstrained Differentiable Multivariate Optimization in Matlab. Retrieved from http://www.cs.ubc. ca/~schmidtm/Software/minFunc.html Selkoe, D. J. (2002). Alzheimer's disease is a synaptic failure. Science, 298(5594), 789–791. https://doi.org/10.1126/science.1074069 Merrin, E. L., Meek, P., Floyd, T. C., & Callaway, E. 3rd (1990). MUSAEUS et al. cognitive decline in dementia due to Alzheimer's disease. Journal of Alzheimer's Disease, 64, 1359–1371. Khanna, A., Pascual-Leone, A., & Farzan, F. (2014). Reliability of rest- ing-state microstate features in electroencephalography. PLoS ONE, 9(12), e114163. Musaeus, C. S., Engedal, K., Hogh, P., Jelic, V., Morup, M., Naik, M., … Andersen, B. B. (2019). Oscillatory connectivity as a diag- nostic marker of dementia due to Alzheimer's disease. Clinical Neurophysiology, 130(10), 1889–1899. https://doi.org/10.1016/j. clinph.2019.07.016 Khanna, A., Pascual-Leone, A., Michel, C. M., & Farzan, F. (2015). Microstates in resting-state EEG: Current status and future direc- tions. Neuroscience and Biobehavioral Reviews, 49, 105–113. Killiany, R. J., Gomez-Isla, T., Moss, M., Kikinis, R., Sandor, T., Jolesz, F., … Albert, M. S. (2000). Use of structural magnetic resonance imag- ing to predict who will get Alzheimer's disease. Annals of Neurology, 47(4), 430–439. https://doi.org/10.1002/1531-8249(20000 4)47:4<430:AID-ANA5>3.0.CO;2-I Musaeus, C. S., Nielsen, M. S., & Hogh, P. (2019). Microstates as disease and progression markers in patients with mild cognitive impairment. Frontiers in Neuroscience, 13, 563. Musaeus, C. S., Nielsen, M. S., Osterbye, N. N., & Hogh, P. (2018). Decreased parietal beta power as a sign of disease progression in pa- tients with mild cognitive impairment. Journal of Alzheimer's Disease, 65(2), 475–487. https://doi.org/10.3233/JAD-180384 Koenig, T., Lehmann, D., Merlo, M. C., Kochi, K., Hell, D., & Koukkou, M. (1999). A deviant EEG brain microstate in acute, neuroleptic-naive schizophrenics at rest. European Archives of Psychiatry and Clinical Neuroscience, 249(4), 205–211. https://doi.org/10.1007/s0040 60050088 Musso, F., Brinkmeyer, J., Mobascher, A., Warbrick, T., & Winterer, G. (2010). Spontaneous brain activity and EEG microstates. A novel EEG/fMRI analysis approach to explore resting-state networks. NeuroImage, 52(4), 1149–1161. https://doi.org/10.1016/j.neuro image.2010.01.093 Koenig, T., Prichep, L., Lehmann, D., Sosa, P. V., Braeker, E., Kleinlogel, H., … John, E. R. (2002). Millisecond by millisecond, year by year: Normative EEG microstates and developmental stages. NeuroImage, 16(1), 41–48. https://doi.org/10.1006/nimg.2002.1070 Nishida, K., Morishima, Y., Yoshimura, M., Isotani, T., Irisawa, S., Jann, K., … Koenig, T. (2013). EEG microstates associated with salience and frontoparietal networks in frontotemporal dementia, schizophrenia and Alzheimer's disease. Clinical Neurophysiology, 124(6), 1106–1114. https://doi.org/10.1016/j.clinph.2013.01.005 Koenig, T., Studer, D., Hubl, D., Melie, L., & Strik, W. K. (2005). Brain connectivity at different time-scales measured with EEG. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 360(1457), 1015–1023. https://doi. org/10.1098/rstb.2005.1649 Okello, A., Koivunen, J., Edison, P., Archer, H. A., Turkheimer, F. E., Nagren, K., … Brooks, D. J. (2009). REFERENCES Greicius, M. D., Srivastava, G., Reiss, A. L., & Menon, V. (2004). Default- mode network activity distinguishes Alzheimer's disease from healthy aging: Evidence from functional MRI. Proceedings of the National Academy of Sciences of United States of America, 101(13), 4637–4642. https://doi.org/10.1073/pnas.0308627101 Andreou, C., Faber, P. L., Leicht, G., Schoettle, D., Polomac, N., Hanganu- Opatz, I. L., … Mulert, C. (2014). Resting-state connectivity in the prodromal phase of schizophrenia: Insights from EEG microstates. Schizophrenia Research, 152(2–3), 513–520. Baron, J. C., Chetelat, G., Desgranges, B., Perchey, G., Landeau, B., de la Sayette, V., & Eustache, F. (2001). In vivo mapping of gray matter loss with voxel-based morphometry in mild Alzheimer's disease. NeuroImage, 14(2), 298–309. https://doi.org/10.1006/ nimg.2001.0848 Gruner, W. (2020). MANCOVAN. Retrieved from https://www.mathw orks.com/matlabcentral/fileexchange/27014-mancovan, MATLAB Central File Exchange. Hogh, P., Madsen Sjo, N., Gade, A., & Waldemar, G. (2004). Temporal lobe hypoperfusion in isolated amnesia with slow onset: A single photon emission computer tomography study. Dementia and Geriatric Cognitive Disorders, 18(1), 15–23. https://doi.org/10.1159/00007 7730 Braak, H., & Braak, E. (1991). Neuropathological stageing of Alzheimer- related changes. Acta Neuropathologica, 82(4), 239–259. https://doi. org/10.1007/BF00308809 Braekhus, A., Ulstein, I., Wyller, T. B., & Engedal, K. (2011). The memory clinic–outpatient assessment when dementia is suspected. Tidsskrift for Den Norske Laegeforening, 131(22), 2254–2257. Ihl, R., Dierks, T., Froelich, L., Martin, E. M., & Maurer, K. (1993). Segmentation of the spontaneous EEG in dementia of the Alzheimer type. Neuropsychobiology, 27(4), 231–236. https://doi. org/10.1159/000118986 Brechet, L., Brunet, D., Birot, G., Gruetter, R., Michel, C. M., & Jorge, J. (2019). Capturing the spatiotemporal dynamics of self-gener- ated, task-initiated thoughts with EEG and fMRI. NeuroImage, 194, 82–92. Kato, T., Knopman, D., & Liu, H. (2001). Dissociation of regional acti- vation in mild AD during visual encoding: A functional MRI study. Neurology, 57(5), 812–816. https://doi.org/10.1212/WNL.57.5.812 14 of 15 Thal, D. R., Rub, U., Orantes, M., & Braak, H. (2002). Phases of A be- ta-deposition in the human brain and its relevance for the develop- ment of AD. Neurology, 58(12), 1791–1800. Winblad, B., Palmer, K., Kivipelto, M., Jelic, V., Fratiglioni, L., Wahlund, L. O., … Petersen, R. C. (2004). Mild cognitive impairment–beyond con- troversies, towards a consensus: Report of the international working group on mild cognitive impairment. Journal of Internal Medicine, 256(3), 240–246. https://doi.org/10.1111/j.1365-2796.2004.01380.x Winblad, B., Palmer, K., Kivipelto, M., Jelic, V., Fratiglioni, L., Wahlund, L. O., … Petersen, R. C. (2004). Mild cognitive impairment–beyond con- troversies, towards a consensus: Report of the international working group on mild cognitive impairment. Journal of Internal Medicine, 256(3), 240–246. https://doi.org/10.1111/j.1365-2796.2004.01380.x Twamley, E. W., Ropacki, S. A., & Bondi, M. W. (2006). Neuropsychological and neuroimaging changes in preclinical Alzheimer's disease. Journal of the International Neuropsychological Society, 12(5), 707–735. https://doi.org/10.1017/S1355617706060863 Yuan, H., Zotev, V., Phillips, R., Drevets, W. C., & Bodurka, J. (2012). Spatiotemporal dynamics of the brain at rest–exploring EEG mi- crostates as electrophysiological signatures of BOLD resting state networks. NeuroImage, 60(4), 2062–2072. https://doi.org/10.1016/j. neuroimage.2012.02.031 Van de Ville, D., Britz, J., & Michel, C. M. (2010). EEG microstate sequences in healthy humans at rest reveal scale-free dynamics. Proceedings of the National Academy of Sciences of the United States of America, 107(42), 18179–18184. https://doi.org/10.1073/pnas.1007841107 Villemagne, V. L., Pike, K. E., Chetelat, G., Ellis, K. A., Mulligan, R. S., Bourgeat, P., … Rowe, C. C. (2011). Longitudinal assessment of Abeta and cognition in aging and Alzheimer disease. Annals of Neurology, 69(1), 181–192. Topographic segmentation of waking EEG in medication-free schizophrenic pa- tients. International Journal of Psychophysiology, 9(3), 231–236. https://doi.org/10.1016/0167-8760(90)90055-I Shulman, K. I. (2000). Clock-drawing: Is it the ideal cognitive screening test? International Journal of Geriatric Psychiatry, 15(6), 548–561. https://doi.org/10.1002/1099-1166(200006)15:6<548:AID-GPS24 2>3.0.CO;2-U Michel, C. M., & Koenig, T. (2018). EEG microstates as a tool for studying the temporal dynamics of whole-brain neuronal networks: A review. NeuroImage, 180(Pt B), 577–593. Stevens, A., & Kircher, T. (1998). Cognitive decline unlike normal aging is associated with alterations of EEG temporo-spatial characteris- tics. European Archives of Psychiatry and Clinical Neuroscience, 248(5), 259–266. https://doi.org/10.1007/s004060050047 Montgomery, S. A., & Asberg, M. (1979). A new depression scale designed to be sensitive to change. British Journal of Psychiatry, 134, 382–389. Strik, W. K., Chiaramonti, R., Muscas, G. C., Paganini, M., Mueller, T. J., Fallgatter, A. J., … Zappoli, R. (1997). Decreased EEG micro- state duration and anteriorisation of the brain electrical fields in mild and moderate dementia of the Alzheimer type. Psychiatry Research, 75(3), 183–191. https://doi.org/10.1016/S0925 -4927(97)00054-1 Morris, J. C., Mohs, R. C., Rogers, H., Fillenbaum, G., & Heyman, A. (1988). Consortium to establish a registry for Alzheimer's disease (CERAD) clinical and neuropsychological assessment of Alzheimer's disease. Psychopharmacology Bulletin, 24(4), 641–652. Musaeus, C. S., Engedal, K., Hogh, P., Jelic, V., Morup, M., Naik, M., … Andersen, B. B. (2018). EEG theta power is an early marker of MUSAEUS et al. 15 of 15 SUPPORTING INFORMATION Additional supporting information may be found online in the pp g y Supporting Information section. Wackermann, J., Lehmann, D., Michel, C. M., & Strik, W. K. (1993). Adaptive segmentation of spontaneous EEG map series into spatially defined microstates. International Journal of Psychophysiology, 14(3), 269–283. https://doi.org/10.1016/0167-8760(93)90041-M Supporting Information section. How to cite this article: Musaeus CS, Engedal K, Høgh P, et al. Changes in the left temporal microstate are a sign of cognitive decline in patients with Alzheimer’s disease. Brain Behav. 2020;00:e01630. https://doi.org/10.1002/brb3.1630 How to cite this article: Musaeus CS, Engedal K, Høgh P, et al. Changes in the left temporal microstate are a sign of cognitive decline in patients with Alzheimer’s disease. Brain Behav. 2020;00:e01630. https://doi.org/10.1002/brb3.1630 Welsh, K., Butters, N., Hughes, J., Mohs, R., & Heyman, A. (1991). Detection of abnormal memory decline in mild cases of Alzheimer's disease using CERAD neuropsychological measures. Archives of Neurology, 48(3), 278–281. https://doi.org/10.1001/archn eur.1991.00530150046016
https://openalex.org/W2599270706	https://europepmc.org/articles/pmc5375149?pdf=render	English	null	Prediction of diuretic response to tolvaptan by a simple, readily available spot urine Na/K ratio	PloS one	2,017	cc-by	5,898	Results Mean weight reduction on day 7 of tolvaptan therapy was −2.9% ± 3.2%, and treatment was effective in 52% of patients. Multivariate analysis revealed that spot urine Na/K ratio 2.5 at baseline was the only factor independently related to therapeutic effect, with an odds ratio of 7.85 (95% confidence interval 2.64–23.40, p = 0.0002). Weight reduction percentage on day 7 was −4.0% ± 2.8% in patients with spot urine Na/K 2.5, which was significantly greater than the 0.7% ± 2.7% loss in those with urine Na/K < 2.5 (p < 0.05). A spot urine Na/ K ratio 2.5 had a sensitivity of 85% and specificity of 60% for predicting effective treatment. No adverse events of treatment led to treatment discontinuation. Data Availability Statement: All relevant data are within the paper. OPEN ACCESS Citation: Komiyama Y, Kurosaki M, Nakanishi H, Takahashi Y, Itakura J, Yasui Y, et al. (2017) Prediction of diuretic response to tolvaptan by a simple, readily available spot urine Na/K ratio. PLoS ONE 12(3): e0174649. https://doi.org/10.1371/ journal.pone.0174649 Prediction of diuretic response to tolvaptan by a simple, readily available spot urine Na/K ratio Yasuyuki Komiyama1,2☯, Masayuki Kurosaki1☯, Hiroyuki Nakanishi1, Yuka Takahashi1, Jun Itakura1, Yutaka Yasui1, Nobuharu Tamaki1, Hitomi Takada2, Mayu Higuchi1, Tomoyuki Gotou1, Youhei Kubota1, Kenta Takaura1, Tsuguru Hayashi1, Wann Oh1, Mao Okada1, Nobuyuki Enomoto2, Namiki Izumi1* 1 Department of Gastroenterology and Hepatology, Musashino Red Cross Hospital, Tokyo, Japan, 2 First Department of Internal Medicine, Faculty of Medicine, University of Yamanashi, Yamanashi, Japan ☯These authors contributed equally to this work. * izumi012@musashino.jrc.or.jp a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 Editor: Han-Chieh Lin, Taipei Veterans General Hospital, TAIWAN Editor: Han-Chieh Lin, Taipei Veterans General Hospital, TAIWAN Tolvaptan was administered to 88 consecutive cirrhotic patients with ascites unresponsive to standard diuretic therapy. An effective treatment response was a 2% reduction in body weight on day 7. The association of patient pretreatment characteristics with therapeutic effects was analyzed. Copyright: © 2017 Komiyama et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. RESEARCH ARTICLE Background Tolvaptan is vasopressin type 2 receptor antagonist that inhibits water reabsorption. It is used in combination with standard diuretics to treat ascites unresponsive to standard diuretic therapy or hyponatremia because of liver cirrhosis. This study evaluated the effec- tiveness and safety of tolvaptan in clinical practice and aimed to determine the factors related to its effectiveness. Conclusions Baseline spot urine Na/K was predictive of an effective response to tolvaptan therapy. It is simple to perform and readily available and might serve as an indicator of optimal timing of tolvaptan administration in patients with inadequate response to conventional Na diuretic therapy. Data Availability Statement: All relevant data are within the paper. Funding: The authors received no specific funding for this work. Funding: The authors received no specific funding for this work. Competing interests: The authors have declared that no competing interests exist. PLOS ONE \| https://doi.org/10.1371/journal.pone.0174649 March 31, 2017 1 / 12 Urinary Na/K ratio in tolvaptan treatment Introduction Hepatic edema and ascites are common complications of decompensated cirrhosis leading to reduced patient quality of life [1]. Hepatic ascites is refractory in about 10% of patients, and does not respond to salt restriction and standard diuretic drugs [2–5]. Tolvaptan (Otsuka Pharmaceutical, Osaka, Japan) is a nonpeptide antagonist of vasopressin type 2 receptor, a novel class of diuretic drugs. It binds to the vasopressin V2 receptors of principle cells in the renal collecting ducts. It inhibits the expression of aquaporin-2 (AQP2), a vasopressin-regu- lated water channel, in the apical membrane of renal collecting duct cells. A reduction of AQP2 water channels in the collecting ducts prevents water reabsorption, and promotes urine excretion [6]. Tolvaptan is used to treat patients with heart failure or hyponatremia [7–9], but following successful Phase 3 trials, tolvaptan was approved in September 2013 in Japan for treatment of refractory ascites [10], and is widely used to treat ascites refractory to other diuretics. However, only a few reports of clinical experience have been published [6, 11, 12, 13]. Consequently, patient factors that influence tolvaptan effectiveness are unclear, and the optimal time to begin tolvaptan administration have not been identified. The guideline of the American Association for the Study of Liver Disease reports a daily urine Na excretion 78 mmol/day as an indicator of good ascites control with Na diuretics [3]. It is not easy to measure daily urine Na excretion in routine outpatient care. However, the spot urine Na/K ratio correlates well with daily urinary Na excretion, and can predict respon- siveness to diuretics [14–16]. Although spot urine Na/K ratio may be effective as a surrogate marker of daily urine Na excretion, the relationship of spot urine Na/K ratio and tolvaptan therapeutic response is unclear. In this study, we focused on spot urine Na/K and its relation- ship with tolvaptan effectiveness. PLOS ONE \| https://doi.org/10.1371/journal.pone.0174649 March 31, 2017 Patients This prospective observational study included 88 consecutive cirrhotic patients with ascites not controlled by moderate dose of diuretics. All patients received tolvaptan treatment at Musashino Red Cross Hospital between August 2013 and December 2015. Patients with a diagnosis of cirrhosis were eligible if they had ascites unresponsive to standard diuretic therapy consisting of spironolactone, furosemide, or both, in addition to sodium (<5g/day) and water (<1000 mL/day) restriction. The dose of diuretics eligible for inclusion in this was spironolac- tone≧25mg and/or furosemide ≧20mg. Basically, diuretic therapy was started with spironolac- tone at dose 25mg or 50mg, then furosemide at dose 20mg or 40mg was added. Addition of tolvaptan was considered if ascites was not controlled by these diuretics. Our policy was to avoid high dose of diuretics because it may cause renal insufficiency. Therefore, tolvaptan ther- apy was preferentially introduced before the appearance of renal dysfunction. The lack of 2 / 12 PLOS ONE \| https://doi.org/10.1371/journal.pone.0174649 March 31, 2017 Urinary Na/K ratio in tolvaptan treatment response to diuretics was defined as the presence of ascites after treatment by furosemide and/ or spironolactone for at least 1 months. Patients with overt hepatic encephalopathy, active gastrointestinal bleeding, spontaneous bacterial peritonitis, stage 4 chronic kidney disease (eGFR < 30 ml/min/1.73 m2), or heart failure were excluded. Spironolactone and furosemide dosage was fixed 7 days before, and continued during, tol- vaptan administration. Diuretics was not discontinued before or during the study period. Tol- vaptan 3.75 mg or 7.5 mg was administered once daily. In the event that the treatment was initiated at a dose of 3.75 mg/day but the response was judged not effective, the dosage was increased to 7.5 mg after few days. In the initial period of this study, we mainly used 3.75mg dose as introduction because we were uncertain of possible adverse events. After confirming the safety of tolvaptan at 7.5mg in several patients, we used 7.5mg dose for most patients. For the baseline data, blood and spot urine test was obtained in the early morning of the day that treatment began, before the administration of diuretics. To avoid hypovolemia and hyperna- tremia, water intake was not restricted during treatment with tolvaptan in accordance with the instruction of the package insert of tolvaptan. The use of albumin infusion was not restricted during the study period. Patients However, to exclude the short-term effect of albumin infusion on diuretic response, patients who received albumin infusion within 1 week prior to the start of tolvaptan was not included. Patients who received albumin infusion within 7 days after the start of tolvaptan due to insufficient diuretic response to tolvaptan were regarded as ineffective. Patients who had past history of large volume paracentesis were included in the present study. In these patients, tolvaptan was started after we confirmed that patients’ weight became stable after paracentesis. To date, there is no established criteria to define response to tolvaptan. Therefore, response to tolvaptan was tentatively defined as weight reduction of 2% at day 7 of treatment. Patients with a weight reduction 2% compared with their baseline weight were recorded as effective cases. Patients who could not continue treatment for 7 days, and patients underwent large vol- ume paracentesis within 7 days after the start of treatment were recorded as ineffective cases. After the patients were stratified into two groups by their response to treatment, the associa- tion of demographic and clinical characteristics with the therapeutic effect was analyzed. Ethical considerations This study protocol conformed to the ethical guidelines of the Declaration of Helsinki and was approved by the institutional ethics committees of the Musashino Red Cross Hospital. Written informed consent to receive tolvaptan treatment and to be included in this study was obtained from each study participant. PLOS ONE \| https://doi.org/10.1371/journal.pone.0174649 March 31, 2017 Baseline characteristics of patients The mean patient age was 69.3 years, and most frequent (53%) underlying liver disease was hepatitis C virus infection. Liver function was Child-Pugh class C in 34 patients (39%); 40 patients (45.5%) had concurrent hepatocellular carcinoma (Table 1). Among 40 patients with HCC, only 1 patient received active treatment for HCC, intra-arterial infusion chemotherapy, during the study period of 7 days. There was no obvious change in the volume of HCC that could cause any improvement of portal hypertension or liver function during the study period. In other patients, active treatment for HCC was not possible because their liver function was too impaired. The median furosemide dose was 20 mg daily; the median spironolactone dose was 50 mg daily. Statistical analysis Statistical analysis was performed with EZR (Saitama Medical Center, Jichi Medical Univer- sity, Saitama Japan), which is a graphical user interface for R (the R Foundation for Statistical Computing, Vienna, Austria). More precisely, a modified version of the R-commander pack- age with additional statistical functions frequently used in biostatistics [17]. Differences in con- tinuous variables were compared with the paired t-test. Differences categorical data values were compared using Fischer’s exact test. Multivariate logistic regression analysis was per- formed on factors that had a significant relationship with treatment effect in univariate analysis. The cut-off value for continuous variables was determined by receiver operating char- acteristic (ROC) curve analysis, or by using the reference normal value. A p-value of <0.05 was considered statistically significant. 3 / 12 PLOS ONE \| https://doi.org/10.1371/journal.pone.0174649 March 31, 2017 Urinary Na/K ratio in tolvaptan treatment https://doi.org/10.1371/journal.pone.0174649.t001 PLOS ONE \| https://doi.org/10.1371/journal.pone.0174649 March 31, 2017 Treatment efficacy: Correction of hyponatremia Twenty-two patients had hyponatremia (serum Na < 135 mEq/L) before tolvaptan therapy. There was a trend in increase of serum Na from 130.3 ± 4.4 mEq/L at baseline to 132.2 ± 4.4 mEq/L on day 7 after tolvaptan (p = 0.08). Baseline hyponatremia was corrected (serum Na > 135 mEq/L) in five of 22 patients (23%), but serum Na did not change in patients with normal serum Na at baseline (138.7 ± 2.4 mEq/L at baseline to 138.9 ± 3.3, p = 0.81 on day 7). Treatment efficacy: Weight reduction The mean weight reduction on day 7 of tolvaptan therapy was 2.9% ± 3.2%. Forty-six patients (52%) achieved a 2% reduction on day 7. Symptoms such as abdominal distension or dys- pnea was improved in 78% of patients with 2% reduction in their weight. Treatment was discontinued before day 7 in two patients who did not respond, but there were no discontinua- tions caused by adverse events. Seven patients had paracentesis before day 7 and were evalu- ated with the patients in the ineffective treatment group. Table 1. Clinical backgrounds of patients. n = 88 Age 69.3±12.0 Gender male/female 57(65%)/31(35%) Child Pugh grade B/C 52(61%)/34(39%) Etiology of cirrhosis(HCV/HBV/ALD/others) 47/3/22/16 Co-exsisting of HCC 40(45.5%) Furosemide (mg/day) 20(0–100) Spironolactone (mg/day) 50(0–100) Prior history of albumin infusion 22(25%) Prior history of paracentesis 18(21%) Albumin (g/dL) 2.6±0.6 Total bilirubin (mg/dL) 1.6±1.1 ALT (IU/L) 32.7±30.2 Creatinin (mg/dL) 1.06±0.81 eGFR (mg/min/1.73m2) 65.0±28.6 Serum Na (mEq/L) 136.6±4.8 Platelet counts (×104/μL) 10.0±6.0 Prothrombin activity (%) 71.7±19.3 CRP (mg/dL) 1.14±1.38 Urine osmolarity (mOsm/L) 394±124 Urine Na (mEq/L) 67.6±31.7 Urine K (mEq/L) 23.0±12.2 Urine Na/K 3.67±2.34 Values are mean ± standard deviation or median (range) HCV: hepatitis C virus, HBV: hepatitis B virus, ALD: alcoholic liver disease, HCC: hepatocellular carcinoma, ALT: alanine aminotransferase Table 1. Clinical backgrounds of patients. Table 1. Clinical backgrounds of patients. Table 1. Clinical backgrounds of patients. 4 / 12 PLOS ONE \| https://doi.org/10.1371/journal.pone.0174649 March 31, 2017 Urinary Na/K ratio in tolvaptan treatment Table 2. Baseline factors and response to tolvaptan. Effective group Weight reduction 2%(n = 46) Ineffective group Weight reduction <2%(n = 42) P-value Age 67.8±13.4 70.9±10.2 0.22 Gender male/female 30/16 27/15 0.99 Child Pugh grade B/C 32/14 26/12 0.5 Etiology (HCV/HBV/ALD/others) 28/0/12/6 19/3/10/10 0.05 Co-exsisting of HCC 17 23 0.13 Furosemide (≦40/ >40mg/day) 37/9 40/2 0.05 Spironolactone (≦50/ >50mg/day) 41/5 42/0 0.06 Albumin (g/dL) 2.6±0.5 2.6±0.6 0.94 Total bilirubin (mg/dL) 1.6±1.0 1.7±1.2 0.7 Serum Creatinin (mg/dL) 0.96±0.64 1.17±0.95 0.22 eGFR (mg/min/1.73m2) 70.3±28.2 59.1±28.3 0.07 Serum Na (mEq/L) 137.5±2.9 135.6±6.1 0.06 CRP (mg/dL) 0.76±0.96 1.51±1.63 <0.05 Platelet counts (×104/μL) 8.7±5.4 11.5±6.3 <0.05 Prothrombin activity (%) 72.6±18.2 70.7±20.6 0.66 Urine osmolarity (mOsm/L) 376±110 415±137 0.14 Urine Na (mEq/L) 76.5±27.5 57.3±33.4 <0.05 Urine K (mEq/L) 19.6±9.6 26.8±13.8 <0.05 Urine Na/K 4.5±2.3 2.7±2.0 <0.05 HCV: hepatitis C virus, HBV: hepatitis B virus, ALD: alcoholic liver disease, HCC: hepatocellular carcinoma, ALT: alanine aminotransferase h //d i /10 13 1/j l 01 4649 002 Table 2. Baseline factors and response to tolvaptan. PLOS ONE \| https://doi.org/10.1371/journal.pone.0174649 March 31, 2017 Factors associated with therapeutic effect https://doi.org/10.1371/journal.pone.0174649.g001 https://doi.org/10.1371/journal.pone.0174649.g001 percentage weight reduction on day 7 was significantly higher in patients with a spot urine Na/ K ratio 2.5 than in those with a ration <2.5 (4.0% ± 2.8% vs −0.7% ± 2.7%, p < 0.05, Fig 2). percentage weight reduction on day 7 was significantly higher in patients with a spot urine Na/ K ratio 2.5 than in those with a ration <2.5 (4.0% ± 2.8% vs −0.7% ± 2.7%, p < 0.05, Fig 2). Factors associated with therapeutic effect Pretreatment characteristics in the effective treatment and ineffective treatment groups were compared (Table 2). There were no significant differences in age, gender, Child-Pugh class, presence or absence of liver cancer, or furosemide/spironolactone dose. When we compare the rate of response between patients with or without prior paracentesis, the rate of response was 33.3% versus 57.1% which was not statistically different (p = 0.11). The effective treatment group had a mean lower C-reactive protein (CRP) level (0.76 ± 0.96 vs. 1.51 ± 1.63 mg/dL, p = 0.01), lower platelet count (8.7 ± 5.4 vs. 11.5 ± 6.3 103/C, p = 0.03), higher mean spot urine Na level (76.5 ± 27.5 vs. 57.3 ± 33.4 mEq/L, p = 0.004), and higher mean spot urine Na/K ratio (4.5 ± 2.2 vs. 2.6 ± 2.0, p < 0.001) than the ineffective treatment group (Table 2). ROC analysis revealed that a spot urine Na/K ratio 2.5 was the best cutoff value to determine effectiveness (Fig 1). The prediction accuracy of spot urine Na/K ratio 2.5 had sensitivity of 85%, specific- ity of 60%, positive predictive value of 70%, and negative predictive value of 78%. Multivariate analysis revealed that spot urine Na/K ratio 2.5 was the only factor independently related to effectiveness, with an odds ratio of 7.85, 95% CI of 2.64–23.40, and p < 0.05 (Table 3). The 5 / 12 PLOS ONE \| https://doi.org/10.1371/journal.pone.0174649 March 31, 2017 Urinary Na/K ratio in tolvaptan treatment Fig 1. ROC curve analysis of spot urine Na/K ratio and prediction of weight loss. The ROC curve shows baseline spot urine Na/K ratio to predict weight reduction 2% on day 7 of tolvaptan therapy. Urine Na/K ratio 2.5 was the best cutoff value to predict effectiveness. https://doi org/10 1371/journal pone 0174649 g001 Fig 1. ROC curve analysis of spot urine Na/K ratio and prediction of weight loss. The ROC curve shows baseline spot urine Na/K ratio to predict weight reduction 2% on day 7 of tolvaptan therapy. Urine Na/K ratio 2.5 was the best cutoff value to predict effectiveness. Fig 1. ROC curve analysis of spot urine Na/K ratio and prediction of weight loss. The ROC curve shows baseline spot urine Na/K ratio to predict weight reduction 2% on day 7 of tolvaptan therapy. Urine Na/K ratio 2.5 was the best cutoff value to predict effectiveness. Factors associated with urine Na/K ratio The relationship between spot urine Na/K ratio and total urine Na excretion per day was ana- lyzed in 23 patients with 24-hour urine collection prior to starting tolvaptan. ROC analysis revealed that the spot urine Na/K ratio was closely associated with urine Na excretion 78 mmol/day with an area under curve of 0.94 (95% CI: 0.85–1.0, Fig 3). For predicting a urine Na excretion 78 mmol/day, spot urine Na/K ratio 2.5 had sensitivity of 67% and specificity of 100%. Underlying factors associated with baseline spot urine Na/K ratio 2.5 included elevated serum Na levels (138.0 ± 3.3 vs. 134.2 ± 5.9, p < 0.001), low CRP level (0.79 ± 1.02 vs. 1.73 ± 1.71, p = 0.002), elevated urine Na level (169.8 ± 90.7 vs. 54.4 ± 43.5, p < 0.001), and Table 3. Multivariate regression analysis assessing the effectiveness of tolvaptan. Odds Ratio (95%CI) P-value Platelet counts <15 (×104/μL) 2.51 (0.65–9.65) 0.18 CRP <1.0 (mg/dL) 1.43 (0.49–4.16) 0.50 Urine Na/K 2.5 7.85 (2.64–23.40) 0.0002 Table 3. Multivariate regression analysis assessing the effectiveness of tolvaptan. 6 / 12 PLOS ONE \| https://doi.org/10.1371/journal.pone.0174649 March 31, 2017 PLOS ONE \| https://doi.org/10.1371/journal.pone.0174649 March 31, 2017 Urinary Na/K ratio in tolvaptan treatment Fig 2. Change in body weight after tolvaptan therapy and baseline spot urine Na/K ratio. The change in mean body weight after tolvaptan therapy is shown over time. The solid line represents patients with baseline spot urine Na/K 2.5, and the dotted line represents those with Na/K < 2.5. The error bars indicate standard deviations. Weight loss on day 7 was significantly greater in patients with spot urine Na/K ratios 2.5 than in those with Na/K < 2.5 (4.0% ± 2.8% vs. −0.7% ± 2.7%, p < 0.05). https://doi.org/10.1371/journal.pone.0174649.g002 Fig 3. ROC curve analysis of urine Na/K ratio and prediction of total urine Na excretion per day. ROC analysis revealed that urine Na/K ratio was closely associated with urinary Na excretion 78 mmol/day, with the area under curve of 0.94 (95% CI: 0.85–1.0). https://doi.org/10.1371/journal.pone.0174649.g003 Urinary Na/K ratio in tolvaptan treatment Fig 2. Change in body weight after tolvaptan therapy and baseline spot urine Na/K ratio. The change in mean body weight after tolvaptan therapy is shown over time. The solid line represents patients with baseline spot urine Na/K 2.5, and the dotted line represents those with Na/K < 2.5. The error bars indicate standard deviations. PLOS ONE \| https://doi.org/10.1371/journal.pone.0174649 March 31, 2017 Safety Temporary hypernatremia (serum Na > 145 mEq/L) was observed in four patients on day 2 after tolvaptan introduction, in three patients on day 3, and in one patient on day 7; all patients recovered without discontinuing tolvaptan therapy. Decrease of estimated glomerular filtra- tion rate (eGFR) >25% was observed in six patients in the effective treatment group, and three in the ineffective treatment group. Two patients in the effective treatment group spontaneously improved during the treatment period; three improved after reducing the tolvaptan and furo- semide dosages. One patient who did not recover had advanced liver failure caused by exacer- bation of the underlying disease. Two of the three patients in the ineffective treatment group had advanced hepatocellular carcinoma, and one had advanced liver cirrhosis with a Child- Pugh score of 12. No patients discontinued treatment because of adverse events, including a patient with liver dysfunction. Urinary Na/K ratio in tolvaptan treatment Table 4. Urine Na/K ratio and background factors. Urine Na/K2.5 Urine Na/K<2.5 p-value n = 56 n = 32 Age 68.0±12.2 71.3±11.6 0.22 Gender male 37 (66.1%) 20 (62.5%) 0.82 Female 19 (33.9%) 12 (37.5%) Etiology of cirrhosis ALD 15 (26.8%) 7 (21.9%) 0.17 HBV 0 (0.0%) 3 (9.4%) HCV 31 (55.4%) 16 (50.0%) Others 10 (17.9%) 6 (18.8%) Child B 38 (67.9%) 20 (62.5%) 0.65 C 18 (32.1%) 12 (37.5%) Co-exsisting of HCC 23 (41.1%) 17 (53.1%) 0.37 Furosemide (≦40/ >40mg/day) 47/9 30/2 0.32 Spironolactone (≦50/ >50mg/day) 52/4 31/1 0.64 Albumin (g/dL) 2.5±0.6 2.6±0.4 0.46 Total bilirubin (mg/dL) 1.5±1.0 1.7±1.2 0.34 Serum Creatinine (mg/dL) 1.13±0.97 0.93±0.36 0.26 Serum Na (mEq/L) 138.0±3.3 134.2±5.9 <0.05 eGFR (mg/min/1.73m2) 64.2±28.1 66.1±29.9 0.76 CRP (mg/dL) 0.79±1.02 1.73±1.71 <0.05 Urine Na (mEq/L) 82.5±25.4 39.8±22.4 <0.05 Urine K (mEq/L) 18.7±8.6 30.9±14.0 <0.05 Urine osmolarity (mOsm/L) 363±105 449±138 <0.05 Platelet counts (×104/μL) 10.0±6.5 10.1±4.8 0.94 Prothrombin activity (%) 72.9±17.2 69.4±22.6 0.42 HCV: hepatitis C virus, HBV: hepatitis B virus, ALD: alcoholic liver disease, HCC: hepatocellular carcinoma Table 4. Urine Na/K ratio and background factors. low urinary osmolality (363.6 ± 105.2 vs. 449.8 ± 138.1, p = 0.002, Table 4). Spironolactone or furosemide dose was not associated with spot urine Na/K ratio. low urinary osmolality (363.6 ± 105.2 vs. 449.8 ± 138.1, p = 0.002, Table 4). Spironolactone or furosemide dose was not associated with spot urine Na/K ratio. low urinary osmolality (363.6 ± 105.2 vs. 449.8 ± 138.1, p = 0.002, Table 4). Spironolactone or furosemide dose was not associated with spot urine Na/K ratio. low urinary osmolality (363.6 ± 105.2 vs. 449.8 ± 138.1, p = 0.002, Table 4). Spironolactone or furosemide dose was not associated with spot urine Na/K ratio. Factors associated with urine Na/K ratio Weight loss on day 7 was significantly greater in patients with spot urine Na/K ratios 2.5 than in those with Na/K < 2.5 (4.0% ± 2.8% vs. −0.7% ± 2.7%, p < 0.05). https://doi org/10 1371/journal pone 0174649 g002 Fig 2. Change in body weight after tolvaptan therapy and baseline spot urine Na/K ratio. The change in mean body weight after tolvaptan therapy is shown over time. The solid line represents patients with baseline spot urine Na/K 2.5, and the dotted line represents those with Na/K < 2.5. The error bars indicate standard deviations. Weight loss on day 7 was significantly greater in patients with spot urine Na/K ratios 2.5 than in those with Na/K < 2.5 (4.0% ± 2.8% vs. −0.7% ± 2.7%, p < 0.05). https://doi.org/10.1371/journal.pone.0174649.g002 Fig 3. ROC curve analysis of urine Na/K ratio and prediction of total urine Na excretion per day. ROC analysis revealed that urine Na/K ratio was closely associated with urinary Na excretion 78 mmol/day, with the area under curve of 0.94 (95% CI: 0.85–1.0). https://doi.org/10.1371/journal.pone.0174649.g003 Fig 3. ROC curve analysis of urine Na/K ratio and prediction of total urine Na excretion per day. ROC analysis revealed that urine Na/K ratio was closely associated with urinary Na excretion 78 mmol/day, with the area under curve of 0.94 (95% CI: 0.85–1.0). https://doi.org/10.1371/journal.pone.0174649.g003 PLOS ONE \| https://doi.org/10.1371/journal.pone.0174649 March 31, 2017 7 / 12 Discussion In this study, tolvaptan was effective in 52% of patients with ascites not controlled by moderate dose of diuretics such as spironolactone and/or furosemide. A spot urine Na/K ration 2.5 8 / 12 PLOS ONE \| https://doi.org/10.1371/journal.pone.0174649 March 31, 2017 Urinary Na/K ratio in tolvaptan treatment prior to administration was predictive of effective tolvaptan treatment. Using this criterion, we were able to include 85% of patients in an effective treatment group. Measurement of spot urine Na/K ratio is easy to do and is readily available, changes during the natural course of liver cirrhosis could be monitored without cost, and thus might be utilized in clinical practice for patient selection or optimize the timing of tolvaptan treatment. Conventional diuretics used to treat ascites include spironolactone and furosemide which are Na diuretics. Spironolactone is an aldosterone antagonist, and is effective in 50% to 90% of cases [18]. Patients with inadequate responses to spironolactone are given escalating doses of furosemide in combination. However, approximately 10% of hepatic ascites are refractory to these standard diuretic drugs [2–5]; moreover high doses of furosemide has been reported to cause renal dysfunction [19, 20] leading to decreased survival [21]. Tolvaptan, an antagonist of vasopressin type 2 receptor, inhibits water reabsorption and promotes the excretion of free water without increasing Na excretion. Its diuretic mechanism is totally different from conventional diuretics, which promote Na excretion into the urine. Clinical trials in Japan confirmed the efficacy of tolvaptan for refractory ascites regardless of serum albumin level [10], and the safely of a 14-day dosage regimen [22]. Tolvaptan was approved for use in combination with Na diuretics for refractory ascites in 2013. This alterna- tive option for refractory ascites is expected to improve the efficacy and safety of treatment. In the latest version of the Japanese guideline, tolvaptan is recommended to be used in patients who are refractory to diuretics, before considering intra venous administration of diuretics, intra venous administration of albumin, large volume paracentesis, TIPS or peritoneal-venous shunting. In fact, after the approval of tolvaptan, early administration of tolvaptan before increasing furosemide or spironolactone to maximal dose is now common in Japan. Although this agent is widely used in clinical practice, few published reports have described its effective- ness in clinical practice. Akiyama et al. PLOS ONE \| https://doi.org/10.1371/journal.pone.0174649 March 31, 2017 Urinary Na/K ratio in tolvaptan treatment urinary osmolality and urine AQP2 occur after tolvaptan administration, thus there are no established pretreatment predictive factors. In this study, we found that baseline spot urine Na/K ratio was independently associated with tolvaptan effectiveness. As measurement of spot urine Na/K ratio is simple, easily done, and readily available, it can be used in clinical practice to identify patients likely to respond to tolvaptan therapy. As the spot urine Na/K ratio changes during the natural course of liver cir- rhosis, it might be used to determine the best time to initiate tolvaptan administration. Consistent with previous reports, the spot urine Na/K ratio was significantly correlated with daily urinary Na excretion [3, 14, 15, 16]. The association of the spot urine Na/K ratio with tolvaptan efficacy indicates that maintenance of Na excretion was required for tolvaptan effectiveness. This seems reasonable because tolvaptan’s mechanism of action includes inhibi- tion of water reabsorption and promotion of free water excretion without increasing Na excre- tion. To achieve maximal effect, both Na excretion and free water excretion are necessary. The present study have some limitations. Patients with TIPS were not included in the pres- ent study because TIPS insertion is not performed in our institute. The dose of diuretics was relatively low to moderate, because our policy was to avoid high dose of diuretics to avoid renal insufficiency. Especially, the dosage of spironolactone was particularly low compared to the recommendation by international societies. Although the response to tolvaptan was sim- ilar between patients with spironolactone <50mg vs. 50mg, it remains unclear whether the similar response to tolvaptan could be achieved by using higher dosage of spironolactone or whether the result of the present study will be applicable in patients receiving a higher dose of diuretics. Therefore, the predictive value of spot urine Na/K ratio for patients under higher dose of diuretics, or those treated by TIPS should be evaluated in the future study. We conclude that the pretreatment spot urine Na/K ratio can indicate the likelihood of effective tolvaptan treatment, and for ascites patients with insufficient response to conven- tional Na diuretic treatment, tolvaptan should be introduced when the urine Na/K is 2.5 to maximize the efficacy. This simple to perform, readily available criterion could serve to indi- cate the optimal timing of tolvaptan administration. Project administration: MK HN. Supervision: MK NI NE. Supervision: MK NI NE. Writing – original draft: Y. Komiyama HN. Writing – review & editing: MK NI NE. Conceptualization: MK HN. Formal analysis: Y. Komiyama HN. Investigation: MK HN Y. Komiyama NI YT JI YY NT HT MH TG Y. Kubota KT TH WO MO. Author Contributions Conceptualization: MK HN. Methodology: MK HN. Methodology: MK HN. Discussion defined a good patient response as a loss of 3 kg body weight loss on day 4 of tolvaptan therapy, and a 46.7% good response rate [12]. Kogiso et al. reported that long- term tolvaptan therapy of 6 months resulted in improvement of the extracellular fluid/total body water ratio in 78.6% of patients [13]. Furthermore, Oki et al. reported that tolvaptan ther- apy was effective in 63.3% of patients based on improvement in subjective symptoms and a weight loss of 2 kg [11]. In this study, 52% of patients experienced a weight loss of 2% with 1 week of tolvaptan therapy even though the existing diuretic regimen was not effective. Col- lectively, tolvaptan was effective in more than half of the patients with refractory ascites, and no patients discontinued treatment because of adverse events. Therefore, we believe that, for refractory hepatic ascites, tolvaptan is an effective treatment option. To date, there is no estab- lished criteria to define response to tolvaptan. Therefore, response to tolvaptan was tentatively defined as weight reduction of 2% at day 7 of treatment in the present study. By this tentative definition, symptoms such as abdominal distension or dyspnea was improved in 78% of patients with 2% reduction in their weight. Therefore, we believe that this tentative definition of response was clinically relevant. It would be very helpful to be able to identify those patients who are not likely to respond to tolvaptan in advance of treatment. Currently factors that predict therapeutic efficacy are unclear. Oki et al. reported a hazard ratio of 20.7 for achieving a 25% reduction in urine osmo- lality when tolvaptan therapy was effective [11]. This finding demonstrates that when tolvaptan is effective to inhibit water reabsorption, it promotes free-water excretion, leading to urine dilution and decrease in osmolality. Nakanishi et al. focused on urine AQP2, demonstrated that urinary AQP2/creatinine decreased after tolvaptan administration, and that the decrease was strongly correlated with decreased urine osmolality [6]. This showed that tolvaptan inhibi- tion of vasopressin V2 receptors can be assessed by urine AQP2. However changes in both PLOS ONE \| https://doi.org/10.1371/journal.pone.0174649 March 31, 2017 9 / 12 References 1. Gines P, Cardenas A, Arroyo V, Rodes J. Management of cirrhosis and ascites. The New England jour- nal of medicine. 2004; 350(16):1646–54. Epub 2004/04/16. https://doi.org/10.1056/NEJMra035021 PMID: 15084697 10 / 12 PLOS ONE \| https://doi.org/10.1371/journal.pone.0174649 March 31, 2017 Urinary Na/K ratio in tolvaptan treatment 2. Arroyo V, Gines P, Gerbes AL, Dudley FJ, Gentilini P, Laffi G, et al. Definition and diagnostic criteria of refractory ascites and hepatorenal syndrome in cirrhosis. International Ascites Club. Hepatology (Balti- more, Md). 1996; 23(1):164–76. Epub 1996/01/01. 3. Runyon BA. Introduction to the revised American Association for the Study of Liver Diseases Practice Guideline management of adult patients with ascites due to cirrhosis 2012. Hepatology (Baltimore, Md). 2013; 57(4):1651–3. Epub 2013/03/07. 4. Perez-Ayuso RM, Arroyo V, Planas R, Gaya J, Bory F, Rimola A, et al. Randomized comparative study of efficacy of furosemide versus spironolactone in nonazotemic cirrhosis with ascites. Relationship between the diuretic response and the activity of the renin-aldosterone system. Gastroenterology. 1983; 84(5 Pt 1):961–8. Epub 1983/05/01. PMID: 6339312 5. Stanley MM, Ochi S, Lee KK, Nemchausky BA, Greenlee HB, Allen JI, et al. Peritoneovenous shunting as compared with medical treatment in patients with alcoholic cirrhosis and massive ascites. Veterans Administration Cooperative Study on Treatment of Alcoholic Cirrhosis with Ascites. The New England journal of medicine. 1989; 321(24):1632–8. Epub 1989/12/14. https://doi.org/10.1056/ NEJM198912143212403 PMID: 2586565 6. Nakanishi H, Kurosaki M, Hosokawa T, Takahashi Y, Itakura J, Suzuki S, et al. Urinary excretion of the water channel aquaporin 2 correlated with the pharmacological effect of tolvaptan in cirrhotic patients with ascites. Journal of gastroenterology. 2016; 51(6):620–7. Epub 2015/11/28. https://doi.org/10.1007/ s00535-015-1143-3 PMID: 26610908 7. Konstam MA, Gheorghiade M, Burnett JC Jr., Grinfeld L, Maggioni AP, Swedberg K, et al. Effects of oral tolvaptan in patients hospitalized for worsening heart failure: the EVEREST Outcome Trial. Jama. 2007; 297(12):1319–31. Epub 2007/03/27. https://doi.org/10.1001/jama.297.12.1319 PMID: 17384437 8. Gheorghiade M, Konstam MA, Burnett JC Jr., Grinfeld L, Maggioni AP, Swedberg K, et al. Short-term clinical effects of tolvaptan, an oral vasopressin antagonist, in patients hospitalized for heart failure: the EVEREST Clinical Status Trials. Jama. 2007; 297(12):1332–43. Epub 2007/03/27. https://doi.org/10. 1001/jama.297.12.1332 PMID: 17384438 9. Schrier RW, Gross P, Gheorghiade M, Berl T, Verbalis JG, Czerwiec FS, et al. Tolvaptan, a selective oral vasopressin V2-receptor antagonist, for hyponatremia. The New England journal of medicine. 2006; 355(20):2099–112. Epub 2006/11/16. https://doi.org/10.1056/NEJMoa065181 PMID: 17105757 10. 20. Gines P, Arroyo V, Quintero E, Planas R, Bory F, Cabrera J, et al. Comparison of paracentesis and diuretics in the treatment of cirrhotics with tense ascites. Results of a randomized study. Gastroenterol- ogy. 1987; 93(2):234–41. Epub 1987/08/01. PMID: 3297907 22. Sakaida I, Yamashita S, Kobayashi T, Komatsu M, Sakai T, Komorizono Y, et al. Efficacy and safety of a 14-day administration of tolvaptan in the treatment of patients with ascites in hepatic oedema. The Journal of international medical research. 2013; 41(3):835–47. Epub 2013/05/21. https://doi.org/10. 1177/0300060513480089 PMID: 23685892 PLOS ONE \| https://doi.org/10.1371/journal.pone.0174649 March 31, 2017 21. Tsien CD, Rabie R, Wong F. Acute kidney injury in decompensated cirrhosis. Gut. 2013; 62(1):131–7. Epub 2012/05/29. https://doi.org/10.1136/gutjnl-2011-301255 PMID: 22637695 References Sakaida I, Kawazoe S, Kajimura K, Saito T, Okuse C, Takaguchi K, et al. Tolvaptan for improvement of hepatic edema: A phase 3, multicenter, randomized, double-blind, placebo-controlled trial. Hepa- tology research: the official journal of the Japan Society of Hepatology. 2014; 44(1):73–82. Epub 2013/04/05. 11. Ohki T, Sato K, Yamada T, Yamagami M, Ito D, Kawanishi K, et al. Efficacy of tolvaptan in patients with refractory ascites in a clinical setting. World journal of hepatology. 2015; 7(12):1685–93. Epub 2015/07/ 04. https://doi.org/10.4254/wjh.v7.i12.1685 PMID: 26140088 12. Akiyama S, Ikeda K, Sezaki H, Fukushima T, Sorin Y, Kawamura Y, et al. Therapeutic effects of short- and intermediate-term tolvaptan administration for refractory ascites in patients with advanced liver cir- rhosis. Hepatology research: the official journal of the Japan Society of Hepatology. 2015; 45 (11):1062–70. Epub 2014/11/29. 13. Kogiso T, Tokushige K, Hashimoto E, Ikarashi Y, Kodama K, Taniai M, et al. Safety and efficacy of long-term tolvaptan therapy for decompensated liver cirrhosis. Hepatology research: the official journal of the Japan Society of Hepatology. 2016; 46(3):E194–200. Epub 2015/07/01. 14. Karatapanis S, Ketikoglou I, Skorda L, et al. The role of spot urine Na+/K+ ratio in the management of ascites in cirrhotic patients. Gut. 2003; 52(Suppl VI)(A53). 15. Stiehm AJ, Mendler MH, Runyon BA. Detection of diuretic-resistance or diuretic-sensitivity by spot urine Na/K ratios in 729 specimens from cirrhotics with ascites: approximately 90 percent accuracy as compared to 24-hr urine Na excretion. Hepatology (Baltimore, Md). 2002; 36(222A). 16. El-Bokl MA, Senousy BE, El-Karmouty KZ, Mohammed Iel K, Mohammed SM, Shabana SS, et al. Spot urinary sodium for assessing dietary sodium restriction in cirrhotic ascites. World journal of gastroenter- ology. 2009; 15(29):3631–5. Epub 2009/08/05. https://doi.org/10.3748/wjg.15.3631 PMID: 19653340 17. Kanda Y. Investigation of the freely available easy-to-use software ’EZR’ for medical statistics. Bone marrow transplantation. 2013; 48(3):452–8. Epub 2012/12/05. https://doi.org/10.1038/bmt.2012.244 PMID: 23208313 18. Boyer TD, Warnock DG. Use of diuretics in the treatment of cirrhotic ascites. Gastroenterology. 1983; 84(5 Pt 1):1051–5. Epub 1983/05/01. PMID: 6832557 19. Bernardi M, De Palma R, Trevisani F, Santini C, Servadei D, Gasbarrini G. Comparative pharmacody- namics of furosemide and muzolimine in cirrhosis. Study on renal sodium and potassium handling and renin-aldosterone axis. Zeitschrift fur Kardiologie. 1985; 74 Suppl 2:129–34. Epub 1985/01/01. PLOS ONE \| https://doi.org/10.1371/journal.pone.0174649 March 31, 2017 11 / 12 Urinary Na/K ratio in tolvaptan treatment 21. Tsien CD, Rabie R, Wong F. Acute kidney injury in decompensated cirrhosis. Gut. References 2013; 62(1):131–7. Epub 2012/05/29. https://doi.org/10.1136/gutjnl-2011-301255 PMID: 22637695 12 / 12
https://openalex.org/W3210716117	https://smic.in.ua/index.php/journal/article/download/354/439	Ukrainian	null	THE EFFICIENCY OF MICROBIAL PREPARATIONS, MACRO-, MICROELEMENTS AND HERBICIDES USE FOR SOYBEAN CULTIVATION	Sìlʹsʹkogospodarsʹka mìkrobìologìâ	2,009	cc-by	3,971	УДК: 579.64:631.52 УДК: 579.64:631.52 Ефективність застосування мікробних препаратів, макро- і мікроелементів та гербіцидів ПРИ вирощуваннІ сої Дерев’янський �� В.П., Власюк О.С. Хмельницька державна сільськогосподарська дослідна станція УААН, вул. Самчики, 1, с. Самчики, Старокостянтинівський район, Хмельницька область, 31182, Україна E-mail: elita@sk.km.ua Вивчали вплив комплексу факторів (обробка насіння та посівів мікробними препаратами, позакореневого внесення макро- і мікроелементів на фоні ґрунтового та післясходового застосування гербіцидів) на ріст і розвиток рослин сої. Виявлено композиції, які дозволяють підвищити кількість бульбочкових бактерій на кореневій системі рослини, зменшити поширення хвороб, підвищити продуктивність, покращити якість продукції. і б б д Ключові слова: соя, мікробні препарати, гербіциди, макро- і мікроелементи, хвороби, продуктивність, якість. Мікробні препарати відіграють все більш значну роль у процесі формування врожайності сільськогосподарських культур. Бактерії, що заселяють коріння, є трофічними посередниками між ґрунтом і рослиною, відповідальними за перетворення складних хімічних сполук у прості й доступні для живлення рослин. Рослина в оточенні повноцінного комплексу мікроорганізмів одержує необхідне кореневе живлення і, як наслідок, повніше реалізує свій генетичний потенціал щодо врожайності [1, 2, 3, 4, 5, 6, 7]. р У більшості ґрунтів сьогодні окремі мікроорганізми, які вважались індикаторами родючості, знаходяться на межі зникнення. При цьому молоде коріння рослин заселяють нетипові мікроорганізми, які конкурують з ними за елементами живлення. Внаслідок цього культури не формують повноцінного урожаю. У більшості ґрунтів сьогодні окремі мікроорганізми, які вважались індикаторами родючості, знаходяться на межі зникнення. При цьому молоде коріння рослин заселяють нетипові мікроорганізми, які конкурують з ними за елементами живлення. Внаслідок цього культури не формують повноцінного урожаю. При інтродукції нових видів культурних рослин, наприклад, сої і козлятника, на нових територіях неможливо забезпечити їх азотне живлення за рахунок “біологічного” азоту без проведення передпосівної бактеризації насіння. Відсутність необхідних азотфіксувальних бактерій у ґрунті зводить значення цих бобових 104 культур як азотнакопичувачів до рівня азотвитратних [18, 19, 20, 21, 22, 23]. У зв’язку з цим виникає потреба в застосуванні агро- прийомів, спрямованих на збільшення кількості агрономічно- цінних мікроорганізмів у ґрунтах, одним з яких є застосування передпосівної інокуляції насіння сільськогосподарських культур. У боротьбі з бур’янами слід враховувати кліматичні зміни в останні роки. Так, різке зростання температури після сівби сої сприяло масовій появі майже усіх типів бур’янів. Навіть теплолюбні паслін, щириця, мишій і куряче просо проростали майже одночасно з більш холодостійкими бур’янами. Появу ж нової хвилі бур’янів спричиняли опади [8, 9, 10, 11, 12]. Відмічається тенденція до щорічного збільшення ураженості посівів хворобами листя, розвиток яких спричинений запасом інфекцій у ґрунтах на фоні недостатнього забезпечення рослин елементами живлення [11, 24, 26]. Стабільне і продуктивне функціонування агроценозів можливе за особливої уваги проблемі захисту рослин від шкідливих організмів (зокрема збудників хвороб), життєдіяльність яких спри- чиняє значні втрати урожаю. Протягом тривалого часу в практиці сільськогосподарського виробництва перевагу віддають хімічному методу захисту рослин. Ефективність застосування мікробних препаратів, макро- і мікроелементів та гербіцидів ПРИ вирощуваннІ сої Дерев’янський �� В.П., Власюк О.С. Хмельницька державна сільськогосподарська дослідна станція УААН, вул. Самчики, 1, с. Самчики, Старокостянтинівський район, Хмельницька область, 31182, Україна E-mail: elita@sk.km.ua Однак, постійно зростаюче застосування пестицидів призводить до забруднення довкілля, появи стійких штамів і популяцій патогенів та шкідників, частота виникнення яких випереджає створення нових хімічних препаратів. У зв’язку з цим актуальність розвитку біологічних методів захисту рослин, які базуються на використанні природних агентів біологічної регуляції шкідливих видів, не викликає сумніву. Метою наших досліджень було вивчення впливу комплексу факторів на рівень ураження хворобами, забур’янення посівів та продуктивність сої, а також, визначення економічно вигідних та екологічно безпечних технологій захисту рослин. Матеріали і методи. Дослідження проводили протягом 2006-2008 років на Хмельницькій державній сільськогосподарській дослідній станції УААН. Ґрунт дослідного поля – чорнозем опідзолений середньо- суглинковий, слабозмитий. Агрохімічні показники (0-30 см): гумус за Тюріним – 3,2-3,6; рН (сольове) – 5,5-6,0; азот легкогідролізова- ний – 12-17 мг на 100 г. ґрунту, рухомий фосфор – 13-18,5; обмін- 105 105 ний калій – 10,0-11,1 мг на 100 г. ґрунту. ний калій – 10,0-11,1 мг на 100 г. ґрунту. ний калій – 10,0-11,1 мг на 100 г. ґрунту. Схема досліду: І. Чинник “А” – захист від бур’янів: 1. Внесення ґрунтового гербіциду Харнес – 3,0 л/га (фон 1); 2. Внесення післясходового гербіциду Півот – 1,0 л/га (фон 2). ІІ. Чинник “В” – обробка насіння перед сівбою суспензіями бактерій з розрахунку 200 тис. клітин на насінину: 1. Без бактеризації насіння; 2. Bradyrhizobium japonicum 614А; 3. Bradyrhizobium japonicum 614А + Bacillus pumilis 1; 4. Bradyrhizobium japonicum 614А + Bacillus subtilis 2. ІІІ. Чинник “С” – обприскування посівів сої суспензією біоактивних препаратів у фазу 3-4 справжніх листків: 1. Без обробки посівів; 2. Хетомік (0,2 л/га); 3. Еколист стандарт (3 л/га); 4. Хетомік (0,2 л/га)+ Еколист стандарт (3 л/га). Усі культури бактерій отримано з колекції корисних ґрунто- вих мікроорганізмів Інституту сільськогосподарської мікробіоло- гії УААН. Сорт сої – Устя. Загальна площа ділянки у досліді – 100 м2, облікова площа Загальна площа ділянки у досліді – 100 м2, облікова площа – 72 м2, повторність триразова, розміщення ділянок систематичне. щ д у д д , щ – 72 м2, повторність триразова, розміщення ділянок систематичне. 72 м2, повторність триразова, розміщення ділянок систематичне Агротехніка у досліді загальноприйнята для західного Лісостепу. Під передпосівну культивацію вносили мінеральні добрива з розрахунку N32P32K32. Норма висіву сої – 900 тис. насінин на 1 га, звичайно-рядковий спосіб сівби (15 см). Агротехніка у досліді загальноприйнята для західного Лісостепу. Під передпосівну культивацію вносили мінеральні добрива з розрахунку N32P32K32. Норма висіву сої – 900 тис. насінин на 1 га, звичайно-рядковий спосіб сівби (15 см). Кліматичні та метеорологічні умови в 2006-2008 роках були сприятливі для вирощування сої. Середньорічна температура по- вітря за вегетаційний період травень-вересень 2006 року складала 18,5 0С, 2007 – 18,7 0С. Сума опадів за 9 місяців становила в 2006 р. – 893 мм, у 2007 р. – 926 мм. Сума опадів за травень-вересень складала в 2006 р. – 695 мм, у 2007 р. – 769,4 мм. Температурний режим квітня 2008 року був у межах норми з дещо більшою кількістю опадів (+170 мм до середньодобової багаторічної). 106 Травень 2008 року був значно теплішим від середньорічних показників з надмірною кількістю опадів. Починаючи з травня створюються досить сприятливі умови для росту і розвитку сої, завдяки високим середньодобовим температурам. Так, середньодобова температура травня була вищою від середньорічного значення на +2,7 0С, червня – на +2,2 0С, липня – на +1,7 0С, серпня – на +2,7 0С. Липень і серпень характеризувалися надмірною кількістю опадів, однак переважна більшість з них мала зливовий характер. ний калій – 10,0-11,1 мг на 100 г. ґрунту. Середньодобові температури вересня були близькими до середньорічних значень з надмірною кількістю опадів (+155,7 мм до середньої багаторічної норми). Дощовими були друга та третя декади. Сума опадів за травень-вересень складала 769,4 мм. Такі ґрунтові та кліматичні умови 2006-2008 років дали можливість в оптимальні строки провести сівбу сої, догляд за посівами і отримати оптимальну врожайність культур. Обліки та спостереження проводились за загальноприйнятими методиками [18, 27]. Результати та їх обговорення. Рівень урожайності сої значною мірою залежить від ефективності заходів захисту її посівів від бур’янів. У зв’язку із загальним зниженням культури землероб- ства та низьким рівнем технічного забезпечення сільськогоспо- дарського виробництва часто допускаються порушення строків та якості проведення технологічних операцій, що спричиняє погіршення умов вирощування сої. Обстеження полів свідчить про зростання засміченості орного шару ґрунту багатьма видами бур’янів. 107 ур У 2006-2008 рр. у польовому досліді вивчали фітотоксичний вплив на кількісний та видовий склад бур’янів ґрунтового та післясходового гербіцидів Харнес та Півот. Найбільш чітко гербіцидну активність препаратів та ефективність способу їх внесення відображають дані кількості, складу та наростання маси бур’янів. Облік, проведений перед збиранням врожаю, показав, що при внесенні в ґрунт Харнесу (3,0 л/га) кількість бур’янів та їх маса були найменшими, порівняно з післясходовою обробкою посівів гербіцидом Півотом (1,0 л/га). Харнес зменшував кількість бур’янів протягом всього періоду вегетації сої. На час збирання врожаю загальна кількість бур’янів була знижена на 96-98 %. Використання Півоту (1,0 л/га) забезпечило зниження рівня загальної забур’яненості на 87-92 %. Обліки, проведенні у фазу сходів та перед збиранням врожаю, свідчать, що ґрунтовий гербі- 107 р цид дещо зменшував густоту стояння рослин сої. Так, при обприскуванні посівів Півотом густота стояння становила 736 тис. рослин на 1 га, тоді як на ділянках, де вносили Харнес, – 710 тис. Проте, при застосуванні Харнесу обробка насіння біопрепаратами підвищила густоту стояння рослин на 11,8-25,4 %, тоді як при внесенні післясходового гербіциду Півот такого явища не спостерігали. Обробка насіння бактеріальними препаратами сприяла збільшенню кількості бульбочок у базальній частині кореня рослин сої. Середня кількість бульбочок однієї інокульованої рослини становила 39-42 од. проти 4-6 од. – у варіанті без інокуляції. У 2008 році передпосівна обробка насіння Bradyrhizobium japonicum 614A + Bacillus pumilis 1 зменшувала поширення пероноспорозу на 22 %, Br. japonicum 614A + B. subtilis 2 – на 27 %, тоді як обробка насіння Br. japonicum 614A + B. pumilis 1 + обробка посівів Хетоміком на фоні внесення ґрунтового гербіциду Харнесу (3,0 л/га) – на 42 %, Br. japonicum 614A + B. ний калій – 10,0-11,1 мг на 100 г. ґрунту. subtilis 2 + обробка посівів Хетоміком на фоні Харнесу – на 51 % порівняно з контролем (без обробки насіння та посівів). Зменшення рівня забур’яненості при застосуванні Харнесу та Півоту, обробка насіння бактеріальними препаратами та обробка посівів Хетоміком з позакореневим підживленням Еколистом стандарт створювали сприятливі умови для росту і розвитку, живлення рослин, підвищували стійкість до пероноспорозу та сприяли формуванню урожайності на 4,8-8,0 ц/га вищої, ніж у контрольному варіанті. Максимальний приріст продуктивності сої досягається при застосуванні Харнесу + обробки насіння Br. japonicum 614A + B. subtilis 2 + обробки посівів Хетоміком з позакореневим підживленням Еколист стандарт (8,0 ц/га або 28,2 %). Аналіз урожайних даних показує, що на обох фонах внесення ґрунтового та післясходового гербіцидів передпосівна обробка насіння Br. japonicum 614A + B. subtilis 2 та обробка насіння Br. japonicum 614A + B. pumilis 1 з позакореневим підживленням Еколист стандарт є ефективнішими порівняно з використанням тільки передпосівної обробки насіннєвого матеріалу. 108 Структурний аналіз, проведений у лабораторних умовах, показує, що на кінець вегетаційного періоду середня висота рослин по досліду дорівнювала 84 см. Мінімальною ця величина була у контролі й становила 68 см. Висота кріплення нижніх бобів у середньому по досліду дорівнює 12,2 см, що відповідає технологічним вимогам збирання комбайном “Нива”. В середньому по досліду на одній рослині бобів налічується 30,2 од., з однієї рослини вихід здорових насінин коливається від 30 до 95 од. (у середньому по досліду – 57 од.), тобто на кожний добре розвинений біб у середньому припадає по 1,3 кондиційних насінин. Маса насінин з однієї рослини, в середньому по досліду, становить 10,2 г, маса 1000 насінин дорівнює 169 г. Комплексна обробка насіння і посівів на фоні внесення Харнесу в найбільшій мірі впливає на формування вегетативних та генеративних органів рослин. Проведений структурний аналіз рослин сої показав, що на висоту рослин усі вказані бактеріальні препарати впливали приблизно однаково. Висота кріплення нижнього бобу залежала від типу бактеріального препарату та способу внесення. Так, у рослин сої, оброблених Br. japonicum 614A + B. subtilis 2 + обробка посівів на фоні внесення Харнесу, висота кріплення нижнього бобу збільшувалася на 13 %, під дією Br. japonicum 614A + B. pumilis – на 12 %. Кількість бобів на одній рослині збільшувалася за оброблення насіння та посівів на 10 %. Кількість насінин на одній рослині була більшою за комплексної обробки препаратами порівняно з контролем. Аналізуючи показники урожайності (табл.), отримані за роки досліджень (2006-2008 рр.), встановлено, що кращим є варіант, де приріст урожайності становив 25,8 %: інокуляція насіння Br. ний калій – 10,0-11,1 мг на 100 г. ґрунту. japonicum 614A + B. subtilis 2 + обробка посівів Хетоміком + позакореневе підживлення Еколист стандарт на фоні внесення ґрунтового гербіциду Харнес. Однією з основних вимог сучасного сільськогосподарського виробництва є зниження витрат на одиницю отриманої продукції. Результати досліджень дозволили встановити, що застосування комплексного мікробного оброблення насіння та посівів на фоні внесення ґрунтового гербіциду підвищувало врожайність насіння на 6,3 ц/га. Вартість приросту продукції при цьому становила 1260 грн/га. Витрати на придбання препаратів та обробку насіння, посівів та внесення гербіцидів становили 323,5 грн/т, на збирання, перевезення та очищення додаткової продукції – 38,0 грн/га. Разом витрати на обробку, збирання та очищення насіння сої становили 369 грн/га. Отже, чистий прибуток становив 898 грн/га, собівар- тість 1 ц – 20,0 грн/ц, рівень рентабельності – 248 %. 109 Таблиця. Вплив біопрепаратів та гербіцидів на продуктивність сої (2006-2008 рр.) Таблиця. Вплив біопрепаратів та гербіцидів на продуктивність сої (2006-2008 рр.) Варіанти досліду Урожайність, ц/га Приріст до контролю фон І фон ІІ 2006 2007 2008 середнє ц/га % ц/га % 1 2 3 4 5 6 7 8 9 Контроль 1 (фон І) внесення Харнесу, без обробки насіння та без обробки посівів 14,2 19,8 20,4 18,1 – – – – Фон І + обробка насіння Br. japonicum 614A 16,0 21,2 21,8 19,7 1,6 8,1 – – Фон І + обробка насіння Br. japonicum 614A + B. pumilis 1 16,5 21,7 22,7 20,3 2,2 10,8 – – Фон І + обробка посівів Br. japonicum 614A + B. subtilis 2 17,7 22,5 23,8 21,3 3,2 15,0 – – Фон І + обробка посівів Хетоміком (без обробки насіння) 16,2 20,3 21,5 19,3 1,2 6,2 – – Фон І + обробка насіння Br. japonicum 614A + B. pumilis 1 + обробка посівів Хетоміком 17,3 23,5 25,6 22,1 4,0 18,1 – – Фон І + обробка насіння Br. japonicum 614A + B. subtilis 2 + обробка посівів Хетоміком 17,6 24,8 26,7 23,0 4,9 21,3 – – Фон І + обробка посівів Еколист стандарт (без обробки насіння) 16,1 20,6 21,8 19,5 1,4 7,2 – – Фон І + обробка насіння Br. japonicum 614A + обробка посівів Еколист стандарт 17,2 21,8 22,9 20,6 2,5 12,1 – – 110 продовження таблиці 1 2 3 4 5 6 7 8 9 Фон І + обробка насіння Br. japonicum 614A + B. pumilis 1 + обробка посівів Еколист стандарт 18,0 22,7 23,7 21,5 3,4 15,8 – – Фон І + обробка насіння Br. japonicum 614A + B. ний калій – 10,0-11,1 мг на 100 г. ґрунту. subtilis 2 + обробка посівів Еколист стандарт 18,4 23,5 25,3 22,4 4,3 19,2 – – Фон І + обробка посівів Хетомік + Еколист стан дарт (без обробки насіння) 16,6 21,4 22,9 20,3 2,2 10,8 – – Фон І + обробка насіння Br. japonicum 614A + об робка посівів Хетоміком + Еколист стандарт 18,1 23,9 25,2 22,4 4,3 19,2 – – Фон І + обробка насіння Br. japonicum 614A + B. pumilis 1 +обробка посівів Хетоміком + Еколист стандарт 18,4 25,9 27,3 23,9 5,8 24,3 – – Фон І + обробка насіння Br. japonicum 614A + B. subtilis 2 + обробка посівів Хетоміком + Еколист стандарт 18,6 26,3 28,4 24,4 6,3 25,8 – – Контроль 2 (фон ІІ) внесення Півоту, без обробки насіння та без обробки посівів 16,0 18,8 19,8 18,2 0,1 0,5 0 0 Фон ІІ + обробка насіння Br. japonicum 614A 16,8 19,4 21,5 19,2 1,1 5,7 1,0 5,2 Фон ІІ + обробка насіння Br. japonicum 614A + B. pumilis 1 17,1 20,1 22,9 20,0 1,9 9,5 1,8 9,0 Фон ІІ + обробка насіння Br. japonicum 614A + B. subtilis 2 16,2 20,9 23,7 20,3 2,2 10,8 2,1 10,3 111 продовження таблиці 1 2 3 4 5 6 7 8 9 Фон ІІ + обробка посівів Хетоміком (без обробки насіння) 15,8 19,0 22,1 19,0 0,9 4,7 0,8 4,2 Фон ІІ + обробка насіння Br. japonicum 614A + об робка посівів Хетоміком 17,4 20,7 21,6 19,9 1,8 9,0 1,7 8,5 Фон ІІ + обробка насіння Br. japonicum 614A + B. pumilis 1 + обробка посівів Хетоміком 18,5 21,6 22,4 20,8 2,7 13,0 2,6 10,3 Фон ІІ + обробка посівів Еколистом (без обробки насіння) 16,9 19,8 20,1 18,9 0,8 4,2 0,7 3,7 Фон ІІ + обробка насіння Br. japonicum 614A + обробка посівів Еколист стандарт 18,1 20,6 21,8 20,2 2,1 10,4 2,0 9,9 Фон ІІ + обробка насіння Br. japonicum 614A + B. pumilis 1 + обробка посівів Еколист стандарт 19,2 21,8 23,4 21,5 3,4 15,8 3,3 15,3 Фон ІІ + обробка насіння Br. japonicum 614A + B. subtilis 2 +обробка посівів Еколистом 20,1 22,4 24,8 22,4 4,3 19,2 4,2 18,8 Фон ІІ + без обробки насіння +обробка посівів Хетоміком + Еколист стандарт 17,4 20,9 21,6 20,0 1,9 9,5 1,8 9,0 Фон ІІ + обробка насіння Br. japonicum 614A + об робка посівів Хетоміком + Еколист стандарт 19,2 21,5 23,2 21,3 3,2 15,0 3,1 14,6 Фон ІІ + обробка насіння Br. ний калій – 10,0-11,1 мг на 100 г. ґрунту. japonicum 614A + B. pumilis 1 + обробка посівів Хетоміком + Еколист стандарт 20,3 22,7 24,8 22,6 4,5 19,9 4,4 19,5 112 продовження таблиці 1 2 3 4 5 6 7 8 9 Фон ІІ + обробка насіння Br. japonicum 614A + B. subtilis 2 + обробка посівів Хетоміком + Еколист стандарт 21,2 23,6 26,3 23,7 5,6 23,6 5,5 23,2 НІР0,5, ц/га фактор А – В – С – взаємодія АВ – АС – ВС – Р, % – 0,38 0,37 0,29 0,51 0,51 0,58 0,32 0,26 0,31 0,31 0,44 0,44 0,54 0,29 0,15 0,22 0,22 0,30 0,30 0,43 0,16 Таким чином, при вирощуванні сої в умовах західного Лісо- степу пропонується використовувати комплекс біологічних препаратів: передпосівну обробку насіння азотфіксувальними бактеріями Br. japonicum 614A і фосфатмобілізувальними бакте- ріями Bacillus subtilis 2, які мають також антагоністичні властивості, та обробку посівів біофунгіцидом Хетомік з позакореневим підживленням препаратом Еколист стандарт. При застосуванні гербіциду Півот забур’яненість посівів знижується на 87-92 %, Харнес – на 96-98 %. Інокуляція насіння мікробними препаратами в поєднанні з обробкою посівів Хетоміком з подальшим позакореневим внесен- ням макро- і мікроелементів дає змогу захистити посіви сої від хвороб та сформувати потрібну густоту рослин. Комплексна обробка насіння та вегетуючих рослин біопрепаратами та макро- і мікроелементами стимулює розвиток вегетативних та генеративних органів рослин, забезпечує приріст урожайності на 25,8 %. Економічні затрати на гербіциди, біопрепарати та макро- і мікроелементи складають 20-22 % від загальних витрат на вирощування культури, а за рахунок зростання урожайності окуповуються у 5-6 разів. 113 113 1. Бровдій В.М. Біологічний захист рослин: Навч. посібник / В.М. Бровдій, В.В. Гулий, В.П. Федоренко. – К.: Світ, 2003. – 352 с. 2. Мікробні препарати у землеробстві. Теорія і практика: Моногра- фія /[В.В. Волкогон, О.В. Надкернична, Т.М. К овалевська, Л.М. Токма- кова та ін.]. – К.: Аграрна наука, 2006. – 312 с. К 3. Воцелко С.К. Роль обробки насіння бобових культур біологіч- ним препаратом ЕПАА у підвищенні імунітету та продуктивності рослин /С.К. Воцелко, С.В. Лапа, Л.А. Д анькевич //Міжвід. темат. наук. зб. С.-г. мікробіологія. – Чернігів, 2007. – Вип. 5. – С. 161-170. 4. К рючкова Л.О. Стимулювання ростових процесів та підвищен- ня стійкості проти хвороб у проростках озимої пшениці під впливом регуляторів росту природного походження /Л.О. К рючкова, Т.І. Маковей- чук //Міжвід. темат. наук. зб. С.-г. мікробіологія. – Чернігів, 2007. – Вип. 5. – С. 153-160. 5. Малиновська І.М. Стан мікробоценозу ризосфери сої за ком- плексного оброблення насіння фосфатмобілізуючими мікроорганізмами і Br. japonicum НТ /І.М. ний калій – 10,0-11,1 мг на 100 г. ґрунту. Малиновська //Агроекол. журн. – 2007. – № 3. – С. 79-83. 6. Д ерев’янський В.П. Поширення хвороб та продуктивність сої / В.П. Д ерев’янський //Карантин і захист рослин. – 2007. – № 5. – С. 11-14. 7. Богач Г.І. Біофунгіциди для обробки насіння /Г.І. Богач, О.Г. Бо- гач //Карантин і захист рослин. – 2007. – № 9. – С. 7-8. р р 8. Іващенко О.О. Геном рослин бур’янів за дії гербіцидів / О.О. Іващенко //Карантин і захист рослин. – 2007. – № 9. – С. 17-18. р р 9. Жеребко В.М. Гербіциди в інтегрованому захисті /В.М. Жереб- ко //Карантин і захист рослин. – 2007. – № 7. – С. 12-14. р р р у р ко //Карантин і захист рослин. – 2007. – № 7. – С. 12-14. 10. Ермантраут Е.Р., Умрихін Н.Л. Ефективність способів контро- лювання забур’яненості посівів цукрових буряків /Е.Р. Ермантраут, р р 10. Ермантраут Е.Р., Умрихін Н.Л. Ефективність способів контро- лювання забур’яненості посівів цукрових буряків /Е.Р. Ермантраут, Н.Л. Умрихін //Цукрові буряки. – 2007. – № 3. – С. 13-14. Н.Л. Умрихін //Цукрові буряки. 2007. № 3. С. 13 14. 11. Роїк М.В. Біологізація технологічних процесів на виробництві цукрових буряків /М.В. Роїк, Н.В. З аіменко, В.О. Борисик, В.І. Пиркін, В.О. Гаврилов //Цукрові буряки. – 2007. – № 3. – С. 15-17. 12. Параментная Л.Н. Симбиотические отношения Rhizobium и бобовых растений при обработке гербицидами /Л.Н. Параментная //Биология ВНШ с.-х. микробиологии. – 1999. – № 32. – С. 11-13. 13. Агрономічний аналіз. Підручник //[М.М. Городній, А.П. Лісо- вий, А.В. Бикін та ін.]. – К.: Арістей, 2005. – 468 с. �� р 14. �� Балакай Г.Г. Соя: Экология, агротехника, переработка /Г.Г. Ба- лакай, О.С. Безуглова. – Ростов-на-Дону: Феникс, 2003. – 160 с.�� 15.�� Mano T. Early events in environmental stress �� /T. Mano //Oxidative stress in plants /Eds. D. Inze, M. Van Montagn. – London: Taybor and Fransis, 2002. – P. 217-245. 114 16. Біологічний азот: Монографія /[В.П. Патика, С.Я. К оць, В.В. Волкогон, О.Ф. Шерстобоєва, Т.М. Мельничук, А.В. К алініченко, І.В. Гриник]; за ред. В.П. Патики. – К.: Світ, 2003. – 424 с. 17. �� Агроэкологическая роль азотфиксирующих микроорганизмов в аллелопатии высших растений /[Патыка В.Ф., Наумов Г.Ф., Подоба Л.В., Николаенко А.Н., Поташова А.Н., Ельникова В.А., Гриник Н.В.]; под ред. В.Ф. Патыки. – К.: Основа, 2004. – 320 с. 18. Рекомендації з ефективного застосування мікробних препара- тів у технології вирощування сільськогосподарських культур /[С.І. Мель- ник, В.А. Жилкін, М.М. ний калій – 10,0-11,1 мг на 100 г. ґрунту. Гаврилюк та ін.]. – К., 2007. – 52 с. 19. Адамень Ф.Ф. Агробиологические особенности возделывания сои в Украине /Ф.Ф. Адамень, В.А. Вергунов, П.Н. Лазер, И.Н. Вергунова. – К.: Аграрная наука, 2006. – 456 с. 20. Іутинська Г.О. Вплив деяких антропогенних факторів на росто- ву активність мікрофлори ґрунту /Іутинська Г.О., Антипчук А.Ф. // Онтогенез рослин, біологічна фіксація молекулярного азоту та азотний метаболізм: Матер. Міжнар. наук. конф. – Тернопіль, 2001. – С. 214-217. ДД 21. К ругова О.Д. Віталист стимулює посіви сої /Кругова О.Д. , Мандровська Н.М., Бублик Л.І., Чернега О.Д. , Журавська Г.С., Гуль- чук Н.Ф. //Карантин і захист рослин. – 2008. – № 7. – С. 19-20. 22. Голодрига О.В. Активність ґрунтової мікрофлори /Голодри- га О.В., Грицаєнко З.М. //Карантин і захист рослин. – 2008. – № 1. – С. 18- 20. 23. К алінчик М.В. Формування ринку сої та продуктів її перероб- ки в Україні /М.В. К алінчик, М.М. Ільчук, О.В. Бєляєва, В.Н. Тимченко, А.В. Пилипченко, В.А. Сонець. – К.: ЗАТ “Ніглава”, 2007. – 200 с. 24. Шендрик Р.Я. Розвиток хвороб цукрових буряків у 2007 році та вірогідність їхньої появи у 2008 році /Шендрик Р.Я., Запольська Н.М. // Цукрові буряки. – 2008. – № 1 (61). – С. 7-8. Цу р ур ( ) 25. Іващенко О.О. Гербологія: напрями досліджень /Іващенко О.О. //Захист рослин. – 2000. – № 2. – С. 3-4. 26. Система заходів посівного комплексу для польових культур: Навч. посібник /[Щербаков В.Я., Лазер П.Н., Яковенко Т.М. та ін.]. – Херсон: Айлант, 2006. – 396 с. 27. Методи випробування і застосування пестицидів //[С.О. Три- бель, Д.Д. Сігарьова, М.П. Секун, О.О. Іващенко та ін.]; за ред. проф. С.О. Трибеля. – К.: Світ, 2001. – 448 с. 27. Методи випробування і застосування пестицидів //[С.О. Три- бель, Д.Д. Сігарьова, М.П. Секун, О.О. Іващенко та ін.]; за ред. проф. С.О. Трибеля. – К.: Світ, 2001. – 448 с. 27. Методи випробування і застосування пестицидів //[С.О. Три- бель, Д.Д. Сігарьова, М.П. Секун, О.О. Іващенко та ін.]; за ред. проф. С.О. Трибеля. – К.: Світ, 2001. – 448 с. 115 115 ЭФФЕКТИВНОСТЬ ПРИМЕНЕНИЯ МИКРОБНЫХ ПРЕПАРАТОВ, МАКРО-, МИКРОЭЛЕМЕНТОВ И ГЕРБИЦИДОВ ПРИ ВЫРАЩИВАНИИ СОИ Деревянский В.П., Власюк О.С. Хмельницкая государственная сельскохозяйственная опытная станция УААН Изучали влияние комплекса факторов (обработка семян и посевов микробными препаратами, внекорневое внесение макро- и микроэлементов на фоне почвенного и послевсходового применения гербицидов) на рост и развитие растений сои. Выявлены композиции, которые позволят увеличить количество клубеньковых бактерий на корневой системе растений, уменьшить распространение болезней, повысить продуктивность, улучшить качество продукции. Ключевые слова: соя, микробные препараты, гербициды, макро- и микроэлементы, болезни, продуктивность, качество. THE EFFICIENCY OF MICROBIAL PREPARATIONS, MACRO-, MICROELEMENTS AND HERBICIDES USE FOR SOYBEAN CULTIVATION Derevyanskiy V.P., Vlasyuk O.S. Hmelinickaya State Agricultural Experimental Station of UAAN The influence of the complex seeds and field treatment of soybean with microbial preparations, as well as top-dressing with macro- and microelements on growing and development of this culture was studied. Revealed efficient compositions lead to the reduction of the diseases spreading, productivity and seed quality improvement. Key words: soybean, microbial preparation, herbicides, macro- and microelements, productivity, seed quality. 116 116
https://openalex.org/W3197864094	https://bmcmusculoskeletdisord.biomedcentral.com/counter/pdf/10.1186/s12891-021-04632-8	English	null	Does meniscectomy have any advantage over conservative treatment in middle-aged patients with degenerative medial meniscus posterior root tear?	BMC musculoskeletal disorders	2,021	cc-by	5,799	Lee et al. BMC Musculoskeletal Disorders (2021) 22:742 https://doi.org/10.1186/s12891-021-04632-8 Lee et al. BMC Musculoskeletal Disorders (2021) 22:742 https://doi.org/10.1186/s12891-021-04632-8 Open Access Does meniscectomy have any advantage over conservative treatment in middle- aged patients with degenerative medial meniscus posterior root tear? Does meniscectomy have any advantage over conservative treatment in middle- aged patients with degenerative medial meniscus posterior root tear? Nam-Hun Lee1†, Hyoung-Yeon Seo2†, Myung-Jin Sung1, Bo-Ram Na1, Eun-Kyoo Song1 and Jong-Keun Seon1* Nam-Hun Lee1†, Hyoung-Yeon Seo2†, Myung-Jin Sung1, Bo-Ram Na1, Eun-Kyoo Song1 and Jon Abstract Background: The best treatment for degenerative medial meniscus posterior root tear (MMPRT) remains controversial. This study aimed to compare the clinical and radiological outcomes of arthroscopic meniscectomy and conservative treatment for degenerative MMPRT. Methods: From January 2007 to December 2014, 146 patients (Meniscectomy group, 90; Conservative group, 56) were evaluated. Clinical outcomes were assessed using the Visual Analog Scale, International Knee Documentation Committee subjective scoring scale, Tegner activity scale, and Lysholm knee scoring scale at the final follow-up. Radiologic outcomes evaluated the progression of osteoarthritis (OA) according to the Kellgren-Lawrence (K-L) classification. We compared the hip-knee-ankle angle (HKAA), medial proximal tibial angle, tibial posterior slope angle, and width of medial joint space. After an average follow-up of 6.3 years, the survivorship was analyzed using the Kaplan–Meier method. Results: All clinical outcomes were significantly improved in both groups after treatment, with no significant differences between the two groups at the final follow-up. The progression of OA according to the K-L classification, HKAA and width of medial joint space was significantly advanced in the meniscectomy group (p = 0.03, 0.04, 0.03, respectively). The 10-year survival rates in the meniscectomy and conservative groups were 87 and 88%, respectively. Conclusions: This study demonstrated that both conservative treatment and meniscectomy provided symptomatic relief. However, it was confirmed that OA progression was more severe in the meniscectomy. We conclude that arthroscopic meniscectomy had no advantage over conservative treatment in terms of clinical outcomes and OA progression in middle-aged patients with MMPRT. Level of evidence: Level III; retrospective comparative study. al meniscus posterior root tear, Arthroscopic meniscectomy, Conservative treatment, Osteoarthritis Keywords: Medial meniscus posterior root tear, Arthroscopic meniscectomy, Conservative treatm rds: Medial meniscus posterior root tear, Arthroscopic meniscectomy, Conservative treatment, Oste * Correspondence: seonbell@jnu.ac.kr Nam-Hun Lee and Hyoung-Yeon Seo contribute to this article equally. 1Department of Orthopaedic Surgery, Chonnam National University Hwasun Hospital and Medical School, 322 Seoyang-ro, Hwasun-gun, Chonnam 58218, Republic of Korea Full list of author information is available at the end of the article * Correspondence: seonbell@jnu.ac.kr Nam-Hun Lee and Hyoung-Yeon Seo contribute to this article equally. 1Department of Orthopaedic Surgery, Chonnam National University Hwasun Hospital and Medical School, 322 Seoyang-ro, Hwasun-gun, Chonnam 58218, Republic of Korea Full list of author information is available at the end of the article Background patients diagnosed with meniscus root tear from January 2007 to December 2014. MMPRT was defined as a radial tear within 9 mm of the posterior bony attachment of the medial meniscus or posterior root avulsion was diag- nosed on MRI by the absence of an identifiable meniscus or high signal replacing the normal dark meniscal signal (“ghost sign”) in the sagittal plane, a vertical linear defect at the root in the coronal plane, and a radial linear de- fect at the posterior insertion in the axial plane [24]. Ini- tial OA grade and OA progression in the medial compartment at the last follow-up were graded accord- ing to the Kellgren-Lawrence (K-L) classification system [25].. The K-L classification was originally described using AP knee radiographs. Each radiograph was assigned a grade from 0 to 4, which they correlated to increasing severity of OA, with Grade 0 signifying no presence of OA and Grade 4 signifying severe OA. Meniscus roots are a vital component of the meniscus as they anchor the meniscus to the tibial plateau and dis- perse axial loads into hoop stresses during loading. Med- ial meniscus posterior root tear (MMPRT) was defined as a radial tear within 9 mm of the posterior bony at- tachment of the medial meniscus or posterior root avul- sion. MMPRT leads to the loss of hoop tension and load transmissibility in the meniscus, which results in a bio- mechanical condition similar to that observed after total meniscectomy [1]. A high incidence (27.8%) of MMPRT has been reported in Asians because of lifestyles, includ- ing frequent squatting and sitting on the floor with legs folded [2]. MMPRT tends to have a worse prognosis be- cause it commonly occurs in patients aged over 50 years, whose meniscal tissue may have degenerated and who may have low healing potential [2, 3]. Treatment options for MMPRT include conservative treatment, meniscectomy, and root repair. Historically, pa- tients with MMPRT have been treated with conservative treatment or a partial meniscectomy [4]. In recent years, there has been increasing interest in root repair because meniscectomy has been reported to increase the risk of osteoarthritis (OA) [5]. Background It is well known that acute trau- matic MMPRT without OA should be repaired whenever possible to restore meniscal hoop tension and to prevent early arthritic progression [6–9]; however, a large propor- tion of meniscal root tears seen in clinical practice involve degenerative MMPRTs in middle-aged or older patients [10–12]. Hence, surgical repair is not always feasible in the population at risk of these tears [10, 13] due to sub- stantial degeneration of the meniscal tissue and concur- rent OA [2, 3, 14]. Therefore, the best treatment for degenerative MMPRT remains controversial [2, 8, 15]. The inclusion criteria were as follows: (1) diagnosis of complete medial meniscus posterior root avulsion or complete radial tear adjacent (within 9 mm) to the medial meniscus posterior root by a musculoskeletal radiologist [24], (2) presentation of clinical symptoms that were correlated with MRI findings, and (3) arthroscopic complete or partial meniscectomy or conservative treatment. The exclusion criteria were as follows: (1) previous or subsequent ligamentous knee injury, such as a high-energy traumatic injury to the root attachment, (2) concomitant traumatic tibial plat- eau fracture, (3) associated with lateral or anterior meniscus tears, (4) subsequent meniscal repair after diagnosis, (5) concomitant high tibial osteotomy caused by varus malalignment, (6) constitutional varus alignment > 5°, (7) presence of grade > III OA based on the K-L classification and severe chondral defect/ injuries, and (8) less than 2 years of clinical follow-up. Out of 255 patients, 146 (meniscectomy group, n = 90; conservative treatment group, n = 56) were included in the study. (Fig. 1). g [ ] Although the short-term clinical results of MMPRT repair have been encouraging [8, 16, 17], meniscectomy for MMPRT has been traditionally used because it is relatively easier than the repair, and symptomatic im- provement can be expected by removing the source of mechanical pain [2, 3, 5]. Even though the results were heterogeneous, recent studies have reported that conser- vative treatment and meniscectomy can be a good op- tion for selected patients with good prognostic factors [18–21], In addition, conservative treatment with exer- cise therapy has also been reported to be a reasonable treatment option for middle-aged patients with early OA [22, 23]. Therefore, we decided to investigate the failure rate, clinical results, and OA progression after arthro- scopic meniscectomy and conservative treatment for de- generative MMPRT in middle-aged patients. Background Conservative treatment included daily nonsteroidal anti-inflammatory drugs for 4–8 weeks and supervised physical therapy twice a week, including activity modifi- cation for at least 6 weeks. The goal of physical therapy was to reduce pain, restore full range of motion, and im- prove knee function. Physical therapy consisted of exer- cises for muscle strength and endurance. Each patient visited a physiotherapy office and followed a standard exercise program twice a week. The indication for arthroscopic meniscectomy was MMPRT in patients with persistent knee pain with mechanical symptoms affecting activities of daily living, despite conservative treatment for 3 months after MMPRT was diagnosed. The treatment modality was chosen by the patients after discussion with surgeon based subjective symptoms. © The Author(s). 2021 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Lee et al. BMC Musculoskeletal Disorders (2021) 22:742 Page 2 of 8 Page 2 of 8 Lee et al. BMC Musculoskeletal Disorders (2021) 22:742 Statistical analysis Statistical analysis was performed using SPSS version 20.0 (SPSS, Chicago, IL). A P-value < 0.05 was consid- ered statistically significant. Pearson’s chi-square test and Fisher’s exact test were used to determine the statis- tical significance of differences in categorical variables. For continuous variables, independent t-test was used for normally distributed variables and the Mann- Whitney U test was used to compare non-normally dis- tributed variables between groups. The Wilcoxon signed-rank test was used to compare the preoperative and last follow-up clinical outcomes in each group. Kaplan–Meier survival analysis was performed to evalu- ate the time-dependent rate of conversion to arthro- plasty, and a hazard ratio was created via a Cox proportional hazards model. Regarding radiological outcomes, we compared the hip-knee-ankle angle (HKAA), medial proximal tibial angle (MPTA), tibial posterior slope angle (TPSA), and width of medial joint space between the groups. We also evaluated the progression of OA in the medial compart- ment of the knee according to the K-L classification sys- tem. To determine the intra- and inter-observer reliabilities of the radiographic outcomes, two investiga- tors performed all the radiographic assessments twice (1-week intervals). Intraclass correlation coefficients were used for inter- and intra-observer reliability assess- ments. After an average follow-up of 6.3 years, the sur- vivorship was analyzed using the Kaplan–Meier method. The endpoint of survival was conversion to TKA or UKA or HTO in the same knee. Methods After obtaining permission from the institutional ethical committee, we retrospectively reviewed our database of Lee et al. BMC Musculoskeletal Disorders (2021) 22:742 Page 3 of 8 Fig. 1 Patient selection flowchart. MMPRT, Medial meniscus posterior root tear; K-L, Kellgren-Lawrence The first visit and follow-up clinical findings were assessed using the Visual Analog Scale (VAS), Inter- national Knee Documentation Committee (IKDC) sub- jective scoring scale, Tegner activity scale, and Lysholm knee scoring scale. If the patients underwent conversion to total knee arthroplasty (TKA), unicompartmental knee arthroplasty (UKA), or high tibial osteotomy (HTO), the final clinical outcomes were assessed just be- fore the conversion. Discussion This retrospective study compared two treatments, arthrosocpic meniscectomy and conservative treatment, for degenerative MMPRT. Although overall improve- ment was observed in the clinical results of both groups without inter-group differences, arthroscopic meniscec- tomy resulted in increased progression of OA in the medial compartment; however, there was no difference in the survival rate after mid-term follow-up. Thus, arthroscopic meniscectomy has no benefit compared to conservative treatment of degenerative MMPRT. The treatment options for MMPRT include conservative treatment, meniscectomy, and surgical repair. Tradition- ally, patients with MMPRT undergo conservative treat- ment or meniscectomy [26]. Meniscectomy can provide symptomatic relief, but in most cases, progression to de- generative OA does occur [3, 5]. Consequently, there has been a recent shift toward meniscal preservation along with surgical repair [4]. It is well known that acute traumatic MMPRT without OA should be repaired whenever possible to restore meniscal hoop tension and to prevent early arthritic progression [8–11]. Unfortu- nately, a large proportion of MMPRT cases seen in clin- ical practice involve degenerative tears in middle-aged or older patients [11, 12]. Hence, surgical repair is not al- ways feasible in the population at risk of these tears [10, 13] due to substantial degeneration of the meniscal The progression of varus in HKAA was significantly advanced in the meniscectomy group (p = 0.04). In addition, the width of medial joint space at the last fol- low up was less in the meniscectomy group(p = 0.03), and OA progression according to the K-L classification was found to have advanced in the meniscectomy group (p = 0.03). Progression to grade III from grade II was ob- served in 45 cases and to grade IV from grade II was ob- served in 17 cases in the meniscectomy group. Progression to grade III from grade II was observed in 26 cases and to grade IV from grade II was observed in three cases in the conservative group. Grade I remained as grade I in both groups. (Table 4). In terms of the survivorship analysis, the Kaplan– Meier survival curve with the percentage of patients free from conversion to TKA, UKA, or HTO is shown in Fig. 2. During the follow-up period, six patients in the meniscectomy group and four patients in the conserva- tive group underwent conversion due to OA progres- sion. Results The demographic data were similar in both groups. With no significant differences between the two groups (Table 1). All inter- and intra-observer intraclass correl- ation coefficients showed good agreement with the reli- ability of radiographic measurement (> 0.85). There were no significant differences between the meniscectomy Lee et al. BMC Musculoskeletal Disorders (2021) 22:742 Page 4 of 8 Table 1 Comparison of the demographics Meniscectomy group (n = 90) Conservative group (n = 56) P-value Sexa (M/F) 31 / 59 17 / 39 0.61 Ageb (y) 55.5 ± 8.6 57.7 ± 8.1 0.13 BMIb (kg/m2) 25.5 ± 3.8 25.4 ± 2.0 0.84 Follow-up durationb (y) 6.4 ± 3.7 6.1 ± 4.0 0.60 aPearson’s chi-square test, bIndependent t-test. Data are presented median ± standard deviation. The P-values reflect the results of inter-group comparisons, with p < 0.05 indicating significance.BMI Body mass index; dependent t-test. Data are presented median ± standard deviation. The P-values reflect the results of inter-group comparisons, with .BMI Body mass index; 87% at 10 years, whereas that for conservative treatment was 98% at 5 years, 88% at 10 years. (p = 0.8). (Fig. 2) The TKA, UKA, or HTO conversion hazard was 116% Higher for the conservative group compared with the meniscectomy group but there was no statistically sig- nificant difference (p = 0.82). and conservative groups in terms of preoperative HKAA (p = 0.76), MPTA (p = 0.23), TPSA (p = 0.73) or width of medial joint space (p = 0.19) (Table 2). On comparing the two groups, the VAS score (p = 0.07), IKDC subjective score (p = 0.18), Tegner activity scale score (p = 0.08), and Lysholm knee score (p = 0.53) showed no significant differences between the two groups at the final follow-up. Because baseline clinical outcomes were statistically different, the degree of im- provement were also compared, and IKDC subjective score (p = 0.19), Tegner activity scale score (p = 0.67), and Lysholm knee score (p = 0.36) showed no significant differences between the two groups. (Table 3). Discussion The overall Kaplan-Meier probability of survival after arthroscopic meniscectomy was 99% at 5 years, Table 2 Comparison of the preoperative variables Meniscectomy group (n = 90) Conservative group (n = 56) P-value HKAAb (varus, °) 2.9 ± 2.5 2.7 ± 2.4 0.76 MPTAb (°) 87.3 ± 3.0 86.6 ± 3.3 0.23 TPSAb (°) 7.0 ± 4.4 6.8 ± 3.5 0.73 K-L gradea (I/II) Width of medial joint space‡ (mm) 3/87 4.7 ± 1.2 5/51 4.6 ± 1.1 0.26 0.19 aPearson’s chi-square test, bIndependent t-test. Data are presented median ± standard deviation. The P-values reflect the results of inter-group comparisons, with p < 0.05 indicating significance.HKAA Hip-knee-ankle angle, MPTA Medial proximal tibial angle, TPSA Tibial posterior slope angle, K-L Kellgren-Lawrence Table 2 Comparison of the preoperative variables aPearson’s chi-square test, bIndependent t-test. Data are presented median ± standard deviation. The P-values reflect the results of inter-group comparisons, with p < 0.05 indicating significance.HKAA Hip-knee-ankle angle, MPTA Medial proximal tibial angle, TPSA Tibial posterior slope angle, K-L Kellgren-Lawrence bIndependent t-test. Data are presented median ± standard deviation. The P-values reflect the results of inter-group comparisons, wit ance.HKAA Hip-knee-ankle angle, MPTA Medial proximal tibial angle, TPSA Tibial posterior slope angle, K-L Kellgren-Lawrence Lee et al. BMC Musculoskeletal Disorders (2021) 22:742 Page 5 of 8 Table 3 Comparison of the clinical outcomes Conservative group (n = 56) Meniscectomy group (n = 90) VAS First visit 5.9 ± 0.8 4.3 ± 1.3 0.00 Last follow-up 4.3 ± 1.5 3.8 ± 1.2 0.07 Improvement 1.7 ± 1.3 0.8 ± 1.2 0.00 P value‡ 0.00 0.01 IKDC subjective scoring scale First visit 26.3 ± 8.3 30.6 ± 9.8 0.00 Last follow-up 33.9 ± 9.3 38.1 ± 8.8 0.18 Improvement 8.6 ± 8.9 8.9 ± 9.3 0.19 P value‡ 0.00 0.01 Tegner activity scale First visit 2.3 ± 0.9 2.7 ± 0.9 0.03 Last follow-up 2.8 ± 1.1 3.1 ± 0.9 0.08 Improvement 0.6 ± 1.0 0.5 ± 0.8 0.67 P value‡ 0.00 0.03 Lysholm knee scoring scale First visit 50.9 ± 8.7 54.1 ± 8.9 0.00 Last follow-up 65.5 ± 9.4 67.0 ± 10.8 0.53 Improvement 14.9 ± 9.1 12.5 ± 9.8 0.36 P value‡ 0.00 0.00 †Mann-Whitney U test for analysis of difference. ‡Wilcoxon signed-rank test for comparison of clinical outcomes between preoperative and last follow up points. Values are presented as means and standard deviations. Discussion The p-values reflect the results of inter-group comparisons, with p < 0.05 indicating significance †Mann-Whitney U test for analysis of difference. ‡Wilcoxon signed-rank test for comparison of clinical outcomes between preoperative and last follow up points. Values are presented as means and standard deviations. The p-values reflect the results of inter-group comparisons, with p < 0.05 indicating significance tissue and concurrent OA [2, 3, 14]. Although the over- all outcomes of surgical repair have been good in some studies [13, 27, 28], Bin et al. reported that partial men- iscectomy can be a good option for selected patients with good prognostic factors and for patients who are not eligible for surgical repair because of the poor qual- ity of their meniscal tissue [1, 28]. In our study, pain and functional outcomes at first visit were significantly worse in the meniscectomy group than in the conservative group. This indicates that the greater the pain intensity, the higher the likelihood of patients choosing surgical treatment over conservative treatment. However, both meniscectomy and conservative treatment resulted in significant improvements in pain and function scores as per the VAS and IKDC scores, respectively, with no inter-group differences after an average follow-up of 6.3 years. The lack of significant differences may be due to the improvement in symptoms, including mechanical pain, with time, regardless of the treatment modality. The lack of differences in clinical outcomes despite greater progression of OA in the meniscectomy group than in the conservative group might be because the follow-up duration was not enough to detect differences in clinical outcomes. In addition, the mean value was Table 4 Comparison of the radiological outcomes Meniscectomy group (n = 90) Conservative group (n = 56) P-value† HKAA† (varus, °) 4.3 ± 2.3 3.6 ± 2.5 0.04 Width of medial joint space† (mm) 3.1 ± 1.1 3.5 ± 1.2 0.03 K-L grade‡ Grade I 3 5 0.03 Grade II 25 22 Grade III 45 26 Grade IV 17 3 †Independent t-test. ‡Fisher’s exact test. The p-values reflect the results of inter-group comparisons, with p < 0.05 indicating significance. HKAA Hip-knee-ankle angle, K-L Kellgren-Lawrence Table 4 Comparison of the radiological outcomes †Independent t-test. ‡Fisher’s exact test. The p-values reflect the results of inter-group comparisons, with p < 0.05 indicating significance. HKAA Hip-knee-ankle angle, K-L Kellgren-Lawrence †Independent t-test. ‡Fisher’s exact test. The p-values reflect the results of inter-group comparisons, with p < 0.05 indicating significance. Discussion HKAA Hip-knee-ankle angle, K-L Kellgren-Lawrence Lee et al. BMC Musculoskeletal Disorders (2021) 22:742 Page 6 of 8 Fig. 2 Kaplan-Meier analysis of joint survival after meniscectomy and conservative treatment a selection bias may be present. Moreover, the baseline pain and functional scores were low in the meniscec- tomy group because patients chose the treatment modal- ity based on their symptoms and treatment characteristics. To overcome this, this study compared the degree of improvement from the baseline level to the final follow-up. Second, the follow-up period was not long enough to detect differences in the survival rate. Third, during the follow-up period, it was not clearly in- dicated whether another conservative treatment that could affect the clinical outcome were performed after the completion of acute treatment. Fourth, the high pro- portion of female patients in our study. Although MMPRT is prevalent in middle-aged female patients, sex and age can affect individual activity. This reduces the extent to which our results can be generalized. Despite these limitations, we tried to only include patients with degenerative MMPRT without significant malalignment and advanced OA to reduce selection bias to ensure ob- jective evaluation of the effectiveness of meniscectomy for MMPRT. lower in the meniscectomy group, and therefore, differ- ences in outcome scores between treatment groups could be clinically significant, although not statistically significant. This may be a result of the analysis being underpowered. The most important finding of this study was that OA progression was more severe in the meniscectomy group than in the conservative group (p = 0.03). Similar to our study, Krych et al. reported that partial meniscectomy for degenerative MMPRT provides no benefit over con- servative treatment in terms of halting arthritic progres- sion [18, 29]. Similarly, early OA development is more likely to occur after meniscectomy than after non- operative treatment [30, 31]. Meniscectomy may in- crease the pressure on the residual meniscus, which may worsen any subsequent articular degeneration [5, 32]. In a study by Han et al. [5], after meniscectomy for MMPR T, progression of OA on radiological examination was noted in 35% of the patients at 5–6 years after surgery. Krych et al. [18] found that 54% of partial meniscectomy patients and 34.6% of non-operative patients showed conversion to TKA at a mean of 54.3 and 30.2 months, respectively. Discussion Contrary to other studies, our study showed that the survival rate was 99% at 5 years and 87% at 10 years after meniscectomy and 98% at 5 years and 88% at 10 years after conservative treatment, possibly because meniscectomy was performed only in patients without significant malalignment or osteoarthritic change. This study has several limitations. First, it was a retrospective investigation of a small, nonrandomized case series; thus, Author details 1 21. Lee JK, Jung M, Yang JH, et al. Repair versus nonrepair of medial meniscus posterior root tear a systematic review of patients’ selection criteria, including clinical and radiographic outcomes. Medicine (Baltimore). 2020; 99(10):e19499. 21. Lee JK, Jung M, Yang JH, et al. Repair versus nonrepair of medial meniscus posterior root tear a systematic review of patients’ selection criteria, including clinical and radiographic outcomes. Medicine (Baltimore). 2020; 99(10):e19499. Author details 1Department of Orthopaedic Surgery, Chonnam National University Hwasun Hospital and Medical School, 322 Seoyang-ro, Hwasun-gun, Chonnam 58218, Republic of Korea. 2Department of Orthopaedic Surgery, Chonnam National University Hospital, 42 Jebongro, Donggu, Gwangju 61469, Republic of Korea. 1Department of Orthopaedic Surgery, Chonnam National University Hwasun Hospital and Medical School, 322 Seoyang-ro, Hwasun-gun, Chonnam 58218, Republic of Korea. 2Department of Orthopaedic Surgery, Chonnam National 22. Lim HC, Bae JH, Wang JH, et al. Non-operative treatment of degenerative posterior root tear of the medial meniscus. Knee Surg Sports Traumatol Arthrosc. 2010;18:535–9. 22. Lim HC, Bae JH, Wang JH, et al. Non-operative treatment of degenerative posterior root tear of the medial meniscus. Knee Surg Sports Traumatol Arthrosc. 2010;18:535–9. University Hospital, 42 Jebongro, Donggu, Gwangju 61469, Republic of Korea. Availability of data and materials The original reports, imaging studies and outpatient clinic records are retained as per normal procedure within the medical records of our institution. If someone wants to request the data from this study, please contact to namhunleeos@gmail.com. 15. Seo HS, Lee SC, Jung KA. Second-look arthroscopic findings after repairs of posterior root tears of the medial meniscus. Am J Sports Med. 2011;39(1): 99–107. 16. Kim SB, Ha JK, Lee SW, et al. Medial meniscus root tear refixation: comparison of clinical, radiologic, and arthroscopic findings with medial meniscectomy. Arthroscopy. 2011;27:346–54. Ethics approval and consent to participate 17. Chung KS, Ha JK, Ra HJ, et al. Pullout fixation of posterior medial meniscus root tears. Am J Sports Med. 2017;45:42–9. The study was conducted with the approval of the institutional review board (IRB) of Chonnam National University Hwasun Hospital. All participants signed written consent to participate in the study. All rights of the patients were protected against any kind of disadvantage and individual matters. No: CNUHH 2020–195. 18. Krych AJ, Johnson NR, Mohan R, Dahm DL, Levy BA, Stuart MJ. Partial meniscectomy provides no benefit for symptomatic degenerative medial meniscus posterior root tears. Knee Surg Sports Traumatol Arthrosc. 2018; 26(4):1117–22. 18. Krych AJ, Johnson NR, Mohan R, Dahm DL, Levy BA, Stuart MJ. Partial meniscectomy provides no benefit for symptomatic degenerative medial meniscus posterior root tears. Knee Surg Sports Traumatol Arthrosc. 2018; 26(4):1117–22. 19. Krych AJ, Reardon PJ, Johnson NR, et al. Non-operative management of medial meniscus posterior horn root tears is associated with worsening arthritis and poor clinical outcome at 5-year follow up. Knee Surg Sports Traumatol Arthrosc. 2017;25(2):383–9. 19. Krych AJ, Reardon PJ, Johnson NR, et al. Non-operative management of medial meniscus posterior horn root tears is associated with worsening arthritis and poor clinical outcome at 5-year follow up. Knee Surg Sports Traumatol Arthrosc. 2017;25(2):383–9. Authors’ contributions NHL: Analysis of data and writing manuscript. HYS: Analysis of data and writing manuscript. MJS: Data collection and analysis. BRN: Data collection and analysis. EKS: Make concepts and design of study. JKS: Make concepts and design of study. Review and correction of draft manuscript. All authors have read and approved the final manuscript. All authors have agreed it for submission to publish in this journal. All authors have agreed to authorship and order of authorship for this manuscript. All authors have agreed that any changes of authors or order of authors can’t be changed if this manuscript is accepted. 11. Koo JH, Choi SH, Lee SA, Wang JH. Comparison of medial and lat meniscus root tears. PLoS One. 2015;10(10):e0141021. 12. Lee DW, Ha JK, Kim JG. Medial meniscus posterior root tear: a comprehensive review. Knee Surg Relat Res. 2014;26(3):125–34. 13. Chung KS, Ha JK, Yeom CH, et al. Comparison of clinical and radiologic results between partial meniscectomy and refixation of medial meniscus posterior root tears: a minimum 5-year follow-up. Arthroscopy. 2015;31(10): 1941–50. 13. Chung KS, Ha JK, Yeom CH, et al. Comparison of clinical and radiologic results between partial meniscectomy and refixation of medial meniscus posterior root tears: a minimum 5-year follow-up. Arthroscopy. 2015;31(10): 1941–50. 14. Park DY, Min BH, Choi BH, et al. The degeneration of meniscus roots is accompanied by fibrocartilage formation, which may precede meniscus root tears in osteoarthritic knees. Am J Sports Med. 2015;43(12):3034–44. 14. Park DY, Min BH, Choi BH, et al. The degeneration of meniscus roots is accompanied by fibrocartilage formation, which may precede meniscus root tears in osteoarthritic knees. Am J Sports Med. 2015;43(12):3034–44. Acknowledgements The authors thank all clinical researchers involved in the research we included in this article. This study was not supported by any company or grant. 9. Ra HJ, Ha JK, Jang HS, Kim JG. Traumatic posterior root tear of the medial meniscus in patients with severe medial instability of the knee. Knee Surg Sports Traumatol Arthrosc. 2015;23(10):3121–6. 9. Ra HJ, Ha JK, Jang HS, Kim JG. Traumatic posterior root tear of the medial meniscus in patients with severe medial instability of the knee. Knee Surg Sports Traumatol Arthrosc. 2015;23(10):3121–6. 10. Chung KS, Ha JK, Ra HJ, Kim JG. Prognostic factors in the midterm results of pullout fixation for posterior root tears of the medial meniscus. Arthroscopy. 2016;32(7):1319–27. Received: 29 December 2020 Accepted: 16 August 2021 23. Neogi DS, Kumar A, Rijal L, et al. Role of nonoperative treatment in managing degenerative tears of the medial meniscus posterior root. J Orthop Traumatol. 2013;14:193–9. 23. Neogi DS, Kumar A, Rijal L, et al. Role of nonoperative treatment in managing degenerative tears of the medial meniscus posterior root. J Orthop Traumatol. 2013;14:193–9. Abbreviations MMPRT M di l MMPRT: Medial meniscus posterior root tear; VAS: Visual Analog Scale; IKDC: International Knee Documentation Committee; K-L: Kellgren-Lawrence; HKAA: Hip-knee-ankle angle; MPTA: Medial proximal tibial angle; TPSA: Tibial posterior slope angle; OA: Osteoarthritis; UKA: Unicompartmental knee arthroplasty; HTO: High tibial osteotomy 7. Bhatia S, Civitarese DM, Turnbull TL, et al. A novel repair method for radial tears of the medial meniscus: biomechanical comparison of transtibial 2- tunnel and double horizontal mattress suture techniques under cyclic loading. Am J Sports Med. 2016;44(3):639–45. 7. Bhatia S, Civitarese DM, Turnbull TL, et al. A novel repair method for radial tears of the medial meniscus: biomechanical comparison of transtibial 2- tunnel and double horizontal mattress suture techniques under cyclic loading. Am J Sports Med. 2016;44(3):639–45. 8. Lee JH, Lim YJ, Kim KB, Kim KH, Song JH. Arthroscopic pullout suture repair of posterior root tear of the medial meniscus: radiographic and clinical results with a 2-year follow-up. Arthroscopy. 2009;25(9):951–8. arthroscopic meniscectomy has no benefit compared to conservative treatment in middle-aged patients with de- generative MMPRT. arthroscopic meniscectomy has no benefit compared to conservative treatment in middle-aged patients with de- generative MMPRT. 5. Han SB, Shetty GM, Lee DH, et al. Unfavorable results of partial meniscectomy for complete posterior medial meniscus root tear with early osteoarthritis: a 5- to 8-year follow-up study. Arthroscopy. 2010;26:1326–32. 5. Han SB, Shetty GM, Lee DH, et al. Unfavorable results of partial meniscectomy for complete posterior medial meniscus root tear with early osteoarthritis: a 5- to 8-year follow-up study. Arthroscopy. 2010;26:1326–32. 5. Han SB, Shetty GM, Lee DH, et al. Unfavorable results of partial meniscectomy for complete posterior medial meniscus root tear with early osteoarthritis: a 5- to 8-year follow-up study. Arthroscopy. 2010;26:1326–32. 6. Ahn JH, Wang JH, Yoo JC, Noh HK, Park JH. A pull out suture for transection of the posterior horn of the medial meniscus: using a posterior trans-septal portal. Knee Surg Sports Traumatol Arthrosc. 2007;15(12):1510–3. 6. Ahn JH, Wang JH, Yoo JC, Noh HK, Park JH. A pull out suture for transection of the posterior horn of the medial meniscus: using a posterior trans-septal portal. Knee Surg Sports Traumatol Arthrosc. 2007;15(12):1510–3. 6. Ahn JH, Wang JH, Yoo JC, Noh HK, Park JH. A pull out suture for transection of the posterior horn of the medial meniscus: using a posterior trans-septal portal. Knee Surg Sports Traumatol Arthrosc. 2007;15(12):1510–3. Competing interests 20. Lee BS, Bin SI, Kim JM, et al. Partial Meniscectomy for degenerative medial meniscal root tears shows favorable outcomes in well-aligned, nonarthritic knee. Am J Sports Med. 2019;47(3):606–11. p g The authors declare that they have no competing interests. Conclusion This study demonstrated that both conservative treat- ment and meniscectomy provided symptomatic relief to patients with degenerative MMPRT without advanced OA and malalignment. However, OA progression was more severe in the meniscectomy group than in the con- servative group, despite the similarity in their survival rates. In light of our findings, we concluded that Page 7 of 8 Page 7 of 8 Page 7 of 8 Lee et al. BMC Musculoskeletal Disorders (2021) 22:742 References Knee ligament injury, surgery and osteoarthrosis: truth or consequences? Acta Orthop Scand. 1994;65(6):605–9. 30. Lohmander LS, Roos H. Knee ligament injury, surgery and osteoarthrosis: truth or consequences? Acta Orthop Scand. 1994;65(6):605–9. 31. Roos H, Adalberth T, Dahlberg L, Lohmander LS. Osteoarthritis of the knee after injury to the anterior cruciate ligament or meniscus: the influence of time and age. Osteoarthr Cartil. 1995;3(4):261–7. 32. Magee T, Shapiro M, Williams D. Prevalence of meniscal radial tears of the knee revealed by MRI after surgery. AJR Am J Roentgenol. 2004;182:931–6. Lee et al. BMC Musculoskeletal Disorders (2021) 22:742 References 24. LaPrade CM, James EW, Cram TR, Feagin JA, Engebretsen L, LaPrade RF. Meniscal root tears: a classification system based on tear morphology. Am J Sports Med. 2015;43:363–9. 1. Allaire R, Muriuki M, Gilbertson L, Harner CD. Biomechanical consequences of a tear of the posterior root of the medial meniscus. Similar to total meniscectomy. J Bone Joint Surg Am. 2008;90(9):1922–31. 25. Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Ann Rheum Dis. 1957;16:494–502. 2. Bin SI, Kim JM, Shin SJ. Radial tears of the posterior horn of the medial meniscus. Arthroscopy. 2004;20:373–8. 2. Bin SI, Kim JM, Shin SJ. Radial tears of the posterior horn of the medial meniscus. Arthroscopy. 2004;20:373–8. 26. Chung KS, Noh JM, Ha JK, et al. Survivorship analysis and clinical outcomes of Transtibial pullout repair for medial meniscus posterior root tears: a 5- to 10-year follow-up study. Arthroscopy. 2018;34(2):530–5. 3. Ozkoc G, Circi E, Gonc U, Irgit K, Pourbagher A, Tandogan RN. Radial tears in the root of the posterior horn of the medial meniscus. Knee Surg Sports Traumatol Arthrosc. 2008;16:849–54. 3. Ozkoc G, Circi E, Gonc U, Irgit K, Pourbagher A, Tandogan RN. Radial tears in the root of the posterior horn of the medial meniscus. Knee Surg Sports Traumatol Arthrosc. 2008;16:849–54. 27. LaPrade RF, Matheny LM, Moulton SG, James EW, Dean CS. Posterior meniscal root repairs: outcomes of an anatomic transtibial pullout technique. Am J Sports Med. 2017;45(4):884–91. 4. Bhatia S, LaPrade CM, Ellman MB, LaPrade RF. Meniscal root tears: significance, diagnosis, and treatment. Am J Sports Med. 2014;42(12): 3016–30. 4. Bhatia S, LaPrade CM, Ellman MB, LaPrade RF. Meniscal root tears: significance, diagnosis, and treatment. Am J Sports Med. 2014;42(12): 3016–30. Page 8 of 8 Lee et al. BMC Musculoskeletal Disorders (2021) 22:742 Lee et al. BMC Musculoskeletal Disorders (2021) 22:742 28. Lee JK, Jung M, Yang JH, et al. Repair versus nonrepair of medial meniscus posterior root tear a systematic review of patients’ selection criteria, including clinical and radiographic outcomes. Medicine (Baltimore). 2020; 99(10):e19499. 29. Yim JH, Seon JK, Song EK, et al. A comparative study of Meniscectomy and nonoperative treatment for degenerative horizontal tears of the medial Meniscus. Am J Sports Med. 2013;41(7):1565–70. 29. Yim JH, Seon JK, Song EK, et al. A comparative study of Meniscectomy and nonoperative treatment for degenerative horizontal tears of the medial Meniscus. Am J Sports Med. 2013;41(7):1565–70. 30. Lohmander LS, Roos H. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
https://openalex.org/W4252432876	https://revistaei.uchile.cl/index.php/REI/article/download/15698/16169	Spanish; Castilian	null	El espacio : ¿patrimonio común de la humanidad?	Estudios Internacionales	2,011	cc-by-sa	5,564	Introducción: Desde que los progresos técnicos han permitido realizar vuelos espaciales e interplanetarios se ha hecho necesario, desde el punto de vista jurídico, distinguir entre el espacio aéreo y el espacio ultraterrestre, cuyos regímenes jurídicos son antinómicos: el espacio aéreo se encuentra sometido a la soberanía del Estado, el espacio ultraterrestre escapa a esta soberanía. Considerando que no se cuestiona la distinción de la naturaleza y estatuto de los dos espacios, no ha.sido posible hasta ahora fijar un criterio físico de delimitación de cada uno de ellos. Los avan- ces científicos no permiten elaborar una solución pacífica aproxi- mada a la realidad física. La práctica ha preferido orientarse hacía un criterio funcional del campo de aplicación de las reglas de nave- gación espacial y aeroespacial, dependiendo su aplicación del objeto de la actividad. Este pragmatismo ha permitido resolver la mayor parte de los problemas, lo que tiene como consecuencia que ningún Estado puede pretender, en virtud de su soberanía territorial, fijar unilateralmente el límite exterior del espacio aéreo de su territorio. El 4 de octubre de 1959, con el lanzamiento del primer Sputnik, se origina una nueva rama del Derecho: el Derecho Espa- cial Ultraterrestre, encargado de establecer una normativa que pueda regular los cambios producidos en el campo de las comuni- caciones, de la observación meteorológica, de la utilización del espacio para fines de investigación científica o de usos militares y de todos aquellos problemas que surgirían día a día en su utili- zación por los diferentes países, de acuerdo a su desarrollo tecno- lógico. [ 3 9 2 ] Jeannette Irigoin B. / El espacio, ¿patrimonio común de la humanidad? 1.- LA COMISIÓN DEL ESPACIO ULTRATERRESTRE DE LAS NACIONES UNIDAS Por dicha resolución se -pedía que el Secretario General llevara [ 3 9 3 ] E S T U D I O S I N T E R N A C I O N A L E S un registro público de información facilitada por los Estados que lanzaran objetos capaces de describir una órbita o alcanzar puntos más distantes. un registro público de información facilitada por los Estados que lanzaran objetos capaces de describir una órbita o alcanzar puntos más distantes. La Comisión ha creado dos Subcomisiones: una sobre los aspectos jurídicos y otra sobre los aspectos científicos y técnicos de la exploración y utilización del espacio ultraterrestre. Todos los países miembros de la Comisión son miembros de pleno dere- cho de las dos Subcomisiones. La Comisión tiene facultades para establecer Grupos de Tra- bajo plenario, tales como el Grupo de Trabajo sobre satélites de transmisión directa, el Grupo de Trabajo sobre servicios de navega- ción mediante satélites y el Grupo de Trabajo sobre teleobserva- ción terrestre mediante satélites. De conformidad con su manda- to, la Comisión presenta informes a la Asamblea General. La ampliación de miembros hasta su composición actual fue debida a una iniciativa de Argentina, del año 1972 (A/AC. 1057 PV. 113, p. 56, 7 de septiembre 1972), reiterada el 19 de octubre de 1972 y fundada extensamente el 28 de junio de 1973 (A/AC. 105/P.V. 123, pp. 36-37). La Resolución 3182 (XXVIII) del 18 de diciembre de 1973 en su párrafo 28 decidió aumentar el número de integrantes de la Comisión con nueve miembros adicionales como máximo, tenien- do en cuenta el principio de la distribución geográfica equitativa. La actual composición de la Comisión del Espacio es: Albania, Alemania (República Federal de), Argentina, Australia, Austria, Bélgica, Brasil, Bulgaria, Canadá, Chad, Checoslovaquia, Chile, Egipto, Estados Unidos de América, Francia, Hungría, India, Indo- nesia, Irán, Italia, Japón, Kenia, Marruecos, México, Mongolia, Nigeria, Pakistán, Polonia, Reino Unido, República Democrática Alemana, Rumania, Sierra Leona, Sudán, Suecia, URSS y Vene- zuela. 1.- LA COMISIÓN DEL ESPACIO ULTRATERRESTRE DE LAS NACIONES UNIDAS La Asamblea General de las Naciones Unidas instituyó, por Reso- lución 1472 (XIV) del 12 de diciembre de 1959, la Comisión sobre la Utilización del Espacio Ultraterrestre con Fines Pacíficos, con el siguiente mandato: • a) Examinar la esfera de la cooperación internacional y estu- diar las medidas prácticas y posibles para realizar los programas de utilización del espacio ultraterrestre con fines pacíficos que pue- dan adecuadamente emprenderse bajo los auspicios de las Naciones Unidas, y en particular las siguientes: A i i i — Asistencia para continuar con carácter permanente las investigaciones acerca del espacio ultraterrestre efectuadas con motivo del Año Geofísico Internacional; — Organización del intercambio y difusión de informaciones relativas a las investigaciones acerca del espacio ultrate- rrestre; — Fomento de los programas nacionales de investigación rela- cionadas con el estudio del espacio ultraterrestre, y presta- ción de toda la ayuda y colaboración posibles para ejecutar los programas; • --• b di l bl j di d l l b) Estudiar los problemas jurídicos que pueda plantear la exploración del espacio ultraterrestre. Dos años después, la Asamblea, en virtud de la Resolución 1721 (XVI), del 20 de diciembre de 1961, amplió a 28 el número de los primeros 24 miembros y pidió a la Comisión que, a través del Secretario General y haciendo uso de sus funciones y de los recursos de la Secretaría: Dos años después, la Asamblea, en virtud de la Resolución 1721 (XVI), del 20 de diciembre de 1961, amplió a 28 el número de los primeros 24 miembros y pidió a la Comisión que, a través del Secretario General y haciendo uso de sus funciones y de los recursos de la Secretaría: a) Mantenga estrecho contacto con las organizaciones guber- namentales y no gubernamentales interesadas en cuestiones rela- tivas al espacio ultraterrestre; b) Organice el intercambio de la información que sobre las actividades relativas al espacio ultraterrestre faciliten voluntaria- mente los gobiernos, procurando que ese intercambio constituya un complemento y no una duplicación de los intercambios técni- cos y científicos que se estén realizando; c) Colabore en el estudio de medidas para fomentar la coope- ración internacional en actividades relativas .al espacio ultrate- rrestre. 1.- LA COMISIÓN DEL ESPACIO ULTRATERRESTRE DE LAS NACIONES UNIDAS Aun cuando no existe todavía una idea o un criterio unificado sobre la naturaleza y el contenido del Derecho del Espacio, pode- mos afirmar que el Tratado del Espacio de 1967 ha significado un extraordinario avance y que, al representar la voluntad unánime de todos los miembros de la comunidad internacional en la orga- nización de las Naciones Unidas, se han originado nuevas figuras y nuevos principios del Derecho Internacional. [ 3 9 4 ] Jeannette Irígoin I El espacio, ¿patrimonio común de la humanidad? Jeannette Irígoin I El espacio, ¿patrimonio común de la humanidad? INaciones Unidas, Asamblea General, Doc. Oficiales A/C. 1/PV., 1974. INaciones Unidas, Asamblea General, Doc. Oficiales A/C. 1/PV., 1974. 2.- EL PATRIMONIO COMÚN DE LA HUMANIDAD La reciente aceptación de la idea del "patrimonio común de la hu- manidad" como un nuevo principio jurídico de derecho interna- cional constituye una expresiva ilustración del "poder de una idea cuyo tiempo ha llegado", al decir de Víctor Hugo. Aun cuando ciertos autores niegan que este concepto tenga un contenido legal preciso, es importante reconocer que "l j ídi ól i ió d l "los conceptos jurídicos no son sólo teorización de las normas y prácticas jurídicas anteriores sino también conceptos creado- res de los que surgen tales normas y prácticas. Si la humanidad se hubiese limitado siempre a aplicar conceptos jurídicos ya existentes no se hubieran desarrollado los sistemas.-jurídicos y el derecho no habría cumplido su adecuada función en el pro- gres o social" } La raíz u origen del concepto o idea de "patrimonio común de la humanidad" se puede encontrar en la doctrina de Francisco de Vitoria, desarrollada más tarde por Andrés Bello bajo la denomina- ción de "patrimonio indivisible de la especie humana", en su obra Principios de Derecho Internacional en 1864. Es necesario reconocer la activa participación de los países miembros de la comunidad latinoamericana respecto a este tema en las negociaciones de la Tercera Conferencia de las Naciones Uni- das sobre el Derecho del Mar, que contribuyó en forma eficaz al logro de las fórmulas finalmente concretadas en la Convención aprobada en dicha Conferencia el 30 de abril de 1982. E f d á fi i l i Este concepto fue usado con carácter oficial anteriormente en la Comisión del Espacio Ultraterrestre de Naciones Unidas en 1967 e incorporado al primer texto de Convenio relativo a la luna el 3 de julio de 1970 y finalmente reconocido en el Acuerdo adoptado por la Asamblea General de las Naciones Unidas el 5 de diciembre de 1979. 2.- EL PATRIMONIO COMÚN DE LA HUMANIDAD Fácilmente se puede deducir que la contribución de la doctrina latinoamericana del Derecho Internacional y la acción de los juris- tas de esta región en reuniones académicas internacionales, así como la de los representantes gubernamentales en las Naciones Unidas y otros foros y organismos internacionales para alcanzar el reconocimiento del principio del "patrimonio común de la huma- nidad" en el Derecho Internacional del Espacio y en el Derecho In- ternacional del Mar, se ha visto además consagrada en documentos e instrumentos relativos al acervo cultural de la humanidad, al de- recho internacional de las Telecomunicaciones, energético inter- [ 3 9 5 ] E S T U D I O S I N T E R N A C I O N A L E S nacional, ambiental internacional, internacional de la'rkdiodiíusión y otros campos del derecho internacional. nacional, ambiental internacional, internacional de la'rkdiodiíusión y otros campos del derecho internacional. Se podría definir el "patrimonio común de la humanidad" como un concepto que abarca todo conjunto de bienes materiales e inmateriales cuya utilización o conservación —sea por el espacio que los comprende o la función que cumplen— incumbe a todo el género humano, es decir, a todos los pueblos cualquiera sea su estatuto jurídico, y debe realizarse con la participación de todos ellos y en su beneficio. La vigencia de este principio ha llenado un vacío jurídico res- pecto de la utilización de determinados espacios, especialmente el espacio ultraterrestre y la Zona Internacional de los Fondos Mari- nos. Este principio implica el derecho intransferible de todos los pueblos y de la persona humana como último destinatario del Derecho, el goce pleno de los beneficios que derivan del patrimo- nio común de la humanidad. El corolario fundamental e indispensable de este principio es el establecimiento de los mecanismos capaces de asegurar efectiva- mente el cabal aprovechamiento del patrimonio común en bene- ficio de toda la humanidad, con especial consideración a las nece- sidades e intereses de los países en desarrollo. 3.- RÉGIMEN JURÍDICO DEL ESPACIO AEREO En la Convención de París, 1919, después de la Primera Guerra Mundial, teniendo-en consideración razones estratégicas y mili- tares, los Estados fijaron normas, adoptando una completa y exclusiva soberanía respecto del espacio aéreo sobre sus terri- torios. El efecto fundamental que se deriva de este principio de sobe- ranía es que el sobrevuelo o aterrizaje en un Estado sólo puede realizarse -mediante un permiso previo. Esta es una diferencia fun- damental con el régimen que existe en la marina mercante, en el q-ue los buques mercantes de todas las banderas pueden navegar por las aguas territoriales de cualquier Estado sin necesidad de permiso previo, y excepto en caso de.prohibiciones específicas no discriminatorias, -están autorizados a entrar en puertos extranje- ros y realizar diversas operaciones de tráfico de pasajeros, correo o carga con cualquier destino. En cambio, el sobrevuelo, el aterri- zaje y los derechos de tráfico de las aeronaves extranjeras depen- [ 3 9 6 ] Jeannette Irigoin B. I El espacio, ¿patrimonio común de la humanidad? den de una concesión del Estado territorial, concesión que debe ser negociada entre los Estados interesados y que puede ser condi- cionada o limitada. La Convención de Chicago de 1944 consagra lo que se deno- mina "las cinco libertades del aire". La primera, es el derecho de los aviones de un Estado de sobre- volar sin hacer escala sobre el territorio de otro Estado. La se- gunda es la libertad de hacer escalas técnicas, que no tengan carác- ter comercial, para el reaprovisionamiento de combustible o repa- raciones en caso de desperfectos. Estas dos libertades se refieren al tránsito de aeronaves. La tercera libertad es el derecho de los aviones de un Estado de desembarcar en el territorio de otro Estado, pasajeros, correspon- dencia o mercadería proveniente del territorio del Estado al que pertenece el avión. La cuarta libertad representa el polo opuesto de la tercera; el derecho de embarcar en un Estado pasajeros, correspondencia o mercadería con destino al territorio de nacionalidad de la aero- nave. Por último, la quinta libertad es el derecho de un avión de embarcar pasajeros, correspondencia o mercadería con destino al territorio de cualquier otro Estado, y el derecho de desembarcar pasajeros, correspondencia o mercadería procedentes de cualquier otro Estado. Estas tres libertades de naturaleza económica, se describen como las libertades de tráfico. 3.- RÉGIMEN JURÍDICO DEL ESPACIO AEREO Como no ha sido posible llegar a un acuerdo multilateral para estas libertades, han debido negociarse en forma bilateral entre el Estado de nacionalidad de la aeronave y el Estado en cuyo territorio se pretenden realizar operaciones de tránsito y de tráfico los "acuerdos de línea" que permiten a los Estados más débiles —al regular o limitar la concesión de la quinta libertad— establecer un régimen de competencia más equitativo, asegurando a sus líneas aeronáuticas cierta participación en el.tráfico interna- cional, conciliando intereses divergentes de los Estados involucra- dos. Las tarifas son controladas en primera instancia por IA.TA (la Asociación de Transportadores Aéreos) existiendo un mecanismo de ajuste en caso que IATA no logre ponerse de acuerdo con su fijación. Los acuerdos bilaterales —11.324 entre 150 Estados— han sido gradualmente más restrictivos en materia de capacidades, fre- cuencias, tipo de aeronaves a ser utilizadas, ingreso de éstas y con- cesión de la quinta libertad para evitar los perjuicios que para los [ 3 9 7 ] E S T U D I O S I N T E R N A C I O N A L E S transportadores nacionales se derivarían de una competencia irrestricta. transportadores nacionales se derivarían de una competencia irrestricta. Gran importancia ha tenido el Acuerdo de Bermudas de 1946 celebrado entre Gran Bretaña y Estados Unidos, quienes acordaron un sistema que regula la quinta libertad con condiciones delibera- damente vagas, que permite una administración flexible de las capacidades ofrecidas. Este modelo se ha aplicado en muchos otros acuerdos, aun cuando en los últimos años-muchos países en desarrollo, ajenos a la formulación de las reglas de Chicago y Bermudas, son partida- rios de restringir el reconocimiento de la quinta libertad de manera de favorecer el desarrollo de sus líneas nacionales. Una de las bases del sistema —que ya ha sido destruida— era que los transportadores nucleados en IATA pudieran controlar indefinidamente las tarifas. Actualmente, ese control ha desapare- cido ante la irresistible y comprensible presión que existe del público para obtener rebajas en el transporte aéreo. Han sur- gido los vuelos no regulares y sin itinerario fijo llamados vuelos "charter", los que han llegado a constituir más del .25 por ciento del total del mercado aéreo. La misma Conferencia de Chicago creó la Organización de la Aviación Civil Internacional —OACI—. 3.- RÉGIMEN JURÍDICO DEL ESPACIO AEREO que ha obtenido positivos resultados en asuntos técnicos relativos a la seguridad de la avia- ción, navegación, certificados aéreos mínimos, etc. Últimamente está haciendo .estudios sobre problemas tales como el control de ruidos, indemnizaciones por accidentes y secuestros. Es una agencia especializada de Naciones Unidas, en la Convención de Tokio de 1963, se incluyeron normas relativas al apoderamiento ilícito de aeronaves y en 1970 se celebró la Convención de La Haya específicamente dirigida a establecer las normas de extradi- ción automática para el delito de piratería aérea. p p En 1971, se celebró la Convención de Montreal —también bajo el auspicio de la OACI— que está orientada a la prevención del sabotaje y actos de violencia contra las aeronaves. El 17 de julio de 1978, los jefes de Estado o de Gobierno de las siete naciones más industrializadas de Occidente acordaron suspender todos los vuelos con destino a aquellos países que eran renuentes a extra- dictar o perseguir criminalmente a aquéllos que hubieren secues- trado una aeronave y/o no la hubieren devuelto. [ 3 9 8 ] [ 3 9 8 ] Jeannette Irigoin B. I El espacio, ¿patrimonio común de la humanidad? 4.- RÉGIMEN JURÍDICO DEL ESPACIO ULTRATERRESTRE Con el lanzamiento del primer satélite construido por el hombre se planteó la interrogante acerca de si era posible admitir que la soberanía estatal se extendiera al espacio ultraterrestre. Desde un punto de vista práctico, el número de territorios nacionales surca- dos por un satélite lanzado a alta velocidad en su órbita alrededor de la Tierra convierte en irrazonable e irreal cualquier idea de con- trol o fiscalización territorial. Es que en realidad serían los pro- pios territorios nacionales los que, como consecuencia de la rota- ción de la tierra, pasarían bajo la órbita fija del satélite espacial. Estos datos básicos de astronomía explican por qué los países involucrados en actividades espaciales —Estados Unidos y la Unión Soviética— se limitaron a difundir públicamente el lanzamiento de astronaves sin dar cuenta de sus propósitos a los Estados cuyos espacios aéreos iban a surcar y sin requerir el consentimiento pre- vio de dichos Estados, constatándose también la ausencia de pro- testa o reclamación de su parte. Esta práctica constituyó un reco- nocimiento implícito de la libertad del espacio ultraterrestre. 4.1.- Los límites del espacio ultraterrestre 4.1.- Los límites del espacio ultraterrestre Aun cuando existen diversas teorías para delimitar el espacio aéreo y el espacio ultraterrestre, en las cuales existen imprecisiones y divergencias, se ha postulado que el espacio ultraterrestre co- mienza por lo menos a partir de la altura más baja en la cual haya • sido colocado en órbita alrededor de la tierra un satélite artificial. Esta posición se fundamenta en la necesidad de impedir que un Estado reclame derechos soberanos hasta una altura superior al perigeo más bajo de satélites ya colocados en órbita, lo cual infrin- giría el principio consagrado en el Tratado de 1967 de la libertad del espacio ultraterrestre. Además, la Convención de Registro de Objetos Lanzados al Espacio Ultraterrestre se ha aplicado en el convencimiento de que los satélites 'artificiales de la Tierra son objetos espaciales, y por lo tanto,, lanzados al Espacio Ultraterres- tre. Aun cuando existen diversas teorías para delimitar el espacio aéreo y el espacio ultraterrestre, en las cuales existen imprecisiones y divergencias, se ha postulado que el espacio ultraterrestre co- mienza por lo menos a partir de la altura más baja en la cual haya • sido colocado en órbita alrededor de la tierra un satélite artificial. sido colocado en órbita alrededor de la tierra un satélite artificial. Esta posición se fundamenta en la necesidad de impedir que un Estado reclame derechos soberanos hasta una altura superior al perigeo más bajo de satélites ya colocados en órbita, lo cual infrin- giría el principio consagrado en el Tratado de 1967 de la libertad del espacio ultraterrestre. Además, la Convención de Registro de Objetos Lanzados al Espacio Ultraterrestre se ha aplicado en el convencimiento de que los satélites 'artificiales de la Tierra son objetos espaciales, y por lo tanto,, lanzados al Espacio Ultraterres- tre. 4.2.- La Tele observación de los recursos terrestres desde el es- pacio. La teleobservación terrestre es un método basado en la emisión y [ 3 9 9 ] E S T U D I O S I N T E R N A C I O N A L E S reflexión de radiaciones electromagnéticas, que permite apreciar la naturaleza y caracteres de los fenómenos existentes tanto sobre como debajo de la superficie de la Tierra por medio de observacio- nes y mediciones efectuadas desde estaciones orbitales? ¿Definición del Grupo de Trabajo de las N.U. sobre Teleobservación Te- rrestre mediante satélites, Doc. N.U. A/AC. 105/III, 14 de febrero de 1973, pág. 2. 3 Véase, Irigoin Barrerme, Jeannette "El Derecho Internacional del Espa- cio y la Cooperación Internacional". Revista Chilena de Derecho, Pontificia Universidad Católica de Chile, Vol. II, Nos 2 - 3, mayo - diciembre 1984 pág. 543 y sgtes. Jeannette Irígoin B, / El espacio, ¿patrimonio común de la humanidad? Jeannette Irígoin B, / El espacio, ¿patrimonio común de la humanidad? 4.3.- Transmisiones directas por televisión. Otra tecnología espacial que se está utilizando consiste en la radio- difusión, particularmente las transmisiones directas por televisión, mediante satélites artificiales de la Tierra situados en el espacio ultraterrestre. Los problemas jurídicos que plantean las transmisiones direc- tas a través de satélites, se refieren especialmente: ) l bi ió l i lt t t d l t i a) la ubicación en el espacio ultraterrestre de las estaciones apropiadas para las transmisiones directas; b) l l d l id d l i b) el control del contenido de los programas que se transmi- tan. ) l l i bl l bi ió d l a) En lo que respecta al primer problema —la ubicación de las estaciones— el instrumento más eficaz en materia de telecomunica- ciones está constituido por las "estaciones de satélites" situadas en la órbita geoestacionaria; sin embargo, el número de satélites que pueden colocarse en dicha órbita es limitado, puesto que cada satélite debe colocarse por lo menos a 2 grados de distancia de otro, de modo de evitar interferencias perjudiciales y riesgos de colisión. a) En lo que respecta al primer problema —la ubicación de las estaciones— el instrumento más eficaz en materia de telecomunica- ciones está constituido por las "estaciones de satélites" situadas en la órbita geoestacionaria; sin embargo, el número de satélites que pueden colocarse en dicha órbita es limitado, puesto que cada satélite debe colocarse por lo menos a 2 grados de distancia de otro, de modo de evitar interferencias perjudiciales y riesgos de colisión. Desde un punto de vista técnico, el límite máximo de estacio- nes que pueden ubicarse en la órbita geoestacionaria asciende a 180 satélites. Además, algunos "canales" en los lugares disponi- bles son más ventajosos que otros por razones técnicas o comercia- les. En efecto, pocos "lugares de estacionamiento" sobre el Océano Indico pueden abarcar en forma conjunta transmisiones dirigidas a la vez a Gran Bretaña y Japón. Por lo tanto, los órganos técnicos de N.U. han declarado que "la órbita geoestacionaria es un recurso natural limitado". 4.1.- Los límites del espacio ultraterrestre Este sistema ofrece aplicaciones prácticas de suma utilidad para la detección de los recursos terrestres como la prospección de yacimientos minerales, ubicación y conservación de recursos hídricos y otras fuentes de energía, predicciones meteorológicas y sísmicas, determinación -del nivel de contaminación del aire y del mar, investigación de la 'vida submarina y dirección de las corrientes marítimas, detección de incendios forestales, migracio- nes de langostas, deriva de icebergs, etc. g De aquí se desprende que los Estados que exploran y utilizan el espacio ultraterrestre están en condiciones de obtener informa- ción acerca de recursos naturales bajo jurisdicción territorial de otros Estados, lo que plantea graves problemas al Derecho Inter- nacional que se refieren a la adquisición y a la divulgación de esa información: a) si un Estado está facultado a obtener esa información sin el consentimiento del Estado teleobservado, a) si un Estado está facultado a obtener esa información sin el consentimiento del Estado teleobservado, b) si puede, sin este consentimiento, divulgar la información obtenida. Como no parece posible ni aceptable imponer un sistema restringido de teleobservación, el principal problema que se ha discutido en la sub Comisión Jurídica de N.U. es el de la divulga- ción de los datos obtenidos. Ha prevalecido un criterio intermedio que concilia el estímulo al desarrollo de la telepbservación y los intereses de otros Estados a quienes satisface -la proposición canadiense de crear ciertas res- tricciones en el manejo de los "análisis y la información procesada referente a recursos naturales" de un Estado teleobservado, con el fin de respetar su carácter confidencial o bien la necesidad priorita- ria de ese Estado de acceder a esa información hasta que ya no sea posible utilizarla en perjuicio de sus intereses. De esta forma el Estado teleobservado se asegura cierta prioridad en el conoci- miento de la información procesada. [ 4 0 0 ] 4.3.- Transmisiones directas por televisión. Esta separación del principio "primero en el tiempo, primero en el derecho", tiene como objeto evitar el monopolio o la prioridad permanente en la asignación de frecuen- cias en favor de las dos potencias espaciales. Se acepta que la asignación de frecuencias terrestres registradas implique el reconocimiento de derechos prioritarios, ya que todos los Estados están en condiciones técnicas y económicas de operar servicios de radiocomunicación no espaciales en un pie de igual- dad. No sucede lo mismo en materia de comunicaciones desde el espacio, las que requieren el lanzamiento y el manejo de objetos espaciales, que en la actualidad sólo pueden ser lanzados por los países que poseen el potencial económico y técnico necesario. N h did l i i l l j ídi No se ha extendido a las estaciones espaciales el status jurídico que garantiza la protección internacional de las frecuencias terres- tres asignadas y debidamente registradas. Se ha establecido que el registro de una frecuencia espacial en la UIT no origina ningún derecho prioritario sobre esta frecuencia, (por lo tanto, a su posi- ción en la órbita) y no deberá considerarse como un obstáculo que impide la instalación posterior de sistemas espaciales pertene- cientes a otros Estados. Esta separación del principio "primero en el tiempo, primero en el derecho", tiene como objeto evitar el monopolio o la prioridad permanente en la asignación de frecuen- cias en favor de las dos potencias espaciales. b) El bl á l i fi b) El problema más grave que se plantea en esta materia se refiere al contenido de los programas que pueden difundirse en otros paí- ses mediante transmisiones directas por medio de satélites colo- cados en el espacio ultraterrestre. Actualmente, las transmisiones vía satélite sólo pueden captarse en lo que se llaman "receptores comunes" que luego deben retransmitir los programas. De este modo, los Estados tienen la fiscalización del contenido de los pro- gramas y aplican leyes y reglamentos nacionales para otorgar licen- cias y controlar todas las formas de transmisión al público. Sin embargo, se ha estimado que en los próximos años estarían dadas las condiciones técnicas y económicas que permitan la recepción directa de emisiones por medio de aparatos privados en cada hogar, sin necesitar el auxilio del receptor común. Esto ha causado preocupación acerca del impacto político y cultural que provocaría la emisión de programas extranjeros exentos del control gubernamental. 4.3.- Transmisiones directas por televisión. Existe preocupación ante la posibi- lidad de apropiación de esos lugares de estacionamiento privile- giados por parte de las dos potencias capaces hoy de colocar saté- lites en órbita.3 S h ll d l l ió d d l ib ió d Se ha llegado a la solución de no acordar a la atribución de frecuencias en el espacio ultraterrestre-el reconocimiento interna- cional que tienen las frecuencias de .radio asignadas en la tierra. Como el espectro de radio-frecuencias en la tierra también Se ha llegado a la solución de no acordar a la atribución de frecuencias en el espacio ultraterrestre-el reconocimiento interna- cional que tienen las frecuencias de .radio asignadas en la tierra. Como el espectro de radio-frecuencias en la tierra también constituye un recurso natural limitado, su utilización está regida por las Convenciones y Reglamentos de Telecomunicaciones Inter- p cional que tienen las frecuencias de .radio asignadas en la tierra. Como el espectro de radio-frecuencias en la tierra también constituye un recurso natural limitado, su utilización está regida por las Convenciones y Reglamentos de Telecomunicaciones Inter- [401] E S T U D I O S I N T E R N A C I O N A L E S nacionales, cuyo cumplimiento está encargado a un órgano com- puesto por expertos independientes —la Junta de Registro de Radio Frecuencias Internacionales— que actúa dentro del marco de la Unión Internacional de Telecomunicaciones (UIT). Se acepta que la asignación de frecuencias terrestres registradas implique el reconocimiento de derechos prioritarios, ya que todos los Estados están en condiciones técnicas y económicas de operar servicios de radiocomunicación no espaciales en un pie de igual- dad. No sucede lo mismo en materia de comunicaciones desde el espacio, las que requieren el lanzamiento y el manejo de objetos espaciales, que en la actualidad sólo pueden ser lanzados por los países que poseen el potencial económico y técnico necesario. No se ha extendido a las estaciones espaciales el status jurídico que garantiza la protección internacional de las frecuencias terres- tres asignadas y debidamente registradas. Se ha establecido que el registro de una frecuencia espacial en la UIT no origina ningún derecho prioritario sobre esta frecuencia, (por lo tanto, a su posi- ción en la órbita) y no deberá considerarse como un obstáculo que impide la instalación posterior de sistemas espaciales pertene- cientes a otros Estados. 4.3.- Transmisiones directas por televisión. El problema está en encontrar una solu- ción que armonice el derecho fundamental del individuo en mate- ria de libertad de información y la soberanía nacional de los Estados. • ' [ 4 0 2 ] Jeannette Irígoin B. / El espacio, ¿patrimonio común de la humanidad? Jeannette Irígoin B. / El espacio, ¿patrimonio común de la humanidad? En el Grupo'de Trabajo de N.U. (1969) surgieron dos opinio- nes contrapuestas: Estados Unidos, partidario del reconocimiento de una libertad total en materia de transmisiones directas de televi- sión, basada en la libertad de pensamiento e información y en la oposición a toda forma de censura previa y a la intromisión de los Gobiernos en el control de los programas y la Unión Soviética que propuso, por su parte, admitir las transmisiones directas de televisión "a los Estados extranjeros únicamente con el consenti- miento expreso-de estos últimos", señalando que el Estado recep- tor que sea objeto de transmisiones televisivas ilegales podrá emplear todos los medios a su alcance para impedirlas, no sólo dentro de su territorio, sino también en el espacio ultraterrestre y en otras regiones más allá de los límites de la jurisdicción nacional de algún Estado. g Esta discusión preocupa a todos los países en los debates de N.U. ante la posibilidad de un sistema de transmisión vía satélite que escape a todas las normas del Estado nacional. Considerando que todos los países han regulado la radiodifusión interna a fin de que no sea culturalmente inaceptable, políticamente tenden- ciosa o difamatoria, no se ve la razón para que las transmisiones de origen extranjero queden exentas de un control similar, expo- niendo a los países a una influencia excesiva de culturas foráneas. Lo que interesa encontrar es un equilibrio entre la libertad del transmisor y la libertad del receptor. y p Algunos -países latinoamericanos han propuesto exigir el consentimiento previo no sólo para proteger los valores políticos, económicos y sociales de los países receptores, sino también para asegurar la participación local en los programas. En los últimos años se han aproximado las oposiciones contra- puestas: el requisito del consentimiento previo ha sido reempla- zado por el de "consultas y acuerdos éntrelos Estados", tratando de conciliarias diversas opiniones. 4 Informe de k Comisión sobre k Utilización del Espacio Ultraterrestre con .Fines Pacíficos, Doc. Of. Asamblea General, 32° sesión, suplemento NO 20 (A/32/30), Anexo 5, par. 1°. satélite no excede los límites autorizados por la UTI4 satélite no excede los límites autorizados por la UTI4 4.3.- Transmisiones directas por televisión. Aunque esta solución de "consultas y acuerdos" ha logrado cierta aceptación, no todos están de acuerdo en sus proyecciones: algunos países sostienen que las consultas deberían ser obligato- rias, en cambio otros (Estados Unidos),-consideran que no. En la reunión de 1979 se manifestaron sustanciales divergencias sobre este punto y también acerca del tratamiento que debe otorgarse a los casos en que la emisión en territorio extranjero no es inten- cional y se debe al desbordamiento de la irradiación de la señal del satélite. En esos casos, no serían necesarias consultas y acuerdos previos, cuando el desbordamiento de la irradiación de la señal del [ 4 0 3 ] E S T U D I O S I N T E R N A C I O N A L E S satélite no excede los límites autorizados por la UTI4 BIBLIOGRAFÍA ternacionales, Universidad de Chile, Santiago, 1984. (320 págs.). É CHRISTOL, C. Q.., "The Geostationa- ry Orbital Position as a Natural Resource of the Space Environ- ment". Netherlands Interna- tional Law Review, XXVI, 1979,1. JIMÉNEZ DE ARECHEGA, EDUARDO El Derecho Internacional Con- temporáneo- Editorial Tecnos, Madrid, 1980. DELEAU, OLIVIER. "La responsabi- lité pour dommages causes par les objets lances dans l'espace extra-atmospherique", Annuaire Frangois de Droit Internatio- nal, 1968, págs. 747'- 755. MARCOFF, M. "Sources du Droit International de l'Espace". Re- cueil des Cours Academie de Droit International deLaHaye, T. II, 1980. INFANTE CAFFI, M. TERESA e IRIGOIN BARRENNE JEANNETTE MATEESCO-MATTE, N. Aerospace. Lazu: Telecommunications Sa- tellites. Me Gilí University, The Instituye of Air and Space Law. Montreal, Canadá, 1982. N N C , . S e IRIGOIN BARRENNE, JEANNETTE (eds.) La utilización del espacio exterior y las comunicaciones; nuevas perspectivas y proble- mas. Instituto de Estudios In- ' [ 4 0 4 ] ' [ 4 0 4 ]
https://openalex.org/W4379051989	https://www.itm-conferences.org/articles/itmconf/pdf/2023/03/itmconf_icdsia2023_02014.pdf	English	null	Time Series Prediction of Wheat Crop based on FB Prophet Forecast Framework	ITM web of conferences	2,023	cc-by	3,812	Time Series Prediction of Wheat Crop based on FB Prophet Forecast Framework Mittal Desai1* and Amisha Shingala1 1CMPICA, CHARUSAT University, CHANGA, 388421, India Abstract. The production of wheat plays an important role in the Indian economy. Wheat yield prediction is significant in trade, industry, and agriculture to increase profitability and better growth for farmers. We propose a prediction model to classify the wheat yield using time series analysis using the FB Prophet algorithm, which is considered as better than most of the other supervised learning models with respect to accuracy. [1]. The study aims to evaluate the predicted growth of wheat yield for the next five years. The dataset is collected by the government agency of India [2], considering the years 1997 to 2022, seasonal data, Gujarat state with four districts, and analysis is done for the Wheat/ Rabi crop. A total of 589 instances are collected from a dataset. We pre-process the data, train the data, and through the testing result set, the experimental result indicates the model achieves the lowest Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE) for the summer wheat prediction (10.03 and 0.39 respectively) when the number of the layer in seasonality is yearly. The study will help the research community and other stakeholders to make plans for the next five years for the sustainable growth of India. * Corresponding author: mittaldesai.mca@charusat.ac.in ITM Web of Conferences 53, 02014 (2023) ICDSIA-2023 ITM Web of Conferences 53, 02014 (2023) ICDSIA-2023 ITM Web of Conferences 53, 02014 (2023) https://doi.org/10.1051/itmconf/20235302014 © The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0 (https://creativecommons.org/licenses/by/4.0/). 1 Introduction The Time series concept is important to understand when there is a need to identify time- based patterns such as how much inventory to maintain, and how to make plans for the next five years for the profitability of the business in the field of agriculture. Crop yield production is one of the important tools that enable the decision-maker to enhance yield and increase productivity for better sustainable growth of India. The agricultural eternity struggles a lot to increase the yield quality so that they can sell their product with high profitability. Therefore, we got the motivation to study the concept which helps farmers to predict and see the yield of their crop before they cultivate and can develop new ideas in the field of yield forecasting. To enhance yield production, we should know the pattern of related data with respect to time. The major components can be the level, trend, and seasonality of the product. To run the components, Facebook has introduced a new open-source forecasting tool- FB Prophet, whose library is available in R and Python languages. The FB Prophet is based on curve fitting techniques whose parameters can be understood and does not require much time to make the prediction [3]. This method is most suitable for seasonal attributes influencing ITM Web of Conferences 53, 02014 (2023) ITM Web of Conferences 53, 02014 (2023) ICDSIA-2023 https://doi.org/10.1051/itmconf/20235302014 factors, various trends and outlier detection. The FB Prophet is used to develop the forecasting needs of the business environment with a time series data set that consists of hourly, daily and monthly data. It also takes care of holiday or break intervals in advance, which we have not considered in our studies. It also does trends, outlier detection, missing data, and so on. factors, various trends and outlier detection. The FB Prophet is used to develop the forecasting needs of the business environment with a time series data set that consists of hourly, daily and monthly data. It also takes care of holiday or break intervals in advance, which we have not considered in our studies. It also does trends, outlier detection, missing data, and so on. 1 Introduction The objective of our study is to do a univariate time series analysis on Rabi crop data of Gujarat state using FB Prophet tool, which helps to enhance our Indian economy by increasing crop yields and losses that can be prohibited by various means of production. The data is collected from authorized government agencies. The data can be analyzed by fitting it to various time series models. 1.1 Background Study Many researchers have carried out work on time series analysis, machine learning, and deep learning models for various problem domains. The table-1 discusses the various researchers work along with methodology, tools, and technology used, data set used and their findings. By analyzing all findings, we found that time series analysis works on a time-based approach in which FB Prophet gets good results which are discussed in the experimental results section. The table-1 below gives the summary of work done in the area of time series prediction from 1997 to 2022. Table 1. Literature review. Dataset Algorithm/Methodolo gy used Conclusion A Comparative Analysis of Crop Yield Prediction using Regression[4] https://data.gov.in/ resources/ district-wise- season-wise-crop- production-statistics- 1997. Dataset taken for state-by- state, district-by-district, year-by-year and season- by season for India. This methodology focuses on predicting crop yield based on data using models like Linear Regression (LR), Lasso Regression (LASSO), Decision Tree Regression (DT), and Random Forest (RF) Regression. The finding from this paper indicates that Regression model was best model with highest accuracy. Wheat Yield Prediction Using Neural Network and Integrated SVM- NN with Regression[5] Data was collected from the Statistic Bureau of Pakistan from 1997 to 2017 for various factors such as soil, fertilizer, yield production etc. The main focus was to forecast the yield of wheat in Sargodha. The researcher has used Neural network, support vector machine and regression techniques to estimate the wheat yield production, Weka software is used for pre- processing of data. The findings of the researcher show good performance of neural network using R-squared value. A Time-Series- Based Yield Forecasting Model Using Stacked LSTM To Predict The Yield Of Paddy In Cauvery Delta Zone In Tamilnadu[6] ICRISAT website 'vdsa.icrisat.ac.in,' which is part of the Village Dynamics in South Asia (VDSA) project. The proposed prediction model used in this paper is Long Short Term Memory(LSTM) – a deep learning algorithm. The model is well trained by deep learning parameters such as epochs, batch size and optimizer for rice dataset which is trained with large dataset using Table 1. Literature review. 2 https://doi.org/10.1051/itmconf/20235302014 ITM Web of Conferences 53, 02014 (2023) ICDSIA-2023 weather, rainfall, humidity and humidity of soil. Time Series Forecasting Model for Supermarket Sales using FB- Prophet[7] They have used a supermarket furniture dataset The proposed research uses an additive model, the Autoregressive integrating moving average(ARIMA) model. 1.1 Background Study The findings concluded that The prophet gives good results with respect to accuracy and scalability for the large data set. Also, transfer learning approach can be used to improve the performance. Crop Yield Prediction Using Hybrid Deep Learning Algorithm for Smart Agriculture[8] ImageNet dataset used. The author has used the hybrid deep learning algorithm. Also the sensor-based camera is used for crop monitoring. For gate implementation, Raspberry pi is used. The system allows the farmer to control the crop yield via cloud-based application, where sensor data is managed by farmers for crop monitoring. Also, to monitor temperature and humidity, RFID sensor is used. Prediction is used to evaluate the quality of the product. Time Series Similarity Analysis Framework in Fresh Produce Yield Forecast Domain[9] The sequential daily yield datasets of three types of FP, including strawberry, raspberry, and blueberry, as well as environmental information related to the Santa Maria region, California, between the years 2011 to 2019, are used to develop and evaluate the models A framework is suggested to investigate the potential similarities between pairs of fresh produce (FP) by focusing on the yield forecasting application Several synthetic time series are generated to evaluate and tune these thresholds. According to the results, the SP is 82% and 52% for strawberry versus raspberry and strawberry versus blueberry, respectively. A Robust and Novel Regression Based Fuzzy Time Series Algorithm for Prediction of Rice Yield[10] historical data of rice yield of University of Agriculture and Technology, Frequency based partitioning has been used as the partition of discourse and actual production as the universe of discourse. Subsequently, Fuzzy Logical Relationships of varying degrees have been executed to effectuate the fuzzification process A Regression Analysis Model has been enforced to accomplish the defuzzification operation 3 https://doi.org/10.1051/itmconf/20235302014 ITM Web of Conferences 53, 02014 (2023) ICDSIA-2023 Mapping winter and summer crops in Uruguay using MODIS time series[11] Four Datasets used for MODIS daily products. It uses spatiotemporal variability and analysis for classification. It also combines NDVI profiles, multi-spectral composites and expert knowledge for high resolution of data. The findings indicate the crop phenology metrics and monitoring inter-annual variability and yield estimation. 2.1 Dataset Description The dataset was collected by the Ministry of Agriculture, India from 1997 to 2019. It consists of parameters such as Area under cultivation(hectare), production (thousand kg) and yield(kg/hectare). We have selected top wheat-producing states such as Ahmedabad, Sabar Khantha, Bhavnagar, and Surendranagar. Using the Principal Component Analysis (PCA) approach of dimensional reduction for a selected state, the relevant features can be determined. The study was conducted to build and train to find the best-fit model using the FB Prophet approach. Comparative research was also conducted in order to estimate the impact of these states on India as a whole. 1.1 Background Study Time Series Forecasting using Facebook Prophet for Cloud Resource Management[12] Time series data points are used It uses Facebook Prophet forecast framework on Microsoft Azure VM workload to predict future resource utilization The result indicates that increase forecasting accuracy with an average percentage change of over 85%. A comparative study for Time Series Forecasting within software 5G networks [13] Four datasets were used for memory usage, operational statistics and outcome for 4 models. Time series forecasting model such as ARIMA (Auto Regressive Integrated Moving Average and FB Prophet techniques. Also Open5GCore tookit is implemented. The prophet model is able to handle outliers and missing values in the time series data, unlike ARIMA which needs regular spaced measurements. ARIMA works well with stationary data. Suicide Trend Analysis and Prediction in India using Facebook Prophet[14] Dataset is taken from MHRD, Govt of India from 2007 to 2016. It contains 197714 instances. A time series algorithm is applied to draw inferences and conclusions; the model is capable to predict trends for ‘n’ number of years. The MAPE and SMAPE error techniques used for accurate measurement in range 0.1 to 0.2 and less than 12. 2.3 Data Pre-processing Here we clean the data in terms of removing unwanted columns, including missing values. It is used to organize the data so that future analysis can be accurate. 2.4 Exploratory data analysis Here we display the collected data and predicted values in terms of graphics and matrices. 2.2 Feature Extraction: Principal Component Analysis(PCA) PCA aims to find the maximum variance in high dimensional data with equal or less than the original area. In our study, we applied to twelve different states of India and we obtained 4 ITM Web of Conferences 53, 02014 (2023) https://doi.org/10.1051/itmconf/20235302014 ITM Web of Conferences 53, 02014 (2023) ICDSIA-2023 maximum output from four states, such as Ahmedabad, Sabar Khantha, Bhavnagar and Surendranagar. maximum output from four states, such as Ahmedabad, Sabar Khantha, Bhavnagar and Surendranagar. 𝑹𝑴𝑺𝑬= 1𝑵𝑵∑𝒊= 1(^𝒚𝒊−𝒚𝒊)2 (1) (1) (2) (2) 2.5 FB Prophet Model Prophet is an open-source software given by Facebook for time series data. It is able to fit data having non-linear trends with yearly, weekly, and daily seasonality. It can pre-process the data. The components are combined in the equation given below [15] y(t) = g(t) + s(t) + h(t) + t g(t): trend of the model, which shows the increase or decrease in the data series. Prophet models the non-periodic changes in the data, by fitting piecewise linear or logistic growth curve over the trend of the time series data. This linear fitting is used so that the model is not affected by outliers or missing data. s(t): seasonality of the model (with Fourier Series), which shows the periodic changes (e.g. weekly/yearly seasonality) in the data because of seasonal factors (based on the year). h(t): holiday effects, which is provided by the user to apply the effects of holidays or big events on the behavior of time series data and t: error term, which stands for any unusual changes that are not fitted by the model. The Prophet model has the advantages of fast training speed and easy interpretation of variables. This model allows data analysts to add their own judgments to the prediction through a set of key model parameters and options. The biggest advantage is that it allows people with certain data analysis experience to convert background knowledge into new parameters and introduce them into the model. The parameter settings of the Prophet model used in this article are shown in Table 2: Table 2. Prophet model parameter setting. Parameter Numerical Value Function Growth Linear Growth mode Seasonality yearly true Holidays None none Seasonality mode and scale 0.5 and 10 Multiplicative Changepoint_prior_scale 0.05 Knee point control parameters Fourier_order 3 None Table 2. Prophet model parameter setting. 5 ITM Web of Conferences 53, 02014 (2023) ICDSIA-2023 ITM Web of Conferences 53, 02014 (2023) https://doi.org/10.1051/itmconf/20235302014 2.6 Model Validation The predictive accuracy was evaluated by a training model on training and testing datasets. The most common measurement of model is RMSE and MAPE whose formula [16] are as shown in equation (1) and (2). 3.1 Error validation and parameter value for training and testing datasets by using FB Prophet for wheat yield prediction The prediction feature of the software model is the main focus of the study. In the decision of techniques implemented to the software, an experiment has been designed to discover the most optimal algorithm which performs with high accuracy and capability of handling increasing data. The prediction performance of model FB Prophet was accessed by Mean Absolute Percentage Error(MAPE), and Root Mean Square Error(RMSE) as shown in Table 3, where District1 indicates Ahmedabad, District2 indicates Sabar Kantha, District3 indicates Bhavnagar and District4 indicates Surendranagar: 6 ITM Web of Conferences 53, 02014 (2023) ITM Web of Conferences 53, 02014 (2023) ICDSIA-2023 https://doi.org/10.1051/itmconf/20235302014 Table 3. MAPE/RMSE indicator. Indicator for Prophet model District1 District2 District3 District4 MAPE 15.460291528 42396 10.0362775376 17408 20.9967201454 2215 20.2110492526 66374 RMSE 0.3907658932 190998 0.41672311043 49757 0.69539317541 48478 0.63715786340 15942 3.2 Graphical representation of model prediction Figure 1-3 represents the testing data vs. predicted graph obtained from training data using FB Prophet model showing its prediction, trends, and seasonality for 1997 to 2022. Ahmedabad FB Prediction Sabarkantha FB Prediction Bhavnagar FB Prediction Surendranagar FB Prediction Fig. 1. FB Prediction for four districts of Gujarat wheat yield prediction. The y-axis of figure represents year-wise wheat yield prediction values and x-axis represents years. Ahmedabad Trend Prediction Sabarkantha Trend Prediction Table 3. MAPE/RMSE indicator. Indicator for Prophet model District1 District2 District3 District4 MAPE 15.460291528 42396 10.0362775376 17408 20.9967201454 2215 20.2110492526 66374 RMSE 0.3907658932 190998 0.41672311043 49757 0.69539317541 48478 0.63715786340 15942 Table 3. MAPE/RMSE indicator. MAPE 15.460291528 42396 10.0362775376 17408 20.9967201454 2215 20.2110492526 66374 RMSE 0.3907658932 190998 0.41672311043 49757 0.69539317541 48478 0.63715786340 15942 3.2 Graphical representation of model prediction Figure 1-3 represents the testing data vs. predicted graph obtained from training data using FB Prophet model showing its prediction, trends, and seasonality for 1997 to 2022. Ahmedabad FB Prediction Sabarkantha FB Prediction Bhavnagar FB Prediction Surendranagar FB Prediction Fig. 1. FB Prediction for four districts of Gujarat wheat yield prediction. The y-axis of figure 3.2 Graphical representation of model prediction 3.2 Graphical representation of model prediction Figure 1-3 represents the testing data vs. predicted graph obtained from training data using FB Prophet model showing its prediction, trends, and seasonality for 1997 to 2022. FB Prophet model showing its prediction, trends, and seasonality for 1997 to 2022. Ahmedabad FB Prediction Sabarkantha FB Prediction Bhavnagar FB Prediction Surendranagar FB Prediction Fig. 1. FB Prediction for four districts of Gujarat wheat yield prediction. The y-axis of figure represents year-wise wheat yield prediction values and x-axis represents years. Ahmedabad Trend Prediction Sabarkantha Trend Prediction Ahmedabad FB Prediction Sabarkantha FB Prediction Bhavnagar FB Prediction Surendranagar FB Prediction Fig. 1. FB Prediction for four districts of Gujarat wheat yield prediction. The y-axis of figure Ahmedabad FB Prediction Sabarkantha FB Prediction Sabarkantha FB Prediction Ahmedabad FB Prediction Sabarkantha FB Prediction Surendranagar FB Prediction Bhavnagar FB Prediction Fig. 1. FB Prediction for four districts of Gujarat wheat yield prediction. The y-axis of figure represents year-wise wheat yield prediction values and x-axis represents years. Fig. 1. FB Prediction for four districts of Gujarat wheat yield prediction. The y-axis of figure represents year-wise wheat yield prediction values and x-axis represents years. Sabarkantha Trend Prediction Ahmedabad Trend Prediction Sabarkantha Trend Prediction Ahmedabad Trend Prediction Ahmedabad Trend Prediction 7 7 ITM Web of Conferences 53, 02014 (2023) ITM Web of Conferences 53, 02014 (2023) ICDSIA-2023 https://doi.org/10.1051/itmconf/20235302014 Bhavnagar Trend Prediction Surendranagar Trend Prediction Fig. 2. Shows the plotting of components trends in wheat yield production using FB Prophet model of all four districts of Gujarat. Bhavnagar Trend Prediction Surendranagar Trend Prediction Surendranagar Trend Prediction Bhavnagar Trend Prediction Fig. 2. Shows the plotting of components trends in wheat yield production using FB Prophet model of all four districts of Gujarat. Ahmedabad Seasonality Prediction Sabarkantha Seasonality Prediction Bhavnagar Seasonality Prediction Surendranagar Seasonality Prediction Fig. 3. Shows the plotting of components’ seasonality in wheat yield production using FB Prophet model of all four districts of Gujarat. Ahmedabad Seasonality Prediction Sabarkantha Seasonality Prediction Sabarkantha Seasonality Prediction Sabarkantha Seasonality Prediction Ahmedabad Seasonality Prediction Ahmedabad Seasonality Prediction Bhavnagar Seasonality Prediction Surendranagar Seasonality Prediction Surendranagar Seasonality Prediction Fig. 3. Shows the plotting of components’ seasonality in wheat yield production using FB Prophet model of all four districts of Gujarat. 4 Conclusion Our paper tries to finalize the FB Prophet model to predict the agricultural crop yield prices for the coming years. The system’s major aim for the farmer fraternity is to get a high amount of crop production and to get maximum profit from their product sales. The study focuses on factors such as trends, and seasonality using the selected model. Compared with other algorithms, the Prophet is more flexible in data processing and it is easier to learn from a small volume of data to obtain an efficient model. 3.3 Comparative study of our model with different models for prediction of wheat yield crop for India Table 4. Comparative study of our model with different models for prediction of wheat yield crop for India Sr. No Methodology used Accuracy 1 A hybrid model of Genetic Algorithm with k-Nearest Neighbour.[22] 98.3% 8 ITM Web of Conferences 53, 02014 (2023) ICDSIA-2023 https://doi.org/10.1051/itmconf/20235302014 2 LSTM[21] 86% 3 PCA - Multivariate Adaptive Regression Splines[9] Average Mean Absolute Error - 21 4 FB Prophet Model- [ current work] Mean Absolute Percentage Error : 10.03 References 1. https://medium.com/analytics-vidhya/facebook-prophet-algorithm-in-time-series- analysis 1. https://medium.com/analytics-vidhya/facebook-prophet-algorithm-in-time-series- analysis 1. https://medium.com/analytics-vidhya/facebook-prophet-algorithm-in-time-series- analysis 1. https://medium.com/analytics-vidhya/facebook-prophet-algorithm-in-time-series- analysis 2. https://aps.dac.gov.in/APY/Public_Report1.aspx 2. https://aps.dac.gov.in/APY/Public_Report1.aspx 3. Y. Ensafi, S.H.Amin, G.Zhang and B.Shah. Time-series forecasting of seasonal items sales using machine learning–A comparative analysis. International Journal of Information Management Data Insights, 2(1), p.100058(2022) 4. S.Rai, J.Nandre and B.R. Kanawade. A Comparative Analysis of Crop Yield Prediction using Regression. In 2022 2nd International Conference on Intelligent Technologies (CONIT) (pp. 1-4). IEEE (2022) 5. N.Ejaz and S.Abbasi. Wheat yield prediction using neural network and integrated svm- nn with regression. Pakistan Journal of Engineering, Technology & Science, 8(2)(2020) 6. M.Geetha, R.C.Suganthe, S.K.Nivetha, R.Anju, R.Anuradha and J.Haripriya. A Time- Series Based Yield Forecasting Model Using Stacked Lstm To Predict The Yield Of Paddy In Cauvery Delta Zone In Tamilnadu. In 2022 First International Conference on Electrical, Electronics, Information and Communication Technologies (ICEEICT) (pp. 1-6). IEEE (2022) 7. B.K.Jha, and S.Pande. Time series forecasting model for supermarket sales using FB- prophet. In 2021 5th International Conference on Computing Methodologies and Communication (ICCMC) (pp. 547-554). IEEE (2021) 8. A.K.Sharma, and A.S. Rajawat. Crop Yield Prediction using Hybrid Deep Learning Algorithm for Smart Agriculture. In 2022 Second International Conference on Artificial Intelligence and Smart Energy (ICAIS) (pp. 330-335). IEEE (2022) 9. F.Jafari, L.Nassar, and F.Karray. Time series similarity analysis framework in fresh produce yield forecast domain. In 2021 IEEE International Conference on Systems, Man, and Cybernetics (SMC) (pp. 2368-2374). IEEE (2021) 10. A.Garg and B.Garg. A robust and novel regression based fuzzy time series algorithm for prediction of rice yield. In 2017 international conference on intelligent communication and computational techniques (ICCT) (pp. 48-54). IEEE (2017) 11. C.Vancutsem, J.F.Pekel, F.Kayitakire, G.Duveiller, M.Meroni, W.Baethgen and Ceccato, P., 2013, August. Mapping winter and summer crops in Uruguay using MODIS 9 9 ITM Web of Conferences 53, 02014 (2023) 2023 ITM Web of Conferences 53, 02014 (2023) ICDSIA-2023 https://doi.org/10.1051/itmconf/20235302014 time series. In 2013 Second International Conference on Agro-Geoinformatics (Agro- Geoinformatics) (pp. 292-295). IEEE (2013) 12. M.Daraghmeh, A.Agarwal, R.Manzano and M.Zaman. Time series forecasting using Facebook Prophet for cloud resource management. In 2021 IEEE International Conference on Communications Workshops (ICC Workshops) (pp. 1-6). IEEE (2021) 13. P. Chakraborty, M. Corici and T.Magedanz. A comparative study for Time Series Forecasting within software 5G networks. In 2020 14th International Conference on Signal Processing and Communication Systems (ICSPCS) (pp. 1-7). IEEE (2020) 14. K.Taunk, P.Singh and R.K.Behera. Suicide Trend Analysis and Prediction in India using Facebook Prophet. 1. https://medium.com/analytics-vidhya/facebook-prophet-algorithm-in-time-series- analysis In 2021 8th International Conference on Computing for Sustainable Global Development (INDIACom) (pp. 670-675). IEEE (2021) 15. S.J. Taylor and B.Letham. Forecasting at scale. The American Statistician, 72(1), pp.37- 45 (2018) 16. B.M.Nayana, K.R.Kumar and C.Chesneau. Wheat Yield Prediction in India Using Principal Component Analysis-Multivariate Adaptive Regression Splines (PCA- MARS). AgriEngineering, 4(2), pp.461-474 (2022) 17. Z.Chen, H.S.Goh, K.L.Sin, K.Lim, N.K.H. Chung, and X.Y.Liew. Automated agriculture commodity price prediction system with machine learning techniques. arXiv preprint arXiv:2106.12747 (2021) 18. R.Alvarez. Predicting average regional yield and production of wheat in the Argentine Pampas by an artificial neural network approach. European Journal of Agronomy, 30(2), pp.70-77 (2009) 19. H. Wang and Z. Ma. Prediction of wheat stripe rust based on support vector machine. In 2011 Seventh International Conference on Natural Computation (Vol. 1, pp. 378-382). IEEE (2011) 20. Y.Çakır, M.Kırcı, and E.O.Güneş. Yield prediction of wheat in south-east region of Turkey by using artificial neural networks. In 2014 The Third International Conference on Agro-Geoinformatics (pp. 1-4). IEEE (2014) 21. J.Hirapara, and P.Vanjara,. A Comparative study of Data Mining Techniques for Agriculture Crop Price Prediction. In 2022 IEEE 7th International conference for Convergence in Technology (I2CT) (pp. 1-6). IEEE (2022) 22. T.Fei, W.Wenbin, L.Dandan, C.Zhongxin, H.Qing, and X.Tian. Yield estimation of winter wheat in North China Plain by using crop growth monitoring system (CGMS). In 2012 First International Conference on Agro-Geoinformatics (Agro- Geoinformatics) (pp. 1-4). IEEE (2012) 23. R.N.Singh, P.Krishnan, V.K.Singh and B.Das. Estimation of yellow rust severity in wheat using visible and thermal imaging coupled with machine learning algorithms. Geocarto International, pp.1-22. (2022) 10
https://openalex.org/W1997598750	https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0017086&type=printable	English	null	A Tumor-Associated Mutation of FYVE-CENT Prevents Its Interaction with Beclin 1 and Interferes with Cytokinesis	PloS one	2,011	cc-by	6,304	Abstract The tumor suppressor activity of Beclin 1 (BECN1), a subunit of class III phosphatidylinositol 3-kinase complex, has been attributed to its regulation of apoptosis and autophagy. Here, we identify FYVE-CENT (ZFYVE26), a phosphatidylinositol 3- phosphate binding protein important for cytokinesis, as a novel interacting protein of Beclin 1. A mutation in FYVE-CENT (R1945Q) associated with breast cancer abolished the interaction between FYVE-CENT and Beclin 1, and reduced the localization of these proteins at the intercellular bridge during cytokinesis. Breast cancer cells containing the FYVE-CENT R1945Q mutation displayed a significant increase in cytokinetic profiles and bi - multinuclear phenotype. Both Beclin 1 and FYVE-CENT were found to be downregulated in advanced breast cancers. These findings suggest a positive feedback loop for recruitment of FYVE-CENT and Beclin 1 to the intercellular bridge during cytokinesis, and reveal a novel potential tumor suppressor mechanism for Beclin 1. Citation: Sagona AP, Nezis IP, Bache KG, Haglund K, Bakken AC, et al. (2011) A Tumor-Associated Mutation of FYVE-CENT Prevents and Interferes with Cytokinesis. PLoS ONE 6(3): e17086. doi:10.1371/journal.pone.0017086 Editor: Andreas Bergmann, University of Texas MD Anderson Cancer Center, United States of America Editor: Andreas Bergmann, University of Texas MD Anderson Cancer Center, United States of America Received November 11, 2010; Accepted January 17, 2011; Published March 24, 2011 Received November 11, 2010; Accepted January 17, 2011; Published March 24, 2011 Copyright: 2011 Sagona et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: A.P.S. is a Ph.D. student supported by ENDOCYTE Research training network of the European Union Marie Curie Program and by the FUGE Program of the Research Council of Norway. I.P.N. is a postdoctoral fellow of European Research Council. K.G.B. is a postdoctoral fellow and K.H. a senior research fellow of the Research Council of Norway. R.I.S. is a group leader supported by the Norwegian Cancer Society, and A.C.B. is a research associate supported by the Research Council of Norway. This work was supported by an Advanced Grant from the European Research Council (to H.S.). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * E-mail: stenmark@ulrik.uio.no Introduction accompanied by cytokinesis arrest and multinuclear phenotype. These results suggest a novel tumor suppressor mechanism for Beclin 1, which is supported by our finding that both Beclin 1 and FYVE-CENT are downregulated in advanced breast cancer. Beclin 1 is a known tumor suppressor protein that regulates apoptosis and autophagy [1–3]. Importantly, Beclin 1 is a subunit of the phosphatidylinositol 3-kinase class III (PI3K-III) complex and interacts directly with VPS34, the catalytic subunit of PI3K class III complex [4–7]. It also serves as a platform for the recruitment of other proteins such as UVRAG (UV radiation resistance-associated gene) [5], BIF-1/Endophilin B1 [8], and ATG14L/Barkor [9,10] with known functions in autophagy and tumor suppression. In addition to its known roles in endocytic and autophagic membrane traffic, it was recently established that the PI3K Class III complex plays a crucial role in cytokinesis [11–13]. More specifically, the phospholipid PtdIns3P, which is produced by VPS34, was found to localize at the intercellular bridge, and depletion of human VPS34 and Beclin 1 resulted in an increased arrest of cells in cytokinesis as well as in an increased amount of binuclear and multinuclear cells [11]. Unsuccessful cytokinesis has been implicated in tumorigenesis but the underlying mechanisms are largely unknown [14]. Antonia P. Sagona1,2, Ioannis P. Nezis1,2, Kristi G. Bache1,2, Kaisa Haglund1,2, Anne Cathrine Bakken1,3, Rolf I. Skotheim1,3, Harald Stenmark1,2* Antonia P. Sagona1,2, Ioannis P. Nezis1,2, Kristi G. Bache1,2, Kaisa Haglund1,2, Anne Cathrine Bakken1,3, Rolf I. Skotheim1,3, Harald Stenmark1,2* 1 Centre for Cancer Biomedicine, Faculty of Medicine, University of Oslo, Oslo, Norway, 2 Department of Biochemistry, Institute for Cancer Research, The Norwegian Radium Hospital, Oslo University Hospital, Oslo, Norway, 3 Department of Cancer Prevention, Institute for Cancer Research, The Norwegian Radium Hospital, Oslo University Hospital, Oslo, Norway PLoS ONE \| www.plosone.org A mutation associated with breast cancer abolishes the interaction between FYVE-CENT and Beclin 1 in a heterozygous cell line, or alternatively, that only the mutant allele is present in the majority of the cells, and the existence of a sub-population of cells which is heterozygous for the mutation. The protein levels of Beclin 1 were comparable in the cell line used as control (HCC-1395) and the mutant FYVE-CENT (HCC- 1954) cells (Figure S1A). Consistent with this, upon siRNA depletion of FYVE-CENT, Beclin 1 protein levels remained the same (Figure S1B). Likewise, FYVE-CENT levels ramained unaffected by depletion of Beclin 1. In contrast, upon depletion of the Beclin 1 interacting protein VPS34, Beclin 1 became downregulated whereas FYVE-CENT protein levels remained the same (Figure S1B). These results show that the FYVE-CENT R1945Q mutation does not affect the protein levels of Beclin 1. The ZFYVE26 gene encoding FYVE-CENT was found mutated in breast cancer samples with a frequency of more than 10% [15]. Since Beclin 1 is a well-known tumor suppressor [2,16], we therefore wanted to test the cell biological consequence of such mutations in the context of FYVE-CENT interaction. To this end we performed a GST pull-down between the C-terminal part of FYVE-CENT (residues 1807–2539) that contains the R1945Q mutation found in breast cancer cell lines [15] and myc-Beclin 1 in HeLa cell lysates. Interestingly we observed that the FYVE-CENT R1945Q mutation abolished the interaction between FYVE- CENT and Beclin 1 (Figure 2C). This was also confirmed in co- immuno-precipitation experiments where endogenous FYVE- CENT R1945Q mutant protein extracted from HCC-1954 breast cancer cells and Beclin 1 did not co-immuno-precipitate with an antibody against endogenous FYVE-CENT, whereas wild-type FYVE-CENT from a control cancer cell line was able to co- immuno-precipitate with Beclin 1 (Figure 2D). We have previously shown that FYVE-CENT interacts with the microtubule-based motor KIF13A and the tetratricopeptide repeat protein TTC19 [11]. KIF13A was found to regulate translocation of FYVE-CENT to the midbody, and the importance of these proteins in cytokinesis is illustrated by the finding that depletion of either FYVE-CENT, KIF13A or TTC19 is sufficient to cause an increased number of cytokinetic profiles and bi- and multinucleate cells [11]. We therefore asked whether the FYVE-CENT R1945Q mutation also interferes with its interaction with KIF13A and TTC19. Interestingly, pull-down assays showed that the R1945Q mutation does not inhibit the interaction of the C-terminus of FYVE-CENT with neither TTC19 nor KIF13A in vitro (Figure 2E). A mutation associated with breast cancer abolishes the interaction between FYVE-CENT and Beclin 1 These data indicate that the FYVE-CENT R1945Q mutation associated with breast cancer specifically abolishes the interaction of FYVE- CENT with Beclin 1. In order to identify any biological consequence of the FYVE- CENT R1945Q mutation, we examined the phenotype of mutant cells by performing immunofluorescence microscopy using the HCC-1954 and HCC-1395 breast cancer cells. We observed that FYVE-CENT R1945Q mutant cells showed an increased population arrested in cytokinesis (16%) compared to the control cells (6%) and also an increased percentage of binuclear- multinuclear profiles (31.5% versus 19%) (Figure 3B–C and Figure S2A–C). In order to examine whether this phenotype is a direct consequence of the FYVE-CENT R1945Q mutation, we tested whether R1945Q mutation can rescue the arrest in cytokinesis observed upon FYVE-CENT depletion. To examine this we back-transfected HeLa cells which were RNAi-depleted for FYVE-CENT with wild type FYVE-CENT C terminus (1807– 2539) or FYVE-CENT C terminus R1945Q mutant. We observed that wild type FYVE-CENT C-terminus could rescue the arrest in cytokinesis and bi-multinuclear phenotype observed upon FYVE- CENT RNAi depletion suggesting that this part of FYVE-CENT entails the minimal functional domains. In contrast, the FYVE- CENT C terminus R1945Q mutant could not rescue the siRNA- induced phenotypes (Figure S3A–C). These data suggest that the FYVE-CENT R1945Q mutation may promote carcinogenesis by interferring with normal cytokinesis. Results FYVE-CENT is a novel Beclin 1 interacting protein We have recently shown that FYVE-CENT is a critical PtdIns3P effector protein that regulates cytokinesis [11]. In order to identify interacting partners of FYVE-CENT, we performed a yeast two-hybrid screen in a human T-lymphocyte library, using the C-terminal part of FYVE-CENT as bait (residues 2120–2539). Using this approach, Beclin 1 was identified as a positive hit (Dataset S1). The interaction of Beclin 1 with FYVE-CENT maps to a region containing the coiled coil domain and the evolutionarily conserved domain of Beclin 1 (Figure 1). To verify this interaction biochemically, we performed a pull-down assay, incubating the C-terminus of FYVE-CENT fused to GST with myc-Beclin 1 expressed in HeLa cell lysates. The pull-down assay showed a positive biochemical interaction (Figure 2A). To further verify this interaction, endogenous FYVE-CENT and Beclin 1 were co-immuno-precipitated with an antibody against FYVE- CENT (Figure 2B), indicating that the two endogenous proteins can form a complex in vivo. Here, we uncover a novel potential tumor suppressor mechanism for Beclin 1. We find that Beclin 1 interacts with FYVE-CENT, a PtdIns3P binding protein involved in cytokine- sis [11]. Further, we show that a tumor-associated mutation of FYVE-CENT abolishes its interaction with Beclin 1, prevents recruitment of Beclin 1 to the intercellular bridge, and is PLoS ONE \| www.plosone.org March 2011 \| Volume 6 \| Issue 3 \| e17086 March 2011 \| Volume 6 \| Issue 3 \| e17086 1 Figure 1. Two-hybrid interactions of Beclin 1 with FYVE-CENT. The figure shows schematically the domain of Beclin 1 that interacts with FYVE-CENT. doi:10.1371/journal.pone.0017086.g001 Beclin 1 Interacts with FYVE-CENT Beclin 1 Interacts with FYVE-CENT Figure 1. Two-hybrid interactions of Beclin 1 with FYVE-CENT. The figure shows schematically the domain of Beclin 1 that interacts with FYVE-CENT. doi:10.1371/journal.pone.0017086.g001 Beclin 1 localizes at the intercellular bridge during cytokinesis, and this localization is abolished in FYVE- CENT R1945Q mutant breast cancer cells Interestingly, Beclin 1 was also found to localize at the intercellular bridge in the control cell line HCC- 1395, whereas in the FYVE-CENT R1945Q mutant breast cancer cell line this localization was significantly reduced (Figure 4A and 4C). Additionally, the localization of FYVE-CENT at the intercellular bridge was partially abolished in the FYVE-CENT R1945Q mutant breast cancer cell line (Figure. 4B and 4C). Taken together, these data suggest that the FYVE-CENT R1945Q mutation prevents localization of Beclin 1 at the intercellular bridge and interferes with proper cytokinesis. found that this protein also localizes at the intercellular bridge (Figure 4A, upper panels). Interestingly, Beclin 1 was also found to localize at the intercellular bridge in the control cell line HCC- 1395, whereas in the FYVE-CENT R1945Q mutant breast cancer cell line this localization was significantly reduced (Figure 4A and 4C). Additionally, the localization of FYVE-CENT at the intercellular bridge was partially abolished in the FYVE-CENT R1945Q mutant breast cancer cell line (Figure. 4B and 4C). Taken together, these data suggest that the FYVE-CENT R1945Q mutation prevents localization of Beclin 1 at the intercellular bridge and interferes with proper cytokinesis. chaperone proteins that would sterically prevent Beclin 1 binding. The R1945Q mutation does not affect the interaction of FYVE- CENT with KIF13A and TTC19, suggesting that it specifically abolishes binding to Beclin-1. The downregulation of FYVE- CENT, BECN 1, KIF13A and TTC19 in advanced breast cancer is consistent with the possibility that these proteins may participate in tumor suppression. We have recently shown that PtdIns3P recruits FYVE-CENT at the midbody during cytokinesis, and that subunits of the PI3K-III complex, including Beclin 1, are required for correct cytokinesis [11]. Our present data suggest a positive-feedback loop model wherein FYVE-CENT can recruit Beclin 1 at the intercellular bridge. Subsequently, Beclin 1 can interact with VPS34, thereby producing more PtdIns3P, which in turn can recruit more FYVE- CENT. This model (Figure 5E) would explain the significant increase in cells arrested in cytokinesis and bi- and multinuclear cells in FYVE-CENT mutant cells and highlight a role for Beclin 1 in cytokinesis. Collectively, our findings reveal a novel regulatory role of the tumor suppressor Beclin 1 and its binding partner FYVE-CENT that has potential implications for carcinogenesis. Cell culture and transfections HeLa cells were grown and transfected as described previously [11]. HCC-1395 (CRL-2324) and HCC-1954 (CRL-2338) cells were purchased from ATCC and grown in RPMI-1640 medium (GIBCO, Invitrogen) supplemented with 10% fetal bovine serum in a 5% CO2 atmosphere at 37uC. Confocal fluorescence microscopy Immunofluorescence microscopy was performed using HeLa, HCC-1395 and HCC-1954 as previously described [11]. The following primary antibodies were used for immunofluorescence studies: rabbit anti-human FYVE-CENT antibody, used in 1:300 dilution, as described before [11], mouse anti-a-tubulin, used in 1:1000 dilution and purchased from SIGMA, rabbit anti-human Beclin 1 and mouse anti-human Aurora B antibody, both used in 1:200 dilution and purchased from Abcam. The secondary antibodies used were goat-anti-mouse Alexa FluorH 488, in 1:500 dilution from Invitrogen and Cy3-labelled goat anti-rabbit antibody, in 1:500 dilution and Cy2-labelled goat anti-mouse antibody, in 1:200 dilution purchased from Jackson Immunor- esearch. Alexa FluorH 594 phalloidin, used in 1:750 dilution, and Hoechst 33342, used at 1 mg/ml, were purchased from Invitrogen. The tumor suppressor activity of Beclin 1 has been attributed to its interactions with proteins that regulate cell death and autophagy [2,3,7]. Our present data suggest an additional mechanism for the tumor suppressor functions of Beclin 1, namely its ability to bind FYVE-CENT and participate in the regulation of cytokinesis. Failure to complete cytokinesis has been implicated in carcinogenesis [14,19,20], and our data demonstrate that the Beclin 1 - FYVE-CENT complex may play important roles in controlling this process. Importantly, mutations in FYVE-CENT associated with breast cancer interfere with its interaction with Beclin 1. It is interesting that loss of this interaction is accompanied by cytokinesis failure, since this suggests a mecha- nism that may contribute to the cancer phenotype of FYVE- CENT mutant cancer cells. The fact that the R1945Q mutation is located outside the minimal interacting part of FYVE-CENT with Beclin 1 may suggest that there are additional interacting surfaces that extend outside the 2120–2539 C-terminal part that was used as bait in the yeast two- hybrid screen. Alternatively, the R1945Q mutation might promote a conformational change in C-terminal folding that could alter its association with Beclin 1, or result in recruitment of Downregulation of FYVE-CENT and Beclin 1 in advanced breast cancer Downregulation of FYVE-CENT and Beclin 1 in advanced breast cancer To further explore the association of FYVE-CENT with breast cancer, we examined its expression pattern in previously published gene expression data [17,18]. We found that the average expression of FYVE-CENT was significantly lower in high vs. low grade breast cancers (Figure 5A). Furthermore, we found that there was a similar significant association between decreased BECN1 mRNA levels and tumor grade (Figure 5B). More specifically, breast cancers of grade 3 had a significantly lower expression mean than grade 1 and 2 tumors. Interestingly, we also observed that the average expression of KIF13A and TTC19 was significantly lower in high vs. low grade breast cancers (Figure 5C and 5D). Altogether, the associations to clinical parameters strengthen the links between FYVE-CENT, Beclin 1 and breast cancer biology. Beclin 1 localizes at the intercellular bridge during cytokinesis, and this localization is abolished in FYVE- CENT R1945Q mutant breast cancer cells In order to examine the biochemical consequences of the cancer-associated R1945Q mutation of FYVE-CENT, we inves- tigated the HCC-1954 breast cancer cell line which contains this mutation [15]. By cDNA sequencing, we confirmed the mutation status of the cell line and also that the mutant gene is indeed expressed (Figure 3A). Interestingly, cDNA sequencing detected almost exclusively the mutant allele and only a weak signal for the wild-type, indicating a preferential expression of the mutant allele We next asked how the interplay between FYVE-CENT and Beclin 1 may regulate cytokinesis. We have recently shown that VPS34, the catalytic subunit of PI3K-III complex, and FYVE- CENT are localized at the intercellular bridge during cytokinesis [11]. Given the interaction of Beclin 1 with FYVE-CENT, we examined the localization of Beclin-1 during cytokinesis, and we PLoS ONE \| www.plosone.org March 2011 \| Volume 6 \| Issue 3 \| e17086 2 Beclin 1 Interacts with FYVE-CENT PLoS ONE \| www.plosone.org 3 March 2011 \| Volume 6 \| Issue 3 \| e17086 PLoS ONE \| www.plosone.org PLoS ONE \| www.plosone.org March 2011 \| Volume 6 \| Issue 3 \| e17086 March 2011 \| Volume 6 \| Issue 3 \| e17086 3 Beclin 1 Interacts with FYVE-CENT Beclin 1 Interacts with FYVE-CENT Figure 2. Beclin 1 interacts with FYVE-CENT. (A) GST pull-down from HeLa cell lysates transiently over-expressing myc-Beclin 1 using recombinant GST-FYVE-CENT C-terminal fusion (2120–2539) protein or GST protein immobilized on glutathione-Sepharose beads. Proteins eluted from the beads were analyzed by SDS-PAGE and immuno-blotting using an anti-myc antibody. Equal amounts of GST-FYVE-CENT C-terminal fusion protein and GST protein were loaded. (B) HeLa cell lysates were subjected to immunoprecipitation (IP) with an antibody against FYVE-CENT. Immunoprecipitated proteins were detected by Western blotting, using anti-Beclin 1 and anti-FYVE-CENT antibodies. (C) HeLa cells transiently over- expressing myc-Beclin 1 were pulled down with recombinant GST-FYVE-CENT C-terminal fusion (1807–2539 or 1807–2539 R1945Q) protein or GST protein immobilized on glutathione-Sepharose beads. (D) HCC-1395 control cells and HCC-1954 FYVE-CENT R1945Q mutant cells were lysed and subjected to immunoprecipitation (IP) with an antibody against FYVE-CENT. Immunoprecipitated proteins were detected by Western blotting, using anti-Beclin 1 and anti-FYVE-CENT antibodies. (E) HeLa cells transiently over-expressing myc-TTC19 or myc-KIF13A were pulled down with recombinant GST-FYVE-CENT C-terminal fusion protein and GST-FYVE-CENT C-terminal R1945Q fusion protein or GST protein immobilized on glutathione-Sepharose beads. doi:10.1371/journal.pone.0017086.g002 found that this protein also localizes at the intercellular bridge (Figure 4A, upper panels). PLoS ONE \| www.plosone.org Immunoblotting To determine the cell-specific distribution of FYVE-CENT, Beclin 1, VPS34, beta-actin and the overexpressed TTC19 and KIF13A-myc tagged constructs, the various cell lines were lysed in lysis buffer (25 mM HEPES pH 7.2, 125 mM potassium acetate, 2.5 mM magnesium acetate, 5 mM EGTA, 1 mM DTT, 0.5% PLoS ONE \| www.plosone.org March 2011 \| Volume 6 \| Issue 3 \| e17086 4 Beclin 1 Interacts with FYVE-CENT Figure 3. A FYVE-CENT R1945Q mutant breast cancer cell line exhibits an increa as bi- and multinuclear cells. (A) Sequencing of cDNA for exons 31 to 33 of FYVE-CENT revealed a G to A substitution at base position 5834 in the HCC1954 cell line. (B). Confocal m lines cells stained with a-tubulin, Alexa FluorH 594 phalloidin and Hoechst. In FYVE-CENT mu arrested in cytokinesis (arrows) compared to the control as well as increase in binuclear-m presentation of quantification of cells arrested at the midbody stage and bi-multinuclear mutant cell line (HCC-1954). Error bars show mean 6 s.d. Control: 3 independent ex experiments, n = 1225 cells. p value for cells arrested at the midbody stage ,0.01. p value doi:10.1371/journal.pone.0017086.g003 Figure 3. A FYVE-CENT R1945Q mutant breast cancer cell line exhibits an increased number of cells arrested in cytokinesis as well as bi- and multinuclear cells. (A) Sequencing of cDNA for exons 31 to 33 of FYVE-CENT from the HCC-1395 and HCC1954 breast cancer cell lines revealed a G to A substitution at base position 5834 in the HCC1954 cell line. (B). Confocal micrographs of HCC-1395 and HCC-1954 breast cancer cell lines cells stained with a-tubulin, Alexa FluorH 594 phalloidin and Hoechst. In FYVE-CENT mutant cells (HCC-1954) there is a significant increase in cells arrested in cytokinesis (arrows) compared to the control as well as increase in binuclear-multinuclear cells (asterisk). Scale bars: 20 mm. (C) Graphic presentation of quantification of cells arrested at the midbody stage and bi-multinuclear cells in control cells (HCC-1395) and FYVE-CENT R1945Q mutant cell line (HCC-1954). Error bars show mean 6 s.d. Control: 3 independent experiments, n = 1142 cells. Mutant cells: 3 independent experiments, n = 1225 cells. p value for cells arrested at the midbody stage ,0.01. p value for binuclear-multinuclear cells ,0.01. doi:10.1371/journal.pone.0017086.g003 Figure 3. A FYVE-CENT R1945Q mutant breast cancer cell line exhibits an increased number of cells arrested in cytokinesis as well as bi- and multinuclear cells. Immunoblotting (A) Sequencing of cDNA for exons 31 to 33 of FYVE-CENT from the HCC-1395 and HCC1954 breast cancer cell lines revealed a G to A substitution at base position 5834 in the HCC1954 cell line. (B). Confocal micrographs of HCC-1395 and HCC-1954 breast cancer cell lines cells stained with a-tubulin, Alexa FluorH 594 phalloidin and Hoechst. In FYVE-CENT mutant cells (HCC-1954) there is a significant increase in cells arrested in cytokinesis (arrows) compared to the control as well as increase in binuclear-multinuclear cells (asterisk). Scale bars: 20 mm. (C) Graphic presentation of quantification of cells arrested at the midbody stage and bi-multinuclear cells in control cells (HCC-1395) and FYVE-CENT R1945Q mutant cell line (HCC-1954). Error bars show mean 6 s.d. Control: 3 independent experiments, n = 1142 cells. Mutant cells: 3 independent experiments, n = 1225 cells. p value for cells arrested at the midbody stage ,0.01. p value for binuclear-multinuclear cells ,0.01. doi:10.1371/journal.pone.0017086.g003 March 2011 \| Volume 6 \| Issue 3 \| e17086 PLoS ONE \| www.plosone.org 5 Beclin 1 Interacts with FY Beclin 1 Interacts with FYVE-CENT PLoS ONE \| www.plosone.org 6 March 2011 \| Volume 6 \| Issue 3 \| e17086 PLoS ONE \| www.plosone.org March 2011 \| Volume 6 \| Issue 3 \| e17086 March 2011 \| Volume 6 \| Issue 3 \| e17086 PLoS ONE \| www.plosone.org PLoS ONE \| www.plosone.org 6 Beclin 1 Interacts with FYVE-CENT Figure 4. The localization of Beclin 1 to the intercellular bridge during cytokinesis is abolished in FYVE-CENT R1945Q mutant breast cancer cells. (A) and (B) Confocal micrographs of HeLa, HCC-1395 and HCC-1954 cells stained with antibodies against Aurora B and Beclin 1 (A) or FYVE-CENT (B), and with Hoechst. Magnifications of the intercellular bridges are shown in the insets. Scale bars: 10 mm. (C) Graphic presentation of quantification of control cells (HCC-1395) and mutant cells (HCC-1954) labeled on the midbody with anti-FYVE-CENT or anti-Beclin 1 antibodies. Error bars show mean 6 s.d. Control cells stained with anti-FYVE-CENT: 4 independent experiments, n = 1769 cells. Mutant cells stained with anti- FYVE-CENT: 4 independent experiments, n = 1781 cells. Control cells stained with anti-Beclin 1: 4 independent experiments, n = 1340. Mutant cells stained with anti-Beclin 1: 4 independent experiments, n = 1521. p value for cells labeled with anti-FYVE-CENT on the midbody: 0.01. p value for cells labeled with anti-Beclin 1 on the midbody: 0.01. Plasmid constructs All the FYVE-CENT constructs used were generated by PCR with the FYVE-CENT cDNA (ORF) (NM_015346.2), which was cloned in a pCMV6-XL4 vector by OriGene Technologies, Inc., as template. Synthetic oligonucleotides were from MWG Biotech. The FYVE-CENT R1945Q mutant was prepared by PCR site- directed mutagenesis. PCR errors were excluded by sequencing. For expression as GST fusion proteins in Escherichia coli BL21 (DE3) cells, the C-terminal part (2120–2539) as well as (1807– 2539) and with mutation (R1945Q) of FYVE-CENT were cloned into pGEX-6P-3 (Pharmacia Amersham). The expression plasmid encoding myc-epitope-tagged mouse KIF13A and the Myc-DDK- tagged ORF clone of Homo sapiens TTC19 (NM_017775.2) were obtained as described previously [11]. Expression in mammalian cells and purification were performed as described previously [11]. RNA interference studies Single deconvoluted siRNAs against FYVE-CENT (cat.no. D- 031136-04), VPS34 (PIK3C3)(cat. no. D-005250-04) and Beclin 1(siRNA 1: cat. no. J-010552-05) were purchased from Dharma- con Research. The siRNA experiments were performed on HeLa cells as described before [11]. HeLa, HCC-1395 and HCC-1954 cells were grown confluent in 10-cm culture dishes and lysed in ice-cold lysis buffer (20 mM HEPES pH 7.2, 2 mM MgCl2, 100 mM NaCl, 0.1 mM EDTA, 0.1% Triton X-100) containing inhibitors (N-ethylmaleimide, mammalian protease inhibitor mixture, phosphatase inhibitor cocktail I and II (Sigma-Aldrich).The lysates were placed on ice and centrifuged at 10,000 g, 4uC and the supernatant was added to the Protein A-coupled magnetic beads (Dynal, Invitrogen) which had been precoupled with rabbit antibody against FYVE- CENT or rabbit IgG as a control, in PBS Tween 20. Antibody Assay of rescuing cytokinesis phenotype in RNAi FYVE- CENT depleted cells HeLa cells were transfected with siRNA (70 nM) against human FYVE-CENT for 72 h. The siRNA-treated cells were then seeded onto coverslips in a 5 cm culture dish and were transfected with myc-tagged C- terminal 1807–2539 and myc-tagged C-terminal 1807–2539 R1945Q FYVE-CENT constructs respectively in three different series of experiments for 36 h. The cells were washed in PBS, stained with anti-myc and anti-a tubulin antibodies and processed in confocal microscopy analysis as described above. The experiment was repeated three times and in total, and 270 back transfected cells were quantified. In parallel, simple depletion experiments using control and FYVE-CENT siRNA were performed in triplicates and quantified using the same stainings and conditions. Immunoblotting doi:10.1371/journal.pone.0017086.g004 Nonidet P40, 1:100 proteinase inhibitor mix (Roche Applied Science). After centrifugation for 5 min at 5,000 g the samples were sonicated for 10 s at 70 volts and incubated for 10 min on ice in lysis buffer. Another centrifugation at 10,000 g separated the supernatant from the pellet and 30 mg of protein of the supernatant was subjected to SDS–PAGE (4–20% gradient) and transferred to Immobilon-P membrane (Millipore) for immuno- blotting. The blot was developed with the Supersignal West Pico Chemiluminescent substrate kit or Supersignal West Femto Maximum Sensitivity Substrate kit (Pierce). The antibodies used for immunoblotting were the following: Rabbit anti-human Beclin 1 antibody used for western blotting and immunoprecipitation, was purchased from Cell Signaling Technology. Rabbit c-Myc polyclonal antibody was purchased from Abcam and the rest antibodies used (anti-FYVE-CENT, anti-VPS34, anti-beta-actin, anti-GST and HRP labeled) were described previously [11]. For quantitative Western blotting, equal amounts of cell lysates (as measured by protein content) from control and mutant cells were loaded in triplicates on a gel for PAGE. The proteins were transferred to a PVDF membrane and stained with antibodies for FYVE-CENT, Beclin1 and b-actin. The bands were detected using LiCore infrared dye secondary antibodies and the Odyssey imaging system. The bands were quantified using the Odyssey quantifying software. coupled magnetic beads and cell lysates were gently mixed for 1 h at 4uC. The beads were then washed with lysis buffer, eluted in 46 sample buffer plus 1 mM DTT at 95uC for 5 min. The eluted proteins were subsequently subjected to SDS–PAGE and immu- noblotting as described previously. GST pull-down assay The GST–FYVE-CENT C-terminus ( amino acid residues 2120–2539), the GST–FYVE-CENT C-terminus (1807–2539) and the GST–FYVE-CENT C-terminus (1807–2539) mutant R1945Q constructs were expressed in BL21 Escherichia coli, purified and GST-pull down assays were performed as described previously [11]. Co-immunoprecipitation analysis Rabbit antibody against FYVE-CENT or rabbit IgG (control) were rotated at RT (room temperature) with Protein A agarose beads for 1 h. Then the beads were washed two times with PBS and two times with 0.2 M triethanolamine, pH 8.2. Crosslinking was performed by rotating the beads in 0.2 M triethanolamine containing 3 mg/ml dimethyl pimelimidate at 4uC overnight. In order to quench the unreacted beads, they were rotated with 10 mM ethanolamine, pH 8.2, at 4uC for 30 min. The beads were washed three times with PBS and were used for immunoprecip- itation. Yeast two-hybrid screening The yeast two-hybrid screening was based on the C terminus (residues 2120–2539) of FYVE-CENT as bait and performed by Hybrigenics S.A Services using a human T cells RP1 (CEMC7) library. Supporting Information Figure S1 FYVE-CENT and Beclin 1 expression in HCC- 1395 and HCC-1954 breast cancer cells. (A) Whole cell lysates from HCC-1395 and HCC-1954 (FYVE-CENT R1945Q mutant) cell lines were analyzed by immunoblotting with the indicated antibodies. Equal amounts of cell lysates (as measured by protein content) from control and mutant cells were loaded in triplicates. The bands were detected using LiCore infrared dye secondary antibodies and the Odyssey imaging system. The bands were quantified using the Odyssey quantifying software, and the numbers resulting from the average of three loadings are shown. (B) Whole cell lysates from HeLa cells transfected with scrambled (scr) or the indicated siRNAs were analyzed by immunoblotting with the indicated antibodies. Experiments were repeated three times and a representative blot is shown. (TIF) Dataset S1 Positive hits from yeast two-hybrid screen- ing with the C-terminus of FYVE-CENT. A list of the interacting proteins with the C-terminus (residues 2120–2539) of ZFYVE26 (FYVE-CENT) were identified in a two-hybrid screen of a human T cells RP1 (CEMC7) cell library. The data are from Hybrigenics S.A, Paris, France. (XLS) Figure S2 A FYVE-CENT R1945Q mutant breast cancer cell line exhibits an increased number of cells arrested in cytokinesis as well as bi- and multinuclear cells. (A–B) Confocal micrographs of HCC-1395 and HCC-1954 breast cancer cells stained with the Aurora B and Hoechst (A), and HCC-1954 breast cancer cells stained with the a-tubulin, Alexa FluorH 594 phalloidin and Hoechst (B). In FYVE-CENT mutant cells (HCC-1954) there is a significant increase in cells arrested in cytokinesis (arrows) compared to the control (A) as well as increase in binuclear-multinuclear cells (B). Scale bars: 10 mm. (C) Graphic Statistical analysis Figure S3 Back-transfection of R1945Q FYVE-CENT mutant C terminus (1807–2539) transgene does not rescue cytokinesis arrest caused by siRNA compared to FYVE-CENT C terminus (1807–2539). Hela cells were tranfected with myc- FYVE-CENT C terminus (1807–2539) and siRNA against FYVE-CENT (A), or myc- FYVE-CENT C terminus (1807–2539) R1945Q mutant and siRNA against FYVE-CENT (B) transgenes simultaneously. The cells expressing myc- FYVE-CENT C terminus (1807–2539) transgene can rescue arrest in cytokinesis compared to the adjacent cells (A) but myc- FYVE-CENT C terminus (1807–2539) R1945Q cannot (B) (arrows). (C), Quantification of the results shown in (A) and (B). Scale bars 10 mm. (TIF) Values are given as means and s.d in all figures. The p values are calculated based on t-test. RNA isolation/cDNA sequencing well plate for each) using the Total RNA Mini Kit (BioRad) according to the manufacturer’s descriptions. One microgram RNA was converted to cDNA using the iScript cDNA Synthesis kit (BioRad). Forward and reverse primers, AGGAGGAAAAT- GAGCTGGTG and CAGCACATCTACCTTGCTGA, were designed with the Primer3 software using default settings, and PCR products were sequenced in forward and reverse using an ABI 3730 DNA Analyzer (Life Technologies). presentation of quantification of cells arrested at the midbody stage and bi-multinuclear cells in FYVE-CENT control (HCC-1395) and R1945Q mutant cell lines (HCC-1954). Error bars show mean 6 s.d. Control: 6 independent experiments, n = 1982 cells. Mutant cells: 6 independent experiments, n = 2001 cells. p value for cells arrested at the midbody stage: 0.001. p value for binuclear-multinuclear cells: 761027. (TIF) Author Contributions Conceived and designed the experiments: APS IPN KGB KH RIS HS. Performed the experiments: APS IPN KGB KH ACB. Analyzed the data: APS IPN RIS. Contributed reagents/materials/analysis tools: HS RIS. Wrote the paper: APS IPN HS. Edited the manuscript: KGB KH ACB RIS. Conceived and designed the experiments: APS IPN KGB KH RIS HS. Performed the experiments: APS IPN KGB KH ACB. Analyzed the data: APS IPN RIS. Contributed reagents/materials/analysis tools: HS RIS. Wrote the paper: APS IPN HS. Edited the manuscript: KGB KH ACB RIS. References 1. Liang XH, Jackson S, Seaman M, Brown K, Kempkes B, et al. (1999) Induction of autophagy and inhibition of tumorigenesis by beclin 1. Nature 402: 672–676. 8. Takahashi Y, Coppola D, Matsushita N, Cualing HD, Sun M, et al. (2007) Bif-1 interacts with Beclin 1 through UVRAG and regulates autophagy and tumorigenesis. Nat Cell Biol 9: 1142–1151. g 9. Itakura E, Kishi C, Inoue K, Mizushima N (2008) Beclin 1 forms two distinct phosphatidylinositol 3-kinase complexes with mammalian Atg14 and UVRAG. Mol Biol Cell 19: 5360–5372. 2. Sinha S, Levine B (2008) The autophagy effector Beclin 1: a novel BH3-only protein. Oncogene 27 Suppl 1: S137–S148. p g pp 3. Lomonosova E, Chinnadurai G (2008) BH3-only proteins in apoptosis and beyond: an overview. Oncogene 27 Suppl 1: S2–19. 10. Sun Q, Fan W, Chen K, Ding X, Chen S, et al. (2008) Identification of Barkor as a mammalian autophagy-specific factor for Beclin 1 and class III phosphatidylinositol 3-kinase. Proc Natl Acad Sci U S A 105: 19211–19216. 4. Kihara A, Kabeya Y, Ohsumi Y, Yoshimori T (2001) Beclin-phosphatidylino- sitol 3-kinase complex functions at the trans-Golgi network. EMBO Rep 2: 330–335. 11. Sagona AP, Nezis IP, Pedersen NM, Liestol K, Poulton J, et al. (2010) PtdIns(3)P controls cytokinesis through KIF13A-mediated recruitment of FYVE-CENT to the midbody. Nat Cell Biol 12: 362–371. 5. Liang C, Feng P, Ku B, Dotan I, Canaani D, et al. (2006) Autophagic and tumour suppressor activity of a novel Beclin1-binding protein UVRAG. Nat Cell Biol 8: 688–699. 12. Thoresen SB, Pedersen NM, Liestol K, Stenmark H (2010) A phosphatidyli- nositol 3-kinase class III sub-complex containing VPS15, VPS34, Beclin 1, UVRAG and BIF-1 regulates cytokinesis and degradative endocytic traffic. Exp Cell Res. 6. Furuya N, Yu J, Byfield M, Pattingre S, Levine B (2005) The evolutionarily conserved domain of Beclin 1 is required for Vps34 binding, autophagy and tumor suppressor function. Autophagy 1: 46–52. 13. Nezis IP, Sagona AP, Schink KO, Stenmark H (2010) Divide and ProsPer: the emerging role of PtdIns3P in cytokinesis. Trends Cell Biol 20: 642–649. 13. Nezis IP, Sagona AP, Schink KO, Stenmark H (2010) Divide and ProsPer: the emerging role of PtdIns3P in cytokinesis. Trends Cell Biol 20: 642–649. 7. Funderburk SF, Wang QJ, Yue Z (2010) The Beclin 1-VPS34 complex–at the crossroads of autophagy and beyond. Trends Cell Biol 20: 355–362. RNA isolation/cDNA sequencing RNA isolation/cDNA sequencing Total RNA was isolated from HCC-1395 (control cells) and HCC-1954 (FYVE-CENT R1945Q mutants cells) (1 well of a 6 PLoS ONE \| www.plosone.org March 2011 \| Volume 6 \| Issue 3 \| e17086 March 2011 \| Volume 6 \| Issue 3 \| e17086 7 Beclin 1 Interacts with FYVE-CENT PLoS ONE \| www.plosone.org 8 March 2011 \| Volume 6 \| Issue 3 \| e17086 March 2011 \| Volume 6 \| Issue 3 \| e17086 March 2011 \| Volume 6 \| Issue 3 \| e17086 PLoS ONE \| www.plosone.org PLoS ONE \| www.plosone.org PLoS ONE \| www.plosone.org 8 Beclin 1 Interacts with FYVE-CENT Figure 5. The average expression of FYVE-CENT and associated genes is significantly lower in high vs. low grade breast cancers. (A)–(D) The gene expression data were derived from the Gene Expression Omnibus GSE1456 and GSE4922 datasets. The expression values were median centred within each of the series separately. The p-values were derived from comparing means by independent samples t-test (SPSS, v.16.0). (E) Proposed model: FYVE-CENT- Beclin 1 interplay in a positive feedback loop manner during cytokinesis. doi:10.1371/journal.pone.0017086.g005 Figure 5. The average expression of FYVE-CENT and associated genes is significantly lower in high vs. low grade breast cancers. (A)–(D) The gene expression data were derived from the Gene Expression Omnibus GSE1456 and GSE4922 datasets. The expression values were median centred within each of the series separately. The p-values were derived from comparing means by independent samples t-test (SPSS, v.16.0). (E) Proposed model: FYVE-CENT- Beclin 1 interplay in a positive feedback loop manner during cytokinesis. doi:10.1371/journal.pone.0017086.g005 well plate for each) using the Total RNA Mini Kit (BioRad) according to the manufacturer’s descriptions. One microgram RNA was converted to cDNA using the iScript cDNA Synthesis kit (BioRad). Forward and reverse primers, AGGAGGAAAAT- GAGCTGGTG and CAGCACATCTACCTTGCTGA, were designed with the Primer3 software using default settings, and PCR products were sequenced in forward and reverse using an ABI 3730 DNA Analyzer (Life Technologies). well plate for each) using the Total RNA Mini Kit (BioRad) according to the manufacturer’s descriptions. One microgram RNA was converted to cDNA using the iScript cDNA Synthesis kit (BioRad). Forward and reverse primers, AGGAGGAAAAT- GAGCTGGTG and CAGCACATCTACCTTGCTGA, were designed with the Primer3 software using default settings, and PCR products were sequenced in forward and reverse using an ABI 3730 DNA Analyzer (Life Technologies). q 17. Pawitan Y, Bjohle J, Amler L, Borg AL, Egyhazi S, et al. (2005) Gene expression profiling spares early breast cancer patients from adjuvant therapy: derived and validated in two population-based cohorts. Breast Cancer Res 7: R953–R964. g q 16. Aita VM, Liang XH, Murty VV, Pincus DL, Yu W, et al. (1999) Cloning and genomic organization of beclin 1, a candidate tumor suppressor gene on chromosome 17q21. Genomics 59: 59–65. 14. Sagona AP, Stenmark H (2010) Cytokinesis and cancer. FEBS Lett 584: 2652–2661. References PLoS ONE \| www.plosone.org March 2011 \| Volume 6 \| Issue 3 \| e17086 9 15. Sjoblom T, Jones S, Wood LD, Parsons DW, Lin J, et al. (2006) The consensus coding sequences of human breast and colorectal cancers. Science 314: 268–274. 18. Ivshina AV, George J, Senko O, Mow B, Putti TC, et al. (2006) Genetic reclassification of histologic grade delineates new clinical subtypes of breast cancer. Cancer Res 66: 10292–10301. 14. Sagona AP, Stenmark H (2010) Cytokinesis and cancer. FEBS Lett 584: 2652–2661. 15. Sjoblom T, Jones S, Wood LD, Parsons DW, Lin J, et al. (2006) The consensus coding sequences of human breast and colorectal cancers. Science 314: 268–274. 16. Aita VM, Liang XH, Murty VV, Pincus DL, Yu W, et al. (1999) Cloning and genomic organization of beclin 1, a candidate tumor suppressor gene on chromosome 17q21. Genomics 59: 59–65. 17. Pawitan Y, Bjohle J, Amler L, Borg AL, Egyhazi S, et al. (2005) Gene expression profiling spares early breast cancer patients from adjuvant therapy: derived and validated in two population-based cohorts. Breast Cancer Res 7: R953–R964. Beclin 1 Interacts with FYVE-CENT 18. Ivshina AV, George J, Senko O, Mow B, Putti TC, et al. (2006) Genetic reclassification of histologic grade delineates new clinical subtypes of breast cancer. Cancer Res 66: 10292–10301. 19. Fujiwara T, Bandi M, Nitta M, Ivanova EV, Bronson RT, et al. (2005) Cytokinesis failure generating tetraploids promotes tumorigenesis in p53-null cells. Nature 437: 1043–1047. 20. Steigemann P, Wurzenberger C, Schmitz MH, Held M, Guizetti J, et al. (2009) Aurora B-mediated abscission checkpoint protects against tetraploidization. Cell 136: 473–484. PLoS ONE \| www.plosone.org March 2011 \| Volume 6 \| Issue 3 \| e17086 10
https://openalex.org/W4287205870	https://zenodo.org/record/5111697/files/TingYi%20-Some%20General%20Solutions%20of%20the%20Linear%20B-H%20Equation-Summary.pdf	English	null	Some General Solutions for Linear Bragg-Hawthorne Equation	Zenodo (CERN European Organization for Nuclear Research)	2,021	cc-by	401	Some General Solutions for Linear Bragg-Hawthorne Equation Ting Yi (tingyi.physics@gmail.com) Abstract: Linear cases of Bragg-Hawthorne equation for steady axisymmetric incompressible ideal flows are systematically discussed. The equation is converted to a more convenient form in spherical coordinate system. A new vorticity decomposition is derived. General solutions for 16 linear cases of the equation are obtained. These solutions can be specified to gain new analytical vortex flows, as examples in the paper demonstrate. A lot of well-known solutions like potential flow past a sphere, Hill’s vortex with and without swirl, are included and extended in these solutions. Special relations between some vortex flows are also revealed when exploring/comparing related solutions. Keywords: axisymmetric flow; Bragg-Hawthorne equation; Grad–Shafranov equation; Eul equations; Beltrami flow; spherical vortex Keywords: axisymmetric flow; Bragg-Hawthorne equation; Grad–Shafranov equation; Euler equations; Beltrami flow; spherical vortex Table of Contains I. Introduction ........................................................................................................................... 1 II. Bragg-Hawthorne equation in spherical coordinates ............................................................ 1 III. Linear cases of Bragg-Hawthorne equation .......................................................................... 2 IV. Vorticity decomposition and flow properties ........................................................................ 3 V. Axisymmetric potential flow (H1C1) ..................................................................................... 4 VI. Axisymmetric non-linear Beltrami flow (H1C2) ................................................................... 5 VII. Beltrami spherical vortex (H1C3) .......................................................................................... 6 VIII. Hill’s vortex and extension (H2C1)........................................................................................ 9 IX. Hill’s vortex with swirl (H2C3) ........................................................................................... 11 X. Other axisymmetric flows (H3Cn and H4Cn) ....................................................................... 13 XI. Summary and Discussion .................................................................................................... 14 References ...................................................................................................................................... 15 Hill’s vortex (without swirl) with hollow Beltrami spherical vortex Fig. 7. Cross sections of core region of H2C1 flow Beltrami spherical vortices 3-layer Beltrami spherical vortex 2-Layer Hill’s Vortex with swirl Non-linear Beltrami flow outside a hyperboloid Fraenkel–Norbury vortex rings Examples of vortex solutions Published 7/16/2021, Physics of Fluids, Full PDF: doi: 10.1063/5.0055228 Examples of vortex solutions Fraenkel–Norbury vortex rings Non-linear Beltrami flow outside a hyperboloid Fraenkel–Norbury vortex rings Non-linear Beltrami flow outside a hyperboloid Hill’s vortex (without swirl) with hollow Beltrami spherical vortex Fig. 7. Cross sections of core region of H2C1 flow Beltrami spherical vortices 3-layer Beltrami spherical vortex 2-Layer Hill’s Vortex with swirl Published 7/16/2021, Physics of Fluids, Full PDF: doi: 10.1063/5.0055228 ctions of core region of H2C1 flow Fig. 7. Cross se Beltrami spherical vortices ctions of core region of H2C1 flow Beltrami spherical vortices 3-layer Beltrami spherical vortex Hill’s vortex (without swirl) with hollow Beltrami spherical vortex 2-Layer Hill’s Vortex with swirl Hill’s vortex (without swirl) with hollow Beltrami spherical vortex 2-Layer Hill’s Vortex with swirl Published 7/16/2021, Physics of Fluids, Full PDF: doi: 10.1063/5.0055228
https://openalex.org/W4385070223	https://www.researchsquare.com/article/rs-3179828/latest.pdf	English	null	Surgical Management of Craniospinal Axis Malignant Peripheral Nerve Sheath Tumors: A Single-Institution Experience and Literature Review	Research Square (Research Square)	2,023	cc-by	5,924	Surgical Management of Craniospinal Axis Malignant Peripheral Nerve Sheath Tumors: A Single-Institution Experience and Literature Review Surgical Management of Craniospinal Axis Malignant Peripheral Nerve Sheath Tumors: A Single-Institution Experience and Literature Review Ajmain Chowdhury University of Iowa, Carver College of Medicine Juan Vivanco-Suarez University of Iowa Hospitals and Clinics Nahom Teferi University of Iowa Hospitals and Clinics Alex Belzer University of Iowa, Carver College of Medicine Hend Al-Kaylani University of Iowa, Carver College of Medicine Meron Challa University of Iowa, Carver College of Medicine Sarah Lee University of Iowa Hospitals and Clinics John Buatti University of Iowa Hospitals and Clinics Patrick Hitchon (  patrick-hitchon@uiowa.edu ) University of Iowa Hospitals and Clinics Surgical Management of Craniospinal Axis Malignant Peripheral Nerve Sheath Tumors: A Single-Institution Experience and Literature Review Surgical Management of Craniospinal Axis Malignant Peripheral Nerve Sheath Tumors: A Single-Institution Experience and Literature Review Ajmain Chowdhury University of Iowa, Carver College of Medicine Juan Vivanco Suarez Surgical Management of Craniospinal Axis Malignant Peripheral Nerve Sheath Tumors: A Single-Institution Experience and Literature Review Ajmain Chowdhury University of Iowa, Carver College of Medicine Juan Vivanco Suarez Conclusions Primary craniospinal MPNSTs are rare and have an aggressive clinical course. Early diagnosis and treatment are essential for managing these tumors. Maximal resection, low-grade pathology, young age (< 30), and adjuvant radiotherapy were associated with improved survival. Results Eight patients satisfied the inclusion criteria (4 male, 4 female). The median age at presentation was 38 years (range 15–67). Most tumors were localized to the spine (75%), and 3 patients had neurofibromatosis type 1. The most common presenting symptoms were paresthesia (50%) and visual changes (13%). The median tumor size was 3 cm, and most tumors were oval-shaped (50%) with well-defined borders (75%). Six tumors were high grade (75%), and gross total resection was achieved in 5 patients, with subtotal resection in the remaining 3 patients. Postoperative radiotherapy and chemotherapy were performed in 6 (75%) and 4 (50%) cases, respectively. Local recurrence occurred in 5 (63%) cases, and distant metastases occurred in 2 (25%). The median overall survival was 26.7 months. Five (63%) patients died due to recurrence. Institutional Setting The study was approved by the University of Iowa Institutional Review Board (IRB). A retrospective review of hospital records was performed for the diagnosis of ‘MPNST’, ‘malignant neoplasms of connective and soft tissue’, and ‘malignant neoplasms of spinal cord/brain’ from January 2005 to May 2023. Informed consent was waived by the IRB for all the subjects (IRB #201902751). This study was conducted at the University of Iowa Hospitals and Clinics. Chart records were obtained from the EPIC (Epic Systems Corporation, Madison, WI) electronic medical record (EMR). Background Malignant peripheral nerve sheath tumor (MPNST) is an exceedingly rare and aggressive tumor, with limited literature on its management. Herein, we present our series of surgically managed craniospinal MPNSTs, analyze their outcomes, and review the literature. Methods We retrospectively reviewed surgically managed primary craniospinal MPNSTs treated at our institution between January 2005 and May 2023. Patient demographics, tumor features, and treatment outcomes were assessed. Neurological function was quantified using the Frankel grade and Karnofsky performance scores. Descriptive statistics, rank-sum tests, and Kaplan-Meier survival analyses were performed. Research Article t peripheral nerve sheath tumors, MPNST, Mesenchymal tumors, Craniospinal axis, Malignant Triton tumor DOI: https://doi.org/10.21203/rs.3.rs-3179828/v1 Version of Record: A version of this preprint was published at World Journal of Surgical Oncology on October 26th, 2023. See the published version at https://doi.org/10.1186/s12957-023-03227-y. Version of Record: A version of this preprint was published at World Journal of Surgical Oncology on October 26th, 2023. See the published version at ttps://doi org/10 1186/s12957 023 03227 y Page 1/11 Page 1/11 Background Malignant peripheral nerve sheath tumors (MPNSTs) are aggressive soft tissue sarcomas arising from peripheral nerves or associated nerve sheaths [1]. These are rare tumors, comprising 5–10% of all soft tissue sarcomas in the United States, with an overall incidence of 0.001% [2]. MPNSTs are characterized by aggressive local invasiveness and high rates of both local recurrence and distant metastases [1]. The term ‘MPNST’ was first coined by the World Health Organization (WHO) [3] in 1990; these tumors were previously referred to as ‘neurofibrosarcoma’, ‘neurogenic sarcoma’, ‘malignant neurolemmoma’, or ‘malignant schwannoma’ [4]. MPNSTs occur most frequently in patients with neurofibromatosis type 1 (NF-1). Patients with NF-1 and plexiform neurofibromas are 18 times more likely to develop MPNSTs, and 20–30% of MPNSTs occur in patients with NF-1 [1]. Radiation exposure, such as prior radiotherapy (RT), is also a risk factor for the development of MPNSTs, with 10% of MPNSTs developing in irradiated patients [1]. Primary intradural MPNSTs, however, can occur in the absence of any predisposing risk factors. MPNSTs occur most frequently in the deep soft tissues of extremities near nerve trunks; however, exact incidence rates by location are difficult to determine [5]. Primary intradural MPNSTs in the central nervous system (CNS) are exceedingly rare, with few reported cases [6–9]. When found in the CNS, MPNSTs are often misdiagnosed on imaging, with more common benign diagnoses such as meningioma or schwannoma often considered [1, 4]. MPNSTs have poor outcomes, with low progression-free survival (PFS) and overall survival (OS) [6, 10]. The five-year OS is as low as 25%, and the PFS ranges from 5 to 32.2 months [6, 9, 11]. Timely diagnosis and management of MPNSTs is key to improving OS given the aggressive nature of these tumors. The natural disease course of MPNST, best treatment options, and associated complications are largely unknown and are currently based on case reports and small case series, owing to its rarity. Additionally, no studies have rigorously described neurological or functional outcomes in surgically managed patients with MPNSTs. To date, we present one of the larger series in the literature on MPNSTs within the craniospinal axis, analyze our treatment algorithm and patient outcomes, and extensively review the associated literature. Statistical Methods Descriptive statistics were used to describe patient demographics, tumor characteristics, clinical course, and treatment factors. Patient demographics included age, sex, and ethnicity. Tumor characteristics included tumor location, radiologic diagnosis, and WHO grade on pathologic diagnosis. Clinical course included presenting symptoms, pre- and postoperative KPS, Frankel grade, tumor recurrence, metastasis, follow-up history, and vital status as of June 2023. Treatment factors included the extent of resection (EOR), reoperation, and use of adjuvant CHE or RT. GraphPad Prism 9 (Dotmatics LLC, San Diego, CA, USA) was used for quantitative analysis. Categorical variables were compared using Fisher’s exact test, and numerical variables were analyzed using the Mann-Whitney-Wilcoxon rank sum test. Survival analyses were performed using Kaplan-Meier estimation. OS was calculated from the date of initial surgery to the date of death reported in patient medical records. PFS was calculated from the date of initial surgery to the date of tumor recurrence found on radiological evaluation. Patients not documented as deceased or having residual tumor or tumor recurrence were censored from the date of the last follow-up for OS and PFS, respectively. The results were considered significant at a p value < 0.05. Data Collection Page 2/11 The EMRs of two hundred twenty-four patients (224) with MPNST were initially reviewed. Eight (8) patients were identified with a diagnosis of primary intradural MPNST of the craniospinal axis. We collected information on patient demographics, clinical characteristics, radiological and pathological The EMRs of two hundred twenty-four patients (224) with MPNST were initially reviewed. Eight (8) patients were identified with a diagnosis of primary intradural MPNST of the craniospinal axis. We collected information on patient demographics, clinical characteristics, radiological and pathological Page 2/11 findings, clinical course, treatment modalities, survival, and functional outcomes. Radiological test results, including computed tomography (CT) and magnetic resonance imaging (MRI), were collected as they pertain to lesion location and appearance, involvement of craniospinal structures, and compression of neural elements. Surgical treatment modalities include craniotomy for intracranial tumor resection and laminectomy/laminoplasty for spinal tumor resection. Any distant metastasis or local tumor recurrence was noted with the corresponding mode of management, which included reoperation, salvage RT, chemotherapy (CHE), or a combination of these treatments. Neurologic status was documented using pre- and postoperative Frankel grading [12]. Karnofsky Performance Scores (KPS) [13] were collected to document functional status. l h d Patient demographics and clinical characteristics A total of 8 patients with a diagnosis of craniospinal MPNST at our institution met the inclusion criteria. The clinical characteristics of all patients are summarized in Table 1. There were 4 male and 4 female patients (sex ratio 0.5), with a median age of 38 years (range 15–67 years). Most patients presented with spinal tumors (6/8, 75%). The most common presenting symptoms were paresthesia/numbness (4/8, 50%), pain (3/8, 37.5%), and weakness (2/8, 25%). Visual changes were noted in 1/2 cranial cases. The median preoperative KPS was 50 (range 30–100). Preoperative Frankel grade was most frequently D in 4 (50%) patients, followed by E in 3 (37.5%) and C in 1 (12.5%). Three (37.5%) patients had NF-1. The duration of symptoms prior to presentation ranged from 3 days to 12 months. Patient demographics and clinical characteristics Page 3/11 Table 1 Summary of the MPNST cases N Age / Sex Clinical presentation Symptom duration Location Size (cm) EOR Grade Adjuvant therapy Recurrence Mets KPS preop/postop Follow- up (mo.) 1 67/M Visual changes, CN VI palsy 4 mo Rt orbital and superior orbital fissure/cavernous sinus 1.0 STR High RT - Lung 50/80 Alive at 69.5 2 58/M Rt flank and LE pain, urinary retention* 1 mo Rt T10-sacrum paraspinal 20.2 GTR High RT 1 - 30/40 Dead at 5.4 3 58/F Lt UE numbness and weakness 12 mo Lt C7-T1 intradural extramedullary 2.5 STR High CHE, RT - - 70/50 Dead at 37.6 4 39/M Lt head, ear, and neck numbness and paresthesia 3 mo Lt C2-C3 extradural extramedullary 5.2 GTR High CHE, RT 3 - 90/100 Dead at 22.1 5 37/F Rt side weakness, confusion, word finding difficulty 3 d Lt frontal lobe 1.7 STR High CHE, RT 3 Lung 30/100 Dead at 26.7 6 35/F Rt lower back pain 3 mo Rt L2-L5 paraspinal (multiple other lesions non- malignant lesions present) 6.2 GTR Low - - - 100/100 Alive at 25.8 7 24/F Rt LE paresthesia, chest pain 1 mo Rt C7-T1 intradural extramedullary 3.0 GTR Low - 2 - 30/100 Alive at 105.2 8 15/M Rt neck pain, Rt UE paresthesia* 5 mo C7 extradural 2.8 STR High CHE, RT 1 - 50/60 Dead at 25.1 * Patient has NF-1 + removal of 3 peripheral tumors (3 MPNSTs); History of NF-1 + multiple superficial neurofibromas; * History of NF-1 + removal of 5 peripheral tumors (3 MPNSTs and 2 plexiform neurofibromas) Abbreviations: CHE (chemotherapy); CN (cranial nerve); d (days); F (female); GTR (gross total resection ); KPS (Karnofsky performance score); LE (lower extremity); Lt (left); M (male); Mets (metastasis); mo (months); radiotherapy (RT); right (Rt); subtotal resection (STR); upper extremity (UE); year (y) fi d Pathology findings On histological hematoxylin-eosin (H&E) examination, the tumors typically presented with spindle-shaped morphology with pleomorphic, hyperchromatic, and atypical nuclei arranged in a fascicular architecture. One patient presented with malignant Triton tumor (MTT) histology (Fig. 2), a high-grade MPNST with focal rhabdomyoblastic differentiation, focal expression of desmin and myogenin, and loss of H3K27me3 expression (Fig. 2e-g). On immunohistochemistry (IHC), S100 was positive in 5 (62.5%) cases, vimentin in 4 (50%), SMA in 2 (25%), desmin in 1 (12.5%), and EMA in 1 (12.5%). According to the WHO classification, 6 (75%) tumors were high grade, and 2 (25%) were low grade [3]. Imaging findings All patients underwent MRI of the neural axis prior to surgery. The radiological characteristics of all patients are presented in Table 2. The median tumor size was 3 cm (range 1.0-20.2 cm). The tumors were oval in 4 (50%) cases and dumbbell or irregular in 2 cases each (25%). Most tumors had well-defined borders (6/8, 75%). Two (25%) tumors were located intracranially. Of the spinal tumors, 2 (33.3%) were cervical, 2 (33.3%) cervicothoracic, 1 (16.7%) lumbar, and 1 (16.7%) thoracolumbosacral (Fig. 1). On T1-weighted imaging, 4 (50%) lesions were hypointense, 3 (37.5%) were isointense, and 1 (12.5%) was hyperintense, while on T2-weighted imaging, 4 (50%) tumors were hyperintense, 3 (37.5%) were isointense, and 1 (12%) had heterogeneous intensity. Only 1 (12.5%) tumor did not show contrast enhancement, and 1 (12.5%) tumor had intramedullary extension. In 3 (37.5%) cases, peripheral tumors were diagnosed and excised before MPNST diagnosis. Based on imaging findings, 5 patients with primary intradural MPNSTs in our study were preoperatively misdiagnosed as having other lesions, including schwannoma (3 cases), meningioma (1 case), and neurofibroma (1 case). Page 4/11 Table 2 Radiological characteristics of the MPNST cases Characteristic No. of cases Size ≤ 3 cm 4 > 3 cm 4 Shape Oval 4 Dumbbell 2 Irregular 2 Border of the tumor Well defined 6 Poorly defined 2 Magnetic resonance imaging findings T1-weighted sequence Hyperintense 1 Isointense 3 Hypointense 4 T2-weighted sequence Hyperintense 4 Isointense 3 Heterogenous 1 Contrast enhancement Yes 7 No 1 Bone erosion Yes 2 No 6 Clinical diagnostic workup, radiologic findings, and histopathology Timely and accurate diagnosis of MPNST is difficult and often requires extensive workup. A thorough history and physical examination, noting the onset and duration of symptoms, are necessary, as rapid progression of symptoms would be concerning for malignancy. It is also important to elicit a past medical or family history of NF-1, schwannomatosis, or prior RT. On physical examination, the typical findings of NF-1, such as café-au-lait spots, Lisch nodules, and cutaneous neurofibromas, should be evaluated. Diligent neurological evaluation of sensory, motor, and gait functions is key to localizing lesions in the craniospinal axis. Following a detailed history and physical examination, further workup primarily consists of imaging with CT and MRI. Contrast-enhanced MRI with gadolinium has the highest resolution and is considered the imaging modality of choice [21]. Radiologic differential diagnoses often considered for MPNSTs include meningiomas, solitary fibrous tumors/hemangiopericytomas, schwannomas, and dural-based metastases. These were considered in radiological reports of our cohort. Certain radiologic features favor a diagnosis of MPNST over benign lesions such as neurofibroma or schwannoma, including size > 5 cm, ill-defined borders, soft tissue edema, lobulation, lack of a target sign, and surrounding bone destruction [21]. The radiological findings of our patient cohort are outlined in Table 2. On histopathological examination, MPNSTs are usually high-grade malignant spindle cell tumors most commonly found in nerves. More specifically, they arise from preexisting Schwann cell tumors (such as plexiform neurofibromas) [22], and H&E microscopy typically reveals a cellular neoplasm with fascicles comprising spindly cells with tapered hyperchromatic nuclei [23]. Mitotic figures and necrosis are common [24] but are notably decreased in low- grade tumors [22]. A marble-like appearance may be seen at low power, with further evaluation under high power revealing alternating hyper and hypocellular areas [22]. MPNSTs also have incredible plasticity and may demonstrate internal cartilage, bone, skeletal or smooth muscle, glandular, epithelioid, and/or perineural differentiation [22, 23, 25]. Macroscopically, these tumors are highly variable in size and are adherent and exophytic with common areas of hemorrhage and necrosis. Our cohort showed many of these histopathological findings. Further analysis using IHC is routinely performed, but given heterogeneous findings, no standard set of diagnostic characteristics exists. The highest yields are for the S100 and SOX10 stains, which are often decreased relative to other neural crest-originating tumors and are often correlated [24, 26]. S100 may be particularly useful for distinguishing MPNST from malignant melanoma [27]. Complications One patient developed cephalic vein thrombosis from their intravenous line, which resolved spontaneously after line removal. One patient developed pneumonia postoperatively, which was managed with antibiotics and resolved after 7 days. No complications were directly attributed to surgical intervention. Management outcomes Given the small sample size in each cohort, these survival observations trended toward significance but were not statistically significant (p < 0.5). Death had occurred in 5 (62.5%) cases at the time of data collection. The most common causes of death were local disease recurrence and increased tumor burden, all attributed to MPNST. Management outcomes The nerve from which the tumor originated was identified in 6 (75%) cases. Gross total resection (GTR) was achieved in 5 (62.5%) cases, and subtotal resection (STR) was performed in the remaining 3 (37.5%). Adjuvant RT after surgery was performed in 6 (75%) patients with a median dose of 60 Gy (range 30–72 Gy). Two patients (25%) had GTR and low-grade tumors on histology and did not receive adjuvant RT. CHE was administered in 4 (50%) cases. Local recurrence was observed in 5 (62.5%) cases, and in 2 (25%) cases, patients presented with distant metastases to the lung. One patient experienced both local recurrence and distant metastasis (1/8, 12.5%). Five out of six (83.3%) patients with local recurrence or distant metastasis received adjuvant RT postoperatively, and half received adjuvant CHE (3/6, 50%). Three patients (3/5, 60%) with local recurrence underwent reoperation. The median postoperative KPS was 90 (range 40–100). The postoperative Frankel grade was E in 5 (63%) patients, D in 2 (25%) and C in 1 (12%). The median PFS was 19.5 months, and the median OS was 26.7 months (Fig. 3a). When stratified by EOR, the median OS was 63.7 months (range 5.4-105.2 months) in patients with GTR and 32.15 months (range 25.1–69.5 months) in patients with STR (Fig. 3b). Similarly, patients without NF-1 mutations, low- grade tumors, and younger age (< 30 years) (median OS 37.6 months; not reached, 65.2 months) were found to have a longer OS than patients with NF-1 mutations, high-grade pathology and older age (age > 30) (median OS 25.1 months, 25.9 months, 26.7 months, respectively) (Fig. 3c-e). Patients who Page 5/11 Page 5/11 received adjuvant chemoradiation had lower OS (25.9 months) than patients who received adjuvant RT alone (37.5 months) (Fig. 3f), likely due to poor prognostic features of advanced disease/metastases. Given the small sample size in each cohort, these survival observations trended toward significance but were not statistically significant (p < 0.5). Death had occurred in 5 (62.5%) cases at the time of data collection. The most common causes of death were local disease recurrence and increased tumor burden, all attributed to MPNST. received adjuvant chemoradiation had lower OS (25.9 months) than patients who received adjuvant RT alone (37.5 months) (Fig. 3f), likely due to poor prognostic features of advanced disease/metastases. Discussion MPNSTs are highly recurrent, aggressive soft tissue sarcomas with a tendency to metastasize [6, 9, 11] and have an incidence of approximately 0.001% in the general population [2]. They are thought to arise from peripheral nerves or their associated nerve sheaths [1]; however, Rubino et al. hypothesized that they originate from the nervi vasorum, which are autonomic peripheral nerves in the adventitial layer of the large and small pial arteries [14]. Primary MPNSTs of CNS origin are even less common and are analogous to the malignant version of schwannomas [14]. A history of NF-1 or prior irradiation are important risk factors in the development of MPNST[1]; however, they do not necessarily compose most cases of MPNST, with only 20–30% of patients having NF-1 and only 10% of patients reporting prior radiation exposure [1]. Sex is not a known risk factor for this tumor [1]. Many of these epidemiological characteristics of MPNST are reflected in our cohort; it was demographically evenly split between males and females, and 3 (37.5%) of our patients had NF- 1. Interestingly, one patient had a history of RT for Hodgkin’s lymphoma and was diagnosed with MTT, a specific subtype of MPNST with an even worse prognosis [15]. Craniospinal axis MPNSTs are exceedingly rare, with approximately 100 cases of intracranial tumors [7, 16, 17] and dozens of spinal tumors [6, 8, 11, 18– 20] reported in the literature, indicating that intracranial location may be more common than spinal location. Our cohort’s composition of craniospinal MPNST locations deviates from this, with 75% of patients having spinal tumors. Patients with primary intradural MPNST often present with insidious neurological symptoms, which are generally attributable to a progressive mass effe on nearby neurovascular structures. When present in the cranium, symptoms include headache, nausea/vomiting, seizures, focal neurological deficits, and/or cranial nerve palsies [1]. When present in the spine, MPNST may cause myelopathic symptoms, pain, motor weakness, sensory deficit/radiculopathy, and bowel/bladder dysfunction [1]. These symptoms were observed in our cohort (Table 1). Clinical diagnostic workup, radiologic findings, and histopathology This deviates from our cohort’s IHC analysis, as 5 (63%) of our patients had a positive S100. The loss of p16 is also typical [24]. Other common traits include loss of neurofibromin expression (which is more common in NF-1-associated than sporadic tumors) [24] and H3K27 trimethylation. Loss of the latter is highly specific for MPNST and is associated with worse survival [28]. Management and outcomes Management and outcomes Page 6/11 Page 6/11 Currently, there is no standard therapeutic approach for MPNSTs. Maximal safe gross total resection with negative margins is recommended when feasible but is often difficult to achieve, as these tumors tend to grow near vital neurovascular structures [1, 4, 17]. Our cohort underwent extensive and variable treatment regimens with multidisciplinary teams involving neurosurgery and oncology, with patients receiving a combination of RT, CHE, or observation postoperatively. The general approach of maximal surgical resection to improve survival was true in our cohort, in which patients who underwent GTR survived longer than those treated with STR (Fig. 3b). Our findings strengthen this approach, and we recommend maximal safe surgical resection for the surgical management of MPNST. There is little available literature on the effect of surgical treatment on neurological and functional outcomes in craniospinal MPNST. Our study found that 87.5% of patients had an improvement in their postoperative functional status as quantified by KPS; however, a larger sample size is warranted prior to making definitive conclusions. Because of the radiation-inducible nature of MPNSTs and increased radiation sensitivity of patients with NF-1, adjuvant RT has not been shown to improve OS for MPNST, with many studies finding that RT may improve local control of disease and lengthen PFS but does not improve OS [1, 4, 6, 17, 29, 30]. Despite this, RT has been shown to improve OS in the management of MTT, as in Case 5 in our cohort [31]. All 6 of the patients in our cohort with high- grade tumor pathology received RT with subsequent improved OS compared to historical cohorts; however, due to the small number of patients in this study, the comparative analysis did not reach significance. MPNST has historically been shown to be poorly responsive to CHE [29, 30], and it is notable that in our cohort, despite not reaching significance, patients treated with adjuvant RT alone had higher survival rates (OS 37.5 months) than patients treated with chemoradiation (25.9 months) (Fig. 3d). Limitations Given the rarity of this pathology and the small number of patients expected from a single-center study, we extensively reviewed the literature to further strengthen our recommendations. The retrospective and nonrandomized nature of this study also decreased the level of evidence. Controlled multicenter large-scale studies are necessary to recommend stronger guidelines. Conclusions The highly aggressive, recurrent, and metastatic characteristics of primary craniospinal axis MPNST, along with its rarity, pose many challenges. Radiation exposure and positive NF-1 status both increase the risk of developing MPNST and worsen prognosis. Initial presenting symptoms are secondary to the mass effect on nearby neural structures. Gadolinium contrast-enhanced MRI is recommended for imaging, along with histopathological analysis, to confirm diagnosis. Maximal tumor resection has consistently been shown to improve survival in patients with MPNST, and RT shows promise as an adjuvant treatment. To date, nonsurgical management of MPNST has not been found to improve outcomes. A multidisciplinary team of neurosurgeons, radiologists, pathologists, and oncologists is essential to optimally diagnose and manage MPNST. Clinical diagnostic workup, radiologic findings, and histopathology We recommend adjuvant RT for the management of MPNST; however, further multicenter reviews and randomized clinical trials are necessary to further strengthen this recommendation. Survival in MPNST is very poor, with 5-year survival rates as low as 25% owing to increased local recurrence and metastasis rates [6, 10]. Local recurrence rates reported range from 31–75%, with a median PFS of 5 to 32.2 months [6, 9, 11] and distal metastasis rates of 22–45% [9, 11]. Negative prognostic factors reported include tumor size over 5 cm, higher tumor grade, positive surgical margin, positive NF-1 status, and Ki-67 score over 20. This was further corroborated in our cohort with patients undergoing GTR of tumor, low grade tumor pathology on histology, and absence of NF-1 mutation trending toward having an improved OS (Fig. 3b, c) [1, 4, 9]. Abbreviations MPNST: Malignant peripheral nerve sheath tumors MPNST: Malignant peripheral nerve sheath tumors WHO: World Health Organization NF-1: Neurofibromatosis type 1 Page 7/11 Page 7/11 CHE: Chemotherapy KPS: Karnofsky Performance Scores EOR: Extent of resection H&E: Hematoxylin-eosin MTT: Malignant Triton tumor IHC: Immunohistochemistry GTR: Gross total resection STR: Subtotal resection Manuscript review and editing: AC, JV, NT, AB, HA, MC, SL, JB, PH All authors read and approved the final manuscript. Acknowledgements Acknowledgements We acknowledge all the authors whose publications are referred in our article. We acknowledge all the authors whose publications are referred in our article. Declarations Ethics approval and consent to participate This study obtained approval from the institutional review board of the University of Iowa and consent was waived by the IRB (#201902751). Consent for publication NA The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. Competing interests No funds were received in support of this work. Authors' contributions Study conception: PH, NT, AC, JV Study design: AC, NT Data acquisition: AC, JV Data analysis: JV, NT Data interpretation: AC, JV, NT Quality assessment: NT Manuscript writing: AC, JV, NT, AB, HA Manuscript review and editing: AC, JV, NT, AB, HA, MC, SL, JB, PH References Systemic Options for Malignant Peripheral Nerve Sheath Tumors. Curr Treat Options Oncol. 2021;22:33. 30. Hassan A, Pestana RC, Parkes A. Systemic Options for Malignant Peripheral Nerve Sheath Tumors. Curr T 31. Li G, Liu C, Liu Y, Xu F, Su Z, Wang Y, et al. Analysis of clinical features and prognosis of malignant triton tumor: A report of two cases and literature review. Oncol Lett. 2015;10:3551-6. Figures References Pathology of Peripheral Nerve Sheath Tumors: Diagnostic Overvi Problems. Acta Neuropathol. 2012;123:295-319. iguez FJ, Folpe AL, Giannini C, Perry A. Pathology of Peripheral Nerve Sheath Tumors: Diagnostic Overview and lems Acta Neuropathol 2012;123:295 319 23. Rodriguez FJ, Folpe AL, Giannini C, Perry A. Pathology of Peripheral Nerve Sheath Tumors: Diagnostic Overview and Update on Selected Diagnostic Problems. Acta Neuropathol. 2012;123:295-319. 4. Pekmezci M, Reuss DE, Hirbe AC, Dahiya S, Gutmann DH, von Deimling A, et al. Morphologic and immunohis peripheral nerve sheath tumors and cellular schwannomas. Mod Pathol. 2015;28:187-200. 24. Pekmezci M, Reuss DE, Hirbe AC, Dahiya S, Gutmann DH, von Deimling A, et al. Morphologic and immunohistochemical features of malignant peripheral nerve sheath tumors and cellular schwannomas. Mod Pathol. 2015;28:187-200. 25. Guo A, Liu A, Wei L, Song X. Malignant peripheral nerve sheath tumors: differentiation patterns and immu our new findings. J Cancer. 2012;3:303-9. 25. Guo A, Liu A, Wei L, Song X. Malignant peripheral nerve sheath tumors: differentiation patterns and immunohistochemical features - a mini-review and our new findings. J Cancer. 2012;3:303-9. 6. Karamchandani JR, Nielsen TO, van de Rijn M, West RB. Sox10 and S100 in the diagnosis of soft-tissue neop Morphol. 2012;20:445-50. 26. Karamchandani JR, Nielsen TO, van de Rijn M, West RB. Sox10 and S100 in the diagnosis of soft-tissue neoplasms. Appl Immunohistochem Mol Morphol. 2012;20:445-50. 27. Nonaka D, Chiriboga L, Rubin BP. Differential expression of S100 protein subtypes in malignant melanoma, and benign and malignant peripheral nerve sheath tumors. J Cutan Pathol. 2008;35:1014-9. 7. Nonaka D, Chiriboga L, Rubin BP. Differential expression of S100 protein subtypes in malignant melanoma, a sheath tumors. J Cutan Pathol. 2008;35:1014-9. 8. Cleven AH, Sannaa GA, Briaire-de Bruijn I, Ingram DR, van de Rijn M, Rubin BP, et al. Loss of H3K27 tri-methyl peripheral nerve sheath tumors and an indicator for an inferior survival. Mod Pathol. 2016;29:582-90. 28. Cleven AH, Sannaa GA, Briaire-de Bruijn I, Ingram DR, van de Rijn M, Rubin BP, et al. Loss of H3K27 tri-methylation is a diagnostic marker for malignant peripheral nerve sheath tumors and an indicator for an inferior survival. Mod Pathol. 2016;29:582-90. 29. Knight SWE, Knight TE, Santiago T, Murphy AJ, Abdelhafeez AH. Malignant Peripheral Nerve Sheath Tumors-A Comprehensive Review of Pathophysiology, Diagnosis, and Multidisciplinary Management. Children (Basel). 2022;9:38. estana RC, Parkes A. Systemic Options for Malignant Peripheral Nerve Sheath Tumors. Curr Treat Options Onco 30. Hassan A, Pestana RC, Parkes A. References 1. Gupta G, Mammis A, Maniker A. Malignant peripheral nerve sheath tumors. Neurosurg Clin N Am. 2008;9:533-43. 2. Hajdu SI. Peripheral nerve sheath tumors. Histogenesis, classification, and prognosis. Cancer. 1993;72:3549-52. 1. Gupta G, Mammis A, Maniker A. Malignant peripheral nerve sheath tumors. Neurosurg Clin N Am. 2008;9:533 2. Hajdu SI. Peripheral nerve sheath tumors. Histogenesis, classification, and prognosis. Cancer. 1993;72:3549- 3. Louis DN, Perry A, Wesseling P, Brat DJ, Cree IA, Figarella-Branger D, et al. The 2021 WHO Classification of Tumors of the Central Nervous System: a summary. Neuro Oncol. 2021;23:1231-51. 3. Louis DN, Perry A, Wesseling P, Brat DJ, Cree IA, Figarella-Branger D, et al. The 2021 WHO Classification of Tu summary. Neuro Oncol. 2021;23:1231-51. Page 8/11 Page 8/11 4. Wanebo JE, Malik JM, Vandenberg SR, Wanebo HJ, Driesen N, Persing JA. Malignant peripheral nerve sheath tumors. A clinicopathologic study of 28 cases. Cancer. 1993;71:1247-53. 5. Tucker T, Wolkenstein P, Revuz J, Zeller J, Friedman JM. Association between benign and malignant peripheral nerve sheath tumors in NF1. Neurology. 2005;65:205-11. 6. Chen J, Zheng Y, Chen Z, Fan F, Wang Y. Clinical presentation and long-term outcome of primary spinal intradural malignant peripheral nerve sheath tumors. Clin Neurol Neurosurg. 2019;185:105484. 7. Scheithauer BW, Erdogan S, Rodriguez FJ, Burger PC, Woodruff JM, Kros JM, et al. Malignant Peripheral Nerve Sheath Tumors of Cranial Nerves and Intracranial Contents: A Clinicopathologic Study of 17 Cases. Am J Surg Pathol. 2009;33:325-38. 8. Wu L, Deng X, Yang C, Xu Y. Spinal intradural malignant peripheral nerve sheath tumor in a child with neurofibromatosis type 2: the first reported case and literature review. Turk Neurosurg. 2014;24:135-9. 9. Ziadi A, Saliba I. Malignant peripheral nerve sheath tumor of intracranial nerve: A case series review. Auris Na 9. Ziadi A, Saliba I. Malignant peripheral nerve sheath tumor of intracranial nerve: A case series review. Auris Nasus Larynx. 2010;37:539-45. 0. Kamran SC, Shinagare AB, Holler Howard SA, Nishino M, Hornick JL, Krajewski KM, et al. Intrathoracic malign 10. Kamran SC, Shinagare AB, Holler Howard SA, Nishino M, Hornick JL, Krajewski KM, et al. Intrathoracic malignant peripheral nerve sheath tumors: imaging features and implications for management. Radiology and Oncology. 2013;47:230-8. 10. Kamran SC, Shinagare AB, Holler Howard SA, Nishino M, Hornick JL, Krajewski KM, et al. Intrathoracic malignant peripheral nerve sheath tumors: imaging features and implications for management. Radiology and Oncology. 2013;47:230-8. imaging features and implications for management. Radiology and Oncology. References 2013;47:230-8. 11. Xu Q, Xing B, Huang X, Wang R, Li Y, Yang Z. Primary malignant peripheral nerve sheath tumor of the cauda equina with metastasis to the brain in a child: case report and literature review. Spine J. 2012;12:e7-e13. 12. Frankel HL, Hancock DO, Hyslop G, Melzak J, Michaelis LS, Ungar GH, et al. The value of postural reduction in the initial management of closed injuries of the spine with paraplegia and tetraplegia. Paraplegia. 1969;7:179-92. 3. Karnofsky DA. The clinical evaluation of chemotherapeutic agents in cancer. Evaluation of chemotherapeutic 4. Rubino F, Eichberg DG, Shah AH, Luther EM, Lu VM, Saad AG, et al. When "Peripheral" Becomes "Central": Prim Intracerebral Nerve Sheath Tumor: A Case Report and a Systematic Review. Neurosurgery. 2021;88:1074-87. 14. Rubino F, Eichberg DG, Shah AH, Luther EM, Lu VM, Saad AG, et al. When "Peripheral" Becomes "Central": Primary and Secondary Malignant Intracerebral Nerve Sheath Tumor: A Case Report and a Systematic Review. Neurosurgery. 2021;88:1074-87. 5. McConnell YJ, Giacomantonio CA. Malignant triton tumors--complete surgical resection and adjuvant radioth J Surg Oncol. 2012;106:51-6. 15. McConnell YJ, Giacomantonio CA. Malignant triton tumors--complete surgical resection and adjuvant radiotherapy associated with improved survival. J Surg Oncol. 2012;106:51-6. 16. Patankar AP, Sheth JH. Intracranial Malignant Nerve Sheath Tumor in the Middle Cranial Fossa: A Rare Case Report with Review of Literature. Asian J Neurosurg. 2019;14:922-6. 17. Ducatman B, Scheithauer B, Piepgras D, Reiman H, Ilstrup D. Malignant peripheral nerve sheath tumors. A clinicopathologic study of 120 cases. Cancer. 1986;57:2006-21. 18. Li Y, Fan F, Xu J, An J, Zhang W. Primary malignant peripheral nerve sheath tumor of the spine with acute hydrocephalus: a rare clinical entity. J Neurosurg Spine. 2014;21:367-71. 19. Chen J, Zheng Y, Chen Z, Fan, Wang Y. Clinical presentation and long-term outcome of primary spinal intradural malignant peripheral nerve sheath tumors. Clin Neurol Neurosurg. 2019;185:105484. 20. Ghosh A, Sastri SB, Srinivas D, Mahadevan A, Anandappa CB, Shankar SK. Malignant triton tumor of cervical spine with hemorrhage. J Clin Neurosci. 2011;18:721-3. 21. Yu YH, Wu JT, Ye J, Chen MX. Radiological findings of malignant peripheral nerve sheath tumor: reports of six cases and review of literature. World J Surg Oncol. 2016;14:142. 2. Belakhoua SM, Rodriguez FJ. Diagnostic Pathology of Tumors of Peripheral Nerve. Neurosurgery. 2021;88:44 22. Belakhoua SM, Rodriguez FJ. Diagnostic Pathology of Tumors of Peripheral Nerve. Neurosurgery. 2021;88:443-56. 3. Rodriguez FJ, Folpe AL, Giannini C, Perry A. Figures Page 9/11 Figure 1 Case 2. Sagittal (a) and axial (b) post-contrast T1-weighted images showing the tumor at the right side extending from T10 to the sacrum. Post-surgical sagittal (c) and axial (d) post-contrast T1-weighted images showing the resection cavity. Figure 1 Figure 1 Case 2. Sagittal (a) and axial (b) post-contrast T1-weighted images showing the tumor at the right side extending from T10 to the sacrum. Post-surgical sagittal (c) and axial (d) post-contrast T1-weighted images showing the resection cavity. Figure 2 Case 5. Sagittal (a) and axial (b) post-contrast T1-weighted images showing the tumor at the left frontal lobe. Post-surgical sagittal (c) and axial (d) post- contrast T1-weighted images showing the resection cavity. Tumor pathologic sample (e) [hematoxylin-eosin 200x magnification] showing hypercellularity, fascicles of hyperchromatic spindled cells with pale cytoplasm, and scattered rhabdomyoblastic cells (arrowheads). Immunohistochemical stains for desmin (f) and myogenin (g). Axial (h) post-contrast T1-weighted images showing final tumor progression before the patient passed. Figure 2 Case 5. Sagittal (a) and axial (b) post-contrast T1-weighted images showing the tumor at the left frontal lobe. Post-surgical sagittal (c) and axial (d) post- contrast T1-weighted images showing the resection cavity. Tumor pathologic sample (e) [hematoxylin-eosin 200x magnification] showing hypercellularity, fascicles of hyperchromatic spindled cells with pale cytoplasm, and scattered rhabdomyoblastic cells (arrowheads). Immunohistochemical stains for desmin (f) and myogenin (g). Axial (h) post-contrast T1-weighted images showing final tumor progression before the patient passed. Page 10/11 Figure 3 Kaplan-Meier survival curves. Overall survival (a) in patients with a diagnosed MPNST. Survival by the following: extent of resection (b), presence of NF-1 mutation (c), tumor grade (d), age (e) and adjuvant chemotherapy/radiation (f). Abbreviations: extent of resection (EOR); gross total resection (GTR); neurofibromatosis type 1 (NF-1); radiotherapy (RT); subtotal resection (STR). Figure 3 Kaplan-Meier survival curves. Overall survival (a) in patients with a diagnosed MPNST. Survival by the following: extent of resection (b), presence of NF-1 mutation (c), tumor grade (d), age (e) and adjuvant chemotherapy/radiation (f). Abbreviations: extent of resection (EOR); gross total resection (GTR); neurofibromatosis type 1 (NF-1); radiotherapy (RT); subtotal resection (STR). Page 11/11
https://openalex.org/W4389108647	https://link.springer.com/content/pdf/10.1007/s11245-023-09963-w.pdf	English	null	The Paradox of the Future: Is it Rational to Feel Emotions for Future Generations?	Topoi	2,023	cc-by	10,104	2 It is absolutely not my intention to simplify or disregard the phe- nomenon of depression, which is very complex. However, for the pur- poses of the present paper—aimed at identifying, in the various cases in which there is an emotional response on the part of the subject, the object toward which it is directed—the case of depression rep- resents a borderline case, since the intense feelings are not directed at anything in particular, and that is indeed one of the reasons why depressed people have an extremely painful experience that is often difficult to heal from (Ratcliffe 2008: ch. 4; Thompson 1995). * Carola Barbero carola.barbero@unito.it Topoi (2024) 43:75–84 https://doi.org/10.1007/s11245-023-09963-w Topoi (2024) 43:75–84 https://doi.org/10.1007/s11245-023-09963-w 1 University of Turin, Department of Philosophy and Education Sciences, Turin, Italy Abstract According to some, there is a problem concerning the emotions we feel toward fictional entities such as Anna Karenina, Werther and the like. We feel pity, fear, and sadness toward them, but how is that possible? “We are saddened, but how can we be? What are we sad about? How can we feel genuinely and involuntarily sad, and weep, as we do know that no one has suffered or died?” (Radford, in: Proceedings of the Aristotelian Society, 1975). This is the paradox of fiction which is based on the assumption that emotions, to be genuine and rational, should be directed toward existent beings. But if beliefs about existence are necessary for us to be rationally moved by something, and such beliefs are lacking when we are moved by fiction (because we do not believe fictional characters and events to be real), then our capacity for an emotional response to fiction is irrational. Consequently, our emotional attitude toward future generations should be considered as irrational as well. But is this really the case? Are there good arguments to consider future and fictional entities as similar from this point of view? Or would it be better to distinguish the two? Is there such a thing as a paradox of the future? If so, how does it relate to the more famous paradox of fiction? Keywords Paradox of fiction · Object theory · Existence · Emotions · Future generations The Paradox of the Future: Is it Rational to Feel Emotions for Future Generations? Carola Barbero1 Accepted: 27 September 2023 / Published online: 29 November 2023 © The Author(s) 2023 1 Introduction These situations differ from each other,1 but what they seemingly have in common is the fact of having to do with affective states (such as fear, sadness, worry, and depres- sion). In all but the first of the above cases, the object toward which the affective response is directed does not exist in spacetime. Anna Karenina does not exist because she is a fic- tional character, Julia’s grandmother does not exist anymore (but she existed once), future generations do not exist yet, and in the case of Martine’s depression, nothing is making her feel down, nonetheless she feels listless and low.2 Take these six different situations: 2 Emotions, Objects, and Existence The third is the thought-theory solution (Carroll 1990; Lamarque 1981), which holds (modifying b2) that our emotional responses to works of fiction have as their object the descriptions, images, and thoughts presented in the work of fiction itself (instead of a belief in the object’s exist- ence). A more literal version of this approach is the shadow-object solution (Charlton 1984), according to which our emotional responses to works of fiction have as their object real people and events, resem- bling in some ways the people and events of fiction. The fourth solu- tion is the non-judgmental one, which argues that it is not necessary for emotional responses to fiction to involve the strong belief in exist- ence typical of judgment (therefore modifying again b2), as a weaker form of belief is sufficient in such cases (Morreal 1993). The fifth is the surrogate belief solution, according to which we respond emo- tionally to fiction because we believe that, in fiction, the characters are really suffering (Neill 1993). The sixth is the irrational solution, whereby emotional responses to fiction are classified as irrational behavior: even though one knows that certain people and situations do not exist, one feels emotions toward them. This is the solution originally proposed by Radford (1975). Rather than a solution, how- ever, it is a sort of explanation of the paradox, which doesn’t solve the problem as much as it proposes to acknowledge and accept it. Finally, the seventh and last one is the make-believe solution (Walton 1978; 1990), according to which the emotions we feel toward works of fic- tion are imaginary emotions (b1 is negated here), which are part of the same game of make-believe in which we are participating. ‘Lord, forgive me all’ she said, feeling it impossible to struggle. A peasant muttering something was work- ing at the iron above her. And the light by which she had read the book filled with troubles, falsehoods, sor- row, and evil, flared up more brightly than ever before, lighted up for her all that had been in darkness, flick- ered, began to grow dim, and was quenched forever” (Tolstoy, Anna Karenina, Ch. XXXI, Part 7). This is one of the most heartbreaking passages in the story. Take these six different situations: (a) Helen is afraid of the lion roaring in front of her. (b) Lisa is reading Lev Tolstoy’s masterpiece, Anna Karenina. When she arrives at the suicide scene at the railway station, she cries. (c) Julia is sad because she misses her grandmother who died 20 years ago. (d) Oriana is sad for her never-born child. The very fact of feeling emotions toward nonexist- ent objects has long been considered by philosophers as somehow problematic, when not even paradoxical, espe- cially when fictional entities are at stake, as happens in (e) Olivia, conscious of the unprecedented climate crisis that is taking place, is worried for future generations. (f) Martine is depressed and cries in bed every night. (f) Martine is depressed and cries in bed every night. 1 For a coherent and impressive account of emotions in all their nuances and facets, see Ben-Ze’ev (2000). This book is an up-to-date compendium on emotions, delving into both psychological and philo- sophical aspects of the subject. It doesn’t rely on strict necessary and sufficient conditions for membership within the category of emotion. (0121 3456789) 3 76 C. Barbero it? Radford sees premise (b2) as plausible and difficult to discard, just as premises (b1) and (b3).3 (b). Why does Lisa cry if there is nothing to cry for? Anna Karenina, as we all know, is a fictional character invented by Tolstoy, so how are we to explain Lisa’s reaction? Is she just irrational (Radford 1975)? Or should we consider her affective response toward Anna Karenina as structur- ally different from the one directed toward a real entity (Walton 1978)? 2 Emotions, Objects, and Existence When the paradox was formulated, the cognitive theory of emotions (according to which emotional experiences are based on the subject’s beliefs or evaluative judgments about the existence of the object of the emotion) was widely accepted. However soon afterward it began to be ques- tioned4: doubts started being raised against (b2), i.e. the fact Let us look more closely at the situation at hand. Lisa is reading ch. XXXI. Anna is at the railway station, and after having crossed herself, she “dropped the red bag and drawing her head back into her shoulders, fell on her hands under the carriage, and lightly, as though she would rise again at once, dropped onto her knees. And at the same instant she was terror-stricken at what she was doing. ‘Where am I? What am I doing? What for?’ She tried to get up, to drop backwards; but something huge and merci- less struck her on the head and rolled her on her back. ‘Lord, forgive me all’ she said, feeling it impossible to struggle. A peasant muttering something was work- ing at the iron above her. And the light by which she had read the book filled with troubles, falsehoods, sor- row, and evil, flared up more brightly than ever before, lighted up for her all that had been in darkness, flick- ered, began to grow dim, and was quenched forever” (Tolstoy, Anna Karenina, Ch. XXXI, Part 7). 3 Levinson (1997: pp. 22–27) summarizes some solutions that have been offered to the paradox of fiction. Since the premises that con- stitute the paradox are mutually incompatible, any solution would inevitably deny (or at least modify) one of the three. The first (negat- ing premise b1) is the non-intentional solution, according to which emotional responses to works of fiction do not fall within the inten- tional and cognitive dimension typical of normal emotions: in these cases, strictly speaking, we would not be dealing with emotions, but rather with less complex states such as moods or automatic reactions, which would not be characterized by the intentional and cognitive elements typical of real emotions. The second solution is the suspen- sion of belief (where premise b3 is negated), originally proposed by Coleridge (1817: ch. XIV). According to it, our emotional responses to works of fiction result from a momentary suspension of belief regarding the fact that the objects of fiction do not exist. 1 3 2 Emotions, Objects, and Existence And what about the deep sorrow experienced by Oriana Fallaci for a child that was never born (Fallaci 1976)? What about Olivia’s concern for the unprecedented climate crisis we are going through and the impact it will have on future generations?5 and the last one defining emotions as distinctive motivational states. Let’s now consider the situation described by (b), where Lisa feels saddened by Anna Karenina’s suicide: there is a belief (that there is a fictional character called “Anna Karenina” who decides to end her days by throwing her- self under a train), a bodily modification (Lisa is crying, her heart is racing), a tendency to do (or not to do) something (for instance, not to go to the railway station when feeling particularly depressed), and an evaluation (Anna Karenina is unable to think clearly when she arrives at the railway station). If compared with the previous one, this situation is different indeed, but not because it is paradoxical whereas the former one is normal and acceptable. Both look like standard situations when considered from a psychological point of view since they both have to do with an emotional reaction stimulated by a specific object/event identified as the reason for that emotional response. This is what one can legitimately conclude from Stecker’s remarks. As Richard Moran (1994) underlines—and the cases just presented show this quite clearly—contrary to what Radford believes, it seems that we can be perfectly rational in feeling emotions toward nonexistent objects. This has been convinc- ingly explained by Robert Stecker (2011) who insists on the fact that even abandoning (b2), the premise according to which believing in the existence of what makes us feel in a certain way is a necessary condition for having certain emotions toward it, we can still discuss valuable topics sur- rounding the debate on the paradox of fiction. But then where does the difference between (a) and (b) lie? If not in the genuineness of the emotion felt, it has to reside in different psychological attitudes. And what is the diversity between these psychological attitudes based on? Evidently on the objects toward which the emotions are respectively directed in (a) and (b). This shows why an onto- logical approach such as the one offered by Object Theory6 could prove to be extremely useful here, because it considers objects not as far as their mode of presentation is concerned (i.e. 2 Emotions, Objects, and Existence While reading, Lisa is well aware, of course, that things are getting worse—Anna and Vronsky have become increasingly bitter toward each other, a combination of boredom and suspicion has destroyed Anna’s mental health, Vronsky is probably cheating on her, and is starting to get tired of the whole situation –, yet she couldn’t imagine such a sudden and tragic end. Then she cries, thinking about how cruel life can be sometimes. Colin Radford (1975) maintains that the fact of feeling emotions for fictional characters like Anna gives rise to a philosophical paradox, the (so-called) paradox of fiction, constituted by three plausible premises that cannot be con- jointly true at the same time: (b1) Lisa feels sad about Anna’s tragic end and she knows Anna is a fictional character; (b1) Lisa feels sad about Anna’s tragic end and she knows Anna is a fictional character; i (b2) Believing in the existence of x (what makes us sad) is a necessary condition for having certain emotions toward x;i 4 Supporters of the idea that Radford was a cognitivist about emo- tions include, among others, Matravers (1998), Hartz (1999), and Wilson (2013). For an account according to which the rise of the paradox is not related to Radford’s commitment to cognitivism, see Friend (2020). Even if the form of cognitivism supposedly main- tained by Radford in his seminal paper has been strongly criticized and eventually dismissed, cognitivism still has supporters today— although they do not agree on what they mean by “cognitive thesis” (belief /rationality/consciousness/understanding) and “cognitive (b3) Lisa does not believe in the existence of fictional characters. Don’t we need to judge/believe that someone suffered or died to be sad for them, exactly as we need to judge/believe that a lion is or can be dangerous to us for being afraid of 1 3 The Paradox of the Future: Is it Rational to Feel Emotions for Future Generations? 77 that emotions involve judgments or beliefs concerning the object’s existence, and hence the paradox itself started to falter. Objections included very naïve yet no less persuasive points. For instance, if it were true that we need to believe in the existence of something to feel emotions toward it, then what about the sadness Julia feels for her grandmother who died 20 years ago (and therefore no longer exists)? 7 Meinong sees metaphysics as determined by “the prejudice in favor of the actual”, whereas Object Theory goes beyond that, being char- acterized as an a priori science dealing with whatever can be known a priori about objects. Object Theory deals with objects as such, it “has to do with the given taken in its entirety” (Meinong 1904: §§ 2, 11). 6 For a historical overview of Object Theory see Nef (1998), Raspa (2002) and Bakaoukas (2003). The most famous Object Theory is undoubtedly Meinong’s (1904), which I consider as a constant refer- ence point in this paper. 8 According to this definition everything that has at least one prop- erty is an object and the criterion for distinguishing what is an object from what is not is the following: Pegasus is an object because the name “Pegasus” stands for something to which certain properties cor- respond (“being a winged horse”, “being Medusa’s and Poseidon’s son”, “being a mythological animal”), while on the contrary wrtgfh is not an object, because “wrtgfh” does not stand for anything. On the problems potentially engendered by such a criterion, see Salmon (1999: pp. 304–308), Kroon (2003: pp. 155–157), and Caplan (2004). 5 As Stecker (2011: p. 298) clearly states: “In the case of pity for those who existed in the past, the attitude is belief that they suffered, the same attitude that we have to those who suffer in the present. In the case of future directed emotions, the proposition is that an event might occur, and the attitude is again belief. Belief in these situations suffices to explain what I feel”. methodology” (should it involve cognitive psychology, evolutive psy- chology, or education?). Convincing cognitive positions have been endorsed especially in film studies by Carroll (2003), Smith (2003), Perrson (2003), and Plantinga (2009). 5 As Stecker (2011: p. 298) clearly states: “In the case of pity for those who existed in the past, the attitude is belief that they suffered, the same attitude that we have to those who suffer in the present. In the case of future directed emotions, the proposition is that an event might occur, and the attitude is again belief. Belief in these situations suffices to explain what I feel”. methodology” (should it involve cognitive psychology, evolutive psy- chology, or education?). Convincing cognitive positions have been endorsed especially in film studies by Carroll (2003), Smith (2003), Perrson (2003), and Plantinga (2009). Footnote 4 (continued) Footnote 4 (continued) 2 Emotions, Objects, and Existence In each instance there are objects to which emotions are directed, even if they are objects of different kinds (and when we analyze the different emotional reactions they engender, their diversity becomes evident). Being terrified by a lion, for example, is different—and therefore elicits a different behavior—from being terrified by a dragon: whereas Helen believes the lion is in front of her, she does not believe the dragon is, and whereas she believes the lion is dangerous to her, she does not believe the dragon is (or could be), even if the dragon is characterized by the property of being danger- ous (Helen knows that fictional entities are ontologically separated from us, hence cannot harm us in any way). That’s why she would run from the lion but not from the dragon. Analogously, ontological considerations will also explain why being sad for one’s dead grandmother (as happens to Julia) and being worried for future generations (as in Olivia’s case) elicit distinct behaviors: a supposedly controlled (and more or less intense) sadness in the first case, together with the conviction that there is nothing to be done but keep the memory of the beloved grandmother alive; an active concern in the second case, in the certainty that the future has to be prepared in the present, and therefore that the best thing is to act and change today if we want to do something for the generations of tomorrow.f Object Theory takes into account all objects, lions as well as Anna Karenina, unborn children as well as future genera- tions,9 i.e. existent objects as well as (variously) nonexist- ent ones. According to this Theory, everything that has at least one property can be considered an object: everything that is not nothing is something. It does not matter whether Anna Karenina is a fictional object we will never meet on the street, whereas the lion is a dangerous animal we can meet, escape from, or befriend (as in the 2018 French movie Mia and The White Lion by G. De Maistre). From this point of view, the definition of what an object is does not include its possible existence. Once we have an object corresponding to a set of proper- ties, we might of course need to know what kind of object it is: is it existing, fictional, past, future, or imaginary? 2 Emotions, Objects, and Existence whether they exist, do not exist, subsist, or simply are) but in a more general way (which is why, according to Alex- ius Meinong, Object Theory has a wider field of inquiry than metaphysics, which deals only with existing things, i.e. with the entirety of the real),7 mostly focusing on the set of properties whose object-correlates they are8 and this independently from their possibly also being objects of a particular kind for someone in some way. i Emotions are generally considered mental states (Dixon 2003; Solomon 2008) connected with a cognitive base, bod- ily changes, a tendency to act, and evaluation. Let’s con- sider the difference between (a) and (b). Take the situation described in (a), where Helen is afraid of a lion: there is a belief (that there exists a dangerous lion), a bodily modifica- tion (Helen is sweating, her heart is racing), a tendency to do something (she wants to run away, call for help, maybe try to close the cage door), and an evaluation (the roaring lion looks angry and is presumably very dangerous). An interesting question concerns the role these elements play in the emotional reaction they provoke: among Helen’s apprais- ing the lion as dangerous, her increased heartbeat, disturb- ing feeling, and tendency to flee, what should we consider as an essential component (Prinz 2004)? Three main ways of answering that question have been identified (Scarantino 2016): the one defining emotions as distinctive feelings, the other regarding emotions as involving a specific evaluation, 8 According to this definition everything that has at least one prop- erty is an object and the criterion for distinguishing what is an object from what is not is the following: Pegasus is an object because the name “Pegasus” stands for something to which certain properties cor- respond (“being a winged horse”, “being Medusa’s and Poseidon’s son”, “being a mythological animal”), while on the contrary wrtgfh is not an object, because “wrtgfh” does not stand for anything. On the problems potentially engendered by such a criterion, see Salmon (1999: pp. 304–308), Kroon (2003: pp. 155–157), and Caplan (2004). 1 3 3 78 C. Barbero longing for her grandmother, Oriana's sorrow over her never- born child, and Olivia’s concern for future generations. 2 Emotions, Objects, and Existence Indeed, the kind of object it is will not only arouse a specific emo- tional response, but also influence our behavior toward it. In relation to Anna Karenina, Lisa knows her emotions are directed toward a fictional literary entity created by Lev Tolstoy in the homonymous novel published in 1877 and accepted (i.e., recognized as such) by a community of read- ers and critics.10 Therefore, all she needs to have a genuine emotional response is to believe in the properties character- izing Anna Karenina together with the events she is involved with. Furthermore, given the properties Anna has—i.e., the property of being desperate and abandoned, the property of being rejected by her friends, and the property of being crushed by a train—it is not difficult to understand why Lisa is brought to tears. Moreover, it is precisely because she rec- ognizes the nature of the object making her feel that way—a fictional object—that Lisa does not even try to prevent Anna from committing suicide at the railway station that day. After all, she knows that we have no causal power over such enti- ties, unless one happens to be the author of the fictional entity. Tolstoy, for instance, could have decided to “save” Anna somehow,11 but Lisa, like the rest of us, unfortunately, cannot. A different case from the situations described in (a)–(e) is (f), concerning Martine’s depression: in that case, there is no object at all to be sad for, there is instead a medical illness affecting her feelings, thoughts, and acts. Martine’s depressed mood—losing interest in activities she once enjoyed, having trouble sleeping, losing energy or experi- encing more fatigue, feeling worthless, finding it difficult to think, concentrate or make decisions—has no direction and no intentional object whatsoever (not even a nonexist- ent one). The same principle applies, from a strictly ontological standpoint, in the cases of Helen’s fear of the lion, Julia’s 10 According to the metaphysical version provided by Thomasson (1999: pp. 139–145), fictional entities are to be seen as abstract arti- facts recognized as such by a literary community. 11 This point is perfectly clear to Annie Wilkies, the protagonist of Stephen King’s novel Misery (1987), who rescues and then takes Paul Sheldon, the author of her favorite book series Misery, prisoner in order to force him to resurrect the character by writing Misery’s Return, a story in which it turns out that the main character was bur- ied alive while comatose. 9 According to the definition of future generations given by Andina (2022: pp. 87–88), “From an ontological point of view at least, future generations are something similar to fictional entities”. Therefore, we could consider kinds of objects we have thus far kept distinct as belonging ontologically to the same category. 11 This point is perfectly clear to Annie Wilkies, the protagonist of Stephen King’s novel Misery (1987), who rescues and then takes Paul Sheldon, the author of her favorite book series Misery, prisoner in order to force him to resurrect the character by writing Misery’s Return, a story in which it turns out that the main character was bur- ied alive while comatose. 9 According to the definition of future generations given by Andina (2022: pp. 87–88), “From an ontological point of view at least, future generations are something similar to fictional entities”. Therefore, we could consider kinds of objects we have thus far kept distinct as belonging ontologically to the same category. 10 According to the metaphysical version provided by Thomasson (1999: pp. 139–145), fictional entities are to be seen as abstract arti- facts recognized as such by a literary community. Footnote 13 (continued) Thought Theory maintains, but—and here the inevitability of the ontological question becomes clear—what are they thoughts and images of? Object Theory says that there is something, a dragon, characterized by, among others, scary properties, which are the ones causing us to feel afraid. This is the parte objecti solution offered by Object Theory. When responding emotionally to something, our emotions always have a formal object which is a property implicitly ascribed by the emotion to its target, in virtue of which the emotion can be seen as intelligible. For example, my fear of a dog construes a number of the dog’s features (its salivating maw, its ferocious bark, its brutality) as being frightening, and it is exactly my perception of the dog as frightening that causes me to feel fear, rather than some other emotion. The formal object associated with a given emotion is essential to the definition of that particular emotion. 13 This solution to the paradox could be seen as complementary to the Thought Theory solution defended by Peter Lamarque (1981) according to which while reading fiction we experience real emo- tions caused by thoughts brought to our mind by the novel (there- fore involving a modification of b2). The idea is that vivid imagining can be a good substitute for belief: indeed, bringing to our mind fic- tional characters and events seems to be enough to feel genuine emo- tions toward them. According to Thought Theory, we are frightened of the dragon (in an intentional sense), not by it (in a causal sense): the objects of our emotion are thoughts (instead of non-existent objects/individuals), and bringing a thought to one’s mind does not mean believing a proposition to be true, but simply entertaining the thought. Distinguishing between the real object of our fear (which is missing in the case of the dragon as it does not exist: that’s why we cannot be frightened by it) and the intentional object (what we are afraid of the dragon, experiencing dragon-fear), Thought Theory is successful in capturing the object of our emotion without being com- pelled to admit emotions caused by nonexistents. Thought Theory satisfactorily solves the paradox of fiction by giving a persuasive answer to questions concerning what happens to people feeling emo- tions toward nonexistents such as dragons, Anna Karenina, and the like. 3 A Solution to the Paradox Let’s go back to the paradox of fiction as set forth by Rad- ford. According to him, there would be a problem with our emotional response to Anna Karenina because in that situ- ation there is seemingly no object (where “object” equals “existing object”) to which our emotions are directed. “We are saddened, but how can we be? What are we sad about? How can we feel genuinely and involuntarily sad, and weep, as we do, knowing as we do that no one has suffered or died?” (Radford 1975: p. 77). By making ontological con- siderations in line with what has been suggested by Object Theory—hence by underlining the plausibility of the distinc- tion between being an object and being an existing object—it 1 3 The Paradox of the Future: Is it Rational to Feel Emotions for Future Generations? 79 directed toward a (fictional/possible/future/past) object is clearly directed toward something. Therefore, one doesn’t have to believe in the existence of Anna Karenina/one’s dead grandmother/one’s never-born child/future generations to be concerned for them. To explain why we act in a specific way rather than in another, i.e., to explain psychological differ- ences and subsequent behavior14, it is enough to disbelieve in their existence (we are well aware that our causal powers, active in reality, are suspended when it comes to the realm of the fictional/possible/future/past). By focusing on the disbe- lief in the actual existence of the object at stake, it is possible to explain not only why such emotions can be considered genuine and rational, but also why they may differ from one another (sadness for Anna Karenina and a never-born child are not the same, since they are directed toward two ontologically different objects), why they motivate different behaviors (while my fear of the dragon has, most likely, no effect on my daily actions, my concern for future genera- tions may and should cause me to behave differently in my present), why they can be appropriate or inappropriate to their objects (I can cry for a while for Anna Karenina, but it would be inappropriate to cry a whole night for her suicide, whereas that would be appropriate in the case of a friend who died yesterday), and why they are involved in distinctive subjective psychological experiences (showing a different phenomenology in each case). 12 According to Object Theory, Anna Karenina can be considered an object for all intents and purposes. This is notoriously the view defended by Meinongian theories of fictional entities (Parsons 1980; Zalta 1983; Barbero 2005; Berto 2011). The debate on the metaphys- ics and ontology of fictional entities goes far beyond the Meinon- gian view: it is articulated by eliminativists (especially Walton 1990) according to whom fictional entities are not objects (or rather they are something only within a game of make-believe); possibilists (Lewis 1978) who see fictional entities as possible objects; artifactualists (Thomasson 1999), who maintain that fictional entities are abstract artifacts similar to laws and games; and finally syncretists (Voltolini 2006), who defend a combination of neo-Meinongianism and artifac- tualism. For a general overview on the debate see Voltolini (2006: ch. 1, 2, 5). (S3) X does believe that there is a fictional/possible/past/ future object exhibiting emotion-inducing properties. (S3) X does believe that there is a fictional/possible/past/ future object exhibiting emotion-inducing properties. Distinguishing between being and existing, Object Theory makes it possible to identify an object (a fictional/possible/ past/future object, in the aforementioned examples) causing a specific emotion (sadness/worry), even if that object does not exist.13 That is how the paradox disappears: an emotion 12 According to Object Theory, Anna Karenina can be considered an object for all intents and purposes. This is notoriously the view defended by Meinongian theories of fictional entities (Parsons 1980; Zalta 1983; Barbero 2005; Berto 2011). The debate on the metaphys- ics and ontology of fictional entities goes far beyond the Meinon- gian view: it is articulated by eliminativists (especially Walton 1990) according to whom fictional entities are not objects (or rather they are something only within a game of make-believe); possibilists (Lewis 1978) who see fictional entities as possible objects; artifactualists (Thomasson 1999), who maintain that fictional entities are abstract artifacts similar to laws and games; and finally syncretists (Voltolini 2006), who defend a combination of neo-Meinongianism and artifac- tualism. For a general overview on the debate see Voltolini (2006: ch. 1, 2, 5). Such an approach is also useful when dealing with rela- tions involving non-existent objects: how else to explain that 3 A Solution to the Paradox can be maintained that some things (like Anna Karenina,12 dead people, never-born children, future generations) may be objects even without being existing ones. From such a perspective the paradox does not arise in the first place: i (S1) X feels sadness/worry for Anna Karenina/their dead grandmother/their never-born child/future generations and X knows that Anna Karenina /their dead grandmother/their never-born child/future generations are fictional/possible/ past/future objects (and therefore that they do not exist); (S2) Believing that there is (and not that there exists) an object exhibiting some of the emotion-inducing properties specific to sadness/worry is a necessary and sufficient condi- tion for experiencing sadness/worry about it (i.e. the emotion has to be directed toward something); 4 A Paradox of the Future? So far, we have tried to explain how the paradox of fiction can be dismantled and why the emotions we feel for fic- tional/past/possible/future objects can be considered genu- ine and rational. Let us now focus specifically on emotions concerning future generations, as seen in (e). (F3) We do believe that future generations will inherit a world where there will be inequality, poverty, migration, and disease due to the climate crisis. Moreover, and differently from what happens to fictional or merely possible entities, future generations are expected to exist in the future. This is quite an important point that Object Theory overlooks, being mostly interested in present- ing objects in their absolute generality and independently from their various ways of being. But this is not something to be sidelined, as it allows us to highlight an important dis- tinguishing feature of certain objects such as future genera- tions over fictional ones such as Anna Karenina and the like. As Andina (2022: p. 94) extensively explains, “[this] is an important point that should be noted, a point which is simi- larly of an ontological nature: while fictional entities are and always will be abstract entities, as they do not and never will exist in space–time, sooner or later future generations—bar- ring unpredictable and certainly undesirable catastrophes— will exist. In other words, as the name itself suggests, future generations are an entity that envisages a passage of status from being potential to being actual. Thus, they are destined to become present generations, which changes things quite considerably”.i As an example, consider the following statement from the main webpage of Save the Children under The Climate Crisis. Climate Change Is a Grave Threat to Children’s Sur- vival15: “Right now, in the U.S. and around the world, chil- dren’s lives are under threat due to climate change. Nearly 710 million children are currently living in countries at the highest risk of suffering the impact of the climate crisis. However, every child will inherit a planet with more fre- quent extreme weather events than ever before”. Do we feel sad and worried about future generations who will inherit a world where frightening events such as floods and hurri- canes will be normal? Do we feel anxious considering how hotter temperatures, air pollution, and violent storms will constitute life-threatening elements for tomorrow’s children? 15 https://www.savethechildren.org/us/what-we-do/emergency-respo nse/climate-change. 16 A historical possibilist account of fictional and, in general, nonex- istent objects is the one provided by Russell (1903) and Lewis (1978). For a discussion of this topic, Orilia (2002: ch. 7). Footnote 13 (continued) That is why it can be considered as a solution parte subjecti, whereas Object Theory, by focusing on ontology, is parte objecti. We may be, and actually are, frightened by thoughts and images as 14 As far as fictional entities are concerned, and as Brock (2007: p. 217) convincingly underlines, we do not experience emotions such as shame, embarrassment or remorse, and this happens precisely because we are aware of the kind of ontological status fictional char- acters have. For instance, to regret our actions toward someone is tan- tamount to somehow believing that there is an existent person toward whom we might have acted differently, but when fictional entities are concerned, we do not believe anything of that sort. We couldn’t have acted differently toward any fictional entity for the simple reason that we do not act toward any such entity (hence we cannot regret or be ashamed of our actions toward any of them). We are ontologically cut-off from having such interactions with fictional entities. 1 3 80 C. Barbero the future has reason to arise, since what we have are the following three premises: Anna Karenina was disappointed in her married life in a similar way to Lady Diana? Or that a given war will lead to a very complicated socio-economic period for future genera- tions? To make sense of such matters, we need to ontologi- cally admit both relata, even those that are not present and existent. (F1) We feel sad/worried/anxious about future genera- tions, even though we know that future generations and the world they will inherit currently do not exist; (F2) Believing that future generations will inherit a world where frightening events will be normal is a nec- essary and sufficient condition for experiencing sadness/ worry/anxiety about them; (F2) Believing that future generations will inherit a world where frightening events will be normal is a nec- essary and sufficient condition for experiencing sadness/ worry/anxiety about them; 5 Empathy for Nonexistents We have seen how, by following the solution offered by Object Theory, we can consider objects in a general way, focusing on the set of properties whose object correlates they are, independently from their possibly also being objects of a particular kind for someone in some way. Thus, the paradox of fiction has been solved thanks to an ontological argument. i But a legitimate question arises at this point: when deal- ing with nonexistent entities such as Anna Karenina, a dead grandmother, a never-born child, or future generations, how can we explain our emotional involvement? When we are sad to the point of tears for Anna Karenina or feel worried for future generations, what happens is that we see both as other people, others than us, hence as something we can feel a sort of “Einfühlung” or empathy for. In fact, the other is constituted by appresentation, as other than myself (Husserl 1950), but also as similar to me: the step from the issues raised by the paradox of fiction to those typical of empathy is a short one. i And what about considering future objects as possi- ble ones? This wouldn’t be a good move either, because whereas possible objects are, say, logically admissible enti- ties in the sense that their characteristics are such that they could exist in the world (even if they do not currently exist), future objects are predicted or expected to exist in the future. Hence, differently from future objects that are definite and determined objects/events and will come to pass, possible objects are somehow “potential” and could in principle be actualized under certain circumstances or conditions, but at the moment are not expected. An example of a possible object is that of a new species arising on Earth, whereas an example of a future event is a solar eclipse, which will occur next time the Moon will pass between the Earth and the Sun, thereby obscuring the view of the Sun from a (bigger or smaller) part of the Earth (this happens approximately every six months). It is therefore important to distinguish possible objects from future ones: “future generations have a form of existence that is destined to move from the pos- sible to the actual, or from potential being to being. 17 According to the seminal definition given by Lipps (1979), then debated by philosophers such as Prandtl (1910), Stein (1989), and Scheler (1954). 4 A Paradox of the Future? Or would it be irrational to feel that way, since the future, by definition, is something that is not present yet? Those main- taining that only the belief in something currently existing justifies our emotional involvement should say that in such cases we are faced with a paradox not too dissimilar to that of fiction set forth by Radford: Hence, even if both future and fictional objects are similar (Andina 2022: p. 94) when considered from an ontological point of view, it may be useful to try to reach a further level of analysis. Are there perhaps good arguments for consid- ering both as possible objects? Put differently, what about considering both fictional and future objects as possibilia? In the debate concerning the ontology of fictional entities, it has been suggested to regard fictional entities as possible ones, i.e., as entities that are nonexisting in the actual world, but located in some other possible world.16 The idea is that, for whatever is told as a fictional story in our world, there is at least one possible world in which that event subsists as a state of affairs. In such a world the story is told as a known fact and is therefore no longer fictional (Lewis 1978). For (F1) We feel sad/worried/anxious about future genera- tions who will inherit a world where frightening events will be normal; (F1) We feel sad/worried/anxious about future genera- tions who will inherit a world where frightening events will be normal; (F2) Believing in the existence of x (what makes us feel worried) is a necessary condition for having certain emo- tions toward x; (F3) We do not believe that future generations and the world they will inherit currently exist. Nonetheless, by adopting the view suggested by Object Theory, I have underlined how it is possible to identify an object causing specific emotions (sadness, worry, anxiety) even if that object does not exist. Therefore, no paradox of 1 3 3 The Paradox of the Future: Is it Rational to Feel Emotions for Future Generations? 81 81 such an account, fictional characters do have the proper- ties attributed to them in the work of fiction. 4 A Paradox of the Future? For instance, Sherlock Holmes is a human being existing in some pos- sible world and has all the properties Conan Doyle’s stories ascribe to him, such as being a brilliant detective, the friend of a man called “Watson”, the enemy of another man called “Professor Moriarty”, and so on. Moreover, since Sherlock Holmes is described as a concrete entity with spatiotemporal existence, he is a concrete entity in the world in which his story is true. A classical objection against such a possibil- ist view is the one raised by Kripke (1980) based on the so-called problem of ontological indeterminacy: how could we choose the right Sherlock Holmes among the many (per- haps infinite) possible concrete individuals possessing all the properties ascribed to Sherlock Holmes in Doyle’s stories, but differing in other crucial ways (height, weight, eye color, etc.)? Moreover, what prevents a single possible world from containing many distinct individuals, each of them satisfying all the relevant properties of Sherlock Holmes? It is far from easy to try to find convincing answers to such questions. 5 Empathy for Nonexistents Thus, we are dealing with abstract artefacts that will become con- crete groups, namely generations that live in a specific time and space. The metaphysically interesting point is that the concrete group (the future generation that occupies cer- tain space–time coordinates) shows a dependency on the abstract artefact that is both historical and genetic, such that the abstract artefact determines the characteristics and pos- sibilities of the concrete group” (Andina 2022: p. 103). The literature on this subject is very extensive and no satisfactory definition of empathy has been reached: indeed, the term “empathy” is used to describe a broad range of psychological abilities—considered essential to what makes humans social beings—that enable us to understand what other people are thinking and feeling, to emotionally con- nect with them, to share their thoughts and feelings, and to be concerned about their well-being. However, we can intuitively consider as empathy the ability to grasp the men- tal states (particularly the emotions) of other people.17 So what happens with the special kind of objects we are con- sidering here? How is it possible to grasp someone’s mental state when that someone does not exist (being fictional or a future entity)? Can we be said to truly perceive those mental states? Or do we merely imagine them? Should empathy for non-existent beings be distinguished from empathy for real people categorically or only by degree? Let’s take three steps back and try to briefly address these questions. According to the supporters of Theory Theory (Carruthers, Smith 1996; Fodor 1987; Gopnik and Wellman 1994), empathy presupposes that the one who feels empa- thy has a sort of folk psychological theory of mind about the person with whom one empathizes. From this point of view, empathy could be seen as a cognitive way of reading and understanding other people’s minds. We would then make law-like generalizations that imply concepts of men- tal states needed to understand others and their reasons for 1 3 82 C. Barbero acting, as well as to predict their future behavior.18 Roughly speaking, if I see somebody crying, I infer that they are sad because I know the connection between the two phenomena, crying and sadness. In other words, since the mind is not observable, what happens—Theory Theory maintains—is that we perceive and interpret others only indirectly. 5 Empathy for Nonexistents This way of understanding human behavior would be similar to the way we usually interpret natural phenomena; therefore, empathic mind reading could be considered almost scien- tific, as it seems possible to use laws and theories to explain and understand the human domain. A potential objection to such a view is that we do not always apply a theory when we meet and try to understand other people, nor do we always cognitively infer concrete behavior from general laws. companions we can see face to face while interacting in shared contexts—by definition we can never perceive them directly; hence, the only way to “reach” them somehow would be through the mediation of a theory or simulation (Currie 2006). Neglecting the difference between existent and nonexistent entities as far as empathy is concerned wouldn’t be a good idea19 since, as I have extensively explained, the respective targets of empathy do not have the same ontological status (Petraschka 2021). Persons exist, whereas fictional characters, past, possible, and future indi- viduals do not. When sitting in our armchair at home watching Hitch- cock’s Psycho, we get scared and are potentially inclined not to shower that night (whereas, of course, we are not motivated to call the police or anything like that because we know the kind of object Marion Crane is—a fictional one—and therefore we know there is nothing we can do to prevent that crime taking place at the Bates Motel). Analogously, when we read Exit West (2017), by Mohsin Hamid,—which tells the story of Nadia and Saeed living between checkpoints, roundups, mortar launches, and shoot- ings, in a place where death appears to be the only hori- zon, and where a strange rumor circulates about mysterious doors that if passed through, by paying and at the risk of one’s life, transport people somewhere else—our emotional engagement and empathic understanding might motivate us to change our opinions on the matter or to donate money to a refugee aid organization. This happens precisely because non-actual worlds (imaginary, fictional, or just not present ones), are always connected to the real world,20 even if they are explicitly defined as nonexistent. 18 For criticism of this position (considered too theoretical and gen- eral), see Zahavi (2014). 5 Empathy for Nonexistents g y g Another approach is the one defended by Simulation the- orists (Goldman 2006; Gallese 2001) according to whom empathy does not imply a theory but rather a simulation mechanism: we imagine how we would feel if we were in that person’s shoes, i.e., we simulate the other’s state in our mind and then arrive at how the other feels by imitating their behavior in our mind, projecting our own mental pro- cess onto theirs. Unlike Theory Theory defenders, always insisting on a third-person perspective, those maintaining Simulation Theory claim that we simulate the other person’s situations from a first-person perspective and use our mental apparatus to generate thoughts, beliefs, desires, as well as emotions. On the one hand (according to Theory Theory) understanding others would mean being able to grasp their reasons (purely observational, third-person perspective), emotions, and desires; on the other (according to Simulation Theory), it would mean looking at things as if we were the others (first-person perspective). Both these theories assume that there is no direct perception of the other and that the only way to access other people is either to try to understand how they feel from how they behave, or to imagine being them and then feeling and thinking from their specific stand- point. Therefore, when trying to empathize with someone else, we infer their state of mind from our own (Maibom 2017). 19 Unless one is only interested in the psychological processes involved in empathy, whereby any ontological difference is irrelevant, as is the case in Currie (1997) and Robinson (2010). 20 For further reasons about whether both fictions and nonfictions should be considered as related to the real world, see Friend (2017). References Routledge, London Carroll N (2003) Engaging the moving image. Yale University Press, New Haven Carruthers P, Smith PK (1996) Theories of theories of mind. Cam- bridge University Press, Cambridgei Charlton W (1984) Feeling for the fictious. Br J Aesth 24:206–216 i Coleridge ST (1817) Biographia literaria, or biographical sketches of my literary life and opinions. In: Stauffer D (ed) Selected poetry and prose of coleridge. New York, Random House, p 1951 Currie G (1997) The paradox of caring. In: Hjort M, Laver S (eds) Emotion and the arts. Oxford University Press, Oxford, pp 63–77 Currie G (1997) The paradox of caring. In: Hjort M, Laver S (eds) Emotion and the arts. Oxford University Press, Oxford, pp 63–77 Currie G (2006) Anne Brontë and the uses of imagination. In: Kieran M (ed) Contemporary debates in aesthetics and the philosophy of art. Blackwell, Oxford, pp 209–221 Dixon T (2003) From passions to emotions: the creation of a secular psychological category. Cambridge University Press, Cambridge Currie G (1997) The paradox of caring. In: Hjort M, Laver S (eds) Emotion and the arts. Oxford University Press, Oxford, pp 63–77 Currie G (2006) Anne Brontë and the uses of imagination. In: Kieran M (ed) Contemporary debates in aesthetics and the philosophy of art. Blackwell, Oxford, pp 209–221 Currie G (2006) Anne Brontë and the uses of imagination. In: Kieran M (ed) Contemporary debates in aesthetics and the philosophy of art. Blackwell, Oxford, pp 209–221 Dixon T (2003) From passions to emotions: the creation of a secular psychological category. Cambridge University Press, Cambridge Fallaci O (1976) Letter to a child never born. Simon & Schuster, New York Fodor JA (1987) Psychosemantics. MIT Press, Cambridgei Friend S (2017) The real foundations of fictional worlds. Australas J Philos 95(1):29–42 Friend S (2020) Fiction and emotion: the puzzle of divergent norms. Br J Aesth 60(4):403–418 Gallagher S (2012) Empathy, simulation, and narrative. Sci Context 25(3):355–381 Funding Open access funding provided by Università degli Studi di Torino within the CRUI-CARE Agreement. Funding Open access funding provided by Università degli Studi di Torino within the CRUI-CARE Agreement. Gallese V (2001) The ‘shared manifold’ hypothesis: from mirror neu- rons to empathy. J Conscious Stud 8:33–50 Goldman AI (2006) Simulating minds: the philosophy, psychology, and neuroscience of mindreading. 21 I would like to thank the editors of this Special Issue, Tiziana Andina and Giulio Sacco, for discussion of earlier versions of this paper: their critiques, suggestions and advises have been extremely useful. I am very grateful also to the two anonymous referees of Topoi for their careful reading, serious objections, and suggestions. References In fact, differently from Anna Karenina, Julia’s dead grandmother, and Pegasus, even if future generations do not exist presently, they will exist in a more or less remote future. That is why, distinguishing future entities from mere possible ones, we should motivate our behavior accordingly. Andina T (2022) A philosophy for future generations. Bloomsbury, London Bakaoukas M (2003) Nothing exists. A history of the philosophy of non-being. Xlibris Corporation, Philadelphia non-being. Xlibris Corporation, Philadelphia Barbero C (2005) Madame bovary: something like a melody. AlboVer- sorio, Milano Ben-Ze’ev A (2000) The subtlety of emotions. MIT Press, Cambridgei We know that Anna Karenina does not exist, as she is a fictional entity created by Tolstoy, therefore we do not even try to save her at the railway station that day. Nonetheless, we feel sad and cry when we think about the madness caused by jealousy and mental obfuscation. Hence, we take Anna Karenina’s story as a reminder not to let a mixture of despair and loneliness play tricks on us. Similarly, when thinking about future generations and the unprecedented climate cri- sis that is taking place, we feel sad and anxious, in a way that is both rational and genuine. But—and here comes the difference between fictional and future objects—we do even more than that. For we are well aware that today, when those generations do not exist (yet), we are preparing the world they will live in tomorrow: that is why our concern for them should be neither sterile nor static, but rather projected for- ward. We must try to figure out what we can do for future generations, even if we know that (because of their onto- logical status) they have not done a thing for us. Moreover, we have seen how it is possible to activate something like empathy even for nonexistents, i.e., a way of understanding others by grasping their mental states. Therefore, it is possi- ble for individuals to be rationally and genuinely emotionally connected to the well-being of future generations, ensur- ing the potential development of transgenerational justice in principle.21 Berto F (2011) Modal meinongianism and fiction: the best of three worlds. Philos Stud 152(3):313–334 Brock S (2007) Fiction, feelings and emotions. Philos Stud 132:211–242i Caplan B (2004) Creatures of fiction, myth and imagination. Am Philos Q 41:331–337 Carroll N (1990) The philosophy of horror, or paradoxes of the hearth. 6 Conclusions I started by analyzing six different situations describing emotional engagement (a–f). In all but one case (f), the situ- ation concerned an emotional response toward an object, although only in one case (a) could the object of the emo- tion be said to exist. I then adopted Object Theory to solve the so-called “paradox of fiction”, explaining how emotions can be considered genuine and rational even when directed toward nonexistents. Such a solution has also proven useful to avoid a possible paradox of the future, since future objects are essentially nonexistent – similarly to past, possible, and Both these theories have been criticized (Gallagher 2012; Zahavi 2014) for being too cognitive and theoretical, and not reflecting our relations with other people considered as embodied minds. Such a remark, even if undoubtedly rel- evant when existent individuals, which are embedded and embodied, are at stake, seems to be less urgent (or even to miss the mark) when nonexistent beings are concerned. And here it is nonexistent beings we are focusing on.i In the case of fictional, past, possible, and future beings— which are not physically and emotionally connected 3 The Paradox of the Future: Is it Rational to Feel Emotions for Future Generations? 83 fictional ones. Nonetheless, I have also emphasized the dif- ferences between future objects and nonexistent ones.f References Oxford University Press, Oxford Open Access This article is licensed under a Creative Commons Attri- bution 4.0 International License, which permits use, sharing, adapta- tion, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Gopnik A, Wellman HM (1994) The theory theory. In: Hirschfeld LA, Gelman SA (eds) Mapping the mind: domain specificity in cog- nition and culture. Cambridge University Press, Cambridge, pp 257–293 Hartz GA (1999) How we can be moved by Anna Karenina, green slime, and a red pony. Philosophy 74(290):557–578 Husserl E (1950) Cartesianische Meditationen und pariser Vorträge, In: Husserliana, vol I, Aia, Nijhoff f Kripke SA (1980) Naming and necessity. Blackwell, Oxford Kroon F (2003) Quantified Negative Existentials. Dialectica 57:149–164i Lamarque P (1981) How can we fear and pity fictions? Br J Aesth 21:291–304 Levinson J (1997) Emotion in response to art: a survey of the terrain. In: Hjort M, Laver S (eds) Emotion and the arts. Oxford Univer- sity Press, New York, pp 20–34i Lewis D (1978) Truth in fiction. Am Philos Q 15:37–46 i Lipps T (1979) Empathy, Inner imitation and sense-feelings. A mod- ern book of esthetics. Holt, Rinehart and Winston, New York, pp 374–382 1 3 84 C. Barbero Russell B (1903) The principles of mathematics. Cambridge University Press, Cambridge Maibom H (2017) The routledge handbook of philosophy of empathy. Routledge, London g Matravers D (1998) Art and emotion. Oxford University Press, Oxford Salmon N (1999) Nonexistence. Noûs 32:277–319 Meinong A (ed) (1904) Über Gegenstandstheorie. In: Untersuchun- gen zur Gegenstandtheorie und Psychologie, Leipzig, Barth, in Gesammelte Abhandlungen, Gesamtausgabe bd. II, Akademische Druck- und Verlagsanstalt, Graz, 1971, pp 481–535; transl. by I. Levi, D.B.Terrell, and R.M. Chisholm, in R. References Chisholm (ed), Realism and the Background of Phenomenology, New York, Free Press, 1960, pp 76–117 Scarantino A (2016) The philosophy of emotions and its impact on affective science. In: Feldman Barrett L, Lewis M, Haviland-Jones JM (eds) Handbook of emotions, 4th edn. Guilford Press, New York, pp 3–48 Scheler M (1954) The nature of sympathy. Routledge & Kegan Paul, London Smith GM (2003) Film structure and the emotion system. Cambridge University Press, Cambridge Moran R (1994) The expression of feeling in imagination. Philos Rev 103:75–106 Solomon RC (2008) The philosophy of emotions. In: Lewis M, Havi- land-Jones JM, Feldman Barrett L (eds) Handbook of emotions, 3rd edn. Guilford Press, New York, pp 3–16i Morreal J (1993) Fear without Belief. J Philos 90:359–366 Nef F (1998) L’objet quelconque. Recherches sur l’ontologie de l’objet. Vrin, Paris Stecker R (2011) Should we still care about the paradox of fiction? Br J Aesthet 51(3):295–308 ll A (1993) Fiction and the emotions. Am Philos Q 30:1–1 Neill A (1993) Fiction and the emotions. Am Philos Q 30:1–13 O ili F (2002) Uli il d d l’ l di F i Stein E (1989) On the problem of empathy. ICS Publishers, Washington Orilia F (2002) Ulisse, il quadrato rotondo e l’attuale re di Francia. Edizioni ETS, Pisa ( ) p p y g Thomasson AL (1999) Fiction and metaphysics. Cambridge University Press, Cambridge Parsons T (1980) Nonexistent objects. Yale University Press, New Haven Thompson T (1995) The beast: a reckoning with depression. Putnam, New Yorkii Perrson P (2003) Understanding cinema: a psychological theory of moving imagery. Cambridge University Press, Cambridgeif Voltolini A (2006) How ficta follow fiction. A syncretistic account of fictional entities. Springer, Dordrechti Petraschka T (2021) How empathy with fictional characters differs from empathy with real person. J Aesthet Art Critic 79(2):227–232i i Walton K (1978) Fearing fictions. J Philos 75:5–27 Plantinga C (2009) Moving viewers: American film and the spectator’s experience. University of California Press, Oakland i Walton K (1990) Mimesis as make-believe: on the foundations of the representational arts. Harvard University Press, Cambridge Wilson C (2013) Grief and the poet. Br J Aesthet 53(1):77–91 Prandtl A (1910) Die Einfühlung. Verlag von Johann Ambrosius Barth, Leipzig Zahavi D (2014) Self and other. Exploring subjectivity, empathy, and shame. Oxford University Press, Oxford Prinz J (2004) Gut reactions: a perceptual theory of emotion. References Oxford University Press, Oxford Zalta E (1983) Abstract objects: an introduction to axiomatic meta- physics. D. Reidel, Dordrecht Radford C (1975) How can we be moved by the fate of Anna Karenina?, In: Proceedings of the Aristotelian Society, vol. 49, pp. 67–80 Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Raspa V (ed) (2002) Teoria dell’oggetto, Parnaso, Trieste 2000f p gg Ratcliffe M (2008) Feelings of being: phenomenology, psychiatry and the sense of reality. Oxford University Press, Oxford Robinson J (2010) Emotion and the understanding of narrative. In: Hagberg GL, Jost W (eds) Companion to the philosophy of lit- erature. Blackwell, Malden, pp 71–92 1 3 3
https://openalex.org/W2075652796	https://zenodo.org/records/2257904/files/article.pdf	German	null	Studien über Styphninsäure	Berichte der Deutschen Chemischen Gesellschaft	1,908	public-domain	1,145	3939 3939 3939 Der Ather bilclet gelbe Nadeln, die in der Warme gut 1011 Alkohol, Benzol, Ather. Aceton. schwer 3-011 Ligroin und kaum 1-011 Tl-asser ge- 16st werden. 0.1286 g Sbst.: 0.2501 g CO,, 0.0396 g HzO. 0.1286 g Sbst.: 0.2501 g CO,, 0.0396 g HzO. g ,, g C11Hg05N2. Ber. C 53.21, H 3.24. huf diegleiche Weise wirde auch der 2.4-Dinitro-a-naphthol- a t h y l a t h e r gewonnen. Er schmilzt liei 90° iind besitzt die ron huf diegleiche Weise wirde auch der 2.4-Dinitro-a-naphthol- a t h y l a t h e r gewonnen. Er schmilzt liei 90° iind besitzt die ron C. A. M a r t i u s '), sowie von H e e r m a n n ?) angegebenen Eigenschnften. 0.1274 g Sbst.: 11.5 ccm N (IS0, 754 mm). C. A. M a r t i u s '), sowie von H e e r m a n n ?) angegebenen Eigenschnften. 0.1274 g Sbst.: 11.5 ccm N (IS0, 754 mm). ClaHloNz05. Ber. N 10.69. Gef. N 10..51. ClaHloNz05. Ber. N 10.69. Gef. N 10..51. 3) Diese Berichte 41, ISi.5 [19OS]. 4) Das Trinitro-resorcin verbindet Cich niitnur 1Mol. Disthylanilin. Das in alkoholischer Lijsung gevonnene Salz bildet gelbe Nadeln, die bei 159'' I) Ztschr. f. Cheni. 1868, S'i. ?) Jonm. f. prakt. Chem. p2] 44, 541 [1891]. 3) Diese Berichte 41, ISi.5 [19OS]. 614. Fritz Ullmann und Walter Bruck: Studien uber Styphninsaure. 614. Fritz Ullmann und Walter Bruck: Studien uber Styphninsaure. yp (3. Mitteilung.) yp (3. Mitteilung.) [Blitted. aus den1 Techn -cheni. Instit. der Kgl. Techn. Hochschule zu Berlin.] (Eingegangen an1 8. November 1908.) [Blitted. aus den1 Techn -cheni. Instit. der Kgl. Techn. Hochschule zu Berlin.] (Eingegangen an1 8. November 1908.) Aus Pikrinsanre uud Trinitrokreaol bildet sich beim Behandeln mit Toluolsulfochlorid -Diiithqlanilin ausschlieSlich das entsprechende Chl~rderivat~). Wir hofften am der S t y p h n i n s l u r e , dem Trinitrore- sorcin , auf gleiche Weise das Dichlortrinitrobenzol zu erhalten. Die cliesbezuglichen Versuche zeigten jedoch. da8 bei der Wechselwirkuug der drei Substanzen nur das E i t h j lanilinsalz des Trinitroresorcin- tolnolsnlfesters entsteht. Es gelang uns nicht, die noch freie Hydroxyl- gruppe zu verestern. Auch dns Diithylanilin ist in der Verbindring sehr fest gebunden, die Substanz lBBt sich z. B. RUS Eisessig umliry- stallisieren. Ilurch Kochen mit verdunuter Snlzsaure gelang es nicht, das Diathylanilin zu entfernen ) da hierbei Zeraetzung tinter Bilduug ron l'uluolsulfochlorid und Styphninslnre eintrat. 1) i a t 11 y 1 an i 1 in - S a lz 4, d e s T r i n i t r o r e so r c i n - t o 111 o 1,ti 1 f e s t e r s, 0 . SOZ. C 7 H; 1) i a t 11 y 1 an i 1 in - S a lz 4, d e s T r i n i t r o r e so r c i n - t o 111 o 1,ti 1 f e s t e r s, 0 . SOZ. C 7 H; 1) i a t 11 y 1 an i 1 in - S a lz 4, d e s T r i n i t r o r e so r c i n - t o 111 o 1,ti 1 f e s t e r s, 0 . SOZ. C 7 H; ; SO2 SO2 Von den vieleu Versiichen, die wir zur Herstellung dieaer Ver- bindnng aiisfiihrten, gab der folgende die best,en Ausbeuteil. 4) Das Trinitro-resorcin verbindet Cich niitnur 1Mol. Disthylanilin. Das in alkoholischer Lijsung gevonnene Salz bildet gelbe Nadeln, die bei 159'' schmelzen. gut yon Alkohol und Benzol und sehr leicht yon Aceton gelost. N (210, 759 mm). Sie sind unloslich in ;ither und Ligroin, werden in der Hitze 0.1299 R Sbst.: 0.2329 g COZ, 0.0562 8 HsO. - 0.1365 g Shst.: 16.6 cc'nl C16H18 OsN+ Ber. C 48.71, H 4.60, N 14.22. Gef. )) 48.89, x 4.83, y 14.10. 3940 12 g Styphninsiure wurden mit 30 g Diathylanilin uncl 19 g Toluolsnl- fochlorid miihrend 4 Stunden auf 800 errr5irmt. Zuerst tritt hesonders beim Riihren vollige I h u n g ein, und nach einiger Zeit erstarrte dann die Masse zu einem Krystallbrei. Die Slasse wird mit 40 ccm Slkohol ausgekocht, nach dem Erkalten filtriert und rnit wenig Alkohol gewaschen. Hierbei \wrdeii 22.8 g (840io der Theorie) an Ester erhalten. g 'Das Rohprodulrt ist so gut \vie rein. Es schmilzt Lei 164O unter Zersetznug nnd behHlt iinch deni Undiisen ails Eisessig den Ychmelz- pnnkt bei. Es bilclet gelbe Krystallblattchen, die unloslich in Ligroin, schwer in heiBeni Alkohol iind Benzol, gut in Aceton und Essigsaure liislich sind. I?urch Kocben mit Alkohol und SnlzsSiure entsteht Styph- ninsaure. BaS04. 0.1303 g Sbst.: 0.2413 R COs, 0.0536 g HsO. - 0.1517 g Sbst.: 0.0644 R C~~H~140ioiT~S. Bey. C 50.34, H 4.41, S 5.84. Gef. )) 50.51, )> 4.60, I 5.83. UbergieBt man obigen Estelr (20 g) mit der gleichen hIenge Aniliu, so farbt sich die Masse rot, erwarmt sich stark und wird flussig, u n nach kurzer Zeit zu einem Krystallbrei zu erstarren. J3eim h i s - kochen rnit verdunnter Salzsaure hinterblieben 11.6 g (1OO0/o) Osy- trinitrophenylamin. Das Rohprodulrt ist sehr rein; es schmilzt 1% O und nach dem Umlijsen aus verdunntem Eisessig bei 162O (korr.). Es bildet orangegelbe Nadeln, die kaum in Ather und Ligroin, leicht in Benzol nnd Aceton und spielend in Eisessig mit gelher Farbe loslich sind. Gleich gefarbt ist die Losung in konzentrierter Schwefelsiiure, wahrend Atznatroi rnit roter Farbe lost. Iiocht man clie alkalische Fliissigkeit, so tritt der Geruch von Anilin anf. g , 0.1230 g Sbst.: 0.2021 g COz, 0.0310 g H20. - 0.1285 g Sbst.: 19 ccni N (16O, 753 mm). ) C I ~ H S O ~ N ~ . Ber. C 44.98, H 2.51, N 17.51. Gef. )> 44.81, )) 2.51, )) 17.31. ) C I ~ H S O ~ N ~ . Ber. C 44.98, H 2.51, N 17.51. Gef. )> 44.81, )) 2.51, )) 17.31. schmelzen. gut yon Alkohol und Benzol und sehr leicht yon Aceton gelost. N (210, 759 mm). Sie sind unloslich in ;ither und Ligroin, werden in der Hitze 0.1299 R Sbst.: 0.2329 g COZ, 0.0562 8 HsO. - 0.1365 g Shst.: 16.6 cc'nl schmelzen. gut yon Alkohol und Benzol und sehr leicht yon Aceton gelost. 3940 Sie sind unloslich in ;ither und Ligroin, werden in der Hitze gut yon Alkohol und Benzol und sehr leicht yon Aceton gelost. N (210, 759 mm). 0.1299 R Sbst.: 0.2329 g COZ, 0.0562 8 HsO. - 0.1365 g Shst.: 16.6 cc'nl C16H18 OsN+ Ber. C 48.71, H 4.60, N 14.22. Gef. )) 48.89, x 4.83, y 14.10. C16H18 OsN+ Ber. C 48.71, H 4.60, N 14.22. Gef. )) 48.89, x 4.83, y 14.10.
https://openalex.org/W3163234789	https://www.frontiersin.org/articles/10.3389/fphar.2021.590201/pdf	English	null	β-Caryophyllene, A Natural Dietary CB2 Receptor Selective Cannabinoid can be a Candidate to Target the Trinity of Infection, Immunity, and Inflammation in COVID-19	Frontiers in pharmacology	2,021	cc-by	22,240	HYPOTHESIS AND THEORY HYPOTHESIS AND THEORY published: 14 May 2021 doi: 10.3389/fphar.2021.590201 Niraj Kumar Jha 1†, Charu Sharma 2†, Hebaallah Mamdouh Hashiesh 3, Seenipandi Arunachalam 3, MF Nagoor Meeran 3, Hayate Javed 4, Chandragouda R. Patil 5, Sameer N. Goyal 6 and Shreesh Ojha 3* 1Department of Biotechnology, School of Engineering & Technology (SET), Sharda University, Greater Noida, India, 2Department of Internal Medicine, College of Medicine and Health Sciences, United Arab Emirates University, Al Ain, United Arab Emirates, 3Department of Pharmacology and Therapeutics, College of Medicine and Health Sciences, United Arab Emirates University, Al Ain, United Arab Emirates, 4Department of Anatomy, College of Medicine and Health Sciences, United Arab Emirates University, Al Ain, United Arab Emirates, 5Department of Pharmacology, Delhi Pharmaceutical Sciences and Research University, New Delhi, India, 6Shri Vile Parle Kelvani Mandal’s Institute of Pharmacy, Dhule, India Correspondence: Shreesh Ojha shreeshojha@uaeu.ac.ae †These authors have contributed equally to this work †These authors have contributed equally to this work Specialty section: This article was submitted to Ethnopharmacology, a section of the journal Frontiers in Pharmacology Received: 31 July 2020 Accepted: 04 February 2021 Published: 14 May 2021 Citation: Jha NK, Sharma C, Hashiesh HM, Arunachalam S, Meeran MFN, Javed H, Patil CR, Goyal SN and Ojha S (2021) β-Caryophyllene, A Natural Dietary CB2 Receptor Selective Cannabinoid can be a Candidate to Target the Trinity of Infection, Immunity, and Inﬂammation in COVID-19. Front. Pharmacol. 12:590201. doi: 10.3389/fphar.2021.590201 Specialty section: This article was submitted to Ethnopharmacology, a section of the journal Frontiers in Pharmacology Specialty section: This article was submitted to Ethnopharmacology, a section of the journal Frontiers in Pharmacology Specialty section: This article was submitted to Ethnopharmacology, a section of the journal Frontiers in Pharmacology Received: 31 July 2020 Accepted: 04 February 2021 Published: 14 May 2021 Received: 31 July 2020 Accepted: 04 February 2021 Published: 14 May 2021 Reviewed by: Reviewed by: Jen-Tsung Chen, National University of Kaohsiung, Taiwan Kuldeep Dhama, Indian Veterinary Research Institute (IVRI), India Reviewed by: Jen-Tsung Chen, National University of Kaohsiung, Taiwan Kuldeep Dhama, Coronavirus disease (COVID-19), caused by novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is an ongoing pandemic and presents a public health emergency. It has affected millions of people and continues to affect more, despite tremendous social preventive measures. Identifying candidate drugs for the prevention and treatment of COVID-19 is crucial. The pathogenesis and the complications with advanced infection mainly involve an immune-inﬂammatory cascade. Therefore, therapeutic strategy relies on suppressing infectivity and inﬂammation, along with immune modulation. One of the most promising therapeutic targets for the modulation of immune-inﬂammatory responses is the endocannabinoid system, particularly the activation of cannabinoid type 2 receptors (CB2R), a G-protein coupled receptor which mediates the anti-inﬂammatory properties by modulating numerous signaling pathways. To pharmacologically activate the CB2 receptors, a naturally occurring cannabinoid ligand, beta-caryophyllene (BCP), received attention due to its potent anti-inﬂammatory, antiviral, and immunomodulatory properties. BCP is recognized as a full selective functional agonist on CB2 receptors and produces therapeutic effects by activating CB2 and the nuclear receptors, peroxisome proliferator-activated receptors (PPARs). BCP is regarded as the ﬁrst dietary cannabinoid with abundant presence across cannabis and non-cannabis plants, including spices and other edible plants. BCP showed tissue protective properties and favorably modulates numerous signaling pathways and inhibits inﬂammatory mediators, including cytokines, chemokines, adhesion molecules, prostanoids, and eicosanoids. Based on its pharmacological properties, molecular mechanisms, and the therapeutic potential of BCP as an immunomodulator, anti-inﬂammatory, organ- protective, and antiviral, we hypothesize that BCP could be a promising therapeutic and/or preventive candidate to target the triad of infection, immunity, and inﬂammation in Correspondence: Shreesh Ojha shreeshojha@uaeu.ac.ae β-Caryophyllene, A Natural Dietary CB2 Receptor Selective Cannabinoid can be a Candidate to Target the Trinity of Infection, Immunity, and Inﬂammation in COVID-19 Niraj Kumar Jha 1†, Charu Sharma 2†, Hebaallah Mamdouh Hashiesh 3, Seenipandi Arunachalam 3, MF Nagoor Meeran 3, Hayate Javed 4, Chandragouda R. Patil 5, Sameer N. Goyal 6 and Shreesh Ojha 3* Niraj Kumar Jha 1†, Charu Sharma 2†, Hebaallah Mamdouh Hashiesh 3, Seenipandi Arunachalam 3, MF Nagoor Meeran 3, Hayate Javed 4, Chandragouda R. Patil 5, Sameer N. Goyal 6 and Shreesh Ojha 3* Edited by: Javier Echeverria, University of Santiago, Chile INTRODUCTION viral infection and immune-inﬂammatory axis (Bahramsoltani and Rahimi, 2020; Basu et al., 2020; Benarba and Pandiella, 2020; Hensel et al., 2020; Mondal et al., 2020; Narkhede et al., 2020). Over the past few months it has been suggested that natural products hold great promise in the management of COVID-19 due to their antiviral, anti-inﬂammatory, and immunomodulator activities (Bahramsoltani and Rahimi, 2020; Basu et al., 2020; Benarba and Pandiella, 2020; Mondal et al., 2020; Narkhede et al., 2020). Thus, identifying candidate compounds which have selectivity against viral components as well as prevent viral entry, enhance immunity and attenuate inﬂammatory factors in host could be important in context to COVID-19. COVID-19, a public health emergency and pandemic, has affected millions of people worldwide and continues to do so, despite numerous preventive measures, and this situation will continue until a vaccine is developed (Huang et al., 2020). The severity of infection varies from patients being asymptomatic to pre-symptomatic to symptomatic with different stages of illness, ranging from mild, moderate, to severe (Yang et al., 2020). The symptoms include fever, dry cough, sore throat, diarrhea, rashes on the skin, face, or toes, shortness of breath, loss of smell, anorexia, fatigue, headache, myalgia, anosmia, and ageusia, identiﬁed as the clinical criteria for diagnosis of COVID-19 (Jin et al., 2020). The majority of deaths are happening due to complications, such as severe pneumonia, acute respiratory distress syndrome (ARDS), shock, sepsis, and resultant multi- organ failure (Huang et al., 2020; Yang et al., 2020). The pathogenesis of COVID-19 emerges as a multifaceted, multi- system, multi-organ disorder, including viremia to overt the activation of immune responses and inﬂammatory processes that result in a dysregulated immune pattern, manifested by a massive rise in the levels of pro-inﬂammatory cytokines, chemokines, and adhesion molecules (Dhama et al., 2020). This causes the onset of a “cytokine storm” or “cytokine release syndrome”, which mainly causes ARDS and further leads to pathogenic effects through a quite ubiquitous target at a multiple-organ level (Dhama et al., 2020; Tang et al., 2020; Vinciguerra et al., 2020). p Many propositions have been made on the possible therapeutic potential of essential oils-derived phytochemicals, including many terpenes or terpeno-alcoholic compounds, in COVID-19 (Asif et al., 2020; Boukhatem and Setzer, 2020; da Silva et al., 2020; Diniz et al., 2021). Citation: Jha NK, Sharma C, Hashiesh HM, Arunachalam S, Meeran MFN, Javed H, Patil CR, Goyal SN and Ojha S (2021) β-Caryophyllene, A Natural Dietary CB2 Receptor Selective Cannabinoid can be a Candidate to Target the Trinity of Infection, Immunity, and Inﬂammation in COVID-19. Front. Pharmacol. 12:590201. doi: 10.3389/fphar.2021.590201 May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 1 Jha et al. Beta-Caryophyllene may be Useful in COVID-19 COVID-19. In line with numerous studies that proposed the potential of cannabinoids in COVID-19, BCP may be a novel candidate compound for pharmaceutical and nutraceutical development due to its unique functional receptor selectivity, wide availability and accessibility, dietary bioavailability, nonpsychoactivity, and negligible toxicity along with druggable properties, including favorable pharmacokinetic and physicochemical properties. Based on reasonable pharmacological mechanisms and therapeutic properties, we speculate that BCP has potential to be investigated against COVID-19 and will inspire further preclinical and clinical studies. Keywords: COVID-19, SARS-CoV-2, beta-caryophyllene, immunomodulators, natural products Keywords: COVID-19, SARS-CoV-2, beta-caryophyllene, immunomodulators, natural products INTRODUCTION TNF-α, tumor necrosis factor alpha; IL, interleukin; PGE-2, prostaglandin E2; NO, nitric oxide; CINC-1, cytokine-induced neutrophil chemoattractant 1; NF-κB, nuclear factor kappa B; IFN-γ, interferon gamma; COX-2, cyclooxygenase-2; VCAM, vascular cell adhesion protein; iNOS, inducible nitric oxide synthase; GPx, glutathione peroxidase; SOD, superoxide dismutase; PPAR-γ, peroxisome proliferator-activated receptor gamma; PGC1-α, peroxisome proliferator-activated receptor gamma coactivator 1-alpha; Nrf2, nuclear factor erythroid 2–related factor 2; FAS, fatty acid synthase; ATGL, adipose triglyceride lipase; ROS, reactive oxygen species; Bax, Bcl-2 associated X protein; Bcl-2, B-cell lymphoma 2; ART1, arginine ADP-ribosyltransferase 1; NFTs, neuroﬁbrillary tangles; Aβ, amyloid beta; CB2R, cannabinoid receptor type 2. FIGURE 1 \| The structure and various polypharmacological properties and therapeutic potential of BCP. TNF-α, tumor necrosis factor alpha; IL, interleukin; PGE-2, prostaglandin E2; NO, nitric oxide; CINC-1, cytokine-induced neutrophil chemoattractant 1; NF-κB, nuclear factor kappa B; IFN-γ, interferon gamma; COX-2, cyclooxygenase-2; VCAM, vascular cell adhesion protein; iNOS, inducible nitric oxide synthase; GPx, glutathione peroxidase; SOD, superoxide dismutase; PPAR-γ, peroxisome proliferator-activated receptor gamma; PGC1-α, peroxisome proliferator-activated receptor gamma coactivator 1-alpha; Nrf2, nuclear factor erythroid 2–related factor 2; FAS, fatty acid synthase; ATGL, adipose triglyceride lipase; ROS, reactive oxygen species; Bax, Bcl-2 associated X protein; Bcl-2, B-cell lymphoma 2; ART1, arginine ADP-ribosyltransferase 1; NFTs, neuroﬁbrillary tangles; Aβ, amyloid beta; CB2R, cannabinoid receptor type 2. additive or preservative, and for aroma in food products and beverages (Gertsch, 2008; Gertsch et al., 2008). BCP is one of the constituents of commonly consumed edible plants, such as cinnamon (Cinnamomum spp.), basil (Ocimum spp.), pepper (Piper spp.), breakfast mint [Perilla frutescens (L.) Britton], coriander (Coriandrum sativum L.), chestnut (Aesculus hippocastanum L.), sage (Salvia ofﬁcinalis L.), cubeb pepper (Piper cubeba L.f.), thyme (Thymus vulgaris L.), myrrh [Myrrhis odorata (L.) Scop.], curry leaves [Murraya koenigii (L.) Spreng.], hops (Humulus lupulus L.), cloves [Syzygium aromaticum (L.) Merr. & L.M. Perry], hemp (Cannabis sativa L.), lavender (Lavandula angustifolia Mill.), oregano (Origanum vulgare L.), and rosemary (Rosmarinus ofﬁcinalis L.), among others. Recently, a majority of the plant derived compounds showed potential in COVID-19 due to their antioxidant and anti- inﬂammatory properties and are of limited occurrence in certain comparison with individual fractions from cannabis (Anil et al., 2021). INTRODUCTION Many of the terpene components present in cannabis are widely consumed in food and used in traditional medicine (Anil et al., 2021). Some of these compounds showed potential to modulate the endocannabinoid system, which represents one of the newest therapeutic targets in regard to regulation of innate and adaptive immunity and immunomodulatory and anti-inﬂammatory properties. The endocannabinoid system is targeted by plant-derived compounds, termed phytocannabinoids, which have gained attention for therapeutic modulation of cannabinoid type-1 receptors (CB1R) and type-2 (CB2R), the components of endocannabinoid system (Oláh et al., 2017). The latest therapeutic strategy in targeting the endocannabinoid system is to activate the CB2R, a G-protein coupled receptor which appears to regulate immunity, inﬂammation, and pain. The activation of CB2R has been shown to exert potent anti- inﬂammatory, immunomodulatory, and organ-protective properties with no psychotropic effects, which are commonly observed with CB1R. Over the past few months, it has been suggested that modulation of the endocannabinoid system by cannabinoids, including cannabidiol, could be useful in prophylaxis and treatment of COVID-19 and may improve prognosis (Costiniuk and Jenabian, 2020; Esposito et al., 2020). Recently, extract of Cannabis sativa containing phytocannabinoids and terpenes were shown to modulate the inﬂammatory mediators in alveolar epithelial cells (A549) in COVID-19-associated inﬂammation and suggested that the phytocannabinoid mix formulation exerted better activity in At present, many drugs are being repurposed for supportive management in COVID-19 based on docking studies, pharmacological rationale, and clinical experiences (Jean and Hsueh, 2020; Kandeel and Al-Nazawi, 2020; Rabaan et al., 2020; Singh et al., 2020). The pathogenesis and the complications developed with the infection mainly involve an immune-inﬂammatory cascade; therefore, the therapeutic strategies focus on reducing inﬂammation and immune modulation of this cascade (Dhama et al., 2020; García, 2020; Scavone et al., 2020; Zhou et al., 2020). Despite recent availability of vaccine for prophylaxis, massive efforts are ongoing for the discovery of novel drugs for the treatment and prevention of COVID-19 (Jean and Hsueh, 2020; Kandeel and Al-Nazawi, 2020; Rabaan et al., 2020; Singh et al., 2020). In parallel with repurposing modern medicines, there are numerous attempts to explore natural products with potential to target the interplay of May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 2 Jha et al. Beta-Caryophyllene may be Useful in COVID-19 FIGURE 1 \| The structure and various polypharmacological properties and therapeutic potential of BCP. INTRODUCTION Many cannabinoids, including cannabidiol, have been suggested for their possible potential as preventive agents or therapeutic adjuvants with other agents in targeting the trinity of infection, inﬂammation, and immunity in COVID-19 (Byrareddy and Mohan, 2020; Costiniuk and Jenabian, 2020; Esposito et al., 2020; Nagarkatti et al., 2020; Sexton, 2020; Raj et al., 2021). Among numerous cannabinoids, beta-caryophyllene (β-Caryophyllene; BCP), a naturally occurring terpene, has received enormous attention in the past few years due to its recognition as a full functional agonist on CB2R which imparts its therapeutic potential by mediating anti-inﬂammatory and immunomodulatory properties (Gertsch et al., 2008). BCP, chemically known as trans-(1R,9S)-8-Methylene-4,11,1 is the ﬁrst dietary cannabinoid of natural origin, with an abundant presence in a variety of spice blends and citrus ﬂavors, as an May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 3 Beta-Caryophyllene may be Useful in COVID-19 Jha et al. genera and species or to a speciﬁc individual plant that may limit supply and demand. However, BCP is unique in terms of wide dietary availability and accessibility across numerous plant genera and species (Sharma et al., 2016). Till date, the presence of BCP has been conﬁrmed in more than two thousand plants, including edible, medicinal, and ornamental plants. BCP is mainly synthesized by plants as a defense mechanism against insects and aphids, and plays a role in pollination. It is usually localized in the aerial parts of the plants including leaves, ﬂowers, spate, inﬂorescence, and buds, with a low presence in the stem, roots, and rhizomes (Sharma et al., 2016). The structure and various polypharmacological properties and therapeutic potential of BCP are depicted in Figure 1. cytokine storm, which is a major reason for death in COVID- 19. The potential of BCP in improving host cellular immunity against infection and its good antiviral and antibacterial activity, along with the antioxidant effects, may further help in controlling the symptoms and the worsening of the disease, secondary infections, complications, progression, and resultant death. BCP has potential to protect from the risk factors, prevent the entry of the virus, and ameliorate organ damage and the pathological manifestation of SARS-CoV-2 on the different organ systems. A scheme of the effect of BCP mediating CB2R activation has been proposed in context of infection, inﬂammation, and immunity in COVID-19 (Figure 2). INTRODUCTION The literature reviewed herein indicates that BCP may be a promising candidate as a preventive and therapeutic agent or adjuvant for COVID-19 given its pharmacological and molecular mechanisms, including its CB2R agonist property, integrating with its antiviral, anti-inﬂammatory, and immunomodulatory properties in numerous experimental studies (Sharma et al., 2016). However, no study has yet directly demonstrated the efﬁcacy of BCP against SARS-CoV-2 infections. But, based on pharmacological properties, a logical approach has been presented on the therapeutic potential of BCP in COVID-19. In a recent molecular docking study, 171 components, including BCP, present in the essential oils of numerous plants were analyzed against SARS-CoV-2 main protease (SARS-CoV- 2 Mpro), SARS-CoV-2 endoribonucleoase (SARS-CoV-2 Nsp15/ NendoU), SARS-CoV-2 ADP-ribose-1″-phosphatase (SARS- CoV-2 ADRP), SARS-CoV-2 RNA-dependent RNA polymerase (SARS-CoV-2 RdRp), the binding domain of the SARS-CoV-2 spike protein (SARS-CoV-2 rS), and human angiotensin−converting enzyme (hACE2) (da Silva et al., 2020). Very recently in an in silico study, BCP was shown to target Mpro (3CLpro), the main protease in SARS-CoV-2 involved in the processing of translating the viral RNA into the viral polyproteins (Narkhede et al., 2020). BCP interacted with the amino acid residues of SARS-CoV-2 via pie-alkyl interactions and showed good afﬁnity along with druggable properties (Narkhede et al., 2020). In another recent study, Muthuramalingam et al., 2020 (Muthuramalingam et al., 2020) carried out a cheminformatics and interactome study using in silico approaches and found that BCP is one of the potential compounds among 259 phytochemicals screened for targeting thirteen COVID-19 immune genes regulating numerous signaling pathways. The study unveiled that 154 compounds interact with COVID-19-associated immune genes. BCP and its derivative, β-caryophyllene oxide, was found to target immune genes, and was suggested useful for designing and developing as a potential agent against COVID-19 (Muthuramalingam et al., 2020). BCP AS A FUNCTIONAL CB2 RECEPTOR AGONIST Gertsch and colleagues ﬁrst recognized BCP as a functional CB2R agonist using numerous model systems, including in silico, in vitro, and in vivo studies (Gertsch et al., 2008). In the molecular docking studies, BCP was observed to interact with CB2R on the same binding sites as that of CP55, 940, a CB2R agonist. It binds well in a hydrophobic sac involving lipophilic amino acid residues and it was suggested that the double bond with conformation E of BCP is vital for the receptor binding (Gertsch et al., 2008). Accumulating experimental studies have demonstrated the CB2R activation mediated effects of BCP in attenuating inﬂammation, oxidative stress, apoptosis, ﬁbrosis, and immune modulation. The CB2R-dependent anti- inﬂammatory mechanism of BCP has been demonstrated in oral mucositis (Picciolo et al., 2020), glioblastoma (Irrera et al., 2020), neuropathic pain (Klauke et al., 2014; Aly et al., 2019), bipolar disorders (Hwang et al., 2020), wound healing (Koyama et al., 2019), interstitial cystitis (Berger et al., 2019), autoimmune encephalomyelitis/multiple sclerosis (Alberti et al., 2017; Askari et al., 2019), neurocognitive disorders (Lindsey et al., 2019; Chávez-Hurtado et al., 2020), arthritis (Irrera et al., 2019), metabolic and neurobehavioral alterations (Youssef et al., 2019), insulin resistance and vascular inﬂammation (Youssef et al., 2019), hyperglycemia (Basha and Sankaranarayanan, 2016), peripheral neuropathy (Segat et al., 2017), atherosclerosis (Zhang et al., 2017), cardiotoxicity (Meeran et al., 2019), osteoporosis (Shan et al., 2017), vascular dementia (Lou et al., 2017), dopaminergic neurodegeneration/ Parkinson’s disease (Javed et al., 2016), Alzheimer’s disease (Cheng et al., 2014), cerebral ischemia-reperfusion (Poddighe et al., 2018), liver ﬁbrosis (Mahmoud et al., 2014), pulmonary The present review scientiﬁcally contemplates the therapeutic prospects of BCP in COVID-19. The possibilities of BCP as a candidate in COVID-19 have been discussed based on reported ﬁndings, particularly immunomodulatory, anti-inﬂammatory, and antiviral properties. Additionally, CB2R activation has been suggested as a possible therapeutic target in COVID-19. Based on the role of CB2R in immune-inﬂammatory mechanisms, we hypothesized that BCP endowed with CB2R agonist properties may potentially limit the severity and progression of COVID-19 by modulating infection, immunity, and inﬂammation. BCP AS A FUNCTIONAL CB2 RECEPTOR AGONIST The potent anti-inﬂammatory activity mediating multiple pathways and mediators of inﬂammation, including the inhibition of pro-inﬂammatory cytokines, chemokines, and adhesion molecules, along with the suppression of macrophage inﬁltration and neutrophil- endothelial cell interaction, might constitute a promising pharmacological and nutritional approach to inhibit the May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 4 Jha et al. Beta-Caryophyllene may be Useful in COVID-19 FIGURE 2 \| The proposed possible mechanisms and potential of BCP in COVID-19. FIGURE 2 \| The proposed possible mechanisms and potential of BCP in COVID-19. (Ciavarella et al., 2020). PPAR-γ agonists were shown to inhibit the replication of numerous viruses, including human immunodeﬁciency virus, respiratory syncytial virus, hepatitis B, and hepatitis C viruses (Skolnik et al., 2002; Du et al., 2017). Further, PPAR-γ agonists have been shown to reduce morbidity and mortality in inﬂuenza A virus infections (Bassaganya-Riera et al., 2010). The activation of PPAR-γ in resident alveolar macrophages was reported to signiﬁcantly ameliorate pulmonary inﬂammation and enhance host recovery following respiratory viral infections (Huang et al., 2019). Following amelioration of the tissue damage, PPAR-γ activation also controls the overproduction of cytokines. Thus, BCP may pause the onset of the cytokine storm from resident macrophages. (Ciavarella et al., 2020). PPAR-γ agonists were shown to inhibit the replication of numerous viruses, including human immunodeﬁciency virus, respiratory syncytial virus, hepatitis B, and hepatitis C viruses (Skolnik et al., 2002; Du et al., 2017). Further, PPAR-γ agonists have been shown to reduce morbidity and mortality in inﬂuenza A virus infections (Bassaganya-Riera et al., 2010). The activation of PPAR-γ in resident alveolar macrophages was reported to signiﬁcantly ameliorate pulmonary inﬂammation and enhance host recovery following respiratory viral infections (Huang et al., 2019). Following amelioration of the tissue damage, PPAR-γ activation also controls the overproduction of cytokines. Thus, BCP may pause the onset of the cytokine storm from resident macrophages. In addition to activation of CB2R, BCP also activates PPAR-α which favorably modulates the lipid metabolism by increasing the ability of hormone nuclear receptors PPAR-α and estrogen- related receptor α (ERRα) to drive the transcription of fatty acid oxidation enzymes by increasing the levels of peroxisome proliferator-activated receptor-gamma coactivator 1α (PGC-1α), as well as stimulating sirtuin 1 (SIRT1) deacetylase activity (Zheng et al., 2013; Wu et al., 2014). IMMUNOMODULATORY PROPERTIES OF BCP ( , ) Since the recognition of BCP as an agonist of CB2R, numerous studies have demonstrated the therapeutic beneﬁts of BCP by suppressing immune-inﬂammatory cascade when CB2R are activated. The activation of CB2R was reported to suppress lung pathology in infants infected with acute respiratory syncytial virus by reducing the levels of cytokines and chemokines (Tahamtan et al., 2018). In HIV patients, the activation of CB2R was shown to impair a productive infection and viral transmission involving a crosstalk/ interaction between CB2R (Costantino et al., 2012) and to inhibit the replication of the virus in monocytes and macrophages (Ramirez et al., 2013). Recently, BCP was shown to modulate systemic and local immunity in an experimental autoimmune encephalomyelitis model (Askari et al., 2019) and the immunomodulatory effect has been attributed to the ability of BCP to inhibit CD4+ and CD8+ T lymphocytes and pro- inﬂammatory cytokines (Alberti et al., 2017). The immunomodulatory activity of BCP was also explained by an enhanced phagocytic capability, following an increased lysosomal activity and nitric oxide production in macrophages (Carvalho et al., 2017). SARS-CoV-2 enters the host cells by binding to ACE2 receptors and the pathogen associated molecular patterns (PAMPs) on the virus alert innate immune cells, the anti-viral effectors, such as T CD8+ cells, NK cells, neutrophils, monocytes, and macrophages about the presence of the invading virus. The innate immune cells, which express pattern recognition receptors (PRRs), such as toll-like receptors (TLRs), retinoic acid-inducible gene I (RIG-I)- like receptors (RLRs), and nucleotide-binding and oligomerization domain (NOD)-like receptors (NLRs), detect PAMPs to achieve a suitable immune response against the invading pathogen (Keam et al., 2020). PRR and PAMP interaction triggers phagocytosis and stimulates the synthesis of pro-inﬂammatory cytokines, such as type I interferon, IFNα/β and type II, IFN-γ, and chemokines, such as CXCL-10 and CCL-2, to onset an antiviral environment (Allegra et al., 2020). In the case of severe infection, the viruses are sensed by monocytes, tissue macrophages, and resident dendritic cells, resulting in an uncontrolled pro-inﬂammatory cytokines (IFN, TNF-α, IL-1β, and IL-6) production, leading to a phenomenon called a “cytokine storm”, which damages the respiratory epithelial cells of the host (Allegra et al., 2020). The immune responses are critical for the eradication of the virus and the resolution of the active disease. BCP AS A FUNCTIONAL CB2 RECEPTOR AGONIST The role of sirtuin in the transcription and replication of viruses is well known and the activation of PPAR-α and lipolysis showed to reduce hepatitis C virus genotype-associated lipid metabolic disorder in liver inﬂammation (Andrade-Silva et al., 2016), intestinal inﬂammation (Bento et al., 2011), acute myocardial infarction (Younis and Mohamed, 2019), acute renal injury (Horváth et al., 2012), diabetic nephropathy (Li et al., 2020), and lipid disorders (Youssef et al., 2019). In the majority of the experimental models involving inﬂammatory states similar to those of human diseases, the principal pharmacological and molecular mechanism observed is the inhibition of pro-inﬂammatory cytokines, NF-κB, adhesion molecules, and chemokines and the subsequent modulation of signaling pathways, mainly involving toll-like receptors, opioid receptors, SIRT1/PGC-1α, AMPK/CREB, MAPK/ERK, Nrf2/ Keap1/HO-1, and the activation of nuclear peroxisome proliferator-activated receptors (PPARs). Cannabinoids are known to interact or crosstalk with a family of PPARs, including three subtypes: PPAR-α, PPAR-β/δ, and PPAR-γ. These subtypes are encoded by distinct genes and are regulated by steroids and lipid metabolites and mainly control lipid and glucose homeostasis and inﬂammatory responses (O’Sullivan, 2016). PPAR-γ agonists, pharmacologically known as thiazolidinediones, are clinically available drugs for use as insulin sensitizers in insulin resistance/type 2 diabetes mellitus. Recently, thiazolidinedione has been suggested for repurposing in COVID-19 due to its potential to attenuate cytokine storms In addition to activation of CB2R, BCP also activates PPAR-α which favorably modulates the lipid metabolism by increasing the ability of hormone nuclear receptors PPAR-α and estrogen- related receptor α (ERRα) to drive the transcription of fatty acid oxidation enzymes by increasing the levels of peroxisome proliferator-activated receptor-gamma coactivator 1α (PGC-1α), as well as stimulating sirtuin 1 (SIRT1) deacetylase activity (Zheng et al., 2013; Wu et al., 2014). The role of sirtuin in the transcription and replication of viruses is well known and the activation of PPAR-α and lipolysis showed to reduce hepatitis C virus genotype-associated lipid metabolic disorder in liver May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 5 Jha et al. Beta-Caryophyllene may be Useful in COVID-19 diseases (Patra et al., 2019). PPAR-α activation was also shown to beneﬁcially inﬂuence inﬂammation in alveolar epithelial cells, suggesting a potentially beneﬁcial role of PPAR-α in ARDS (Hecker et al., 2015). Thiazolidinediones have shown numerous adverse effects, such as weight gain, osteoporosis, heart failure, stroke, and an increased risk of urinary cancer. BCP AS A FUNCTIONAL CB2 RECEPTOR AGONIST Since BCP is natural, non-toxic, and devoid of the adverse effects of synthetic cannabinoids, it could be a safer alternative over synthetics. Together, the role of BCP as a PPAR-γ, as well as a PPAR-α agonist, seems promising in the regulation of the lipid and glucose metabolism, along with additional regulatory roles on cell proliferation and differentiation, vascular homeostasis, and inﬂammation, and the immune systems. Thus, BCP may be possibly useful to control the orchestrated immune- inﬂammatory events in COVID-19. CB2R has received enormous attention, since these novel therapeutic agents would have no psychotropic effects, as is the case with CB1R. Human CB2R were ﬁrst cloned in 1993 from the promyelocytic leukemia cell line HL-60 (Munro et al., 1993) and the ﬁrst CB2R-deﬁcient mouse was generated in 2000 (Buckley et al., 2000); therefore, CB2R represent the relatively newest therapeutic targets. Mice deﬁcient in CB2R showed an increased susceptibility and vulnerability to inﬂuenza infection, demonstrating that CB2R are important in immunoregulation in respiratory viral infections (Kapellos et al., 2019). The activation of CB2R exerted potent immunomodulation, mediating cell death induction, cytokine suppression, and inhibition of cell proliferation, along with the stimulation of regulatory T cell induction and anti-inﬂammatory cytokines (Rieder et al., 2010; Karmaus et al., 2012). However, few studies notably demonstrate that CB2R may modulate susceptibility to the experimental cerebral malaria through a CCL17-dependent mechanism (Alferink et al., 2016). Frontiers in Pharmacology \| www.frontiersin.org IMMUNOMODULATORY PROPERTIES OF BCP CB2R represents an important receptor target for immune regulation and is predominantly expressed by immune cells of the immune system, such as B cells, T cells, CD8+ lymphocytes, CD4+ lymphocytes, NK cells, neutrophils, macrophages, basophils, eosinophils, platelets, mast cells, dendritic cells, microglia, and astrocytes (Howlett and Abood, 2017). The CB2R are well expressed in several organs, including the liver, spleen, thymus, brain, lungs, kidneys, tonsils, nasal epithelium, and PBMC, which are present in the pancreas, uterus, and reproductive tissues (Cabral et al., 2015). Both cannabinoid receptors, CB1R and CB2R, play an important role in the modulation of the immune system, potentially inducing immunosuppression (Cabral et al., 2015; Hernández-Cervantes et al., 2017). The therapeutic targeting of Further, BCP exerted a potent immunomodulatory effect by simultaneously inhibiting both Th1 cytokines, including IL-2 and IFN-γ, and Th2 cytokines, including IL-4, IL-5, and IL-10, in primary splenocytes (Ku and Lin, 2013). Also, BCP is present in many plants, such as Chrysanthemum indicum L. (Hwang and Kim, 2013), Pterodonem arginatus Vogel (Alberti et al., 2014), Myracrodruon urundeuva Allemão (Carvalho et al., 2017), Schizonepeta tenuifolia (Benth.) Briq. (Ng et al., 2018), and copaiba oil (Urasaki et al., 2020), in which it exerts immunomodulatory activity. Taken together, the studies demonstrate that CB2R play a key role in balancing the immune response and BCP by activating CB2R, holding promise in the therapeutic modulation of immune- inﬂammatory changes in patients with SARS-CoV-2 infection. In COVID-19, the use of immunomodulators is receiving attention and being regarded as a “sub-etiological treatment” in the absence of an effective antiviral drug. Additionally, BCP May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 6 Jha et al. Beta-Caryophyllene may be Useful in COVID-19 FIGURE 3 \| The immunomodulatory mechanisms and organ-protective effects of BCP. FIGURE 3 \| The immunomodulatory mechanisms and organ-protective effects of BCP. (Allegra et al., 2020). Mounting evidence demonstrates that BCP exerts potent anti-inﬂammatory properties in all body organs, including the liver, kidneys, brain, heart, pancreas, and blood, and suppresses systemic inﬂammation by inhibiting proinﬂammatory cytokines in macrophages and other inﬂammatory mediators, as well as signaling pathways (Yamaguchi and Levy, 2020). BCP was shown to exhibit a CB2R-dependent anti-inﬂammatory property by inhibiting lipopolysaccharide/endotoxin (LPS)-induced phosphorylation of kinases ERK1/2 and JNK1/2 in macrophages, since it is recognized as a CB2R agonist and a dietary cannabinoid (Gertsch et al., 2008). IMMUNOMODULATORY PROPERTIES OF BCP Macrophages in the lungs express CB2R, which, upon further activation by a CB2R agonist, reduced the release of pro-inﬂammatory cytokines (such as IL-6) and angiogenic factors (Staiano et al., 2016). was shown to reduce ACE activity, which may be useful as ACE2 receptors, as the gateways of the virus entry, play a role from the entry of the virus to viremia and from respiratory distress to sepsis (Adefegha et al., 2017; Ajiboye et al., 2019). The immunomodulatory role of BCP is represented in Figure 3. Given the role of immunomodulators on the modulation of the hyperimmune-inﬂammatory response in COVID-19 patients, BCP may be a potential candidate for modulating immunity in patients at risk of infection and chronic metabolic and/or degenerative diseases, as well as in preventing the development and reducing the severity of COVID-19. ANTI-INFLAMMATORY PROPERTIES OF BCP Many pathways were shown to be responsible for the anti- inﬂammatory activity in macrophages, including inhibiting the Ras-MAPK pathway, JNK pathway, TNF-α translation, and the inhibition of proinﬂammatory cytokines, including TNF-α (Gertsch et al., 2008; Rajesh et al., 2008). TNF-α triggers the activation of Ras, p38 MAPK, ERK1/2, SAPK/JNK, HMGB1/ TLR4, and Akt pathways, and ultimately, the expression of proinﬂammatory cytokines, cellular proliferation, and migration (Gertsch et al., 2008; Rajesh et al., 2008). Additionally, numerous studies have also demonstrated the Cytokines regulate both inﬂammation and the immunopathology of viral infection. The massive production of proinﬂammatory cytokines is the key element that leads to an acute systemic hyperinﬂammatory state, to a cytokine storm syndrome, determining the intensity and severity of symptoms, eliciting the onset of acute respiratory distress, involving the extrapulmonary system, and increasing the risk of multiple organ failure and mortality during SARS-CoV-2 infection May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 7 Beta-Caryophyllene may be Useful in COVID-19 Jha et al. anti-inﬂammatory effects of BCP by activating CB2R and their subsequent pathways (Youssef et al., 2019). The CB2R-dependent anti-inﬂammatory effect of BCP has been demonstrated in inﬂammatory states of the heart (Meeran et al., 2019), liver (Cho et al., 2015; Arizuka et al., 2017; Varga et al., 2018), intestines (Cho et al., 2015), kidneys (Horváth et al., 2012; Hammad et al., 2018), lungs (Andrade-Silva et al., 2016), brain (Fontes et al., 2017; Yang et al., 2017; Askari et al., 2019; Askari and Shaﬁee-Nick, 2019), pancreas (Basha and Sankaranarayanan, 2016), urinary bladder (Berger et al., 2019), joints (Ruﬁno et al., 2015; Irrera et al., 2019; D’Ascola et al., 2019), skin (Koyama et al., 2019), oral cavity, and blood (Brito et al., 2019). BCP also showed anti-inﬂammatory effects mediating histaminergic and arachidonic acid pathways (Oliveira-Tintino et al., 2018). Though, evidence supports that CB2R activation has anti- inﬂammatory effects, it has yet to be targeted to treat human disease. should be assessed for speciﬁcity, afﬁnity, dose-response, and kinetics in experimental studies. The binding of these compounds limits the availability of the substrate, modiﬁes conﬁguration of active sites, and prevents dimerization, viral entry, and/or viral replication. ANTI-INFLAMMATORY PROPERTIES OF BCP The role of cannabinoids against virus replication, maturation, transmission, and entry in particular has been demonstrated in in silico and in vitro studies (Khodadadi et al., 2020; Mohammed et al., 2020; Salles et al., 2020; Wang et al., 2020; Raj et al., 2021). It has become apparent that agents which have antiviral properties corroborated with anti- inﬂammatory and immunomodulatory properties are important to target the trinity of infection, inﬂammation, and immunity in context of COVID-19. To tackle SARS-CoV-2, the identiﬁcation of viral protease appears as a striking therapeutic target to limit the replication of SARS-CoV-2 and many of the compounds are being investigated for their potential to target replication by inhibiting viral components such as Mpro (3CLpro), PLpro and spike proteins. The protease of SARS-CoV-2 emerged as an attractive target to inhibit the replication of the virus. Recently, BCP was shown to target SARS-CoV-2 virus via pie-alkyl interactions to PHE 294 of SARS-CoV-2 with an afﬁnity of -7.2 in an in silico docking study (Narkhede et al., 2020). BCP is present in many plants, such as Campomanesia phaea (O.Berg) Landrum (Lorençoni et al., 2020), Pterodon pubescens (Benth.) Benth. (Basting et al., 2019), Ocimumm icranthum Willd. (de Pinho et al., 2012), Mosla dianthera (Buch.-Ham. ex Roxb.) Maxim. (Wu et al., 2012), Cordia verbenacea A. DC. (Basting et al., 2019), Duguetia furfuracea (A.St.-Hil.) Saff. (Saldanha et al., 2019), Cinnamomum osmophloeum Kaneh. (Tung et al., 2008), Croton campestris A.St.-Hil., A. Juss. & Cambess. (Oliveira-Tintino et al., 2018), Pinus spp.(Basholli- Salihu et al., 2017), and Copaiba oil (Ames-Sibin et al., 2018), and has been considered responsible for their anti-inﬂammatory effects by suppressing proinﬂammatory cytokines and other inﬂammatory mediators. The antiviral properties of BCP or BCP-containing plants have been summarized in Table 1. The antiviral properties against herpes simplex virus type 1 (HSV-1) from the essential oils of many plants have been attributed to their chemical constituents, including BCP (Astani et al., 2011). The IC50 and TC50 for BCP were found to be 0.25 and 35 μg/ml, respectively (Astani et al., 2011). In plaque reduction assays, BCP exerted a concentration- dependent antiviral effect, with a selectivity index (ratio of TC50/ IC50) of 140. BCP showed 98% reduction in infectivity, comparable to acyclovir, a standard antiviral drug. ANTI-INFLAMMATORY PROPERTIES OF BCP The authors suggested that BCP has potential to inactivate the herpes virus and may affect the structure of the virion envelope, which is essential for adsorption or entry into the host cells (Astani et al., 2011). BCP was shown to inhibit both Herpes Simplex Virus-2 (HSV-2) and acyclovir-resistant strain infections with a similar or lower selectivity index, compared to the BCP-rich essential oil of Salvia desoleana Atzei & V. Picci. However, BCP was not found to inhibit HSV-1 infection (Loizzo et al., 2008; Cagno et al., 2017). The selectivity index value >1 suggests that the compound has inhibitory action on viral replication and has low cytotoxicity on the host cells, thus a high selectivity index demonstrates better action of the compound. In the absence of guidelines on acceptability or appropriateness of selectivity index, values greater than 10 are considered better candidates for antiviral actions. The extracts rich in BCP displayed a very high selectivity index that indicates potent antiviral activity with negligible cytotoxicity on the host cells. The therapeutic efﬁcacy and safety may also have different implications and should be taken in account considering the severity of viral infections and its onset, whether acute or chronic. BCP was tested in cell-based assays and showed a selectivity index of 71.1 in inhibiting the replication of the dengue virus (DENV- 2). BCP acts as a viricidal by interfering with the very early steps of the viral replication cycle and in silico data showed that BCP Frontiers in Pharmacology \| www.frontiersin.org ANTIVIRAL PROPERTIES OF BCP The role of plant-based natural products are well explored for their antiviral properties and are gaining attention for their therapeutic potential in COVID-19 (Mahmud et al., 2020; ul Qamar et al., 2020). The antiviral role of plant-derived compounds in inhibiting replication and blocking entry of viruses, including coronaviruses, in the host cells has been well reviewed elsewhere (Wen et al., 2007; Dhama et al., 2018; Hensel et al., 2020). The antiviral potential of many plant-derived compounds against SARS-CoV-2 has been recently demonstrated in in silico, in vitro, and in vivo studies (Bahramsoltani and Rahimi, 2020; Basu et al., 2020; Benarba and Pandiella, 2020; Mondal et al., 2020; ul Qamar et al., 2020). Many of the compounds showed targeting of SARS-CoV-2 using bioinformatic tools such as in silico analysis, molecular docking, or molecular farming to enhance the production of recombinant proteins including vaccines and antibodies (Rosales-Mendoza et al., 2020). In search of antiviral compounds, a library of plant-derived constituents containing 32,297 phytochemicals have been screened in molecular docking and results displayed that nine compounds, including myricitrin, methyl rosamarinate, licoleafol, and amaranthin, may curb the activity of 3CLpro enzymes in SARS-CoV-2 (ul Qamar et al., 2020). The inhibitory activity on the proteases and other molecular targets May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 8 Beta-Caryophyllene may be Useful in COVID-19 Jha et al. TABLE 1 \| The antiviral activities of β-Caryophyllene (BCP) or BCP containing plants. TABLE 1 \| The antiviral activities of β-Caryophyllene (BCP) or BCP containing plants. Sources BCP (%) Viral targets References Mosla dianthera (Buch. -Ham. ex Roxb.) Maxim 14.49 Inﬂuenza virus A (IVA) Wu et al. (2012) Glechon spathulata Benth 14.2 Human Herpes Virus Type 1 (HSV-1) Venturi et al. (2015) Glechon marifolia Benth 32.2 HSV-1 Venturi et al. (2015) Illicium verum Hook.f - HSV-1 Astani et al. (2011) Buddleja cordobensis Griseb 16.5 DENV-2, JUNV and HSV-1 Duschatzky et al. (2005) Cinnamomum zeylanicum Blume 0.5–6.7 Inﬂuenza type A (H1N1) Setzer (2016) Eupatorium patens D. Don ex Hook. and Arn 14.1 HSV-1 García et al. (2003) Gaillardia megapotamica (Spreng.) Baker 6.7 DENV-2, JUNV and HSV-1 Duschatzky et al. (2005) Hyptis mutabilis (Rich.) Briq 10.9 Human Herpes Virus Type 2 (HSV-2) Brand et al. (2015) Jungia polita Griseb. 8.1 DENV-2, JUNV and HSV-1 Duschatzky et al. (2005) Lavandula angustifolia Mill 5.1 H1N1 Setzer (2016) Lepechinia vulcanicola J.R.I. Wood 8.7 HSV-1, HSV-2 Brand et al. ANTIVIRAL PROPERTIES OF BCP (2015) Lippia turbinata Griseb. 6.4 HSV-1 García et al. (2003) Melissa ofﬁcinalis L 14.2 HSV-2 Allahverdiyev et al. (2004) Ocimum campechianum Mill 13.0 HSV-2 Brand et al. (2015) Thymus capitatus (L.) Hoffmanns. and Link 2.9 Cytopathogenic murine norovirus Moussaoui et al. (2013) Thymus vulgaris L 7.0 HSV-1 Schnitzler et al. (2007) Zataria multiﬂora Boiss. 3.0 Real time PCR (H9N2 subtype of AIV) Shayeganmehr et al. (2018) speciﬁcally targets the dengue virus proteins. BCP was also found useful in Epstein-Barr virus-associated diseases. BCP has been recognized in the essential oil of Waldheimia glabra (Decne.) Regel, popularly known as ‘Ghaan-Poe’, is used for inﬂuenza in Tibetan medicine (Manzo et al., 2016; De et al., 2017). The essential oil showed antiviral activity against inﬂuenza virus H3N2 in an in vitro assay and was found comparable to ribavirin, a standard antiviral drug (Manzo et al., 2016). margarita (Lour.) Swingle, commonly known as Kumquats, which belongs to the citrus family, contained BCP and was shown effective against avian inﬂuenza-A virus (H5N1). BCP- containing essential oil of Schizonepeta tenuifolia (Benth.) Briq. was shown to inhibit norovirus replication through the induction of antiviral interferon production during virus replication by inducing the expression of both type I and type II interferons and increasing the transcription of interferon-β in infected RAW 264.7 cells via an increased phosphorylation of interferon regulatory factor 3, a critical transcription regulator for type I interferon production (Ng et al., 2018). Very recently, BCP on oral supplementation showed antiviral and immunomodulatory potential in an in vivo viral model of Newcastle disease virus (Hassanin et al., 2020). Further, the anti-inﬂammatory activity of essential oil was evidenced by the inhibition of NO production in LPS-stimulated macrophages and was found to be more potent than the standard drug dexamethasone. BCP was found in the essential oil of Teucrium pseudochamaepitys Georgi, an important Tunisian ﬂora element that is used in traditional medicine for its antiviral activity against an enterovirus, Coxsackie 4 (CV-B4), known for causing myocarditis and CNS pathologies (Hammami et al., 2015). The essential oil showed potent antioxidant properties. The BCP-containing essential oil of Glechon spathulata Spreng. and Glechon marifolia Benth. are traditionally used in viral infections for their viricidal activity against HSV-1 strain KOS, VR733 (ATCC), or 29-R (ACVres) (Venturi et al., 2015). The essential oil of Glechon spathulata Spreng. exhibited activity against all strains and Glrchon marifolia Benth. ANTIVIRAL PROPERTIES OF BCP was found to be active against two strains, KOS and VR733. HSV-1 was more susceptible to the oil of Glechon spathulata Spreng. than that of Glechon marifolia Benth. The viral titer was reduced by up to 2 log10 for KOS and VR-733 strains. BCP-containing essential oil of Mosla dianthera (Buch.- Ham. ex Roxb.) Maxim., a herb popularly used in respiratory illnesses, showed antiviral activity in mice infected with inﬂuenza virus A (Wu et al., 2012). It exerted potent antioxidant, anti- inﬂammatory, and antiviral effects, as evidenced by the reduced serum levels of IFN-γ and IL-4, viral titer in the lungs, amelioration of pneumonia, and an increased endogenous antioxidant level in the lung tissues. The ﬁndings were suggestive of its possible use in inﬂuenza and viral pneumonia (Wu et al., 2012). The essential oil obtained from Fortunella In COVID-19 patients, the prevalence of coinfections has been reported and the co-pathogens may be bacteria, such as Streptococcus pneumoniae, Staphylococcus aureus, Klebsiella pneumoniae, Mycoplasma pneumoniae, Chlamydia pneumonia, Legionella pneumophila, and Acinetobacter baumannii, fungi, such as Candida species and Aspergillus ﬂavus, or viruses, such as inﬂuenza, coronavirus, rhinovirus/enterovirus, parainﬂuenza, metapneumovirus, inﬂuenza B virus, and human immunodeﬁciency virus (Lai et al., 2020). Additionally, the antibacterial and antifungal effects of BCP have been reported in a number of studies (Schmidt et al., 2010; Rather et al., 2012; Dahham et al., 2015; Nieto-Bobadilla et al., 2015; Yang et al., 2015; Okoh et al., 2019;). Taken together, the antiviral and antibacterial activities, BCP may be a promising agent for secondary infections, as well as the viral infections. ANTIOXIDANT PROPERTIES OF BCP Besides the immune-inﬂammatory changes, macrophages and neutrophils can produce numerous reactive oxygen species (ROS), including H2O2, (O2 −), (•OH), which further activates many signaling pathways and the onset of inﬂammation and cell May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 9 Beta-Caryophyllene may be Useful in COVID-19 Jha et al. death in many organs, including the lungs (Imai et al., 2008). Oxidative stress and the subsequent activation of NF-kB-toll-like receptor signaling pathways, triggered by viral pathogens such as SARS-CoV-2, are believed to amplify the host inﬂammatory response that results in acute lung injury (Saleh et al., 2020). Additionally, the hyper inﬂammatory/oxidative state may lead to the dysfunction of mitochondria, the hub of cellular oxidative homeostasis, and cause platelet damage, which, upon interaction with coagulation cascades, aggravates the clotting events and thrombus formation. and glutathione depletion, free radical scavenging, and augmenting the endogenous antioxidant defense in the tissues of different organs, such as the heart (Ojha et al., 2016; Baldissera et al., 2017; Meeran et al., 2019), brain (Choi et al., 2013; Ojha et al., 2016; Tian et al., 2016), intestine (Bento et al., 2011), liver (Arizuka et al., 2017; Baldissera et al., 2017; Varga et al., 2018), stomach (Tambe et al., 1996), kidneys (Horváth et al., 2012; Hammad et al., 2018), pancreas (Basha and Sankaranarayanan, 2016), and blood (Youssef et al., 2019), which may aid the protective, as well as the adaptative, responses against viral infections and drugs. Mitochondrial oxidative stress may contribute to microbiota dysbiosis, altering the coagulation pathways and fueling the inﬂammatory/oxidative response, leading to a vicious cycle of events (Saleh et al., 2020). Oxidative stress further primes endothelial cells to acquire a pro-thrombotic and pro- inﬂammatory phenotype, predisposing patients to thromboembolic and vasculitic events and disseminated intravascular coagulopathy (Panfoli, 2020). Nrf2, a transcription factor which regulates the redox balance and the expression of genes involved in immunity and inﬂammation, is believed to defend against SARS-CoV-2 (McCord et al., 2020). The suppressed redox status of a cell enhances its susceptibility to oxidative stress, which may lead to cell death and viral release (Khomich et al., 2018). SARS-CoV-2 infections can lead to alterations of the redox balance in infected cells through the modulation of NAD+ biosynthesis and PARP function, along with altering the proteasome and mitochondrial function in cells, thereby leading to enhanced cell stress responses that further exacerbate inﬂammation. BCP MAY BE PROSPECTIVE IN COVID-19 ASSOCIATED SEPSIS SARS-CoV-2 infections may lead to sepsis and to subsequent multi-organ failure. Sepsis involves both the inﬂammatory response and immune suppression in response to an infection (Mira et al., 2017). CB2R plays a vital role in neutrophil/leukocyte recruitment, thereby suppressing infection and inﬂammation during sepsis (He et al., 2019). However, CB2R was also shown to contribute to septic immune dysfunction and mortality (Csóka et al., 2009). In a recent review the role of CB2R as a therapeutic target has been suggested based on the reports from preclinical animal models or in vitro cultured cells (He et al., 2019). The authors suggested that due to the lack of clinical evidence and the ambiguous underlying mechanisms, the clinical application of CB2R stimulation in sepsis is yet to be conﬁrmed further (He et al., 2019). In many recent studies speciﬁc CB2R synthetic agonists, including HU-308 (Liu et al., 2020), GW405833 (Zhou et al., 2020), JWH133 (Çakır et al., 2020), and natural agonist, BCP (Brito et al., 2019), have been shown to ameliorate lung tissue damage, inhibiting oxidative stress, release of inﬂammatory mediators, recruitment of leucocytes and bacteremia, and improve survival in different preclinical models of sepsis. CB2R agonists were reported to ameliorate leukocyte adhesion to the endothelium, oxidative stress, systemic inﬂammatory mediators, microcirculatory dysfunction, bacteremia, and lung injury, along with an improvement in survival in experimental models of sepsis (Sardinha et al., 2014; Toguri et al., 2015). CB2R activation speciﬁcally mitigated septic lung injury by suppressing inﬂammatory mediators and augmenting autophagy (Liu et al., 2014; Liu et al., 2020). In an experimental model of polymicrobial BCP is present in Ocimum sanctum L. (Kamyab and Eshraghian, 2013), Pinus spp.(Xie et al., 2015), Salvia ofﬁcinalis L. (El-Hosseiny et al., 2016), Citrus limoni (L.) Osbeck (Oboh et al., 2014), Stachys pilifera Benth. (Sadeghi et al., 2020), Pistacia lentiscus L. (Mohamed et al., 2018), Eplingiella fruticose (Salzm. ex Benth.) Harley & J.F.B. Pastore (Beserra-Filho et al., 2019), Lantana montevidensis (Spreng.) Briq. (de Oliveira et al., 2019), Azadirachta indica A. Juss. (Okoh et al., 2019), Rosmarinus ofﬁcinalis L. (Mohamed et al., 2016), Aquilaria crassna Pierre ex Lecomte (Dahham et al., 2015), and Copaiaba oil (Ames-Sibin et al., 2018) and has been shown to augment the levels of endogenous antioxidants, exerting ferric reducing properties, a Fe2+ chelation, and radicals scavenging activity in DPPH, FRAP, ORAC, ABTS, •OH, and NO assays (Oboh et al., 2014; Pant et al., 2014). ANTIOXIDANT PROPERTIES OF BCP ROS production can increase IL-6 production and lipid peroxidation, resulting in cell damage (Nasi et al., 2020). Virus-induced inﬂammation and oxidative stress could be the common mechanisms responsible for the cardiovascular, pulmonary, renal, and neurological symptoms in COVID-19 patients (Nuzzo and Picone, 2020). BCP was found to exert protective effects in renal cells by suppressing ROS generation, NADPH oxidase 2/4 expression, and by controlling cell proliferation and inﬂammation by inhibiting proinﬂammatory cytokines, Nrf2/HO-1 and NF-κB/Nrf2 signaling pathways (Li et al., 2020). BCP has been shown superior to probucol, α-humulene, α-tocopherol (Calleja et al., 2013), and synthetic CB2R agonist, JWH133 (Klauke et al., 2014). Also, BCP was shown to correct neurobehavior (anxiety, depression, and memory deﬁcit), and neurochemical (oxidative, inﬂammatory, and neurotrophic factor) alterations in diet-induced obese rats (Youssef et al., 2019). Taken together, it is evident that BCP attenuated the oxidative stress and subsequent inﬂammation in organ dysfunction and metabolic disorders, favorably modulated redox signaling pathways (Baldissera et al., 2017; Varga et al., 2018), which are akin to the pathophysiology of SARS-CoV-2 infection. BCP MAY BE PROSPECTIVE IN COVID-19 ASSOCIATED NEUROLOGICAL MANIFESTATIONS SARS-CoV-2 infection has been reported to increase the susceptibility of patients affected by coronary artery disease and risk factors of atherosclerotic cardiovascular disease to develop adverse outcomes and lead to death (Vinciguerra et al., 2020). SARS-CoV-2 mediating ACE2 receptors infect endothelial cells, which regulate inﬂammation, vasomotor tone, and hemostatic balance. Pathological conditions associated with atherosclerotic progression, such as heart failure, coronary heart disease, hypertension, and diabetes mellitus, are the predictive factors for severity and susceptibility during SARS-CoV-2 infection (Vinciguerra et al., 2020). The pathogenesis involves endothelial dysfunction, altered vasopermeability, and formation of pulmonary microthrombi subsequent to inﬂammation, hypoxia, oxidative stress, mitochondrial dysfunction, and DNA damage. Patients with preexisting pulmonary vascular diseases also appear to have an increased risk of morbidity and mortality (Potus et al., 2020). SARS-CoV-2 is considered to be neurovirulent and neuroinvasive, in parallel with adherence to endothelial cells and cardiomyocytes (Sweid et al., 2020). Ischemic stroke, venous thrombosis, and intracerebral hemorrhage are the reported neurological manifestations of SARS-CoV-2 infection (Jiménez-Ruiz et al., 2020). The pathophysiology of ischemic stroke or cerebral hemorrhage includes an increased level of inﬂammatory cytokines in the brain, subsequent to the activation of microglia, astrocytes, and adhesion molecules, along with leukocyte recruitment and an impaired blood brain barrier. CB2R are upregulated during the inﬂammatory activation and CB2R agonists have been shown to be effective in acute ischemia and hemorrhagic stroke (Capettini et al., 2012). BCP has been shown to exert a protective role on neurological deﬁcit and neuroinﬂammation in experimental models, including middle cerebral artery occlusion induced-cerebral ischemia by suppressing the oxidative stress, inﬂammatory mediators, apoptosis, and reduction in brain edema, as well as preservation of tight junction proteins and repair of blood brain barrier (Zhang et al., 2017; Tian et al., 2019). BCP exerted its protective effects mediating CB2R activation (Choi et al., 2013) and its associated mechanisms, including the downregulation of TLR4 pathways to suppress inﬂammation and polarizing microglial phenotype from M1 to M2 (Tian et al., 2019), PI3K/Akt signaling pathway to suppress apoptosis (Zhang et al., 2017), an upregulation of the modulation of AMPK/CREB signaling (Choi et al., 2013), and the upregulation of Nrf2/HO-1 pathway to suppress oxidative stress and apoptosis (Lou et al., 2016). Atherosclerosis is considered as an ideal pathogenetic substrate for high viral replication ability, leading to adverse outcomes, as found in patients with cardiovascular factors. BCP MAY BE PROSPECTIVE IN COVID-19 ASSOCIATED SEPSIS BCP also enhances tolerance against stress, augments chaperons, and improves the antioxidant power (Srivastava et al., 2016). BCP mitigates the oxidative stress by counteracting ROS generation, inhibiting lipid peroxidation May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 10 Jha et al. Beta-Caryophyllene may be Useful in COVID-19 sepsis, CB2R activation decreased the histopathological damage in the brain, heart, lungs, and liver by reducing the levels of caspase-3, p-NF-κB, TNF-α, IL-1β, and IL-6 in these tissues, as well as in the serum, and improved the anti-inﬂammatory cytokine IL-10 levels (Çakır et al., 2020). BCP also attenuated neuronal necrosis, receptor-interaction protein kinase-1 (RIPK1), receptor-interaction protein kinase-3 (RIPK3) expression, and mixed lineage kinase domain-like protein (MLKL) phosphorylation in cerebral ischemia by inhibiting high-mobility group box 1 (HMGB1)-toll-like receptor 4 (TLR4) signaling pathways and proinﬂammatory cytokines. HMGB1, which is released by macrophages and monocytes in response to high levels of proinﬂammatory cytokines, plays a critical role in allowing innate immune cells to respond to both infection and injury. After its release, HMGB1 binds to its receptor for an advanced glycation of the end- products, which further activates MAPK and NF-κB, resulting in an overgeneration of various cytokines, causing a massive neutrophil inﬁltration into the lungs, and subsequent acute lung injury. The agents that target the release of HMGB1 are suggested to be useful in reducing mortality by preventing the progression from respiratory distress to sepsis (Wyganowska-Swiatkowska et al., 2020). Given the protective role of BCP on redox homeostasis and on the immune-inﬂammatory cascade in acute cerebrovascular disorders, it holds therapeutic promise for neurological manifestations of SARS-CoV-2. To model sepsis, many of the experimental models rely on LPS-induced macrophages, which involve the activation and release of inﬂammatory mediators, including cytokines (Brito et al., 2019). BCP was reported to reduce the level of leukocytes, cytokines TNF-α, IL-6, IL-12, and IFN-γ, and increase the levels of IL-4 and IL-5 (Brito et al., 2019). BCP was shown to suppress inﬂammatory mediators and exert inhibitory effects on macrophages (Tung et al., 2008; Yamaguchi and Levy, 2019; Yamaguchi and Levy, 2020). Although the role of CB2R in sepsis has mixed reports, BCP has been shown to be beneﬁcial in sepsis via CB2R activation and the off target effects cannot be excluded (Meza and Lehmann, 2018). Additionally, BCP is known to have a better safety proﬁle over synthetic cannabinoids. BCP MAY BE PROSPECTIVE IN COVID-19 ASSOCIATED SEPSIS Given the association of SARS-CoV-2 infections and sepsis-induced life- threatening organ dysfunction, BCP may be a promising candidate for COVID-19 associated sepsis. BCP MAY BE PROSPECTIVE IN COVID-19 ASSOCIATED AIRWAY INFLAMMATION BCP containing essential oil of Croton sonderianus Müll. Arg. was found to exert myorelaxant activity in rat airway smooth muscles, which is suggestive of its potential in bronchospasm (Pinho-da- Silva et al., 2010). During viral infections, the activation of selective CB2R by agonists was shown to suppress leukocyte migration into the site of inﬂammation (Tahamtan et al., 2018). CB2 agonists in HIV-1 infection also reduced infection in BCP MAY BE PROSPECTIVE IN COVID-19 ASSOCIATED NEUROLOGICAL MANIFESTATIONS SARS- CoV-2 may aggravate atherosclerosis due to an excessive and aberrant plasmatic concentration of cytokines (Vinciguerra et al., 2020). Atherosclerosis involves vascular inﬂammation, characterized by a narrowed vascular lumen in the entire tunica intima and a reduced elasticity of the arterial walls. CB2R activation mitigated endothelial cell activation, transendothelial migration of monocytes, and monocyte/ neutrophil-endothelial adhesion, and suppressed the proliferation and migration of human coronary vascular smooth muscle cells induced by TNF-α (Rajesh et al., 2007). A May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 11 Beta-Caryophyllene may be Useful in COVID-19 Jha et al. pneumonia causing pathogen, Chlamydia pneumoniae, provokes atheroma lesions by releasing heat shock proteins, which, by activating Hsp60 on endothelial cells, increase vascular smooth muscle cell proliferation. BCP was found to inhibit Hsp60- induced vascular smooth muscle cell proliferation and its potential in atherosclerosis has been suggested (Fukuoka et al., 2004). BCP also ameliorated acute myocardial injury by improving cardiac function, reducing infarct, restoring myocyte enzymes, and suppressing inﬂammation by inhibiting HSP-60/TLR/MyD88/NFκB signaling pathways (Younis and Mohamed, 2019). BCP was found to counteract drug-induced cardiomyopathy by attenuating inﬂammation, oxidative stress, and apoptosis by activating CB2R (Meeran et al., 2019). BCP mitigated hypercholesterolemia, dyslipidemia, and vascular inﬂammation, reduced atherogenic and coronary risk index, and corrected lipid metabolism by inhibiting proatherogenic vascular cell adhesion molecule 1 (VCAM-1) and restoring vascular eNOS/iNOS expression by maintaining the NO levels, mediating the activation of CB2 and PPAR-γ receptors in a high- fat diet and fructose-induced obesity (Baldissera et al., 2017; Harb et al., 2018; Youssef et al., 2019). N-acetylglucosaminidase activity and the levels of mRNA expression of TNF-α, IL-1β, IFN-γ, chemokines, and the activation of extracellular signal-regulated kinase 1/2, NF-κB, IκB-kinase α/β, cAMP response element binding, and the expression of caspase-3 and Ki-67. BCP increased IL-4 levels and forkhead box P3 mRNA expression in the colon (Cho et al., 2007; Bento et al., 2011). In macrophages challenged with LPS, BCP reduced the levels of cytokines, such as TNF-α, keratinocyte- derived chemokines, and MIP-2. Recently, in patients infected with SARS-CoV-2, an inﬂammatory response in the gut is evidenced by diarrhea and increased IL-6 and fecal calprotectin levels, showing the activation of neutrophils (Effenberger et al., 2020). Additionally, diarrhea appears as one of the most frequent symptoms in patients infected with SARS-CoV-2 (D’Amico et al., 2020). BCP MAY BE PROSPECTIVE IN COVID-19 ASSOCIATED NEUROLOGICAL MANIFESTATIONS Given the role of BCP in suppressing intestinal inﬂammation (Cho et al., 2007; Bento et al., 2011) and diarrhea (Nieto-Bobadilla et al., 2015), BCP may hold great therapeutic promise for COVID-19. BCP MAY BE PROSPECTIVE IN COVID-19 ASSOCIATED AIRWAY INFLAMMATION ) Furthermore, in addition to correcting the lipid proﬁle, BCP- mediating CB2R-dependent mechanism inhibited leukocyte- endothelial attachment, neutrophil recruitment, and macrophage inﬁltration, inducing VCAM-1 to mediate the JAK2/STAT1/IRF-1 pathway (Zhang et al., 2017). BCP is one of the most important components of Copaiba oil, popularly used in Brazil for respiratory and cardiovascular illnesses. The nano- capsules of copaiba oil were shown to attenuate monocrotaline- induced pulmonary arterial hypertension in rats by counteracting the oxidative stress and inﬂammation, and by improving the cardiac function (Campos et al., 2017). One of the major clinical features and reasons for death in COVID-19 patients is respiratory distress syndrome, that also leads to acute cardiac injury (Huang et al., 2020). The potential of BCP on pulmonary vasculature is also promising and can be useful in reducing the risk of cardiopulmonary complications. The available studies are clearly suggestive of the therapeutic beneﬁts of BCP in atherosclerosis, acute myocardial infarction, dyslipidemia, obesity, and fatty liver and could be important in preventing the worsening of the condition in COVID-19 patients. In many reports, vaccination of Bacillus Calmette-Guérin (BCG), a live attenuated vaccine of Mycobacterium bovis strain, is believed to provide protection against SARS-CoV-2 infection. BCG vaccination is believed to be associated with the induction of trained immunity, a kind of epigenetic reprogramming of innate immune cell types (Goodridge et al., 2016). In vaccinated individuals, monocytes and/or natural killer cells exhibit an upregulation of surface markers of activation and synthesis of cytokines, such as IL-1β, IL-6, IFN-γ, and TNF-α, in response to infection compared to non- vaccinated individuals; this helps in the faster clearance of pathogens, including inﬂuenza (Arts et al., 2018). BCP was found to ameliorate pulmonary inﬂammation in a mice model of Mycobacterium Bovis BCG-induced pulmonary inﬂammation by suppressing neutrophil accumulation, suppressing CXCL1/KC, LTB4, IL-12, and NO production, and mediating the CB2R activation (Andrade-Silva et al., 2016). Additionally, BCP was also found to exert spasmolytic effects on the tracheal smooth muscle in the isolated organs (Pinho-da- Silva et al., 2012). BCP produced antispasmodic effects on the isolated tracheal smooth muscle of rats by inhibiting voltage- dependent L-type Ca2+ channels. BCP did not affect Ca2+ release from the intracellular storage. Further, the inhibitory effect on epithelial COX and a balance between relaxant and constrictor prostanoids exerted by BCP suggested that it may be useful in asthma-like conditions (Pinho-da-Silva et al., 2012). BCP MAY BE PROSPECTIVE IN COVID-19 ASSOCIATED INTESTINAL INFLAMMATION BCP was found to reduce the number of enterobacteria in the luminal and mucosal components, improving the clinical course of an intestinal inﬂammation in the mice model of colitis (Nieto- Bobadilla et al., 2015). BCP ameliorated intestinal inﬂammation in the animal models by mediating the activation of CB2 and the PPAR-γ pathway (Cho et al., 2007; Bento et al., 2011). It suppressed MPO activity and reduced the serum levels of protein and mRNA of IL-6 by 55% (Cho et al., 2007). IL-6 signaling pathway appears as one of the potential therapeutic targets for COVID-19. BCP also suppressed May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 12 Beta-Caryophyllene may be Useful in COVID-19 Jha et al. primary CD4+ T cells, as well as CXCR4-activation-mediated G-protein activity and the phosphorylation of MAPK (Costantino et al., 2012). The CB2 selective property of BCP is reasonably speculated as a basis for its potential to inhibit virus replication, bacterial growth, and to regulate neutrophil recruitment, thus regulating inﬂammation. constitute metabolic syndrome, are one of the risk factors of NAFLD (Friedman et al., 2018; Prins and Olinga, 2020). BCP showed a cholesterol-lowering effect by inhibiting the activity of hepatic hydroxy-methylglutaryl coenzyme A reductase in experimental models of hypercholesterolemia (Arizuka et al., 2017; Baldissera et al., 2017; Harb et al., 2018). Besides correcting the lipid metabolism, BCP also increased high density lipoprotein and attenuated liver injury and ﬁbrosis, restored liver function enzymes and improved antioxidants (Harb et al., 2018). The acute viral respiratory infections may increase the chances of secondary bacterial infections due to a compromised host immune response and thereby worsen the condition. SARS-CoV- 2 was also reported to cause secondary bacterial infection (Dong et al., 2020). BCP is present in many plants, such as Artemisia capillaris Thunb. (Yang et al., 2015), Juniperus rigida var. hibernica Pshenn. (Meng et al., 2016), Lavandula coronopifolia Poir. (Ait Said et al., 2015), Juglans regia L. (Rather et al., 2012), Mosla dianthera (Buch.-Ham. Ex Roxb.) Maxim. (Wu et al., 2012), Thymbra spicata L. (Saidi et al., 2012), and Lantana camara subsp. glandulosissima (Hayek) R.W. Sanders (Tesch et al., 2011), which have been shown to exert inhibitory activity against respiratory pathogens and many virus, fungi, bacteria, and parasites in experimental studies and in human isolates. BCP MAY BE PROSPECTIVE IN COVID-19 ASSOCIATED INTESTINAL INFLAMMATION BCP also attenuated chronic and binge alcohol-induced liver injury and inﬂammation by attenuating the pro-inﬂammatory phenotypic `M1` switch of Kupffer cells and by decreasing the expression of vascular adhesion molecules intercellular adhesion molecule 1, E-selectin, and P-selectin, as well as the neutrophil inﬁltration, and corrected hepatic metabolic dysregulation (Varga et al., 2018). BCP inhibited palmitate- inducible lipid accumulation in human HepG2 hepatocytes by activating AMPK mediating CB2R-dependent Ca2+ signaling pathway (Kamikubo et al., 2016). Mechanistically, BCP regulated hepatic lipid and glucose metabolism by modulating adenosine monophosphate (AMP)-activated protein kinase (AMPK), the main cellular energy sensor (Xu et al., 2018). Considering its hepatoprotective roles, BCP could be promising in conditions of liver injury associated with SARS- CoV-2 infection. BCP MAY BE PROSPECTIVE IN COVID-19 ASSOCIATED LIVER DYSFUNCTION Liver impairment has been reported in patients with SARS-CoV- 2 infection (Feng et al., 2020; Sun et al., 2020) and it is believed to be due to systemic inﬂammation caused by a cytokine storm or pneumonia-associated hypoxia, and the drug regimens containing acetaminophen (Zhang et al., 2020). ACE2 receptors in the bile duct epithelial cells are expressed twenty times more than in hepatocytes and this plausibly explains that SARS-CoV-2 infection may cause bile duct epithelial cell damage (Lee et al., 2020). BCP was reported to ameliorate liver ﬁbrosis in a bile duct ligation induced model, suppressing inﬂammation and apoptotic cell death by mediating the activation of CB2R (Mahmoud et al., 2014). BCP has also been reported to ameliorate drug induced liver injuries, such as ketoprofen- induced liver injury (Kelany and Abdallah, 2016), carbon tetrachloride-induced liver injury (Calleja et al., 2013), and D-galactosamine and lipopolysaccharide-induced liver failure by suppressing inﬂammation and mediating TLR4 and RAGE signaling pathways (Cho et al., 2015). BCP was partially attributed to the hepatoprotective effects of many plants, such as Ocimum sanctum L. (holy basil) (Kamyab and Eshraghian, 2013). Frontiers in Pharmacology \| www.frontiersin.org BCP MAY BE PROSPECTIVE IN COVID-19 ASSOCIATED RENAL INJURIES Acute kidney injury is one of the major complications in patients with SARS-CoV-2 infection (Cheng et al., 2020). ACE2 receptors located on the apical membrane and tubular cells facilitate viral entry and the infection elicits inﬂammatory responses that cause acute kidney injury (Fanelli et al., 2020; Soleimani, 2020). BCP ameliorated acute kidney injury in experimental models by attenuating renal impairment and tubular injury, suppressing renal inﬂammatory mediators, oxidative stress, apoptotic cell death, and preserving renal morphology via activation of CB2R (Horváth et al., 2012; Hammad et al., 2018). BCP is present in many plants, such as Stachys pilifera Benth. (Sadeghi et al., 2020), Salvia ofﬁcinalis L. (Koubaa et al., 2019), Rosmarinus ofﬁcinalis L. (Mohamed et al., 2016), and Pluchea indica (L.) Less. (Sirichaiwetchakoon et al., 2020), and has been shown to be responsible for the renoprotective effects against drug induced-acute kidney injury, as well as diabetic and chronic kidney diseases, by restoring the renal function and suppressing oxidative stress, inﬂammation, and apoptosis. BCP also attenuated renal inﬂammation and oxidative stress by regulating NF-κB/Nrf2 signaling pathways in diabetic kidney diseases (Li et al., 2020). Given the increased risk of renal dysfunction in COVID-19 and the worsening of conditions in patients with chronic kidney or diabetic kidney disease, BCP may be a valuable candidate in preventing renal dysfunction in patients with COVID-19. SARS-CoV-2 infection also increases vulnerability in patients with non-alcoholic fatty liver disease (NAFLD), a chronic liver disease characterized by hepatic steatosis (fatty liver), inﬂammation and hepatocyte damage (steatohepatitis), and lipotoxicity (Prins and Olinga, 2020). The expression of ACE2 is increased in cholangiocytes and hepatocytes during chronic liver damage and was increased in a diet-induced experimental model of NAFLD (Prins and Olinga, 2020). Metabolic perturbations, such as obesity, insulin resistance, hyperglycemia, dyslipidemia, and systemic hypertension, which May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 13 Beta-Caryophyllene may be Useful in COVID-19 Jha et al. TISSUE PROTECTIVE EFFECTS OF BCP MAY BE PROSPECTIVE IN COVID-19 ASSOCIATED ORGAN INJURIES inﬂammation, depression, anxiety, addiction, epilepsy, cancer, and microbial infections (Russo, 2011). Given the impact of COVID-19 on organ functions and considering the organ-protective effect of BCP, it is reasonable to hypothesize that the organ-protective activity of BCP will be beneﬁcial in COVID-19. Besides the lungs, the main site of virus entry and injury, SARS- CoV-2 infection may also affect other organs or organ systems, including the hepatic, renal, neurological, cardiovascular, musculoskeletal, gastrointestinal, hematological, olfactory, gustatory, ophthalmic, and cutaneous systems (Lai et al., 2020). Cardiac manifestations of SARS-CoV-2 involve endothelial damage, an altered lipid proﬁle, endotoxemia, catecholamine, hypoperfusion, unstable hemodynamics, and drug-induced toxicity. BCP showed protective effects against catecholamine-induced myocardial injury and drug-induced cardiotoxicity by improving hemodynamics and alleviating endotoxemia by suppressing inﬂammation, oxidative stress, and apoptosis via activation of CB2R (Meeran et al., 2019). SAFETY AND TOXICITY OF BCP The United States Food and Drug Administration (USFDA) included BCP in the list of compounds regarded as Generally Recognized as Safe (GRAS) for its use as an additive and preservative in food products and beverages. BCP was shown to modulate the expression of drug metabolizing enzymes (phase I and II) in cell lines, rodents, and human liver microsomes, which may inﬂuence the bioavailability and efﬁcacy of concomitantly administered drugs (Ambroˇz et al., 2019). BCP was shown to have a chemopreventive effect and is free from genotoxicity (Álvarez-González et al., 2014), mutagenicity (Di Giacomo et al., 2016), and clastogenicity (Di Sotto et al., 2010). The clinical manifestations of COVID-19 range from mild to severe with extensive involvement of the lungs, from pneumonia to ARDS, acute liver injury, acute cardiac injury, and neurological manifestations that may lead to multi-organ failure with a poor prognosis (Wang et al., 2020; Zhu et al., 2020). Severe lung disease with extensive alveolar damage and progressive respiratory failure leads to deadly outcomes (Yang et al., 2020). The fatalities are higher in older people with cardiometabolic diseases, cancer, immunocompromised patients, or patients with comorbidities. BCP was found to ameliorate renal dysfunction in acute and chronic kidney injury and diabetic kidneys. G aco o et a ., 0 6), a d c astoge c ty ( Sotto et a ., 0 0). BCP exerted synergistic and/or additive actions with many drugs including azithromycin (Zhang et al., 2020), atovaquone (Zhang et al., 2020), metaxolone (Yamaguchi and Levy, 2020), imipramine (Askari et al., 2019), ﬂuoxetine (Askari and Shaﬁee- Nick, 2019), docosahexaenoic acid (Brito et al., 2019), curcumin (Srivastava et al., 2016; D’Ascola et al., 2019), baicalein (Yamaguchi and Levy, 2016), catechin (Yamaguchi and Levy, 2016) and vitamins, which are suggested to be useful for repurposing for COVID-19. In many experimental studies, BCP was found to be better than the standard modern drugs such as phenylbutazone (Basile et al., 1988), probucol (Calleja et al., 2013), tocopherol (Calleja et al., 2013), ribavirin (Wu et al., 2012), atorvastatin (Campos et al., 2015), glibenclamide (Basha and Sankaranarayanan, 2016), and pioglitazone (Youssef et al., 2019). BCP delivered by inhalation was found to be bioavailable in the saliva and appears safe and tolerable (Tarumi and Shinohara, 2020). SAFETY AND TOXICITY OF BCP BCP was convincingly shown to mitigate drugs or xenobiotics-induced organ injuries; for example, it was found to improve the therapeutic efﬁcacy of immunosuppressive drugs and reduce their side effects, such as myelosuppression and hepatotoxicity in experimental arthritis (El-Sheikh et al., 2019). BCP studied at the therapeutic doses was found devoid of organ toxicity in the experimental studies. BCP was found to be effective in liver failure by suppressing liver necrosis, ﬁbrosis, and restoring liver function, mediating CB2R activation. BCP has been shown to be neuroprotective in models of cerebral ischemia, dopaminergic neurodegeneration, seizures, dementia, neurocognitive disorders, depression, anxiety, and encephalitis. BCP improved systemic inﬂammation and oxidative status with no hepatotoxicity, as with nonsteroidal anti-inﬂammatory drugs (Ames-Sibin et al., 2018). It also reduced nausea, epigastric pain, and diarrhea, and improved gastrointestinal activity (Patra et al., 2010). BCP was also found to promote wound healing by modulating numerous signaling pathways (Parisotto-Peterle et al., 2020). Hematological abnormalities, including lymphopenia and leukopenia, have been reported in COVID-19 patients (Ding et al., 2020). The occurrence of leukopenia induced by chemotherapeutic drugs in an experimental model has been shown to be prevented by BCP (Campos et al., 2015). Frontiers in Pharmacology \| www.frontiersin.org CLINICAL EFFICACY AND SAFETY OF BCP BCP administered orally at a dose of 126 mg/day was evaluated in patients with peptic ulcer in a randomized double-blind, placebo- controlled trial (Shim et al., 2019). BCP improved dyspepsia symptoms by reducing Helicobacter pylori infections, improving nausea and epigastric pain, and mediating the inhibition of proinﬂammatory cytokines (Shim et al., 2019). BCP (3%) was evaluated in nineteen women for 20 min using an odor exposure device and was found to improve the libido and vaginal sensation during intercourse in women by improving the salivary Upon oral administration, BCP was found to be bioavailable in almost every organ, including the liver, kidneys, heart, lungs, and blood (Pant et al., 2019). BCP was shown to modulate stress-related genes, provide resistance against stress, improve life span, reduce ageing, and was considered one of the best adaptogenic compounds to enhance the tolerance against stress. The interactions between phytocannabinoids and terpenoids have been suggested to exert synergy for the therapeutic beneﬁts in pain, May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 14 Beta-Caryophyllene may be Useful in COVID-19 Jha et al. testosterone concentrations with no effect on estrogen (Tarumi and Shinohara, 2020). antiviral effects. BCP is one of the main compounds identiﬁed in a large number of dietary plants and is widely accessible and well-studied for its therapeutic beneﬁts. However, the safety and efﬁcacy of BCP still needs to be established in preclinical and clinical trials for its evidence-based use and application in humans. BCP is a functional agonist of CB2R and devoid of psychotropic effects, which makes BCP a fascinating candidate molecule for further investigation. BCP was administered to diabetes patients with diabetes- related complications; painful distal symmetric polyneuropathy was found to relieve polyneuropathy with an increased amplitude and a reduction of pain, with good tolerance and no adverse effects (Semprini et al., 2018). Recently, in a placebo-controlled clinical study, patients with hand arthritis applied BCP- containing copaiba oil topically and BCP was found to be safe, well tolerated, and beneﬁcial in reducing pain and inﬂammation (Bahr et al., 2018). g A majority of the experimental research carried out on the therapeutic beneﬁts of phytochemicals are based on ethnopharmacological usage of the particular plant rich in these. Many have also been evaluated for their antiviral properties in addition to their anti-inﬂammatory and immunoregulatory roles in numerous immune-related disease models. CLINICAL EFFICACY AND SAFETY OF BCP Since the emergence of COVID-19, the repurposing of drugs began ﬁrst with target identiﬁcation and continues to be used in the screening of druggable agents against viral infections. It is noteworthy to state that until recent years, the antiviral potential of natural products were shown to be effective in in vitro studies, whereas the SARS-CoV emerged in 2003 (Kim et al., 2014; Park et al., 2017). But none of them have been evaluated meticulously enough to translate their effects to humans despite their potential efﬁcacy in preclinical studies. This is due to many reasons, including the lack of an integrated approach. A recent report suggested that if an integrated and rigorous approach could have been followed since the emergence of SARS-CoV, we may have progressed to clinical studies and developed some useful agents in the process of drug discovery and development, which involves the testing of druggable compounds from laboratory to clinics (Pandey et al., 2020). It can be proposed that the phytochemicals should be investigated and validated in the preclinical models of COVID-19 despite strong evidence for their anti-inﬂammatory, immunomodulatory, and antiviral properties. DOSAGE FORMS AND PHARMACEUTICAL DEVELOPMENT OF BCP Many formulations containing BCP have been developed, including Amukkara Choornam (Patra et al., 2010), CIN-102, a coated pellets and matrix mini-tablet (Nieto-Bobadilla et al., 2015), and PipeNig®-FL, a high standardized content of BCP (Geddo et al., 2019). BCP is highly lipophilic, less soluble in water and, upon exposure to air, it easily oxidizes. To overcome its low bioavailability, many novel drug delivery systems have been developed. Various kinds of formulations, such as liposomes, nanoemulsions, nanoﬁbers, microemulsions, nanoparticles, micelles, phospholipid complexes, nanocarriers, nanocomposites, hydrogels, and matrix formulations using cyclodextrin, have been developed to enhance the solubility, stability, and release pattern of BCP (Santos et al., 2018). Novel formulations will pave the way for the pharmaceutical development of BCP and may aid in improving its clinical usage. LIMITATIONS y, p p In the present manuscript, the possible role of BCP in COVID- 19 has been proposed based on the previously reported potent pharmacological activity of BCP against infection, inﬂammation, and immunity in experimental models of human diseases other than SARS-CoV-2. Many authors proposed the hypotheses that CB2R, an important constituent in endocannabinoid system, may play a role in targeting the trinity of infection, inﬂammation, and immune dysregulation (Nagoor Meeran et al., 2020). Given the role of CB2R activation in attenuating inﬂammation, viral replication, and favorable modulation of immune systems, BCP endowed with the CB2R selective agonist property has been pharmacologically reasoned to be a candidate for its possible use as preventive agent or therapeutic adjunct in COVID-19. There are reports of long-term complications in some patients even after recovery from COVID-19. Thus, given the tissue protective effects and effect on numerous tissues remodeling effects, BCP could be a candidate to be investigated for possible use in improving prognosis and combating the long-term complications in COVID-19. Taking into consideration the safety of BCP in humans, dietary use, and efﬁcacy of BCP in various disease models in experimental studies, BCP may be a valuable agent to be investigated further for COVID-19. Since the emergence of COVID-19, a signiﬁcant number of natural products, including plant extracts and phytochemicals, have been proposed for their possible use as a preventive agent or as an adjuvant in COVID-19 (Asif et al., 2020; Boukhatem and Setzer, 2020; da Silva et al., 2020; Diniz et al., 2021). Though, given their pleiotropic and immunomodulatory nature, the role of phytocannabinoids are reasonably suggested useful, but caution should be exercised. The potential application of cannabinoids in COVID-19 management can’t be overlooked until proof-of-concept studies become available (Pastor et al., 2020; Anil et al., 2021). The cannabinoids shown to possess potent immunomodulatory, anti-inﬂammatory, and antimicrobial properties are proposed for their use in COVID-19. However, few of the phytocannabinoids have been screened in molecular docking studies for their potential activity against viral targets using the in-silico tools. The role of phytocannabinoids is believed to be delicate given their action on inﬂammation and immune modulation and the possibility of the unfavorable effects in acute infection due to risk of immunosuppression (Sexton, 2020). LIMITATIONS Among numerous cannabinoids, BCP has been shown to be more convincing in terms of its immunomodulatory, anti-inﬂammatory, and May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 15 Beta-Caryophyllene may be Useful in COVID-19 Jha et al. with dietary use. However, the suggestion on the possible use of BCP in COVID-19 remains inconclusive until the in-silico observations could be conﬁrmed in the experimental studies and further proof of the concept studies. The available reports clearly demonstrate that the progression and complications of COVID-19 involves cytokine storm, therefore, cannabinoids activating CB2R may inhibit cytokine storm due to their additional organ-protective effects. However, until now there has been no clear evidence available on the antiviral activity of BCP on SARS-CoV-2. There is no preclinical or clinical data available on whether BCP can protect against COVID-19 or may be useful in treatment of COVID-19. The recent availability of animal models could be important in evaluating its preclinical efﬁcacy. There is a lack of clinical data and rigorous pharmacokinetics in humans. Therefore, preclinical evaluation, including duration of use and dose, is suggested. The safety and interaction with concomitant drugs, as well as the heterogeneity of the target population, should also be considered before the possible use of BCP whether in prevention or as adjunct treatment. Nonetheless, there is an opportunity for further investigation to investigate the possible use against COVID-19. Considering the safety evidenced in numerous experimental and few clinical studies, further studies are encouraged to recommend the clinical usage and pharmaceutical development of BCP. The polypharmacological properties, including receptor selectivity provide rationale and its drug-like properties, provide more realism for its future in drug discovery and development. Additionally, the antioxidant, anti-stress, and longevity potential provide a nutritional basis of its use to boost the immunity and suppress overt oxidative stress and subsequent hyperinﬂammatory states. Considering the recognition of safety status by USFDA and its favorable pharmacokinetic and physicochemical properties, BCP itself or plants containing a high amount of BCP may be important for nutritional or dietary usage. BCP and the plants containing BCP as a major ingredient may be candidates for developing novel antiviral and immunomodulator therapies for coronaviruses. However, further research is needed to address novel drug discovery employing the chemical scaffold or pharmacophores of BCP. CONCLUSION BCP is a unique molecule in various ways, such as being dietary, devoid of psychotropic effects, possessing negligible toxicity, wide availability in plants, oral bioavailability, a druggable property, and functional receptor selectivity. BCP interacts or binds to different receptors, including CB2, PPAR-α, and PPAR-γ, opioid, histaminergic, TRPV, and TLR, and has enzyme inhibitory activities, including amylase, lipase, α-glucosidase, HMG-CoA reductase, acetylcholinesterase, secretase, cyclooxygenase, and nitric oxide synthase. Taken together, the receptor selectivity made it a distinctive candidate with a pharmacological rationale for pharmaceutical development, more than an antioxidant molecule, which is common with natural products-based nutraceuticals. Integrating the potent anti-inﬂammatory, immunomodulator, and antiviral properties of BCP and its potential beneﬁts in pathological features of cardiovascular, neurological, gastrointestinal, hematological, renal, ocular, and cutaneous systems, which are the common accompaniments of SARS-CoV-2 infection, the beneﬁts of BCP and plants rich in BCP may be important for COVID- 19. The candidature of BCP in COVID-19 treatment may appear somewhat speculative but cannot be overlooked as it possesses favorable physiochemical and druggable properties AUTHOR CONTRIBUTIONS SO conceptualized the study and hypothesis. NJ, CS, HH, and HJ performed literature search. NJ draw the schemes and drafted the artwork. SA drafted the tables. NJ, CS, HJ, CP, SG, and SO contributed signiﬁcantly in editing the manuscript. All authors read, edited and approved the manuscript. LIMITATIONS The natural dietary availability and CB2 receptors mediated functional properties and selectivity are suggestive of developing BCP-based nutraceuticals and pharmaceuticals as candidate compounds for COVID-19 and other coronavirus diseases. The opinion of the authors on possible candidature for possible use of BCP in COVID-19 is solely based on available literature on the effects of BCP against infection, immunity, and inﬂammation in corroboration with the in-silico studies. The authors do not promote the use of BCP in any form for COVID-19 until clear evidence becomes available from proof of the concept studies. ACKNOWLEDGMENTS The authors are grateful to the United Arab Emirates University, UAE for providing facilities. The authors are grateful to the United Arab Emirates University, UAE for providing facilities. Coronopifolia Essential Oil against Antibiotic-Resistant Bacteria. Nat. Prod. Res. 29, 582–585. doi:10.1080/14786419.2014.954246 REFERENCES doi:10.3389/fphar.2020.571434 Allahverdiyev, A., Duran, N., Ozguven, M., and Koltas, S. (2004). Antiviral Activity of the Volatile Oils of Melissa Ofﬁcinalis L. Against Herpes Simplex Virus Type- 2. Phytomedicine 11, 657–661. doi:10.1016/j.phymed.2003.07.014 Baldissera, M. D., Souza, C. F., Grando, T. H., Doleski, P. H., Boligon, A. A., Stefani, L. M., et al. (2017). Hypolipidemic Effect of β-Caryophyllene to Treat Hyperlipidemic Rats. Naunyn-Schmiedeberg’s Arch. Pharmacol. 390, 215–223. doi:10.1007/s00210-016-1326-3 Allegra, A., Gioacchino, M. D., Tonacci, A., Musolino, C., and Gangemi, S. (2020). Immunopathology of SARS-CoV-2 Infection: Immune Cells and Mediators, Prognostic Factors, and Immune-Therapeutic Implications. IJMS 21, 4782. doi:10.3390/ijms21134782 Baldissera, M. D., Souza, C. F., Grando, T. H., Stefani, L. M., and Monteiro, S. G. (2017). β-Caryophyllene Reduces Atherogenic Index and Coronary Risk Index in Hypercholesterolemic Rats: The Involvement of Cardiac Oxidative Damage. Chem.-Biol. Interact. 270, 9–14. doi:10.1016/j.cbi.2017.04.008 Álvarez-González, I., Madrigal-Bujaidar, E., and Castro-García, S. (2014). Antigenotoxic Capacity of Beta-Caryophyllene in Mouse, and Evaluation of Its Antioxidant and GST Induction Activities. J. Toxicol. Sci. 39, 849–859. doi:10.2131/jts.39.849 Basha, R. H., and Sankaranarayanan, C. (2016). β-Caryophyllene, A Natural Sesquiterpene Lactone Attenuates Hyperglycemia Mediated Oxidative and Inﬂammatory Stress in Experimental Diabetic Rats. Chem.-Biol. Interact. 245, 50–58. doi:10.1016/j.cbi.2015.12.019 Aly, E., Khajah, M. A., and Masocha, W. (2019). β-Caryophyllene, a CB2-Receptor- Selective Phytocannabinoid, Suppresses Mechanical Allodynia in a Mouse Model of Antiretroviral-Induced Neuropathic Pain. Molecules 25, 106. doi:10.3390/molecules25010106 Basholli-Salihu, M., Schuster, R., Hajdari, A., Viernstein, H., Mustafa, B., Mueller, M., et al. (2017). Phytochemical Composition, Anti-inﬂammatory Activity and Cytotoxic Effects of Essential Oils From Three Pinus Spp. Pharm. Biol. 55, 1553–1560. doi:10.1080/13880209.2017.1309555 Ambroˇz, M., ˇSmatová, M., ˇSadibolová, M., Pospíˇsilová, E., Hadravská, P., Kaˇsparová, M., et al. (2019). Sesquiterpenes α-Humulene and β-Caryophyllene Oxide Enhance the Efﬁcacy of 5-ﬂuorouracil and Oxaliplatin in Colon Cancer Cells. Acta Pharm. 69, 121–128. doi:10.2478/ acph-2019-0003 Basile, A. C., Sertié, J. A., Freitas, P. C., and Zanini, A. C. (1988). Anti-inﬂammatory Activity of Oleoresin From Brazilian Copaifera. J. Ethnopharmacol. 22, 101–109. doi:10.1016/0378-8741(88)90235-8 Ames-Sibin, A. P., Barizão, C. L., Castro-Ghizoni, C. V., Silva, F. M. S., Sá- Nakanishi, A. B., Bracht, L., et al. (2018). β-Caryophyllene, the Major Constituent of Copaiba Oil, Reduces Systemic Inﬂammation and Oxidative Stress in Arthritic Rats. J. Cell. Biochem. 119, 10262–10277. doi:10.1002/jcb. 27369 Bassaganya-Riera, J., Song, R., Roberts, P. C., and Hontecillas, R. (2010). PPAR-γ Activation as an Anti-inﬂammatory Therapy for Respiratory Virus Infections. Viral Immunol. 23, 343–352. doi:10.1089/vim.2010.0016 Basting, R. T., Spindola, H. M., Sousa, I. M., Trigo, J. REFERENCES R., de Carvalho, J. E., and Foglio, M. A. (2019). Pterodon Pubescens and Cordia Verbenacea Association Promotes a Synergistic Response in Antinociceptive Model and Improves the Anti-inﬂammatory Results in Animal Models. Biomed. Pharmacother. 112, 108693. doi:10.1016/j.biopha.2019.108693 Andrade-Silva, M., Correa, L. B., Candéa, A. L. P., Barbosa, H. S., Rosas, E. C., Henriques, M. G., et al. (2016). The Cannabinoid 2 Receptor Agonist β-Caryophyllene Modulates the Inﬂammatory Reaction Induced by Mycobacterium Bovis BCG by Inhibiting Neutrophil Migration. Inﬂamm. Res. 65, 869–879. doi:10.1007/s00011-016-0969-3 Basu, A., Sarkar, A., and Maulik, U. (2020). Molecular Docking Study of Potential Phytochemicals and Their Effects on the Complex of SARS-CoV2 Spike Protein and Human ACE2. Sci. Rep. 10, 10. doi:10.1038/s41598-020-74715-4 Anil, S. M., Namdar, D., Belausov, E., Shoval, I., Mani, K. A., Mechrez, G., et al. (2021). Cannabis Compounds Exhibit Anti-inﬂammatory Activity In Vitro in COVID-19-Related Inﬂammation in Lung Epithelial Cells and Pro- inﬂammatory Activity in Macrophages. Sci. Rep. 11, 11. doi:10.1038/s41598- 021-81049-2 Benarba, B., and Pandiella, A. (2020). Medicinal Plants as Sources of Active Molecules Against COVID-19. Front. Pharmacol. 11, 1189. doi:10.3389/ fphar.2020.01189 Bento, A. F., Marcon, R., Dutra, R. C., Cola, M., Pereira Leite, D. F., Calixto, J. B., et al. (2011). β-Caryophyllene Inhibits Dextran Sulfate Sodium-Induced Colitis in Mice Through CB2 Receptor Activation and PPARγ Pathway. Am. J. Pathol. 178, 1153–1166. doi:10.1016/j.ajpath.2010.11.052 Arizuka, N., Murakami, T., and Suzuki, K. (2017). The Effect of β-Caryophyllene on Nonalcoholic Steatohepatitis. J. Toxicol. Pathol. 30, 263–273. doi:10.1293/ tox.2017-0018 Arts, R. J. W., Wang, S.-Y., Oosting, M., Kumar, V., Xavier, R. J., Wijmenga, C., et al. (2018). BCG Vaccination Protects Against Experimental Viral Infection in Humans Through the Induction of Cytokines Associated With Trained Immunity. Cell Host Microbe 23, 89–100. doi:10.1016/j.chom.2017.12.010 Berger, G., Chinnadurai, A., Westhofen, R., Hagn, G., Cox, A., Kelly, M., et al. (2019). Experimental Cannabinoid 2 Receptor Activation by Phyto-Derived and Synthetic Cannabinoid Ligands in LPS-Induced Interstitial Cystitis in Mice. Molecules 24, 4239. doi:10.3390/molecules24234239 Asif, M., Saleem, M., Saadullah, M., Yaseen, H. S., and Al Zarzour, R. (2020). COVID-19 and Therapy With Essential Oils Having Antiviral, Anti- inﬂammatory, and Immunomodulatory Properties. Inﬂammopharmacology 28, 1153–1161. doi:10.1007/s10787-020-00744-0 Beserra-Filho, J. I. A., Santos, J. R., de Oliveira-Melo, A. J., Menezes, P. D. P., Duarte, M. C., de Souza Araújo, A. A., et al. (2019). Eplingiella Fruticosa Leaf Essential Oil Complexed With β-Cyclodextrin Produces a Superior neuroprotective and Behavioral Proﬁle in a Mice Model of Parkinson’s Disease. Food Chem. Toxicol. REFERENCES Coronopifolia Essential Oil against Antibiotic-Resistant Bacteria. Nat. Prod. Res. 29, 582–585. doi:10.1080/14786419.2014.954246 Ajiboye, B. O., Ojo, O. A., Fatoba, B., Afolabi, O. B., Olayide, I., Okesola, M. A., et al. (2019). In Vitro Antioxidant and Enzyme Inhibitory Properties of the n-Butanol Fraction of Senna podocarpa (Guill. And Perr.) Leaf. J. Physiol. Pharmacol. 31. doi:10.1515/jbcpp-2019-0123 Adefegha, S. A., Olasehinde, T. A., and Oboh, G. (2017). Essential Oil Composition, Antioxidant, Antidiabetic and Antihypertensive Properties of Two Afromomum Species. J. Oleo Sci. 66, 51–63. doi:10.5650/jos. ess16029 Alberti, T. B., Marcon, R., Bicca, M. A., Raposo, N. R. B., Calixto, J. B., and Dutra, R. C. (2014). Essential Oil from Pterodon emarginatus Seeds Ameliorates Experimental Autoimmune Encephalomyelitis by Alberti, T. B., Marcon, R., Bicca, M. A., Raposo, N. R. B., Calixto, J. B., and Dutra, R. C. (2014). Essential Oil from Pterodon emarginatus Seeds Ameliorates Experimental Autoimmune Encephalomyelitis by Ait Said, L., Zahlane, K., Ghalbane, I., El Messoussi, S., Romane, A., Cavaleiro, C., et al. (2015). Chemical Composition and Antibacterial Activity of Lavandula Ait Said, L., Zahlane, K., Ghalbane, I., El Messoussi, S., Romane, A., Cavaleiro, C., et al. (2015). Chemical Composition and Antibacterial Activity of Lavandula May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 16 Jha et al. Beta-Caryophyllene may be Useful in COVID-19 Modulating Th1/Treg Cell Balance. J. Ethnopharmacol. 155, 485–494. doi:10.1016/j.jep.2014.05.044 Autoimmune Encephalomyelitis (EAE). Int. Immunopharmacol. 77, 105923. doi:10.1016/j.intimp.2019.105923 Autoimmune Encephalomyelitis (EAE). Int. Immunopharmacol. 77, 105923. doi:10.1016/j.intimp.2019.105923 Astani, A., Reichling, J., and Schnitzler, P. (2011). Screening for Antiviral Activities of Isolated Compounds From Essential Oils. Evid.-Based Complementary Altern. Med. 2011, 1–8. doi:10.1093/ecam/nep187 Alberti, T. B., Barbosa, W., Vieira, J., Raposo, N., and Dutra, R. (2017). (-)-β-Caryophyllene, a CB2 Receptor-Selective Phytocannabinoid, Suppresses Motor Paralysis and Neuroinﬂammation in a Murine Model of Multiple Sclerosis. IJMS 18, 691. doi:10.3390/ijms18040691 Bahr, T., Allred, K., Martinez, D., Rodriguez, D., and Winterton, P. (2018). Effects of a Massage-like Essential Oil Application Procedure Using Copaiba and Deep Blue Oils in Individuals With Hand Arthritis. Complement. Therap. Clin. Pract. 33, 170–176. doi:10.1016/j.ctcp.2018.10.004 Alferink, J., Schmidt, K., Ruland, C., Lundt, R., Kemter, A., Dlugos, A., et al. (2016). Cannabinoid Receptor 2 Modulates Susceptibility to Experimental Cerebral Malaria through a CCL17-Dependent Mechanism. J. Biol. Chem. 291, 19517–19531. doi:10.1074/jbc.M116. 746594 Bahramsoltani, R., and Rahimi, R. (2020). An Evaluation of Traditional Persian Medicine for the Management of SARS-CoV-2. Front. Pharmacol. 11, 571434. REFERENCES 124, 17–29. doi:10.1016/j.fct. 2018.11.056 Askari, V. R., and Shaﬁee-Nick, R. (2019). Promising Neuroprotective Effects of β-Caryophyllene Against LPS-Induced Oligodendrocyte Toxicity: A Mechanistic Study. Biochem. Pharmacol. 159, 154–171. doi:10.1016/j.bcp. 2018.12.001 Boukhatem, M. N., and Setzer, W. N. (2020). Aromatic Herbs, Medicinal Plant- Derived Essential Oils, and Phytochemical Extracts as Potential Therapies for Coronaviruses: Future Perspectives. Plants 9, 800. doi:10.3390/ plants9060800 Askari, V. R., and Shaﬁee-Nick, R. (2019). The Protective Effects of β-Caryophyllene on LPS-Induced Primary Microglia M1/M2 Imbalance: A Mechanistic Evaluation. Life Sci. 219, 40–73. doi:10.1016/j.lfs.2018.12.059 Brand, Y. M., Roa-Linares, V. C., Betancur-Galvis, L. A., and Stashenko, E. (2015). Antiviral Activity of Colombian Labiatae and Verbenaceae Family Essential Oils and Monoterpenes on Human Herpes Viruses. J. Essent. Oil Res. 28, 130–137. doi:10.1080/10412905.2015.1093556 Askari, V. R., Baradaran Rahimi, V., Tabatabaee, S. A., and Shaﬁee-Nick, R. (2019). Combination of Imipramine, A Sphingomyelinase Inhibitor, and β-Caryophyllene Improve Their Therapeutic Effects on Experimental May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 17 Beta-Caryophyllene may be Useful in COVID-19 Jha et al. Choi, I. Y., Ju, C., Anthony Jalin, A. M. A., Prather, P. L., and Kim, W.-K. (2013). Activation of Cannabinoid CB2 Receptor-Mediated AMPK/CREB Pathway Reduces Cerebral Ischemic Injury. Am. J. Pathol. 182, 928–939. doi:10.1016/j. ajpath.2012.11.024 Brito, L. F., Oliveira, H. B. M., Júnior, M. N. S., Souza, E. P., Souza Nascimento, F., Amorim, A. T., et al. (2019). Anti-Inﬂammatory Activity of β-Caryophyllene Combined With Docosahexaenoic Acid in a Model of Sepsis Induced by Staphylococcus aureus in Mice. J. Sci. Food Agric. 99, 5870–5880. doi:10. 1002/jsfa.9861 Choi, I.-Y., Prather, P. L., and Kim, W.-K. (2013). Activation of Cannabinoid CB2 Receptor-Mediated AMPK/CREB Pathway Reduces Cerebral Ischemic Injury. Am. J. Pathol. 182, 928–939. doi:10.1016/j.ajpath.2012.11.024 Buckley, N. E., McCoy, K. L., Mezey, E., Bonner, T., Zimmer, A., Felder, C. C., et al. (2000). Immunomodulation by Cannabinoids is Absent in Mice Deﬁcient for the Cannabinoid CB(2) Receptor. Eur. J. Pharmacol. 396, 141–149. doi:10.1016/ s0014-2999(00)00211-9 Ciavarella, C., Motta, I., and Valente, S. (2020). Pharmacological (Or Synthetic) and Nutritional Agonists of PPAR-γ as Candidates for Cytokine Storm Modulation in COVID-19 Disease. Molecules 25, 2076. doi:10.3390/ molecules25092076 Byrareddy, S. N., and Mohan, M. (2020). SARS-CoV2 Induced Respiratory Distress: Can Cannabinoids Be Added to Anti-viral Therapies to Reduce Lung Inﬂammation? Brain Behav. Immun. 87, 120–121. doi:10.1016/j.bbi. 2020.04.079 Costantino, C. M., Yewdall, A. W., Devi, L. A., and Chen, B. K. (2012). REFERENCES Cannabinoid Receptor 2-mediated Attenuation of CXCR4-Tropic HIV Infection in Primary CD4+ T Cells. PLOS ONE 7, e33961. doi:10.1371/ journal.pone.0033961 Cabral, G. A., Ferreira, G. A., and Jamerson, M. J. (2015). Endocannabinoids and the Immune System in Health and Disease. Handbook Exp. Pharmacol. 231, 185–211. doi:10.1007/978-3-319-20825-1_6 Costiniuk, C. T., and Jenabian, M.-A. (2020). Acute Inﬂammation and Pathogenesis of SARS-CoV-2 Infection: Cannabidiol as a Potential Anti- inﬂammatory Treatment? Cytokine Growth Factor Rev. 53, 63–65. doi:10. 1016/j.cytogfr.2020.05.008 Cagno, V., Ballero, M., Civra, A., Donalisio, M., Bicchi, C., Lembo, D., et al. (2017). In Vitro Anti-Herpes Simplex Virus-2 Activity of Salvia desoleana Atzei & V. Picci Essential Oil. PLOS ONE 12, e0172322. doi:10.1371/journal.pone.0172322 Csóka, B., Rajesh, M., Federici, S., Deitch, E. A., Bátkai, S., Pacher, P., et al. (2009). CB2 Cannabinoid Receptors Contribute to Bacterial Invasion and Mortality in Polymicrobial Sepsis. PLOS ONE 4, e6409. doi:10.1371/journal. pone.0006409 Çakır, M., Tekin, S., Okan, A., Çakan, P., and Do˘ganyi˘git, Z. (2020). The Ameliorating Effect of Cannabinoid Type 2 Receptor Activation on Brain, Lung, Liver and Heart Damage in Cecal Ligation and Puncture-Induced Sepsis Model in Rats. Int. Immunopharmacol. 78, 105978. doi:10.1016/j.intimp.2019. 105978 D’Ascola, A., Irrera, N., Ettari, R., Bitto, A., Pallio, G., Mannino, F., et al. (2019). Exploiting Curcumin Synergy With Natural Products Using Quantitative Analysis of Dose-Effect Relationships in an Experimental In Vitro Model of Osteoarthritis. Front. Pharmacol. 10, 1347. doi:10.3389/fphar.2019.01347 Calleja, M. A., Vieites, J. S., Montero-Meléndez, T., Torres, M. I., Faus, M. J., Gil, A., et al. (2013). The Antioxidant Effect of β-Caryophyllene Protects Rat Liver From Carbon Tetrachloride-Induced Fibrosis by Inhibiting Hepatic Stellate Cell Activation. Br. J. Nutr. 109, 394–401. doi:10.1017/S0007114512001298 da Silva, J. K. R., Figueiredo, P. L. B., Byler, K. G., and Setzer, W. N. (2020). Essential Oils as Antiviral Agents, Potential of Essential Oils to Treat SARS-CoV-2 Infection: An In-Silico Investigation. IJMS 21, 3426. doi:10.3390/ijms21103426 Oils as Antiviral Agents, Potential of Essential Oils to Treat SARS-CoV-2 Infection: An In-Silico Investigation. IJMS 21, 3426. doi:10.3390/ijms21103426 Campos, M. I., CamposVieira, C. N., and Aarestrup, B. J. V. (2015). Atorvastatin and Trans-Caryophyllene for the Prevention of Leukopenia in an Experimental Chemotherapy Model in Wistar Rats. Mol. Clin. Oncol. 3, 825–828. doi:10.3892/ mco.2015.544 Dahham, S., Tabana, Y., Iqbal, M., Ezzat, M., Majid, A., Majid, A., et al. (2015). The Anticancer, Antioxidant and Antimicrobial Properties of the Sesquiterpene β-Caryophyllene From the Essential Oil of Aquilaria crassna. Molecules 20, 11808–11829. REFERENCES doi:10.3390/molecules200711808 Campos, C., de Castro, A. L., Ortiz, V. D., Barboza, T. E., Pereira, C., Apel, M., et al. (2017). Effect of Free and Nanoencapsulated Copaiba Oil on Monocrotaline- Induced Pulmonary Arterial Hypertension. J. Cardiovasc. Pharmacol. 69, 79–85. doi:10.1097/FJC.0000000000000442 D’Amico, F., Baumgart, D. C., Danese, S., and Peyrin-Biroulet, L. (2020). Diarrhea During COVID-19 Infection: Pathogenesis, Epidemiology, Prevention, and Management. Clin. Gastroenterol. Hepatol. 18, 1663–1672. doi:10.1016/j.cgh. 2020.04.001 Capettini, L. S. A., da SilvaSavergnini, R. F., Santos, R. A. S., Mach, F., and Montecucco, F. (2012). Update on the Role of Cannabinoid Receptors After Ischemic Stroke. Mediators Inﬂamm. 2012, 1–8. doi:10.1155/2012/824093 De, J., Lu, Y., Ling, L., Peng, N., and Zhong, Y. (2017). Essential Oil Composition and Bioactivities of Waldheimia glabra (Asteraceae) from Qinghai-Tibet Plateau. Molecules 22, 460. doi:10.3390/molecules22030460 Carvalho, C. E. S., Carvalho, T. P., Moura, A. K. S., Rodrigues, K. A. F., Carneiro, S. M. P., Arcanjo, D. D. R., et al. (2017). Anti-Leishmania Activity of Essential Oil of Myracrodruon urundeuva (Engl.) Fr. All.: Composition, Cytotoxity and Possible Mechanisms of Action. Exp. Parasitol. 175, 59–67. doi:10.1016/j. exppara.2017.02.012 de Oliveira, M. R. C., da Silva, B. A. F., Oliveira Brito Pereira Bezerra, A., Oliveira Tintino, C. D. M., Afonso Pereira de Oliveira, V., Boligon, A. A., et al. (2019). GC-MS Chemical Characterization and In Vitro Evaluation of Antioxidant and Toxic Effects Using Drosophila melanogaster Model of the Essential Oil of Lantana montevidensis (Spreng) Briq. Medicina 55, 194. doi:10.3390/ medicina55050194 Chávez-Hurtado, P., González-Castañeda, R. E., and Viveros-Paredes, J. M. (2020). β-Caryophyllene Reduces DNA Oxidation and the Overexpression of Glial Fibrillary Acidic Protein in the Prefrontal Cortex and Hippocampus of D-Galactose-Induced Aged BALB/c Mice. J. Med. Food 23, 515–522. doi:10. 1089/jmf.2019.0111 de Pinho, J. P., Silva, A., Pinheiro, B., de Carvalho Bayma, J., Lahlou, S., da Cunha Sousa, P., et al. (2012). Antinociceptive and Antispasmodic Effects of the Essential Oil of Ocimum micranthum: Potential Anti-inﬂammatory Properties. Planta Med. 78, 681–685. doi:10.1055/s-0031-1298372 Cheng, Y., Dong, Z., and Liu, S. (2014). β-Caryophyllene Ameliorates the Alzheimer-Like Phenotype in APP/PS1 Mice Through CB2 Receptor Activation and the PPARγ Pathway. Pharmacology 94, 1–12. doi:10.1159/ 000362689 Dhama, K., Tiwari, R., Rana, R., Khurana, S. K., Khan, R. U., Alagawany, M., et al. (2018). Medicinal and Therapeutic Potential of Herbs and Plant Metabolites/ Extracts Countering Viral Pathogens - Current Knowledge and Future Prospects. Curr. Drug. Metab. 19, 236–263. REFERENCES Immunoregulatory Role of Cannabinoids During Infectious Disease. Neuroimmunomodulation 24, 183–199. doi:10.1159/000481824 Horváth, B., Mukhopadhyay, P., Kechrid, M., Tanchian, G., Wink, D. A., Gertsch, Esposito, G., Lu, J., Corpetti, C., and Sarnelli, G. (2020). The Potential of Cannabidiol in the COVID-19 Pandemic. Br. J. Pharmacol. 177, 4967–4970. doi:10.1111/bph.15157 Horváth, B., Mukhopadhyay, P., Kechrid, M., Tanchian, G., Wink, D. A., Gertsch, J., et al. (2012). β-Caryophyllene Ameliorates Cisplatin-Induced nephrotoxicity in a Cannabinoid 2 Receptor-dependent Manner. Free Radic. Biol. Med. 52, 1325–1333. doi:10.1016/j.freeradbiomed.2012.01.014 Fanelli, V., Stallone, G., Ronco, C., and Castellano, G. (2020). Acute Kidney Injury in SARS-CoV-2 Infected Patients. Crit. Care 24, 155. doi:10.1186/s13054-020- 02872-z Howlett, A. C., and Abood, M. E. (2017). CB 1 and CB 2 Receptor Pharmacology. Adv. Pharmacol. 80, 169–206. doi:10.1016/bs.apha.2017.03.007 Feng, G., Zheng, K. I., Yan, Q.-Q., Targher, G., Byrne, C. D., Poucke, S. V., et al. (2020). COVID-19 and Liver Dysfunction: Current Insights and Emergent Therapeutic Strategies. J. Clin. translational Hepatol. 8, 1–7. doi:10.14218/ JCTH.2020.00018 Huang, S., Jiang, L., Zhang, R., Peebles, R. S., Mack, M., Kaplan, M. H., et al. (2019). PPAR-γ in Macrophages Limits Pulmonary Inﬂammation and Promotes Host Recovery Following Respiratory Viral Infection. J. Virol. 93, e00019–e00030. doi:10.1128/JVI.00030-19 Fontes, L. B. A., Dias, D., Aarestrup, B. J. V., Da Silva Filho, A. A., and Corrˆea, J. (2017). β-Caryophyllene Ameliorates the Development of Experimental Autoimmune Encephalomyelitis in C57BL/6 Mice. Biomed. Pharmacother. 91, 257–264. doi:10.1016/j.biopha.2017.04.092 Huang, C., Wang, Y., Li, X., Ren, L., Zhao, J., Hu, Y., et al. (2020). Clinical Features of Patients Infected With 2019 Novel Coronavirus in Wuhan, China. Lancet 395, 497–506. doi:10.1016/S0140-6736(20)30183-5 Huang, X., Wei, F., and Wen, L. (2020). Epidemiology and Clinical Characteristics of COVID-19. Arch. Iran Med. 23, 268–271. doi:10.34172/aim.2020.09 Friedman, S. L., Neuschwander-Tetri, B. A., and Rinella, M. (2018). Mechanisms of NAFLD Development and Therapeutic Strategies. Nat. Med. 24, 908–922. doi:10.1038/s41591-018-0104-9 Hwang, E.-S., and Kim, G. H. (2013). Safety Evaluation of Chrysanthemum indicum L. Flower Oil by Assessing Acute Oral Toxicity, Micronucleus Abnormalities, and Mutagenicity. Prev. Nutr. Food Sci. 18, 111–116. doi:10. 3746/pnf.2013.18.2.111 Fukuoka, K., Okuda, H., Kitayama, T., Shirai, M., Komai, K., Komemushi, S., et al. (2004). Inhibitory Actions of Several Natural Products on Proliferation of Rat Vascular Smooth Muscle Cells Induced by Hsp60 from Chlamydia pneumoniae J138. J. Agric. Food Chem. 52, 6326–6329. doi:10.1021/jf0351164 Hwang, E.-S., Kim, H.-B., Han, G., Han, S.-Y., Lee, E.-A., Yoon, J.-H., et al. (2020). REFERENCES doi:10.2174/ 1389200219666180129145252 Cheng, Y., Wang, Z., Dong, L., Li, J., Yao, Y., Ge, S., et al. (2020). Kidney Disease is Associated With In-Hospital Death of Patients With COVID-19. Kidney Int. 97, 829–838. doi:10.1016/j.kint.2020.03.005 Dhama, K., Patel, S. K., Tiwari, R., Malik, Y. S., Singh, R., Sah, R., et al. (2020). An Update on SARS-CoV-2/COVID-19 With Particular Reference to its Clinical Pathology, Pathogenesis, Immunopathology and Mitigation Strategies. Trav. Med. Infect. Dis. 37, 101755. doi:10.1016/j.tmaid.2020.101755 Cho, J. Y., Chang, H.-J., Lee, S.-K., Hwang, J.-K., and Chun, H. S. (2007). Amelioration of Dextran Sulfate Sodium-Induced Colitis in Mice by Oral Administration of β-caryophyllene, a Sesquiterpene. Life Sci. 80, 932–939. doi:10.1016/j.lfs.2006.11.038 Cho, H.-I., Hong, J.-M., Hwan Kwak, J., Lee, D.-U., Kook Lee, S., Lee, S.-M., et al. (2015). β-Caryophyllene Alleviates D-Galactosamine and Lipopolysaccharide- Induced Hepatic Injury Through Suppression of the TLR4 and RAGE Signaling Pathways. Eur. J. Pharmacol. 764, 613–621. doi:10.1016/j.ejphar.2015.08.001 Di Giacomo, S., Mazzanti, G., and Di Sotto, A. (2016). Mutagenicity of Cigarette Butt Waste in the Bacterial Reverse Mutation Assay: The Protective Effects of β-Caryophyllene and β-Caryophyllene Oxide. Environ. Toxicol. 31, 1319–1328. doi:10.1002/tox.22136 Cho, J. Y., Kim, H. Y., Kim, S.-K., Lee, H. J., and Chun, H. S. (2015). β-Caryophyllene Attenuates Dextran Sulfate Sodium-Induced Colitis in Mice via Modulation of Gene Expression Associated Mainly with Colon Inﬂammation. Toxicol. Rep. 2, 1039–1045. doi:10.1016/j.toxrep.2015.07.018 Di Sotto, A., Mazzanti, G., Carbone, F., Hrelia, P., and Maffei, F. (2010). Inhibition by β-Caryophyllene of Ethyl Methanesulfonate-Induced Clastogenicity in Cultured Human Lymphocytes. Mutat. Res. 699, 23–28. doi:10.1016/j. mrgentox.2010.04.008 May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 18 Beta-Caryophyllene may be Useful in COVID-19 Jha et al. Hammad, F. T., Ojha, S., Azimullah, S., and Lubbad, L. (2018). Does β-Caryophyllene Protect Against Renal Dysfunction Following Ischemia- Reperfusion Injury in the Rat? Int. J. Physiol. Pathophysiol. Pharmacol. 10, 163–171. Ding, X., Chen, M., Kang, Y., Lou, J., and Liu, Z. (2020). Dynamic Proﬁle and Clinical Implications of Hematological Parameters in Hospitalized Patients With Coronavirus Disease 2019. Clin. Chem. Lab. Med. 58, 1365–1371. doi:10. 1515/cclm-2020-0411 Diniz, L. R. L., Perez-Castillo, H. A., and de Sousa, D. P. (2021). Bioactive Terpenes and Their Derivatives as Potential SARS-CoV-2 Proteases Inhibitors From Molecular Modeling Studies. Biomolecules 11, 74. doi:10.3390/biom11010074 Hammami, S., Faidi, K., Dhaouadi, H., Aouni, M., Aouni, M., and Joshi, R. (2015). Essential Oil Composition, Antioxidant, Cytotoxic and Antiviral Activities of Teucrium pseudochamaepitys Growing Spontaneously in Tunisia. REFERENCES Molecules 20, 20426–20433. doi:10.3390/molecules201119707 Dong, X., Cao, Y., Lu, X., Du, H., Yan, Y., Akdis, C. A., et al. (2020). Eleven Faces of Coronavirus Disease 2019. Allergy 75, 1699–1709. doi:10.1111/all.14289 Harb, A. A., Bustanji, Y. K., and Abdalla, S. S. (2018). Hypocholesterolemic Effect of β-Caryophyllene in Rats Fed Cholesterol and Fat Enriched Diet. J. Clin. Biochem. Nutr. 62, 230–237. doi:10.3164/jcbn.17-3 Du, L., Ma, Y., Liu, M., and Tang, H. (2017). Peroxisome Proliferators Activated Receptor (PPAR) Agonists Activate Hepatitis B Virus Replication In Vivo. Virol. J. 14, 96. doi:10.1186/s12985-017-0765-x Hassanin, O., Abdallah, F., and Gallal, A. A. (2020). In Vitro and In Vivo Experimental Trials to Assess the Modulatory Inﬂuence of β-Caryophyllene on NDV Replication and Immunopathogenesis. Comp. Immunol. Microbiol. Infect. Dis. 73, 101547. doi:10.1016/j.cimid.2020.101547 Duschatzky, C. B., Possetto, M. L., Talarico, L. B., Michis, F., Almeida, N. V., de Lampasona, M. P., et al. (2005). Evaluation of Chemical and Antiviral Properties of Essential Oils from South American Plants. Antivir. Chem. Chemother. 16, 247–251. doi:10.1177/095632020501600404 He, Q., Zhan, J., and Zhang, Z. (2019). Cannabinoid Receptor 2: A Potential Novel Therapeutic Target for Sepsis? Acta Clin. Belg. 74, 70–74. doi:10.1080/ 17843286.2018.1461754 Effenberger, M., Nairz, M., Seifert, M., Hilbe, R., Seiwald, S., Scholl-Buergi, S., et al. (2020). Faecal Calprotectin Indicates Intestinal Inﬂammation in COVID-19. Gut 69, 1543–1544. doi:10.1136/gutjnl-2020-321388 Hecker, M., Behnk, R. E., Vadász, I., Herold, S., Seeger, W., and Mayer, K. (2015). PPAR-α Activation Reduced LPS-Induced Inﬂammation in Alveolar Epithelial Cells. Exp. Lung Res. 41, 393–403. doi:10.3109/01902148.2015.1046200 Hecker, M., Behnk, R. E., Vadász, I., Herold, S., Seeger, W., and Mayer, K. (2015). PPAR-α Activation Reduced LPS-Induced Inﬂammation in Alveolar Epithelial El-Hosseiny, L. S., Alqurashy, N. N., and Sheweita, S. A. (2016). Oxidative Stress Alleviation by Sage Essential Oil in Co-Amoxiclav Induced Hepatotoxicity in Rats. Int. J. Biomed. Sci. 12, 71–78. Hensel, A., Kayser, O., Hempel, G., and Kraft, K. (2020). Challenges at the Time of COVID-19: Opportunities and Innovations in Antivirals From Nature. Planta Med. 86, 659–664. doi:10.1055/a-1177-4396 El-Sheikh, S. M. A., Abd El-Alim, A. E.-A. F., Galal, A. A. A., El-Sayed, R. G., and El-Naseery, N. I. (2019). Anti-Arthritic Effect of β-Caryophyllene and its Ameliorative Role on Methotrexate And/or Leﬂunomide-Induced Side Effects in Arthritic Rats. Life Sci. 233, 116750. doi:10.1016/j.lfs.2019. 116750 Hernández-Cervantes, R., Méndez-Díaz, M., Prospéro-García, O., and Morales- Montor, J. (2017). Immunoregulatory Role of Cannabinoids During Infectious Disease. Neuroimmunomodulation 24, 183–199. doi:10.1159/000481824 Montor, J. (2017). REFERENCES Gastrointestinal and Liver Manifestations in Patients With COVID-19. J. Chin. Med. Assoc. 83, 521–523. doi:10.1097/JCMA.0000000000000319 Li, H., Wang, D., Chen, Y., and Yang, M. (2020). β-Caryophyllene Inhibits High Glucose-Induced Oxidative Stress, Inﬂammation and Extracellular Matrix Accumulation in Mesangial Cells. Int. Immunopharmacol. 84, 106556. doi:10.1016/j.intimp.2020.106556 Jiménez-Ruiz, A., García-Grimshaw, M., and Ruiz-Sandoval, J. L. (2020). Neurological Manifestations of COVID-19. Gac. Med. Mex. 156, 156. doi:10. 24875/GMM.M20000380 Jin, Y., Chen, S., Zhang, W., and Duan, G. (2020). Virology, Epidemiology, Pathogenesis, and Control of COVID-19. Viruses 12, 372. doi:10.3390/ v12040372 Lindsey, L. P., Abney, S. E., Uchakina, O. N., Khusial, R. D., Akil, A., and Murnane, K. S. (2019). The Cannabinoid Receptor 2 Agonist, β-Caryophyllene, Improves Working Memory and Reduces Circulating Levels of Speciﬁc Proinﬂammatory Cytokines in Aged Male Mice. Behav. Brain Res. 372, 112012. doi:10.1016/j.bbr. 2019.112012 Kamikubo, R., Kai, K., Tsuji-Naito, K., and Akagawa, M. (2016). β-Caryophyllene Attenuates Palmitate-Induced Lipid Accumulation Through AMPK Signaling by Activating CB2 Receptor in Human HepG2 Hepatocytes. Mol. Nutr. Food Res. 60, 2228–2242. doi:10.1002/mnfr.201600197 Liu, M. W., Su, M., Wang, Y., Qin, L., Liu, X., Tian, M., et al. (2014). Effect of Melilotus Extract on Lung Injury by Upregulating the Expression of Cannabinoid CB2 Receptors in Septic Rats. BMC Complement. Altern. Med. 14, 94. doi:10.1186/1472-6882-14-94 Kamyab, A., and Eshraghian, A. (2013). Anti-Inﬂammatory, Gastrointestinal and Hepatoprotective Effects of Ocimum sanctum Linn: An Ancient Remedy With New Application. Inﬂamm. Allergy Drug Targets 12, 378–384. doi:10.2174/ 1871528112666131125110017 Liu, A. P., Yuan, Q. H., He, Q. W., Chen, K., Liu, Q. S., Li, Z., et al. (2020). Cannabinoid Receptor 2 Activation Alleviates Septic Lung Injury by Promoting Autophagy via Inhibition of Inﬂammatory Mediator Release. Cell Signal. 69, 109556. doi:10.1016/j.cellsig.2020.109556 Kandeel, M., and Al-Nazawi, M. (2020). Virtual Screening and Repurposing of FDA Approved Drugs against COVID-19 Main Protease. Life Sci. 251, 117627. doi:10.1016/j.lfs.2020.117627 Kapellos, T. S., Hussain, M. T., Rainger, G. E., Greaves, D. R., and Iqbal, A. J. (2019). Cannabinoid Receptor 2 Deﬁciency Exacerbates Inﬂammation and Neutrophil Recruitment. FASEB J. 33, 6154–6167. doi:10.1096/fj.201802524R Loizzo, M. R., Lampronti, I., Menichini, F., Gambari, R., Cinatl, J., and Doerr, H. W. (2008). Phytochemical Analysis and In Vitro Evaluation of the Biological Activity Against Herpes Simplex Virus Type 1 (HSV-1) of Cedrus libani A. Rich. Phytomedicine 15, 79–83. doi:10.1016/j.phymed.2007.03.013 Karmaus, P. W., Chen, W., Kaplan, B. L., and Kaminski, N. E. (2012). REFERENCES Δ9- Tetrahydrocannabinol Suppresses Cytotoxic T Lymphocyte Function Independent of CB1 and CB2, Disrupting Early Activation Events. J. Neuroimmune Pharmacol. 7, 843–855. doi:10.1007/s11481-011-9293-4 Lorençoni, M. F., Figueira, M. M., Endringer, D. C., Scherer, R., Barth, T., Vilela Bertolucci, S. K., et al. (2020). Chemical Composition and Anti-inﬂammatory Activity of Essential Oil and Ethanolic Extract of Campomanesia phaea (O. Berg.) Landrum Leaves. J. Ethnopharmacol. 252, 112562. doi:10.1016/j.jep.2020. 112562 Keam, S., Megawati, S. K., Dhama, K., and Harapan, H. (2020). Immunopathology and Immunotherapeutic Strategies in Severe Acute Respiratory Syndrome Coronavirus 2 Infection. Rev. Med. Virol. 30, e2123. doi:10.1002/rmv.2123 Lou, J., Cao, G., Li, R., Liu, J., Dong, Z., and Xu, L. (2016). β-Caryophyllene Attenuates Focal Cerebral Ischemia-Reperfusion Injury by Nrf2/HO-1 Pathway in Rats. Neurochem. Res. 41, 1291–1304. doi:10.1007/s11064-016-1826-z Kelany, M. E., and Abdallah, M. A. (2016). Protective Effects of Combined β-Caryophyllene and Silymarin Against Ketoprofen-Induced Hepatotoxicity in Rats. Can. J. Physiol. Pharmacol. 94, 739–744. doi:10.1139/cjpp-2015-0607 Lou, J., Teng, Z., Zhang, L., Ma, L., Wang, F., Tian, X., et al. (2017). β-Caryophyllene/Hydroxypropyl-β-Cyclodextrin Inclusion Complex Improves Cognitive Deﬁcits in Rats With Vascular Dementia Through the Cannabinoid Receptor Type 2 -Mediated Pathway. Front. Pharmacol. 8, 2. doi:10.3389/fphar.2017.00002 Khodadadi, H., Salles, É. L., Costigliola, V., Yu, J. C., Vaibhav, K., Hess, D. C., et al. (2020). Cannabidiol Modulates Cytokine Storm in Acute Respiratory Distress Syndrome Induced by Simulated Viral Infection Using Synthetic RNA. Cannabis Cannabinoid Res. 5, 197–201. doi:10.1089/can.2020.0043 Mahmoud, M. F., El Swefy, S., Hasan, R. B., and Ibrahim, A. (2014). Role of Cannabinoid Receptors in Hepatic Fibrosis and Apoptosis Associated With Bile Duct Ligation in Rats. Eur. J. Pharmacol. 742, 118–124. doi:10.1016/j.ejphar. 2014.08.021 Khomich, O., Kochetkov, S., Bartosch, B., and Ivanov, A. V. (2018). Redox Biology of Respiratory Viral Infections. Viruses 10, 392. doi:10.3390/v10080392 Kim, D. W., Seo, K. H., Oh, J.-W., Cho, J. K., Lee, K. H., and Park, K. H. (2014). Phenolic Phytochemical Displaying SARS-CoV Papain-Like Protease Inhibition From the Seeds of Psoralea corylifolia. J. Enzyme Inhib. Med. Chem. 29, 59–63. doi:10.3109/14756366.2012.753591 Mahmoud, M. F., Swefy, S. E., and Hasan, R. A. (2014). Role of Cannabinoid Receptors in Hepatic Fibrosis and Apoptosis Associated With Bile Duct Ligation in Rats. Eur. J. Pharmacol. 742, 118–124. doi:10.1016/j.ejphar.2014. 08.021 Klauke, A.-L., Zimmer, A. M., Gertsch, J., and Zimmer, A. (2014). The Cannabinoid CB2 Receptor-Selective Phytocannabinoid Beta-Caryophyllene Exerts Analgesic Effects in Mouse Models of Inﬂammatory and Neuropathic Pain. Eur. Neuropsychopharmacol. 24, 608–620. doi:10.1016/j.euroneuro.2013. REFERENCES Antidepressant-Like Effects of β-Caryophyllene on Restraint Plus Stress- Induced Depression. Behav. Brain Res. 380, 112439. doi:10.1016/j.bbr.2019. 112439 J138. J. Agric. Food Chem. 52, 6326–6329. doi:10.1021/jf0351164 García, C. C., Duschatzky, C., and Damonte, E. B. (2003). Virucidal Activity of Essential Oils From Aromatic Plants of San Luis, Argentina. Phytother. Res. 17, 1073–1075. doi:10.1002/ptr.1305 Imai, Y., Kuba, G. G., Perkmann, T., van Loo, G., Ermolaeva, M., Veldhuizen, R., et al. (2008). Identiﬁcation of Oxidative Stress and Toll-like Receptor 4 Signaling as a Key Pathway of Acute Lung Injury. Cell 133, 235–249. doi:10. 1016/j.cell.2008.02.043 García, L. F. (2020). Immune Response, Inﬂammation, and the Clinical Spectrum of COVID-19. Front. Immunol. 11, 1441. doi:10.3389/ﬁmmu.2020.01441 Geddo, F., Querio, G., Maffei, M. E., Bovolin, P., and Gallo, M. P. (2019). PipeNig- FL, A Fluid Extract of Black Pepper (Piper Nigrum L.) With a High Standardized Content of Trans-β-Caryophyllene, Reduces Lipid Accumulation in 3T3-L1 Preadipocytes and Improves Glucose Uptake in C2C12 Myotubes. Nutrients 11, 2788. doi:10.3390/nu11112788 Irrera, N., Mazzon, E., Mannino, F., Squadrito, V., Arcoraci, V., Minutoli, L., et al. (2019). β-Caryophyllene Mitigates Collagen Antibody Induced Arthritis (CAIA) in Mice Through a Cross-Talk Between CB2 and PPAR-γ Receptors. Biomolecules 9, 326. doi:10.3390/biom9080326 Gertsch, J., Leonti, M., Raduner, S., Chen, J.-Z., Xie, X.-Q., Altmann, K.-H., et al. (2008). Beta-caryophyllene is a Dietary Cannabinoid. Proc. Natl. Acad. Sci. 105, 9099–9104. doi:10.1073/pnas.0803601105 Irrera, N., Mannino, F., Arcoraci, V., Rottura, M., Ieni, A., Minutoli, L., et al. (2020). β-Caryophyllene Inhibits Cell Proliferation through a Direct Modulation of CB2 Receptors in Glioblastoma Cells. Cancers 12, 1038. doi:10.3390/ cancers12041038 Gertsch, J. (2008). Antiinﬂammatory Cannabinoids in Diet - Towards a Better Understanding of CB2 Receptor Action? Commun. Integr. Biol. 1, 26–28. doi:10.4161/cib.1.1.6568 Javed, H., Azimullah, S., and Haque, M. E. (2016). Cannabinoid Type 2 (CB2) Receptors Activation Protects Against Oxidative Stress and Neuroinﬂammation Associated Dopaminergic Neurodegeneration in Rotenone Model of Parkinson’s Disease. Front. Neurosci. 10, 321. doi:10. 3389/fnins.2016.00321 Goodridge, H. S., Ahmed, S. S., Curtis, N., Levy, O., Netea, M. G., Pollard, A. J., et al. (2016). Harnessing the Beneﬁcial Heterologous Effects of Vaccination. Nat. Rev. Immunol. 16, 392–400. doi:10.1038/nri.2016.43 May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 19 Jha et al. Beta-Caryophyllene may be Useful in COVID-19 Jean, S. S., and Hsueh, P.-R. (2020). Old and Re-Purposed Drugs for the Treatment of COVID-19. Expert Rev. Anti-Infect. Ther. 18, 843–847. doi:10.1080/ 14787210.2020.1771181 Lee, I.-C., Huo, T.-I., and Huang, Y.-H. (2020). REFERENCES Sesquiterpene Rich Essential Oil From Nepalese bael Tree (Aegle Marmelos (L.) Correa) as Potential Antiproliferative Agent. Fitoterapia 138, 104266. doi:10.1016/j. ﬁtote.2019.104266 Munro, S., Thomas, K. L., and Abu-S, M. (1993). Molecular Characterization of a Peripheral Receptor for Cannabinoids. Nature 365, 61–65. doi:10.1038/ 365061a0 Parisotto-Peterle, J., Falkembach, M. C., da Silva Marques, M., Horn, A. P., dos Santos, M. K., da Veiga, V. F., et al. (2020). Healing Activity of Hydrogel Containing nanoemulsiﬁed β-Caryophyllene. Eur. J. Pharm. Sci. 148, 105318. doi:10.1016/j.ejps.2020.105318 Muthuramalingam, P., Karthika, C., Gowrishankar, S., Pandian, S. K., Ramesh, M., and Chen, J.-T. (2020). Global Multi-Omics and Systems Pharmacological Strategy Unravel the Multi-Targeted Therapeutic Potential of Natural Bioactive Molecules Against COVID-19: An In Silico Approach. Genomics 112, 4486–4504. doi:10.1016/j.ygeno.2020.08.003 Park, J.-Y., Yuk, H. J., Ryu, H. W., Kim, K. S., Park, K. H., Ryu, Y. B., et al. (2017). Evaluation of Polyphenols from Broussonetia papyrifera as Coronavirus Protease Inhibitors. J. Enzyme Inhib. Med. Chem. 32, 504–512. doi:10.1080/ 14756366.2016.1265519 Nagarkatti, P., Miranda, K., and Nagarkatti, M. (2020). Use of Cannabinoids to Treat Acute Respiratory Distress Syndrome and Cytokine Storm Associated With Coronavirus Disease-2019. Front. Pharmacol. 11, 589438. doi:10.3389/ fphar.2020.589438 Pastor, F. P., Isorna Folgar, M., Carvalho, N., Carvalho, F., Horcajadas, F. A., et al. (2020). Therapeutic Cannabis and COVID-19: Between Opportunism and Infoxication. Adicciones 32, 167–172. doi:10.20882/adicciones.1603 Nagoor Meeran, M. F., Sharma, C., and Ojha, S. (2020). CB2 Receptor-Selective Agonists as Candidates for Targeting Infection, Inﬂammation, and Immunity in SARS-CoV -2 Infections. Drug Dev. Res. 82, 7–11. doi:10. 1002/ddr.21752 Patra, K. C., Singh, B., and Pareta, S. (2010). A Validated HPTLC Method for Determination Oftrans-Caryophyllene from Polyherbal Formulations. Nat. Prod. Res. 24, 1933–1938. doi:10.1080/14786419.2010.497147 Patra, T., Sasaki, R., Meyer, K., Ray, R. B., and Ray, R. (2019). Transforming Growth Factor β Acts as a Regulatory Molecule for Lipogenic Pathways Among Hepatitis C Virus Genotype-Speciﬁc Infections. J. Virol. 93, e00811–e00819. doi:10.1128/JVI.00811-19 Narkhede, R. R., Pise, A. V., and Cheke, R. S. (2020). Recognition of Natural Products as Potential Inhibitors of COVID-19 Main Protease (Mpro): In-Silico Evidences. Nat. Prod. Bioprospect. 10, 297–306. doi:10.1007/s13659-020- 00253-1 Picciolo, G., Oteri, G., Irrera, N., and Squadrito, F. (2020). β-Caryophyllene Reduces the Inﬂammatory Phenotype of Periodontal Cells by Targeting CB2 Receptors. Biomedicines 8, 164. doi:10.3390/biomedicines8060164 Nasi, A., Melief, C., Rockberg, J., Arens, R., Kouretas, D., Sjölin, J., et al. (2020). REFERENCES 19, 123. doi:10.1186/s12906-019-2538-0 Mira, J. C., Gentile, L. F., Brakenridge, S. C., Mohr, A. M., Moore, F. A., and Moldawer, L. L. (2017). Sepsis Pathophysiology, Chronic Critical Illness, and Persistent Inﬂammation-Immunosuppression and Catabolism Syndrome. Crit. Care Med. 45, 253–262. doi:10.1097/CCM.0000000000002074 Oláh, A., Szekanecz, Z., and Bíró, T. (2017). Targeting Cannabinoid Signaling in the Immune System: "High"-Ly Exciting Questions, Possibilities, and Challenges. Front. Immunol. 8, 1487. doi:10.3389/ﬁmmu.2017.01487 Mohamed, W. A., Abd-Elhakim, Y. M., and Farouk, S. M. (2016). Protective Effects of Ethanolic Extract of Rosemary Against Lead-Induced Hepato-Renal Damage in Rabbits. Exp. Toxicol. Pathol. 68, 451–461. doi:10.1016/j.etp.2016.07.003 Oliveira-Tintino, C. D. M., Pessoa, R. T., Fernandes, M. N. M., da Silva, B. A. F., de Oliveira, M. R. C., Tintino, S. R., et al. (2018). Anti-inﬂammatory and Anti- edematogenic Action of the Croton campestris A. St.-Hil (Euphorbiaceae) Essential Oil and the Compound β-Caryophyllene in In Vivo Models. Phytomedicine 41, 82–95. doi:10.1016/j.phymed.2018.02.004 Mohamed, K., Sharifudin, S. M., and Duduku, K. (2018). NiO nanoparticles Induce Cytotoxicity Mediated through ROS Generation and Impairing the Antioxidant Defense in the Human Lung Epithelial Cells (A549): Preventive Effect of Pistacia lentiscus Essential Oil. Toxicol. Rep. 5, 480–488. doi:10.1016/j. toxrep.2018.03.012 y j y O’Sullivan, S. E. (2016). An Update on PPAR Activation by Cannabinoids. Br. J. Pharmacol. 173, 1899–1910. doi:10.1111/bph.13497 Pandey, A., Khan, M. K., Hamurcu, M., and Gezgin, S. (2020). Natural Plant Products: A Less Focused Aspect for the COVID-19 Viral Outbreak. Front. Plant Sci. 11, 568890. doi:10.3389/fpls.2020.568890 Mohammed, A., Singh, N., Cai, G., Putluri, N., Nagarkatti, P., and Nagarkatti, M. (2020). Δ9-Tetrahydrocannabinol Prevents Mortality From Acute Respiratory Distress Syndrome Through the Induction of Apoptosis in Immune Cells, Leading to Cytokine Storm Suppression. IJMS 21, 6244. doi:10.3390/ ijms21176244 Panfoli, I. (2020). Potential Role of Endothelial Cell Surface Ectopic Redox Complexes in COVID-19 Disease Pathogenesis. Clin. Med. 20, e146–e147. doi:10.7861/clinmed.2020-0252 Mondal, P., Natesh, J., Salam, A., and Meeran, S. M. (2020). Traditional Medicinal Plants Against Replication, Maturation and Transmission Targets of SARS- CoV-2: Computational Investigation. J. Biomol. Struct. Dyn., 1–18. doi:10.1080/ 07391102.2020.1842246 Pant, A., Akhoon, B. A., and Pandey, R. (2014). Beta-Caryophyllene Modulates Expression of Stress Response Genes and Mediates Longevity in Caenorhabditis elegans. Exp. Gerontol. 57, 81–95. doi:10.1016/j.exger.2014.05.007 Moussaoui, N. E., Sanchez, G., and Khay, E. O. (2013). Antibacterial and Antiviral Activities of Essential Oils of Northern Moroccan Plants. Biotechnol. J. Int. 3 (3a), 318–331. doi:10.9734/BBJ/2013/3596 Pant, P., Maggi, F., Gyawali, R., and Dall’Acqua, S. (2019). REFERENCES Reactive Oxygen Species as an Initiator of Toxic Innate Immune Responses in Retort to SARS-CoV-2 in an Ageing Population, Consider N-Acetylcysteine as Early Therapeutic Intervention. Toxicol. Rep. 7, 768–771. doi:10.1016/j.toxrep. 2020.06.003 Pinho-da-Silva, L., de Morais, S. M., Coelho-de-Souza, A. N., Lahlou, S., and Leal- Cardoso, J. H. (2010). Croton sonderianus Essential Oil Samples Distinctly Affect Rat Airway Smooth Muscle. Phytomedicine 17, 721–725. doi:10.1016/j. phymed.2010.01.015 Ng, Y. C., Kim, Y. W., Lee, J.-S., and Jung Song, M. (2018). Antiviral Activity of Schizonepeta tenuifolia Briquet Against Noroviruses via Induction of Antiviral Interferons. J. Microbiol. 56, 683–689. doi:10.1007/s12275-018-8228-7 Pinho-da-Silva, L., Ceccatto, V. M., Coelho-de-Souza, A. N., Santos Cruz, J., and Leal-Cardoso, J. H. (2012). Trans-Caryophyllene, A Natural Sesquiterpene, Causes Tracheal Smooth Muscle Relaxation through Blockade of Voltage- dependent Ca2+ Channels. Molecules 17, 11965–11977. doi:10.3390/ molecules171011965 Nieto-Bobadilla, M. S., Tesse, N., Willart, J.-F., Dubreuil, L., Siepmann, J., and Neut, C. (2015). Controlled Delivery of a New Broad Spectrum Antibacterial Agent Against Colitis: In Vitro and In Vivo Performance. Eur. J. Pharmaceutics Biopharmaceutics 96, 152–161. doi:10.1016/j.ejpb.2015.07.012 Poddighe, L., Boi, M., Lisai, S., Murru, E., Muredda, L., Collu, M., et al. (2018). Acute Administration of Beta-Caryophyllene Prevents Endocannabinoid System Activation during Transient Common Carotid Artery Occlusion and Reperfusion. Lipids Health Dis. 17, 23. doi:10.1186/s12944-018-0661-4 Nuzzo, D., and Picone, P. (2020). Potential Neurological Effects of Severe COVID- 19 Infection. Neurosci. Res. 158, 1–5. doi:10.1016/j.neures.2020.06.009 Oboh, G., Olasehinde, T. A., and Ademosun, A. O. (2014). Essential Oil From Lemon Peels Inhibit Key Enzymes Linked to Neurodegenerative Conditions and Pro-Oxidant Induced Lipid Peroxidation. J. Oleo Sci. 63, 373–381. doi:10. 5650/jos.ess13166 Potus, F., Breton-Gagnon, E., Lajoie, A. C., Boucherat, O., Bonnet, S., and Provencher, S. (2020). Novel Insights on the Pulmonary Vascular Consequences of COVID-19. Am. J. Physiol. Lung Cell Mol. Physiol. 319, L277–L288. doi:10.1152/ajplung.00195.2020 Ojha, S., Javed, H., and Azimullah, S. (2016). β-Caryophyllene, a Phytocannabinoid Attenuates Oxidative Stress, neuroinﬂammation, Glial Activation, and Salvages Dopaminergic Neurons in a Rat Model of Parkinson Disease. Mol. Cell. Biochem 418, 59–70. doi:10.1007/s11010-016-2733-y Prins, G. H., and Olinga, P. (2020). Potential Implications of COVID-19 in Non- Alcoholic Fatty Liver Disease. Liver Int. 40, 2568. doi:10.1111/liv.14484 Okoh, S. O., Okoh, O. O., and Okoh, A. I. (2019). Inhibitory Effects of Azadirachta indica Secondary Metabolites Formulated Cosmetics on Some Infectious Rabaan, A. A., Yatoo, M. I., Patel, S. K., Pathak, M., Malik, Y. S., Dhama, K., et al. (2020). REFERENCES 10.008 Mahmud, S., Uddin, M. A. R., Rahman, M. E., Shehab, M. N., Islam, A., Alom, M. W., et al. (2020). Molecular Docking and Dynamics Study of natural Compound for Potential Inhibition of Main Protease of SARS-CoV-2. J. Biomol. Struct. Dyn. 2020, 1–9. doi:10.1080/07391102.2020.1796808 Koubaa, F. G., Abdennabi, R., and Salah, A. S. B. (2019). Microwave Extraction of Salvia ofﬁcinalis Essential Oil and Assessment of its GC-MS Identiﬁcation and Protective Effects Versus Vanadium-Induced Nephrotoxicity in Wistar Rats Models. Arch. Physiol. Biochem. 125, 404–413. doi:10.1080/13813455.2018. 1478427 Manzo, A., Dallavalle, S., Catalano, E., Scarì, G., and Giorgi, A. (2016). Screening of the Chemical Composition and Bioactivity of Waldheimia glabra (Decne.) Regel Essential Oil. J. Sci. Food Agric. 96, 3195–3201. doi:10.1002/jsfa.7499 the Chemical Composition and Bioactivity of Waldheimia glabra (Decne.) Regel Essential Oil. J. Sci. Food Agric. 96, 3195–3201. doi:10.1002/jsfa.7499 McCord, J. M., Hybertson, B. M., and Gao, B. (2020). Nrf2 Activator PB125 as a Potential Therapeutic Agent against COVID-19. Antioxidants 9, 518. doi:10. 3390/antiox9060518 Koyama, S., Novotny, M. V., Davis, K., Kao, C. C., Matsunami, H., and Mescher, A. (2019). Beta-Caryophyllene Enhances Wound Healing Through Multiple Routes. PLOS ONE 14, e0216104. doi:10.1371/journal.pone.0216104 Meeran, M. F. N., Adeghate, E., and Ojha, S. (2019). β-Caryophyllene, A Natural Bicyclic Sesquiterpene Attenuates Doxorubicin-Induced Chronic Cardiotoxicity via Activation of Myocardial Cannabinoid Type-2 (CB2) Receptors in Rats. Chem.-Biol. Interact. 304, 158–167. doi:10.1016/j.cbi.2019. 02.028 Ku, C.-M., and Lin, J.-Y. (2013). Anti-inﬂammatory Effects of 27 Selected Terpenoid Compounds Tested Through Modulating Th1/Th2 Cytokine Secretion Proﬁles Using Murine Primary Splenocytes. Food Chem. 141, 1104–1113. doi:10.1016/j.foodchem.2013.04.044 Lai, C.-C., Ko, W.-C., Lee, P.-I., and Hsueh, P.-R. (2020). Extra-Respiratory Manifestations of COVID-19. Int. J. Antimicrob. Agents 56, 106024. doi:10. 1016/j.ijantimicag.2020.106024 Meng, X., Li, D., Zhou, D., Liu, Q., and Fan, S. (2016). Chemical Composition, Antibacterial Activity and Related Mechanism of the Essential Oil From the Leaves of Juniperus rigida Sieb. et Zucc Against Klebsiella pneumoniae. J. Ethnopharmacol. 194, 698–705. doi:10.1016/j.jep.2016.10.050 Lai, C.-C., Wang, C.-Y., and Hsueh, P.-R. (2020). Co-Infections Among Patients With COVID-19: The Need for Combination Therapy With non-anti-SARS- CoV-2 Agents? J. Microbiol. Immunol. Infect. 53, 505–512. doi:10.1016/j.jmii. 2020.05.013 Meza, A., and Lehmann, C. (2018). Betacaryophyllene - A Phytocannabinoid as Potential Therapeutic Modality for Human Sepsis? Med. Hypotheses 110, 68–70. doi:10.1016/j.mehy.2017.10.025 May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 20 Jha et al. Beta-Caryophyllene may be Useful in COVID-19 Pathogens and Oxidative Stress Radicals. BMC Complement. Altern. Med. REFERENCES Cannabis in the Time of Coronavirus Disease 2019: The Yin and Yang of the Endocannabinoid System in Immunocompetence. J. Altern. Complement. Med. 26, 444–448. doi:10.1089/acm.2020.0144 Shan, J., Chen, L., and Lu, K. (2017). Protective Effects of Trans-Caryophyllene on Maintaining Osteoblast Function. IUBMB Life 69, 22–29. doi:10.1002/iub. 1584 Rieder, S. A., Chauhan, A., Singh, U., and Nagarkatti, P. (2010). Cannabinoid-induced Apoptosis in Immune Cells as a Pathway to Immunosuppression. Immunobiology 215, 598–605. doi:10.1016/j. imbio.2009.04.001 Sharma, C., Al Kaabi, J. M., Nurulain, S., Amjad Kamal, M., and Ojha, S. (2016). Polypharmacological Properties and Therapeutic Potential of β-Caryophyllene: A Dietary Phytocannabinoid of Pharmaceutical Promise. CPD 22, 3237–3264. doi:10.2174/1381612822666160311115226 Rosales-Mendoza, S., Márquez-Escobar, V. A., and Arévalo-Villalobos, J. I. (2020). What Does Plant-Based Vaccine Technology Offer to the Fight against COVID-19? Vaccines 8, 183. doi:10.3390/vaccines8020183 Shayeganmehr, A., Marandi, M. V., Karimi, V., Barinau, A., and Ghalyanchilangeroudi, A. (2018). Zataria multiﬂora Essential Oil Reduces Replication Rate of Avian Inﬂuenza Virus (H9N2 Subtype) in Challenged Broiler Chicks. Br. Poult. Sci. 59, 389–395. doi:10.1080/00071668.2018.1478064 Ruﬁno, A. T., Salgueiro, L., Cavaleiro, C., and Mendes, A. F. (2015). Evaluation of the Anti-inﬂammatory, Anti-catabolic and Pro-anabolic Effects of E-Caryophyllene, Myrcene and Limonene in a Cell Model of Osteoarthritis. Eur. J. Pharmacol. 750, 141–150. doi:10.1016/j.ejphar.2015.01.018 Shim, H. I., Park, Y. S., Kim, N., and Lee, D. H. (2019). Inhibitory Effects of β-Caryophyllene on Helicobacter pylori Infection: A Randomized Double- Blind, Placebo-Controlled Study. Korean J. Gastroenterol. 74, 199. doi:10. 4166/kjg.2019.74.4.199 Russo, E. B. (2011). Taming THC: Potential Cannabis Synergy and Phytocannabinoid-Terpenoid Entourage Effects. Br. J. Pharmacol. 163, 1344–1364. doi:10.1111/j.1476-5381.2011.01238.x Singh, T. U., Kumar, D., and Singh, R. K. (2020). Drug Repurposing Approach to Fight COVID-19. Pharmacol. Rep. 72, 1479–1508. doi:10.1007/s43440-020- 00155-6 Sadeghi, H., Sadati, I., Abbaszadeh-Goudarzi, K., Asfaram, A., and Doustimotlagh, A. H. (2020). Antioxidant and Protective Effect of Stachys pilifera Benth Against Nephrotoxicity Induced by Cisplatin in Rats. J. Food Biochem. 44. doi:10.1111/ jfbc.13190 Sirichaiwetchakoon, K., Lowe, G. M., Kupittayanant, S., and Eumkeb, G. (2020). Pluchea indica (L.) Less. Tea Ameliorates Hyperglycemia, Dyslipidemia, and Obesity in High Fat Diet-Fed Mice. Evidence-Based Complement. Altern. Med. 2020, 1–12. doi:10.1155/2020/8746137 Saidi, M., Ghafourian, S., Zarin-Abaadi, M., Movahedi, K., and Sadeghifard, N. (2012). In Vitro Antimicrobial and Antioxidant Activity of Black Thyme (Thymbra spicata L.) Essential Oils. Roum. Arch. Microbiol. Immunol. 71, 61–69. Skolnik, P. R., Rabbi, M. F., Mathys, J.-M., and Greenberg, A. S. (2002). REFERENCES SARS-CoV-2/COVID-19 and Advances in Developing Potential May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 21 Jha et al. Beta-Caryophyllene may be Useful in COVID-19 Therapeutics and Vaccines to Counter This Emerging Pandemic. Ann. Clin. Microbiol. Antimicrob. 19, 40. doi:10.1186/s12941-020-00384-w Schmidt, E., Bail, S., Friedl, S. M., and Jirovetz, L. (2010). Antimicrobial Activities of Single Aroma Compounds. Nat. Prod. Commun. 5, 1365–1368. doi:10.1177/ 1934578x1000500906 Raj, V., Park, J. G., Kim, T., Ham, J., and Lee, J. (2021). Assessment of Antiviral Potencies of Cannabinoids Against SARS-CoV-2 Using Computational and In Vitro Approaches. Int. J. Biol. Macromol. 168, 474–485. doi:10.1016/j.ijbiomac. 2020.12.020 Schnitzler, P., Koch, C., and Reichling, J. (2007). Susceptibility of Drug-Resistant Clinical Herpes Simplex Virus Type 1 Strains to Essential Oils of Ginger, Thyme, Hyssop, and Sandalwood. AAC 51, 1859–1862. doi:10.1128/AAC. 00426-06 Rajesh, M., Liaudet, L., Huffman, J. W., Csiszar, A., Ungvari, Z., Mackie, K., et al. (2007). CB2-Receptor Stimulation Attenuates TNF-α-Induced Human Endothelial Cell Activation, Transendothelial Migration of Monocytes, and Monocyte-Endothelial Adhesion. Am. J. Physiol.Heart Circulat. Physiol. 293, H2210–H2218. doi:10.1152/ajpheart.00688.2007 Segat, G. C., Manjavachi, M. N., Matias, D. O., Freitas, C. S., Costa, R., and Calixto, J. B. (2017). Antiallodynic Effect of β-Caryophyllene on Paclitaxel-Induced Peripheral neuropathy in Mice. Neuropharmacology 125, 207–219. doi:10.1016/ j.neuropharm.2017.07.015 Rajesh, M., Mackie, K., and Pacher, P. (2008). CB2 Cannabinoid Receptor Agonists Attenuate TNF-α-Induced Human Vascular Smooth Muscle Cell Proliferation and Migration. Br. J. Pharmacol. 153, 347–357. doi:10.1038/sj.bjp.0707569 Semprini, R., Martorana, A., Ragonese, M., and Motta, C. (2018). Observational Clinical and Nerve Conduction Study on Effects of a Nutraceutical Combination on Painful Diabetic Distal Symmetric Sensory-Motor Neuropathy in Patients With Diabetes Type 1 and Type 2. Minerva Med. 109, 358–362. doi:10.23736/S0026-4806.18.05710-5 Ramirez, S. H., Reichenbach, N. L., Merkel, S. F., Wang, X., Ho, W.-Z., and Persidsky, Y. (2013). Attenuation of HIV-1 Replication in Macrophages by Cannabinoid Receptor 2 Agonists. J. Leukoc. Biol. 93, 801–810. doi:10.1189/jlb. 1012523 Setzer, W. N. (2016). Essential Oils as Complementary and Alternative Medicines for the Treatment of Inﬂuenza. Am. J. Essent. Oils Nat. Prod. 4 (4), 16–22. Rather, M. A., Dar, B. A., Dar, M. Y., Shah, W. A., Bhat, B. A., Ganai, B. A., et al. (2012). Chemical Composition, Antioxidant and Antibacterial Activities of the Leaf Essential Oil of Juglans regia L. and its Constituents. Phytomedicine 19, 1185–1190. doi:10.1016/j.phymed.2012.07.018 Sexton, M. (2020). REFERENCES Stimulation of Peroxisome Proliferator-Activated Receptors α and γ Blocks HIV-1 Replication and TNFα Production in Acutely Infected Primary Blood Cells, Chronically Infected U1 Cells, and Alveolar Macrophages From HIV-Infected Subjects. J. Acquir. Immune Deﬁc. Syndr. 31, 1–10. doi:10.1097/00126334- 200209010-00001 Saldanha, A. A., Santos, H. B., Silva, D. B., Carollo, C. A., Oliveira, F. M., Lopes, D., et al. (2019). Chemical Composition and Evaluation of the Anti- inﬂammatory and Antinociceptive Activities of Duguetia furfuracea Essential Oil: Effect on Edema, Leukocyte Recruitment, Tumor necrosis Factor Alpha Production, iNOS Expression, and Adenosinergic and Opioidergic Systems. J. Ethnopharmacol. 231, 325–336. doi:10.1016/j.jep. 2018.11.017 Soleimani, M. (2020). Acute Kidney Injury in SARS-CoV-2 Infection: Direct Effect of Virus on Kidney Proximal Tubule Cells. IJMS 21, 3275. doi:10.3390/ ijms21093275 Saleh, J., Peyssonnaux, C., and Singh, K. K. (2020). Mitochondria and Microbiota Dysfunction in COVID-19 Pathogenesis. Mitochondrion 54, 1–7. doi:10.1016/j. mito.2020.06.008 Srivastava, S., Pant, A., Trivedi, S., and Pandey, R. (2016). Curcumin and β-caryophellene Attenuate Cadmium Quantum Dots Induced Oxidative Stress and Lethality in Caenorhabditis elegans Model System. Environ. Toxicol. Pharmacol. 42, 55–62. doi:10.1016/j.etap.2016.01.001 Salles, É. L., Paffaro, V. A., Costigliola, V., Yu, J. C., Hess, D. C., Dhandapani, K. M., et al. (2020). Cannabidiol (CBD) Modulation of Apelin in Acute Respiratory Distress Syndrome. J. Cell. Mol. Med. 24, 12869–12872. doi:10.1111/jcmm. 15883 Staiano, R. I., Piscitelli, F., Orlando, P., Secondo, A., Granata, F., Lepore, M. T., et al. (2016). Human Lung-Resident Macrophages Express CB1 and CB2 Receptors Whose Activation Inhibits the Release of Angiogenic and Lymphangiogenic Factors. J. Leukoc. Biol. 99, 531–540. doi:10.1189/jlb.3HI1214-584R Santos, P. S., Oliveira, T. C., Júnior, L. M. R., Figueiras, A., and Nunes, L. C. C. (2018). β-Caryophyllene Delivery Systems: Enhancing the Oral Pharmacokinetic and Stability. Curr. Pharm. Des. 24, 3440–3453. doi:10. 2174/1381612824666180912151412 Sun, J., Aghemo, A., Forner, A., and Valenti, L. (2020). COVID-19 and Liver Disease. Liver Int. 40, 1278–1281. doi:10.1111/liv.14470 Sweid, A., Tjoumakaris, S. I., Gooch, M. R., Herial, N. A., Zarzour, H., Romo, V., et al. (2020). Cerebral Ischemic and Hemorrhagic Complications of Coronavirus Disease 2019. Int. J. Stroke 15, 733–742. doi:10.1177/ 1747493020937189 Sardinha, J., Kelly, M. E. M., Zhou, J., and Lehmann, C. (2014). Experimental Cannabinoid 2 Receptor-Mediated Immune Modulation in Sepsis. Mediators Inﬂamm. 2014, 1–7. doi:10.1155/2014/978678 Scavone, C., Rafaniello, C., Zoccoli, A., Berrino, L., Racagni, G., Rossi, F., et al. (2020). Current Pharmacological Treatments for COVID-19: What’s next? Br. J. Pharmacol. 177, 4813–4824. doi:10.1111/bph.15072 Tahamtan, A., NayeriSamieipoor, F. REFERENCES S., Izadi, A., Rashidi-Nezhad, A., and Tavakoli-Yaraki, M. (2018). Effects of Cannabinoid Receptor Type 2 in May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 22 Jha et al. Beta-Caryophyllene may be Useful in COVID-19 Respiratory Syncytial Virus Infection in Human Subjects and Mice. Virulence 9, 217–230. doi:10.1080/21505594.2017.1389369 Respiratory Syncytial Virus Infection in Human Subjects and Mice. Virulence 9, 217–230. doi:10.1080/21505594.2017.1389369 Wyganowska-Swiatkowska, M., Nohawica, M., Grocholewicz, K., and Nowak, G. (2020). Inﬂuence of Herbal Medicines on HMGB1 Release, SARS-CoV-2 Viral Attachment, Acute Respiratory Failure, and Sepsis. A Literature Review. IJMS 21, 4639. doi:10.3390/ijms21134639 Tambe, Y., Tsujiuchi, H., Honda, G., Ikeshiro, Y., and Tanaka, S. (1996). Gastric Cytoprotection of the Non-Steroidal Anti-Inﬂammatory Sesquiterpene, β-Caryophyllene. Planta Med. 62, 469–470. doi:10.1055/s-2006-957942 Xie, Q., Liu, Z., and Li, Z. (2015). Chemical Composition and Antioxidant Activity of Essential Oil of Six Pinus taxa Native to China. Molecules 20, 9380–9392. doi:10.3390/molecules20059380 Tang, Y., Liu, J., Zhang, D., Xu, Z., Ji, J., and Wen, C. (2020). Cytokine Storm in COVID-19: The Current Evidence and Treatment Strategies. Front. Immunol. 11, 1708. doi:10.3389/ﬁmmu.2020.01708 Xu, G., Huang, K., and Zhou, J. (2018). Hepatic AMP Kinase as a Potential Target for Treating Nonalcoholic Fatty Liver Disease: Evidence From Studies of Natural Products. CMC 25, 889–907. doi:10.2174/ 0929867324666170404142450 Tarumi, W., and Shinohara, K. (2020). Olfactory Exposure to β-Caryophyllene Increases Testosterone Levels in Women’s Saliva. Sex. Med. 8, 525–531. doi:10. 1016/j.esxm.2020.06.001 Yamaguchi, M., and Levy, R. M. (2016). The Combination of β-Caryophyllene, Baicalin and Catechin Synergistically Suppresses the Proliferation and Promotes the Death of RAW267.4 Macrophages In Vitro. Int. J. Mol. Med. 38, 1940–1946. doi:10.3892/ijmm.2016.2801 Tesch, N. R., Mora, F., and Rojas, L. (2011). Chemical Composition and Antibacterial Activity of the Essential Oil of Lantana camara Var. Moritziana. Nat. Prod. Commun. 6, 1031–1034. doi:10.1177/ 1934578x1100600727 Tian, X., Peng, J., Zhong, J., Pang, J., Lou, J., Li, M., et al. (2016). β-Caryophyllene Protectsin Vitroneurovascular Unit against Oxygen-Glucose Deprivation and Re-oxygenation-induced Injury. J. Neurochem. 139, 757–768. doi:10.1111/jnc. 13833 Yamaguchi, M., and Levy, R. (2019). The Combination of Catechin, Baicalin and β-Caryophyllene Potentially Suppresses the Production of Inﬂammatory Cytokines in Mouse Macrophages In Vitro. Exp. Ther. Med. 17, 4312–4318. doi:10.3892/etm.2019.7452 Tian, X., Liu, H., Xiang, F., Xu, L., and Dong, Z. (2019). β-Caryophyllene Protects against Ischemic Stroke by Promoting Polarization of Microglia toward M2 Phenotype via the TLR4 Pathway. Life Sci. 237, 116915. doi:10.1016/j.lfs.2019. 116915 Yamaguchi, M., and Levy, R. M. (2020). REFERENCES Metaxalone Suppresses Production of Inﬂammatory Cytokines Associated with Painful Conditions in Mouse Macrophages RAW264.7 Cells In Vitro: Synergistic Effect with β-Caryophyllene. CMM 20, 643–652. doi:10.2174/ 1566524020666200217102508 Toguri, J. T., Kelly, M. E. M., and Lehmann, C. (2015). Cannabinoid 2 Receptor Activation Reduces Leukocyte Adhesion and Improves Capillary Perfusion in the Iridial Microvasculature during Systemic Inﬂammation. CH 61, 237–249. doi:10.3233/CH-151996 Yang, C., Hu, D. H., and Feng, Y. (2015). Antibacterial Activity and Mode of Action of the Artemisia capillaris Essential Oil and Its Constituents Against Respiratory Tract Infection-Causing Pathogens. Mol. Med. Rep. 11, 2852–2860. doi:10.3892/mmr.2014.3103 Tung, Y.-T., Chua, M. T., Wang, S.-Y., and Chang, S-T. (2008). Anti-inﬂammation Activities of Essential Oil and Its Constituents From Indigenous cinnamon (Cinnamomum osmophloeum) Twigs. Bioresour. Technol. 99, 3908–3913. doi:10.1016/j.biortech.2007.07.050 Yang, M., Lv, Y., Tian, X., An, R., Zhang, Q., Li, M., et al. (2017). Neuroprotective Effect of β-Caryophyllene on Cerebral Ischemia- Reperfusion Injury via Regulation of Necroptotic Neuronal Death and Inﬂammation: In Vivo and In Vitro. Front. Neurosci. 11, 583. doi:10.3389/ fnins.2017.00583 ul Qamar, M. T., Alqahtani, S. M., and Alamri, M. A. (2020). Structural Basis of SARS-CoV-2 3CL Pro and Anti-COVID-19 Drug Discovery from Medicinal Plants. J. Pharm. Anal. 10, 313–319. doi:10.1016/j.jpha.2020.03.009 Yang, X., Yu, Y., and Xu, J.(2020). Clinical Course and Outcomes of Critically Ill Patients with SARS-CoV-2 Pneumonia in Wuhan, China: A Single-Centered, Retrospective, Observational Study. Lancet Respir. Med. 8, 475–481. doi:10. 1016/S2213-2600(20)30079-5 Urasaki, Y., Hill, D. K., and Le, T. T. (2020). Fast-Acting and Receptor-Mediated Regulation of Neuronal Signaling Pathways by Copaiba Essential Oil. IJMS 21, 2259. doi:10.3390/ijms21072259 Younis, N. S., and Mohamed, M. E. (2019). β-Caryophyllene as a Potential Protective Agent Against Myocardial Injury: The Role of Toll-Like Receptors. Molecules 24, 1929. doi:10.3390/molecules24101929 Varga, Z. V., Matyas, C., Erdelyi, K., Nieri, D., Chicca, A., Nemeth, B. T., et al. (2018). β-Caryophyllene Protects Against Alcoholic Steatohepatitis by Attenuating Inﬂammation and Metabolic Dysregulation in Mice. Br. J. Pharmacol. 175, 320–334. doi:10.1111/bph.13722 Youssef, D. A., El-Fayoumi, H. M., and Mahmoud, M. F. (2019). Beta- Caryophyllene Alleviates Diet-Induced neurobehavioral Changes in Rats: The Role of CB2 and PPAR-γ Receptors. Biomed. Pharmacother. 110, 145–154. doi:10.1016/j.biopha.2018.11.039 Venturi, C. R., Montanha, J. A., Bordignon, S. A. L., Roehe, P. M., Fuentefria, A. M., and Henriques, A. T. (2015). Chemical Analysis and In Vitro Antiviral and Antifungal Activities of Essential Oils From Glechon spathulata and Glechon marifolia. Pharm. Biol. 53, 682–688. doi:10.3109/13880209.2014. 936944 Youssef, D. A., El-Fayoumi, H. REFERENCES M., and Mahmoud, M. F. (2019). Beta- caryophyllene Protects against Diet-Induced Dyslipidemia and Vascular Inﬂammation in Rats: Involvement of CB2 and PPAR-γ Receptors. Chem. Biol. Interact. 297, 16–24. doi:10.1016/j.cbi.2018.10.010 Vinciguerra, M., Romiti, S., Fattouch, K., De Bellis, A., and Greco, E. (2020). Atherosclerosis as Pathogenetic Substrate for Sars-Cov2 Cytokine Storm. JCM 9, 2095. doi:10.3390/jcm9072095 Zhang, Q., An, R., Tian, X., Li, M., Lou, J., Xu, L., et al. (2017). β-Caryophyllene Pretreatment Alleviates Focal Cerebral Ischemia-Reperfusion Injury by Activating PI3K/Akt Signaling Pathway. Neurochem. Res. 42, 1459–1469. doi:10.1007/s11064-017-2202-3 Wang, B., Ilnytskyy, Y., Kovalchuk, I., and Kovalchuk, O. (2020). In Search of Preventative Strategies: novel High-CBD Cannabis sativa Extracts Modulate ACE2 Expression in COVID-19 Gateway Tissues. Aging 12, 22425–22444. doi:10.18632/aging.202225 Zhang, Z., Yang, C., Dai, X., Ao, Y., and Li, Y. (2017). Inhibitory Effect of Trans-Caryophyllene (TC) on Leukocyte-Endothelial Attachment. Toxicol. Appl. Pharmacol. 329, 326–333. doi:10.1016/j. taap.2017.06.016 Wang, D., Liu, X., Zhang, J., Wang, B., Xiang, H., Cheng, Z., et al. (2020). Clinical Characteristics of 138 Hospitalized Patients With 2019 Novel Coronavirus- Infected Pneumonia in Wuhan, China. JAMA 323, 1061. doi:10.1001/jama. 2020.1585 Zhang, C., Shi, L., and Wang, F. S. (2020). Liver Injury in COVID-19: Management and Challenges. Lancet Gastroenterol. Hepatol. 5, 428–430. doi:10.1016/S2468- 1253(20)30057-1 Wen, C.-C., Kuo, Y.-H., Wang, S.-Y., Liu, H.-G., Lee, C.-K., Chang, S.-T., et al. (2007). Speciﬁc Plant Terpenoids and Lignoids Possess Potent Antiviral Activities Against Severe Acute Respiratory Syndrome Coronavirus. J. Med. Chem. 50, 4087–4095. doi:10.1021/jm070295s Zhang, Y., Bai, C., Shi, W., Alvarez-Manzo, H., and Zhang, Y. (2020). Identiﬁcation of Essential Oils Including Garlic Oil and Black Pepper Oil With High Activity against Babesia duncani. Pathogens 9, 466. doi:10.3390/ pathogens9060466 Wu, Q. F., Shen, Z.-H., Ying, H.-Z., and Yu, C.-H. (2012). Chemical Compositions and Anti-Inﬂuenza Activities of Essential Oils From Mosla dianthera. J. Ethnopharmacol. 139, 668–671. doi:10.1016/j.jep.2011.11.056 Zheng, X., Sun, T., and Wang, X. (2013). Activation of Type 2 Cannabinoid Receptors (CB2R) Promotes Fatty Acid Oxidation through the SIRT1/PGC-1α Pathway. Biochem. Biophys. Res. Commun. 436, 377–381. doi:10.1016/j.bbrc. 2013.05.108 Wu, C., Lee, H.-S., Hwang, K.-Y., and Lee, S.-J. (2014). Trans-Caryophyllene is a Natural Agonistic Ligand for Peroxisome Proliferator-Activated Receptor-α. Bioorg. Med. Chem. Lett. 24, 3168–3174. doi:10.1016/j.bmcl.2014.04.112 Natural Agonistic Ligand for Peroxisome Proliferator-Activated Receptor-α. Bioorg. Med. Chem. Lett. 24, 3168–3174. doi:10.1016/j.bmcl.2014.04.112 May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 23 Jha et al. Zhou, G., Chen, S., and Chen, Z. (2020). Advances in COVID-19: The Virus, The Pathogenesis, and Evidence-Based Control and Therapeutic Strategies. Front. Med. 14, 117–125. doi:10.1007/s11684- 020-0773-x Zhou, H., Ma, J., Mao, C., Wang, J., and Gui, H. (2020). Cannabinoid Receptor 2 Promotes the Intracellular Degradation of HMGB1 via the Autophagy- Lysosome Pathway in Macrophage. Int. Immunopharmacol. 78, 106007. doi:10.1016/j.intimp.2019.106007 Zhu, Z., Shi, L., and Wang, Y. (2020). Pathological Findings of COVID-19 Associated With Acute Respiratory Distress Syndrome. Lancet Respir. Med. 8, 420–422. doi:10.1016/S2213-2600(20)30076-X Frontiers in Pharmacology \| www.frontiersin.org REFERENCES Beta-Caryophyllene may be Useful in COVID-19 Conﬂict of Interest: The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Copyright © 2021 Jha, Sharma, Hashiesh, Arunachalam, Meeran, Javed, Patil, Goyal and Ojha. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. May 2021 \| Volume 12 \| Article 590201 Frontiers in Pharmacology \| www.frontiersin.org 24
https://openalex.org/W4234107831	https://peerj.com/articles/1577v0.2/submission	Latin	null	Peer Review #2 of "The effect of repeated laser stimuli to ink-marked skin on skin temperature—recommendations for a safe experimental protocol in humans (v0.1)"	null	2,016	cc-by	7,475	Manuscript to be reviewed Victoria J Madden, Mark J Catley, Luzia Grabherr, Francesca Mazzola, Mohammad Shohag, G Lorimer Moseley Background: Nd:YAP laser is widely used to investigate the nociceptive and pain systems, generating perpetual and laser-evoked neurophysiological responses. A major procedural concern for the use of Nd:YAP laser stimuli in experimental research is the risk of skin damage. The absorption of Nd:YAP laser stimuli is greater in darker skin, or in pale skin that has been darkened with ink, prompting some ethics boards to refuse approval to experimenters wishing to track stimulus location by marking the skin with ink. Some research questions, however, require laser stimuli to be delivered at particular locations or within particular zones, a requirement that is very difficult to achieve if marking the skin is not possible. We thoroughly searched the literature for experimental evidence and protocol recommendations for safe delivery of Nd:YAP laser stimuli over marked skin, but found nothing. Methods: We designed an experimental protocol to define safe parameters for the use of Nd:YAP laser stimuli over skin that has been marked with black dots, and used thermal imagine to assess the safety of the procedure at the forearm and the back. Results: Using thermal imaging and repeated laser stimulation to ink-marked skin, we demonstrated that skin temperature did not increase progressively across the course of the experiment, and that the small change in temperature seen at the forearm was reversed during the rest periods between blocks. Furthermore, no participant experienced skin damage due to the procedure. Conclusion: This protocol offers parameters for safe, confident and effective experimentation using repeated Nd:YAP laser on skin marked with ink, thus paving the way for investigations that depend on it. PeerJ reviewing PDF \| (2015:09:6973:1:1:NEW 15 Dec 2015) PeerJ reviewing PDF \| (2015:09:6973:1:1:NEW 15 Dec 2015) The effect of repeated laser stimuli to ink-‐ 1 marked skin on skin temperature – 2 recommendations for a safe experimental 3 protocol in humans 4 5 Authors: 6 Victoria J Maddena, torymadden@gmail.com 7 Mark J Catleya, mark.catley@unisa.edu.au 8 Luzia Grabherra, luziagrabherr@gmx.ch 9 Francesca Mazzolaa, mazzolavincenzi@alice.it 10 Mohammad Shohaga, sg_0480@yahoo.com 11 G Lorimer Moseleya,b, lorimer.moseley@gmail.com 12 13 aSansom Institute for Health Research, University of South Australia, Adelaide, Australia. 14 bNeuroscience Research Australia, Sydney, Australia. 15 16 Address for correspondence: Body in Mind research group, GPO Box 2471, University of 17 South Australia, Adelaide 5001. Email: lorimer.moseley@gmail.com. Tel: +61 8 8302 1416. 18 Fax: N/A. Institutional URL: www.unisa.edu.au 19 20 Manuscript to be review 22 PeerJ reviewing PDF \| (2015:09:6973:1:1:NEW 15 Dec 2015) Introduction 23 Laser stimulation is an important tool for investigating the nociceptive and pain systems, 24 because it allows for the selective activation of nociceptive neurons with a brief, tightly 25 controlled stimulus. Although CO2 lasers have been widely used, the skin discolouration 26 they produce (Lefaucheur et al. 2012) and difficulties with using them in small spaces and at 27 awkward angles have prompted a shift in favour of solid state lasers (e.g. Cruccu et al. 28 2003). Solid state lasers such as Neodymium Yttrium–Aluminium–Perovskite (Nd:YAP) 29 lasers may be preferable due to their greater flexibility: the machines tend to be less 30 cumbersome, and transmission of the laser through a flexible fibre-‐optic cable means that 31 these stimuli can be delivered inside an MRI tunnel (Perchet et al. 2008). 32 Many experimental protocols require repeated stimulus application (for example, to 33 measure thresholds, assess acuity, etc.). However, when laser is used for repetitive 34 stimulation there is a risk of progressive temperature increase if the same skin area is 35 heated too often. Progressive increases in skin temperature over the course of a procedure 36 may (a) elevate the risk of skin damage and (b) compromise the reliability of the outcome 37 being studied – for example, perceptual acuity, perceptual intensity or ERP amplitude. This 38 risk may be elevated in people with chronic pain because there are emerging data that 39 suggest spatially defined dysfunction in thermoregulation and cortical stimulus processing in 40 these groups (e.g. Moseley et al. 2012a; Moseley et al. 2012b; Moseley et al. 2012c). To 41 reduce this risk, the location of the stimulus is usually shifted slightly between trials to 42 prevent heating of the same skin area in successive trials. This shifting of location is also 43 thought to reduce sensitisation or habituation to the stimulus (Iannetti et al. 2003; Wiech et 44 al. 2010). 45 This risk of temperature increase and consequent skin damage presents an obvious problem 46 because, to our knowledge, there are no published data outlining the number or frequency 47 of Nd:YAP laser stimuli that is safe. A further issue is the quandary presented when 48 researchers want to accurately localise stimulation sites: it is generally considered that 49 Manuscript to be review PeerJ reviewing PDF \| (2015:09:6973:1:1:NEW 15 Dec 2015) PeerJ reviewing PDF \| (2015:09:6973:1:1:NEW 15 Dec 2015) Manuscript to be reviewed Using black pens for skin markings is common practice in other experiments that 56 deliver stimuli to human skin and would be a convenient and sensible option for studies that 57 use laser stimuli, if such markings can be used without increasing skin temperature enough 58 to compromise reliability or cause skin damage. 59 An additional ‘safety’ check that is commonly used in studies involving painful stimuli is to 60 calibrate the stimulus intensity to individual participants according to pain rating (Mancini et 61 al. 2011; Wager et al. 2006; Weiss et al. 1997). This is a reasonable approach for preventing 62 undue suffering that could be caused by the experimental pain percept. However, pain is 63 influenced by a multitude of factors, including attention (Villemure & Bushnell 2002), 64 salience (Wiech et al. 2010), emotion (Wiech & Tracey 2009), task demands (Petrovic et al. 65 2000) and expectations (Atlas & Wager 2012). Consequently, stimulus calibration according 66 to pain percept may not be a useful strategy to ensure safety. The finding that people with 67 chronic pain have alterations in perceptual acuity (Wand et al. 2011) suggests that the 68 stimulus energy-‐percept mismatch may be even more pronounced in people with pain, 69 emphasising that experimental procedures for studying pain must be tested for safety in 70 those with and those without pain. 71 Here we present data from an experimental protocol in which multiple stimuli were applied 72 to the black-‐ink-‐marked skin of 15 people with chronic low back pain and 13 healthy control 73 participants, and thermal imaging was used to evaluate skin temperature before and after 74 stimulation blocks. We quantified the effect of repeated stimulation on skin temperature to 75 ascertain the safety of an inter-‐stimulus interval and block-‐rest-‐block protocol when the 76 locations of stimuli have been marked with black ink. 77 Methods 78 Manuscript to be reviewe localise stimulation sites. Furthermore, marking stimulation sites with a colour that is 54 difficult to see makes experimenter fatigue more likely, and could thus compromise 55 accuracy. Using black pens for skin markings is common practice in other experiments that 56 deliver stimuli to human skin and would be a convenient and sensible option for studies that 57 use laser stimuli, if such markings can be used without increasing skin temperature enough 58 to compromise reliability or cause skin damage. Manuscript to be reviewed Manuscript to be reviewed Laser stimulation is an important tool for investigating the nociceptive and pain systems, 24 because it allows for the selective activation of nociceptive neurons with a brief, tightly 25 controlled stimulus. Although CO2 lasers have been widely used, the skin discolouration 26 they produce (Lefaucheur et al. 2012) and difficulties with using them in small spaces and at 27 awkward angles have prompted a shift in favour of solid state lasers (e.g. Cruccu et al. 28 2003). Solid state lasers such as Neodymium Yttrium–Aluminium–Perovskite (Nd:YAP) 29 lasers may be preferable due to their greater flexibility: the machines tend to be less 30 cumbersome, and transmission of the laser through a flexible fibre-‐optic cable means that 31 these stimuli can be delivered inside an MRI tunnel (Perchet et al. 2008). 32 Many experimental protocols require repeated stimulus application (for example, to 33 measure thresholds, assess acuity, etc.). However, when laser is used for repetitive 34 stimulation there is a risk of progressive temperature increase if the same skin area is 35 heated too often. Progressive increases in skin temperature over the course of a procedure 36 may (a) elevate the risk of skin damage and (b) compromise the reliability of the outcome 37 being studied – for example, perceptual acuity, perceptual intensity or ERP amplitude. This 38 risk may be elevated in people with chronic pain because there are emerging data that 39 suggest spatially defined dysfunction in thermoregulation and cortical stimulus processing in 40 these groups (e.g. Moseley et al. 2012a; Moseley et al. 2012b; Moseley et al. 2012c). To 41 reduce this risk, the location of the stimulus is usually shifted slightly between trials to 42 prevent heating of the same skin area in successive trials. This shifting of location is also 43 thought to reduce sensitisation or habituation to the stimulus (Iannetti et al. 2003; Wiech et 44 al. 2010). 45 PeerJ reviewing PDF \| (2015:09:6973:1:1:NEW 15 Dec 2015) localise stimulation sites. Furthermore, marking stimulation sites with a colour that is 54 difficult to see makes experimenter fatigue more likely, and could thus compromise 55 accuracy. Manuscript to be reviewed 59 Manuscript to be reviewed An additional ‘safety’ check that is commonly used in studies involving painful stimuli is to 60 calibrate the stimulus intensity to individual participants according to pain rating (Mancini et 61 al. 2011; Wager et al. 2006; Weiss et al. 1997). This is a reasonable approach for preventing 62 undue suffering that could be caused by the experimental pain percept. However, pain is 63 influenced by a multitude of factors, including attention (Villemure & Bushnell 2002), 64 salience (Wiech et al. 2010), emotion (Wiech & Tracey 2009), task demands (Petrovic et al. 65 2000) and expectations (Atlas & Wager 2012). Consequently, stimulus calibration according 66 to pain percept may not be a useful strategy to ensure safety. The finding that people with 67 chronic pain have alterations in perceptual acuity (Wand et al. 2011) suggests that the 68 stimulus energy-‐percept mismatch may be even more pronounced in people with pain, 69 emphasising that experimental procedures for studying pain must be tested for safety in 70 those with and those without pain. 71 An additional ‘safety’ check that is commonly used in studies involving painful stimuli is to 60 calibrate the stimulus intensity to individual participants according to pain rating (Mancini et 61 al. 2011; Wager et al. 2006; Weiss et al. 1997). This is a reasonable approach for preventing 62 undue suffering that could be caused by the experimental pain percept. However, pain is 63 influenced by a multitude of factors, including attention (Villemure & Bushnell 2002), 64 Here we present data from an experimental protocol in which multiple stimuli were applied 72 to the black-‐ink-‐marked skin of 15 people with chronic low back pain and 13 healthy control 73 participants, and thermal imaging was used to evaluate skin temperature before and after 74 stimulation blocks. We quantified the effect of repeated stimulation on skin temperature to 75 ascertain the safety of an inter-‐stimulus interval and block-‐rest-‐block protocol when the 76 locations of stimuli have been marked with black ink. 77 Methods 78 Participants 79 We recruited adult volunteers using flyers and word of mouth. Volunteers were not eligible 80 if they had sensation problems, diagnosed peripheral vascular disease, diabetes mellitus, or 81 psychological or neurological problems, or unusual skin conditions (e.g. dermographism) 82 that might compromise skin safety with laser application. Manuscript to be reviewed Volunteers with skin too dark for 83 PeerJ reviewing PDF \| (2015:09:6973:1:1:NEW 15 Dec 2015) possible erythema to be noticed were also ineligible. Participants had to be pain-‐free or 84 have chronic back pain, defined as pain between the levels of T12 and S1 (with or without 85 concurrent leg pain)(Merskey & Bogduk 1994). Volunteers with back pain were excluded if 86 they had concurrent neck or arm pain, were pregnant or less than six months post-‐partum, 87 or had had spinal surgery. Participants were remunerated at AU$20/hr for their 88 involvement. All procedures conformed to the Helsinki Declaration and were approved by 89 the institutional ethics committee. 90 Experimental procedure 91 Participants received laser stimuli to the skin of the forearm and the back. Prior to 92 stimulation, the hair on the dorsal forearm and the back was trimmed using clippers, and a 93 template was used to draw two dot-‐grids onto the skin with a black Artline®200 FINE 0.4 pen. 94 Dots were spaced 4mm apart in the proximal-‐distal (forearm) or medial-‐lateral (back) 95 direction, and 5mm apart in the radial-‐ulnar (forearm) or cephalo-‐caudad (back) direction. 96 The forearm grid measured 48mm x 20mm; the back grid measured 80mm x 20mm (see 97 Figures 1 and 2). 98 Stimuli 99 Laser stimuli were delivered using an Nd:YAP laser (Deka: Stimul 1340, wavelength 1340 nm, 100 spot diameter measured at 3.5mm, pulse width 6ms), which has previously been shown to 101 activate Aδ nociceptors specifically (Iannetti et al. 2006). The intensity of laser stimulus for 102 each participant was established, on the basis of an ascending staircase procedure, before 103 testing began. The operator delivered laser stimuli of changing intensity, asking participants 104 to describe what they felt. Once a painful pinprick was elicited, participants rated the pain 105 intensity on a 0-‐10 numerical rating scale with anchors of “no pain” (0) and “worst pain ever 106 felt” (10). The operator aimed to identify the intensity that consistently elicited pinprick 107 pain of 2-‐3 out of 10, and this was used for the testing procedure. If a participant reported a 108 diminution or an increase of pain intensity during the procedure, the operator adjusted the 109 laser stimulus intensity accordingly, with a maximum permitted laser intensity of 2.0J. Manuscript to be reviewed This 110 level was established during piloting as a limit that would prevent skin damage, and thus 111 became an additional exclusion criterion (if a laser stimulus intensity of 2.0J produced less 112 than 2/10 pinprick pain). Participants received stimuli in pairs. Three blocks of 14 stimulus 113 pairs were delivered to the forearm and three blocks of 16 pairs to the back, with at least 30 114 PeerJ reviewing PDF \| (2015:09:6973:1:1:NEW 15 Dec 2015) Manuscript to be review possible erythema to be noticed were also ineligible. Participants had to be pain-‐free or 84 have chronic back pain, defined as pain between the levels of T12 and S1 (with or without 85 concurrent leg pain)(Merskey & Bogduk 1994). Volunteers with back pain were excluded if 86 they had concurrent neck or arm pain, were pregnant or less than six months post-‐partum, 87 or had had spinal surgery. Participants were remunerated at AU$20/hr for their 88 involvement. All procedures conformed to the Helsinki Declaration and were approved by 89 the institutional ethics committee. 90 Manuscript to be reviewed seconds between pairs and 3-‐8 minutes between blocks (See Figure 3). The interval 115 between pairs was selected to reduce perceptual habituation, and the break between 116 blocks was chosen to allow participants relief from the sustained concentration required by 117 the perceptual acuity task. Blocks were randomly ordered. Each stimulus was delivered 118 over a dot in the dot-‐grid. The two stimuli of each pair were delivered 4-‐44mm (at the 119 forearm) or 4-‐76mm (at the back) apart. This approach was part of a wider plan to evaluate 120 perceptual acuity at each location. It was impossible for any one location to be stimulated 121 twice without a break of at least 120 seconds. The operator monitored skin colour visually, 122 so as to detect any localised erythema that could indicate undue skin heating. 123 Outcome 124 We used infrared thermal imaging (FLIR SC620 camera, FLIR systems, Oregon, USA; 125 sensitivity <40 mK, field of view = 24 x 18°) to record the average skin temperatures within 126 the demarcated zones before and after each stimulation block. This camera produced an 127 image that is colour-‐coded by infrared radiation. The ThermaCAM Researcher Professional 128 software (version 2.9, FLIR systems, Oregon, USA) allows the user to delineate a certain area 129 and calculate absolute temperature parameters for that area. Manuscript to be reviewed In this study, the demarcated 130 stimulation zone (forearm grid 48mm x 20mm; back grid 80mm x 20mm) was delineated 131 and the calculation was made as an average for that area. Skin condition was visually 132 monitored throughout the procedure. 133 Statistical Analysis 134 Temperature data were analysed using two analyses of variance (ANOVA): (1) temperatures 135 before and after each block were compared using repeated-‐measures ANOVA with Time 136 (before/after block) and Site (forearm/back) and Block (1/2/3) as within-‐subject factors, and 137 Group (patient/healthy) as the between-‐subject factor, and (2) forearm skin temperatures 138 before each block were compared using repeated-‐measures ANOVA with Block (1/2/3) as 139 the only factor. Planned comparisons were used to investigate significant effects, and 140 Bonferroni adjustments were applied to correct for multiple comparisons. Alpha was set at 141 0.05. Where the assumption of sphericity was violated, adjusted values are reported, with 142 Manuscript to be review seconds between pairs and 3-‐8 minutes between blocks (See Figure 3). The interval 115 between pairs was selected to reduce perceptual habituation, and the break between 116 blocks was chosen to allow participants relief from the sustained concentration required by 117 the perceptual acuity task. Blocks were randomly ordered. Each stimulus was delivered 118 over a dot in the dot-‐grid. The two stimuli of each pair were delivered 4-‐44mm (at the 119 forearm) or 4-‐76mm (at the back) apart. This approach was part of a wider plan to evaluate 120 perceptual acuity at each location. It was impossible for any one location to be stimulated 121 twice without a break of at least 120 seconds. The operator monitored skin colour visually, 122 so as to detect any localised erythema that could indicate undue skin heating. 123 Manuscript to be reviewed We used infrared thermal imaging (FLIR SC620 camera, FLIR systems, Oregon, USA; 125 sensitivity <40 mK, field of view = 24 x 18°) to record the average skin temperatures within 126 the demarcated zones before and after each stimulation block. This camera produced an 127 image that is colour-‐coded by infrared radiation. The ThermaCAM Researcher Professional 128 software (version 2.9, FLIR systems, Oregon, USA) allows the user to delineate a certain area 129 and calculate absolute temperature parameters for that area. Manuscript to be reviewed In this study, the demarcated 130 stimulation zone (forearm grid 48mm x 20mm; back grid 80mm x 20mm) was delineated 131 and the calculation was made as an average for that area. Skin condition was visually 132 monitored throughout the procedure. 133 PeerJ reviewing PDF \| (2015:09:6973:1:1:NEW 15 Dec 2015) Results 144 In this experiment, each participant received 84 stimuli at the forearm and 96 stimuli at the 145 back, delivered over a period of ∼70 minutes. 146 We recruited 29 adult volunteers, one of whom was excluded for dermographism. Of the 28 147 adults tested, 15 of them had chronic low back pain (mean age ± SD = 43 ± 17 years; 8 148 female) and 13 were pain-‐free (mean age ± SD = 49 ± 14 years; 8 female). 149 Overall, skin temperature was greater at the back than at the forearm (main effect of Site, 150 F(1,26) = 9.23, p = .005). Skin temperature increased over each block (main effect of Time, 151 F(1,26) = 19.45, p < .001) and the extent of the increase was greater in the forearm than it 152 was in the back (Time x Site interaction, F(1,26) = 30.028, p < .001). Comparisons of the Time 153 x Site interaction showed that forearm skin temperature increased over a block (p < .001, 154 mean change ± SE = 0.40 ± 0.07°C, 95% CI 0.27 -‐ 0.54) and that back skin temperature did 155 not change over a block (p = .098, mean change ± SD = -‐0.08 ± 0.05°C, 95% CI -‐0.02 – 0.18). 156 There was no main effect of Group (p = 0.938) or Block (p = 0.626) on skin temperature and 157 there were no other significant interactions. 158 Despite the small change in forearm temperature over each block, there was no difference 159 between skin temperatures at the start of each block (no effect of Block, p = .248), 160 Manuscript to be review PeerJ reviewing PDF \| (2015:09:6973:1:1:NEW 15 Dec 2015) Manuscript to be reviewed Manuscript to be reviewed In this experiment, each participant received 84 stimuli at the forearm and 96 stimuli at the 145 back, delivered over a period of ∼70 minutes. 146 We recruited 29 adult volunteers, one of whom was excluded for dermographism. Of the 28 147 adults tested, 15 of them had chronic low back pain (mean age ± SD = 43 ± 17 years; 8 148 female) and 13 were pain-‐free (mean age ± SD = 49 ± 14 years; 8 female). 149 Overall, skin temperature was greater at the back than at the forearm (main effect of Site, 150 F(1,26) = 9.23, p = .005). Skin temperature increased over each block (main effect of Time, 151 F(1,26) = 19.45, p < .001) and the extent of the increase was greater in the forearm than it 152 was in the back (Time x Site interaction, F(1,26) = 30.028, p < .001). Comparisons of the Time 153 x Site interaction showed that forearm skin temperature increased over a block (p < .001, 154 mean change ± SE = 0.40 ± 0.07°C, 95% CI 0.27 -‐ 0.54) and that back skin temperature did 155 not change over a block (p = .098, mean change ± SD = -‐0.08 ± 0.05°C, 95% CI -‐0.02 – 0.18). 156 There was no main effect of Group (p = 0.938) or Block (p = 0.626) on skin temperature and 157 there were no other significant interactions. 158 Despite the small change in forearm temperature over each block, there was no difference 159 between skin temperatures at the start of each block (no effect of Block, p = .248), 160 indicating that the rest periods between blocks were sufficient for the forearm skin 161 temperature to return to baseline. 162 None of our participants had localised erythema sufficient to warrant termination of the 163 procedure, none had any visible indications of skin damage on completion of the 164 experiment, and none reported adverse effects afterwards. 165 None of our participants had localised erythema sufficient to warrant termination of the 163 procedure, none had any visible indications of skin damage on completion of the 164 experiment, and none reported adverse effects afterwards. 165 Discussion 166 In this study, Nd:YAP laser stimuli were applied to the ink-‐marked skin of 28 participants 167 without resultant skin damage or lasting heating of the skin. Manuscript to be reviewed Those pilot subjects would not have been excluded on the basis of 189 the other exclusion criteria. We therefore recommend a conservative upper limit to the 190 intensity of laser to be used in an experiment, so as to identify and exclude participants who 191 may be at risk of skin burn during a procedure in which the laser stimulus intensity is 192 determined according to subjective ratings. Our results also indicate that calibration of 193 stimulus intensity according to pain report was not dangerous in this group of participants 194 with and without chronic back pain. Considering that people with chronic pain may display 195 a greater energy-‐percept mismatch than their healthy counterparts (Wand et al. 2011), 196 there is reason to expect that report-‐based calibration of stimulus intensity may be unsafe. 197 However, this study did not find evidence of skin damage with the stimuli used, which may 198 be attributable to the 2J safety limit on stimulus energy. 199 The third factor that ensured participants’ safety was the use of break periods, which 200 allowed for recovery of skin temperature towards a baseline level between stimulation 201 blocks. The fourth factor that ensured the safety of this design was that the ink markings 202 used in this study were small dots made with a 0.4mm-‐diameter pen. As such, the 203 Manuscript to be review These results demonstrate that it is possible to use Nd:YAP laser stimuli repeatedly over ink-‐ 174 marked skin without inducing skin damage, provided that strict safety parameters are 175 observed. 176 There were four factors that ensured the safety of this design. The first factor was strict 177 exclusion criteria. We excluded participants with neurological or sensation problems, and 178 we ensured that skin tone was pale enough for the operator to detect any laser-‐induced 179 change in skin colour. Although Nd:YAP does not produce the same discoloured skin lesions 180 that CO2 laser is known to, whether it produces lesions of deeper layers of the skin is 181 unknown (Iannetti et al. 2006). We therefore considered the visual monitoring of skin 182 colour a necessary precaution in this study, and planned to exclude any volunteer in whom 183 this would not be possible. 184 Manuscript to be reviewed The second factor was the selection of a conservative upper limit to the intensity of laser to 185 be used in the experiment. Manuscript to be reviewed No participant reported skin 168 damage due to the procedure. The thermal imaging data showed no overall increase in skin 169 temperature across the course of the experiment. Skin temperature at the back did not 170 change within blocks. Skin temperature at the forearm increased within each block, but 171 that increase was reversed during the rest periods between blocks such that skin 172 temperatures at the start of each forearm block did not differ. 173 In this study, Nd:YAP laser stimuli were applied to the ink-‐marked skin of 28 participants 167 without resultant skin damage or lasting heating of the skin. No participant reported skin 168 damage due to the procedure. The thermal imaging data showed no overall increase in skin 169 temperature across the course of the experiment. Skin temperature at the back did not 170 change within blocks. Skin temperature at the forearm increased within each block, but 171 that increase was reversed during the rest periods between blocks such that skin 172 temperatures at the start of each forearm block did not differ. 173 These results demonstrate that it is possible to use Nd:YAP laser stimuli repeatedly over ink-‐ 174 marked skin without inducing skin damage, provided that strict safety parameters are 175 observed. 176 There were four factors that ensured the safety of this design. The first factor was strict 177 exclusion criteria. We excluded participants with neurological or sensation problems, and 178 we ensured that skin tone was pale enough for the operator to detect any laser-‐induced 179 change in skin colour. Although Nd:YAP does not produce the same discoloured skin lesions 180 that CO2 laser is known to, whether it produces lesions of deeper layers of the skin is 181 unknown (Iannetti et al. 2006). We therefore considered the visual monitoring of skin 182 colour a necessary precaution in this study, and planned to exclude any volunteer in whom 183 this would not be possible. 184 The second factor was the selection of a conservative upper limit to the intensity of laser to 185 be used in the experiment. This limit was based on pilot work in our lab, during which two 186 participants (both males with Fitzpatrick skin type III, 50 and 24 years old) developed minor 187 skin lesions after repeated application of stimuli over 2J, despite reporting less than 2/10 188 pain at such intensities. Manuscript to be reviewed This limit was based on pilot work in our lab, during which two 186 participants (both males with Fitzpatrick skin type III, 50 and 24 years old) developed minor 187 skin lesions after repeated application of stimuli over 2J, despite reporting less than 2/10 188 pain at such intensities. Those pilot subjects would not have been excluded on the basis of 189 the other exclusion criteria. We therefore recommend a conservative upper limit to the 190 intensity of laser to be used in an experiment, so as to identify and exclude participants who 191 may be at risk of skin burn during a procedure in which the laser stimulus intensity is 192 determined according to subjective ratings. Our results also indicate that calibration of 193 stimulus intensity according to pain report was not dangerous in this group of participants 194 with and without chronic back pain. Considering that people with chronic pain may display 195 a greater energy-‐percept mismatch than their healthy counterparts (Wand et al. 2011), 196 there is reason to expect that report-‐based calibration of stimulus intensity may be unsafe. 197 However, this study did not find evidence of skin damage with the stimuli used, which may 198 be attributable to the 2J safety limit on stimulus energy. 199 The third factor that ensured participants’ safety was the use of break periods, which 200 allowed for recovery of skin temperature towards a baseline level between stimulation 201 blocks. The fourth factor that ensured the safety of this design was that the ink markings 202 used in this study were small dots made with a 0.4mm-‐diameter pen. As such, the 203 blackened skin area constituted a very small fraction of the area over which the laser 204 ( ) in a differential heating effect over the area of surface exposed to the beam. Blackened skin 206 is expected to absorb Nd:YAP laser more superficially, leading to quicker absorption and 207 dissipation of heat, and more selective activation of Aδ fibres than would be expected in un- 208 blackened skin (Leandri et al. 2006). In this study, the rate at which skin temperature 209 recovered from laser stimulation was likely linked to the comparatively small area over 210 which the skin was blackened. Any future work based on this report will need to consider 211 these four factors in order to achieve equivalent safety. Manuscript to be reviewed 212 Limitations 213 This report provides safe parameters for Nd:YAP laser stimulation using a 3.5mm spot 214 diameter. Further work would be required to determine safe parameters for other spot 215 sizes, because the area over which the stimulus is delivered affects the extent of skin 216 heating. If researchers require moment-‐by-‐moment information on skin temperature 217 changes, real-‐time thermal imaging will be necessary. We used imaging before and after 218 blocks, and are therefore unable to provide data on skin temperature between stimuli 219 within a block. 220 A reasonable concern about this report is that we did not personally reassess participants’ 221 skin condition in the days following the procedure. It is possible that participants developed 222 delayed signs of skin damage, particularly considering that any strongly absorbent regions of 223 tissue could be subject to greater risk of heat-‐induced damage due to an unequal heating 224 effect. However, participants were explicitly asked to report any signs of skin damage, and 225 we received no such reports. Furthermore, the intensity limit of 2J was established on the 226 basis of pilot testing that showed no skin damage in pilot participants, according to visual 227 assessments made immediately after the procedure and in the following days. We are 228 therefore confident that the parameters presented here did not result in visible skin 229 damage in this group of 28 participants. 230 Manuscript to be review in a differential heating effect over the area of surface exposed to the beam. Blackened skin 206 is expected to absorb Nd:YAP laser more superficially, leading to quicker absorption and 207 dissipation of heat, and more selective activation of Aδ fibres than would be expected in un-‐ 208 blackened skin (Leandri et al. 2006). In this study, the rate at which skin temperature 209 recovered from laser stimulation was likely linked to the comparatively small area over 210 which the skin was blackened. Any future work based on this report will need to consider 211 these four factors in order to achieve equivalent safety. 212 Manuscript to be reviewed blackened skin (Leandri et al. 2006). In this study, the rate at which skin temperature 209 recovered from laser stimulation was likely linked to the comparatively small area over 210 which the skin was blackened. PeerJ reviewing PDF \| (2015:09:6973:1:1:NEW 15 Dec 2015) Manuscript to be reviewed Any future work based on this report will need to consider 211 these four factors in order to achieve equivalent safety. 212 Limitations 213 This report provides safe parameters for Nd:YAP laser stimulation using a 3.5mm spot 214 diameter. Further work would be required to determine safe parameters for other spot 215 sizes, because the area over which the stimulus is delivered affects the extent of skin 216 heating. If researchers require moment-‐by-‐moment information on skin temperature 217 changes, real-‐time thermal imaging will be necessary. We used imaging before and after 218 blocks, and are therefore unable to provide data on skin temperature between stimuli 219 within a block. 220 A reasonable concern about this report is that we did not personally reassess participants’ 221 This procedure offers parameters for safe and effective experimentation using Nd:YAP laser 232 and black ink skin markings when delivering stimuli to the back and the forearm of healthy 233 participants and participants with chronic back pain. 234 235 PeerJ reviewing PDF \| (2015:09:6973:1:1:NEW 15 Dec 2015) Acknowledgements 236 The authors are grateful to Dr Tasha R Stanton for sharing her extensive knowledge about 237 Nd:YAP laser stimuli. Part of this work was presented at the 15th World Congress on Pain 238 (2014). 239 250 References 251 Atlas LY, and Wager TD. 2012. How expectations shape pain. Neuroscience Letters 520:140-‐ 252 148. http://dx.doi.org/10.1016/j.neulet.2012.03.039 253 Cruccu G, Pennisi E, Truini A, Iannetti GD, Romaniello A, Le Pera D, De Armas L, Leandri M, 254 Manfredi M, and Valeriani M. 2003. Unmyelinated trigeminal pathways as assessed by laser 255 stimuli in humans. Brain 126:2246-‐2256. 10.1093/brain/awg227 256 Iannetti GD, Truini A, Romaniello A, Galeotti F, Rizzo C, Manfredi M, and Cruccu G. 2003. 257 Evidence of a Specific Spinal Pathway for the Sense of Warmth in Humans. J Neurophysiol 258 89:562-‐570. 10.1152/jn.00393.2002 259 Iannetti GD, Zambreanu L, and Tracey I. 2006. Similar nociceptive afferents mediate 260 psychophysical and electrophysiological responses to heat stimulation of glabrous and hairy 261 skin in humans. The Journal of Physiology 577:235-‐248. 10.1113/jphysiol.2006.115675 262 Leandri M, Saturno M, Spadavecchia L, Iannetti GD, Cruccu G, and Truini A. 2006. 263 Measurement of skin temperature after infrared laser stimulation. Clinical Neurophysiology 264 36:207-‐218. http://dx.doi.org/10.1016/j.neucli.2006.08.004 265 Lefaucheur JP, Ahdab R, Ayache SS, Lefaucheur-‐Ménard I, Rouie D, Tebbal D, Neves DO, and 266 Ciampi de Andrade D. 2012. Manuscript to be reviewed Pain-‐related evoked potentials: A comparative study between 267 electrical stimulation using a concentric planar electrode and laser stimulation using a CO2 268 laser. Neurophysiologie Clinique/Clinical Neurophysiology 42:199-‐206. 269 http://dx.doi.org/10.1016/j.neucli.2011.12.003 270 Mancini F, Longo MR, Iannetti GD, and Haggard P. 2011. A supramodal representation of the 271 body surface Neuropsychologia 49:1194‐ 1201 272 Manuscript to be review Acknowledgements 236 The authors are grateful to Dr Tasha R Stanton for sharing her 237 Nd:YAP laser stimuli. Part of this work was presented at the 15 238 (2014). 239 250 References 251 Atlas LY, and Wager TD. 2012. How expectations shape pain. N 252 148. http://dx.doi.org/10.1016/j.neulet.2012.03.039 253 Cruccu G, Pennisi E, Truini A, Iannetti GD, Romaniello A, Le Per 254 Manfredi M, and Valeriani M. 2003. Unmyelinated trigeminal p 255 stimuli in humans. Brain 126:2246-‐2256. 10.1093/brain/awg22 256 Iannetti GD, Truini A, Romaniello A, Galeotti F, Rizzo C, Manfre 257 Evidence of a Specific Spinal Pathway for the Sense of Warmth 258 89:562-‐570. 10.1152/jn.00393.2002 259 Iannetti GD, Zambreanu L, and Tracey I. 2006. Similar nocicept 260 psychophysical and electrophysiological responses to heat stim 261 skin in humans. The Journal of Physiology 577:235-‐248. 10.111 262 Leandri M, Saturno M, Spadavecchia L, Iannetti GD, Cruccu G, 263 Measurement of skin temperature after infrared laser stimulat 264 36:207-‐218. http://dx.doi.org/10.1016/j.neucli.2006.08.004 265 Lefaucheur JP, Ahdab R, Ayache SS, Lefaucheur-‐Ménard I, Rou 266 Ciampi de Andrade D. 2012. Pain-‐related evoked potentials: A 267 electrical stimulation using a concentric planar electrode and l 268 laser. Neurophysiologie Clinique/Clinical Neurophysiology 42:1 269 http://dx.doi.org/10.1016/j.neucli.2011.12.003 270 Mancini F, Longo MR, Iannetti GD, and Haggard P. 2011. A sup 271 body surface. Neuropsychologia 49:1194-‐1201. 272 Ma Manuscript to be reviewed Manuscript to be reviewed Manuscript to be reviewed Iannetti GD, Zambreanu L, and Tracey I. 2006. Similar nociceptive afferents mediate 260 psychophysical and electrophysiological responses to heat stimulation of glabrous and hairy 261 skin in humans. The Journal of Physiology 577:235-‐248. 10.1113/jphysiol.2006.115675 262 Leandri M, Saturno M, Spadavecchia L, Iannetti GD, Cruccu G, and Truini A. 2006. 263 Measurement of skin temperature after infrared laser stimulation. Clinical Neurophysiology 264 36:207-‐218. http://dx.doi.org/10.1016/j.neucli.2006.08.004 265 Lefaucheur JP, Ahdab R, Ayache SS, Lefaucheur-‐Ménard I, Rouie D, Tebbal D, Neves DO, and 266 Ciampi de Andrade D. 2012. Pain-‐related evoked potentials: A comparative study between 267 electrical stimulation using a concentric planar electrode and laser stimulation using a CO2 268 laser. Neurophysiologie Clinique/Clinical Neurophysiology 42:199-‐206. 269 Lefaucheur JP, Ahdab R, Ayache SS, Lefaucheur-‐Ménard I, Rouie D, Tebbal D, Neves DO, and 266 Ciampi de Andrade D. 2012. Pain-‐related evoked potentials: A comparative study between 267 electrical stimulation using a concentric planar electrode and laser stimulation using a CO2 268 laser. Neurophysiologie Clinique/Clinical Neurophysiology 42:199-‐206. 269 Lefaucheur JP, Ahdab R, Ayache SS, Lefaucheur-‐Ménard I, Rouie D, Tebbal D, Neves DO, and 266 Ciampi de Andrade D. 2012. Pain-‐related evoked potentials: A comparative study between 267 electrical stimulation using a concentric planar electrode and laser stimulation using a CO2 268 laser. Neurophysiologie Clinique/Clinical Neurophysiology 42:199-‐206. 269 Mancini F, Longo MR, Iannetti GD, and Haggard P. 2011. A supramodal representation of the 271 body surface. Neuropsychologia 49:1194-‐1201. 272 Mancini F, Longo MR, Iannetti GD, and Haggard P. 2011. A supramodal representation of the 271 body surface. Neuropsychologia 49:1194-‐1201. 272 Mancini F, Longo MR, Iannetti GD, and Haggard P. 2011. A supramodal representation of the 271 body surface. Neuropsychologia 49:1194-‐1201. 272 Merskey H, and Bogduk N. 1994. Classification of chronic pain: descriptions of chronic pain 274 syndromes and definitions of pain terms: IASP press Seattle. 275 Merskey H, and Bogduk N. 1994. Classification of chronic pain: descriptions of chronic pain 274 syndromes and definitions of pain terms: IASP press Seattle. 275 PeerJ reviewing PDF \| (2015:09:6973:1:1:NEW 15 Dec 2015) Moseley GL, Gallace A, and Iannetti GD. 2012a. Spatially defined modulation of skin 276 temperature and hand ownership of both hands in patients with unilateral complex regiona 277 pain syndrome. Brain 135:3676-‐3686. 10.1093/brain/aws297 278 Moseley GL, Gallace A, and Spence C. 2012b. Manuscript to be reviewed Bodily illusions in health and disease: 279 Physiological and clinical perspectives and the concept of a cortical ‘body matrix’. 280 Neuroscience & Biobehavioral Reviews 36:34-‐46. 281 http://dx.doi.org/10.1016/j.neubiorev.2011.03.013 282 Moseley GL, Gallagher L, and Gallace A. 2012c. Neglect-‐like tactile dysfunction in chronic 283 back pain. Neurology 79:327-‐332. 10.1212/WNL.0b013e318260cba2 284 Perchet C, Godinho F, Mazza S, Frot M, Legrain V, Magnin M, and Garcia-‐Larrea L. 2008. 285 Evoked potentials to nociceptive stimuli delivered by CO2 or Nd:YAP lasers. Clinical 286 Neurophysiology 119:2615-‐2622. http://dx.doi.org/10.1016/j.clinph.2008.06.021 287 Petrovic P, Petersson KM, Ghatan PH, Stone-‐Elander S, and Ingvar M. 2000. Pain-‐related 288 cerebral activation is altered by a distracting cognitive task. Pain 85:19-‐30. 289 http://dx doi org/10 1016/S0304‐ 3959(99)00232‐ 8 290 Manuscript to be revie Moseley GL, Gallace A, and Iannetti GD. 2012a. Spatially defined modulation of skin 276 temperature and hand ownership of both hands in patients with unilateral complex regional 277 pain syndrome. Brain 135:3676-‐3686. 10.1093/brain/aws297 278 Moseley GL, Gallace A, and Spence C. 2012b. Bodily illusions in health and disease: 279 Physiological and clinical perspectives and the concept of a cortical ‘body matrix’. 280 Neuroscience & Biobehavioral Reviews 36:34-‐46. 281 http://dx.doi.org/10.1016/j.neubiorev.2011.03.013 282 Moseley GL, Gallagher L, and Gallace A. 2012c. Neglect-‐like tactile dysfunction in chronic 283 back pain. Neurology 79:327-‐332. 10.1212/WNL.0b013e318260cba2 284 Perchet C, Godinho F, Mazza S, Frot M, Legrain V, Magnin M, and Garcia-‐Larrea L. 2008. 285 Evoked potentials to nociceptive stimuli delivered by CO2 or Nd:YAP lasers. Clinical 286 Neurophysiology 119:2615-‐2622. http://dx.doi.org/10.1016/j.clinph.2008.06.021 287 Petrovic P, Petersson KM, Ghatan PH, Stone-‐Elander S, and Ingvar M. 2000. Pain-‐related 288 cerebral activation is altered by a distracting cognitive task. Pain 85:19-‐30. 289 http://dx.doi.org/10.1016/S0304-‐3959(99)00232-‐8 290 Villemure C, and Bushnell MC. 2002. Cognitive modulation of pain: how do attention and 291 emotion influence pain processing? Pain 95:195-‐199. 292 Wager TD, Matre D, and Casey KL. 2006. Placebo effects in laser-‐evoked pain potentials. 293 Brain, Behavior, and Immunity 20:219-‐230. http://dx.doi.org/10.1016/j.bbi.2006.01.007 294 Wand BM, Parkitny L, O’Connell NE, Luomajoki H, McAuley JH, Thacker M, and Moseley GL. 295 2011. Cortical changes in chronic low back pain: Current state of the art and implications for 296 clinical practice Manual Therapy 16:15‐ 20 http://dx doi org/10 1016/j math 2010 06 008 297 Manuscript to be review http://dx.doi.org/10.1016/j.neubiorev.2011.03.013 282 Villemure C, and Bushnell MC. 2002. Cognitive modulation of pain: how do attention and 291 emotion influence pain processing? Pain 95:195-‐199. 292 Villemure C, and Bushnell MC. 2002. Cognitive modulation of pain: how do attention and 291 emotion influence pain processing? Pain 95:195-‐199. 292 Wager TD, Matre D, and Casey KL. 2006. Placebo effects in laser-‐evoked pain potentials. 293 Brain, Behavior, and Immunity 20:219-‐230. http://dx.doi.org/10.1016/j.bbi.2006.01.007 294 Wand BM, Parkitny L, O’Connell NE, Luomajoki H, McAuley JH, Thacker M, and Moseley GL. 295 2011. Cortical changes in chronic low back pain: Current state of the art and implications for 296 clinical practice. Manual Therapy 16:15-‐20. http://dx.doi.org/10.1016/j.math.2010.06.008 297 Weiss T, Kumpf K, Ehrhardt J, Gutberlet I, and Miltner WHR. 1997. A bioadaptive approach 298 for experimental pain research in humans using laser-‐evoked brain potentials. Neuroscience 299 Letters 227:95-‐98. http://dx.doi.org/10.1016/S0304-‐3940(97)00320-‐0 300 Wiech K, Lin C-‐s, Brodersen KH, Bingel U, Ploner M, and Tracey I. 2010. Anterior Insula 301 Integrates Information about Salience into Perceptual Decisions about Pain. The Journal of 302 Neuroscience 30:16324-‐16331. 10.1523/jneurosci.2087-‐10.2010 303 Wiech K, and Tracey I. 2009. The influence of negative emotions on pain: Behavioral effects 304 Wager TD, Matre D, and Casey KL. 2006. Placebo effects in laser-‐evoked pain potentials. 293 Brain, Behavior, and Immunity 20:219-‐230. http://dx.doi.org/10.1016/j.bbi.2006.01.007 294 Wand BM, Parkitny L, O’Connell NE, Luomajoki H, McAuley JH, Thacker M, and Moseley GL. 295 2011. Cortical changes in chronic low back pain: Current state of the art and implications for 296 clinical practice. Manual Therapy 16:15-‐20. http://dx.doi.org/10.1016/j.math.2010.06.008 297 Weiss T, Kumpf K, Ehrhardt J, Gutberlet I, and Miltner WHR. 1997. A bioadaptive approach 298 for experimental pain research in humans using laser-‐evoked brain potentials. Neuroscience 299 Letters 227:95-‐98. http://dx.doi.org/10.1016/S0304-‐3940(97)00320-‐0 300 Wiech K, Lin C-‐s, Brodersen KH, Bingel U, Ploner M, and Tracey I. 2010. Anterior Insula 301 Integrates Information about Salience into Perceptual Decisions about Pain. The Journal of 302 Neuroscience 30:16324-‐16331. 10.1523/jneurosci.2087-‐10.2010 303 Wiech K, and Tracey I. 2009. The influence of negative emotions on pain: Behavioral effects 304 and neural mechanisms. NeuroImage 47:987-‐994. 305 http://dx.doi.org/10.1016/j.neuroimage.2009.05.059 306 307 PeerJ reviewing PDF \| (2015:09:6973:1:1:NEW 15 Dec 2015) Wager TD, Matre D, and Casey KL. 2006. Placebo effects in laser-‐evoked pain potentials. 293 Brain, Behavior, and Immunity 20:219-‐230. http://dx.doi.org/10.1016/j.bbi.2006.01.007 294 Wand BM, Parkitny L, O’Connell NE, Luomajoki H, McAuley JH, Thacker M, and Moseley GL. 295 2011. Manuscript to be reviewed 1 Figure 1: Dot-grids as drawn onto forearm (a) and back (b). Two dot-grids are pictured at each site: the second was used for tactile stimulation (data not presented here). PeerJ reviewing PDF \| (2015:09:6973:1:1:NEW 15 Dec 2015) http://dx.doi.org/10.1016/j.neubiorev.2011.03.013 282 Cortical changes in chronic low back pain: Current state of the art and implications for 296 clinical practice. Manual Therapy 16:15-‐20. http://dx.doi.org/10.1016/j.math.2010.06.008 297 Weiss T, Kumpf K, Ehrhardt J, Gutberlet I, and Miltner WHR. 1997. A bioadaptive approach 298 for experimental pain research in humans using laser-‐evoked brain potentials. Neuroscience 299 Letters 227:95-‐98. http://dx.doi.org/10.1016/S0304-‐3940(97)00320-‐0 300 Wiech K, Lin C-‐s, Brodersen KH, Bingel U, Ploner M, and Tracey I. 2010. Anterior Insula 301 Integrates Information about Salience into Perceptual Decisions about Pain. The Journal of 302 Neuroscience 30:16324‐ 16331 10 1523/jneurosci 2087‐ 10 2010 303 Wiech K, and Tracey I. 2009. The influence of negative emotions on pain: Behavioral effects 304 and neural mechanisms. NeuroImage 47:987-‐994. 305 PeerJ reviewing PDF \| (2015:09:6973:1:1:NEW 15 Dec 2015) Manuscript to be reviewed Manuscript to be reviewed 2 Diagram depicting experimental procedure and main results. CLBP: people with chronic low back pain. PeerJ reviewing PDF \| (2015:09:6973:1:1:NEW 15 Dec 2015) Manuscript to be reviewed PeerJ reviewing PDF \| (2015:09:6973:1:1:NEW 15 Dec 2015)
https://openalex.org/W3086517559	https://agroparistech.hal.science/hal-02983540/document	English	null	Adsorption of Phenolic Compounds from an Aqueous By-product of Sunflower Protein Extraction/Purification by Macroporous Resins	HAL (Le Centre pour la Communication Scientifique Directe)	2,020	cc-by	9,313	To cite this version: Tuong Le Thi, Arnaud Aymes, Xavier Framboisier, Irina Ioannou, Romain Kapel. Adsorption of Phenolic Compounds from an Aqueous By-product of Sunflower Protein Extraction/Purification by Macroporous Resins. Journal of Chromatography & Separation Techniques, 2020, 11 (6), pp.435. 10.35248/2157-7064.20.11.435. hal-02983540 Adsorption of Phenolic Compounds from an Aqueous By-product of Sunflower Protein Extraction/Purification by Macroporous Resins Tuong Le Thi, Arnaud Aymes, Xavier Framboisier, Irina Ioannou, Romain by Macroporous Resins Tuong Le Thi, Arnaud Aymes, Xavier Framboisier, Irina Ioannou, Romain Kapel Adsorption of Phenolic Compounds from an Aqueous By-product of Sunflower Protein Extraction/Purification by Macroporous Resins Correspondence to: Correspondence to: Romain Kapel, Laboratory Reactions And Chemical Engineering, University of Lorraine, CNRS, LRGP, F-54500, Nancy, France, Tel: +33766863934; E-mail: romain.kapel@univ-lorraine.fr Received: August 25 , 2020; Accepted: September 1, 2020; Published: September 15, 2020 Citation: Tuong Thi Le, Aymes A, Framboisier X, Ioannou I, Kapel R (2020) Adsorption of phenolic compounds from an aqueous by-product of sunflower protein extraction/ purification by macroporous resins. J Chromatogr Sep Tech. 11:435. DOI: 10.35248/2157-7064.20.11.435 Copyright: © 2020 Tuong Thi Le, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. ABSTRACT In this study, adsorption of phenolic compounds from an aqueous by-product of sunflower protein isolate production was investigated. Phenolic compounds in this by-product (ultrafiltration permeate of protein purification step) were almost exclusively CGA (mainly 5-CQA isomer). Five different macroporous resins including XAD4, XAD7, XAD16, XAD1180, and HP20 were screened for CGA capture. XAD16 had the best massic adsorption capacities (15.32 ± 0.04 mg/g), while XAD7 had a better surface adsorption capacity (0.027 ± 0.00146 mg/m²). CGA adsorption on both resins followed the pseudo-second-order kinetic model with a similar intra diffusional pattern. Adsorption isotherms of the two resins better fitted the Langmuir model with Qmax for XAD7 and XAD16 of 0.054 and 0.040 (mg/m²), respectively. The adsorption process of phenolic compounds revealed to be exothermic, physical adsorption, and spontaneous. Better adsorption results were observed at 25°C. Maximal CGA desorption ratio was observed from 70% (v/v) ethanol. The high values were reached with both the resins (88.09 ± 0.13 and 86.16 ± 0.32% for XAD7 and XAD16, respectively). The CGA purity in the desorption phase was surprisingly high (77.56 ± 0.99% and 74.59 ± 0.12% for XAD7 and XAD16, respectively). Keywords: Sunflower meal; Macroporous resin; Adsorption study; Chlorogenic acid. “meal” composed of up to 50% of proteins [9,10], which contains seeds phenolic compounds. Beside CGA, these compounds are mainly CGA isomers, CGA dimers, and caffeic acid. “meal” composed of up to 50% of proteins [9,10], which contains seeds phenolic compounds. Beside CGA, these compounds are mainly CGA isomers, CGA dimers, and caffeic acid. HAL Id: hal-02983540 https://agroparistech.hal.science/hal-02983540v1 Submitted on 30 Oct 2020 L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. J OPEN ACCESS Freely available online OPEN A Journal of Chromatography & Separation Techniques Research Article Adsorption of Phenolic Compounds from an Aqueous By-product of Sunflower Protein Extraction/Purification by Macroporous Resins Tuong Thi Le1,2, Aymes A1, Framboisier X1, Ioannou I1,3, Kapel R1* 1Laboratory Reactions And Chemical Engineering, University of Lorraine, CNRS, LRGP, F-54500, Nancy, France; 2Laboratory Stress Immunity Pathogens, University of Lorraine, EA7300, 9 Avenue de la foret de haye, 54500 Vandoeuvre-les-Nancy, France; 3URD Industrial Agro-Biotechnologies, AgroParisTech, CEBB, 3 rue des Rouges Terres, 51110 Pomacle, France Sunflower protein extraction/purification process The process was carried out in two stages as described in [27]. The first one was a solid/liquid extraction from the sunflower meal. The second was a protein purification by ultrafiltration (UF). The extraction step was achieved using a 0.5 M NaCl solution at a solid/liquid ratio of 1:9 (w/w). The pH of the slurry was adjusted at pH 7.5 with HCl 1 M and stirred at room temperature for 30 min. Then the slurry was centrifugated (Thermo Scientific Lynx 6000 centrifuge, USA) at 15000 g and 20°C for 30 min. The purification step was achieved with an Akta Flux 6 ultrafiltration apparatus (GE Healthcare Life Science, USA). The membrane used was a polyethersulfone hollow fiber membrane of 3 kDa cut-off (4800 cm² area, UFP-3, C-6A, GE Healthcare, USA). The transmembrane pressure was set at 1.5 bar and the feed rate at 1.5 L/min. In a first stage, six diavolumes (DV, DV = diafiltration solution volume/ initial volume) of 0.5 M NaCl solution was used to flush proteins from micro-solutes (low molar weight carbohydrate, minerals, non- protein nitrogen, and phenolic compounds). Then, NaCl was removed by three diavolumes of deionized water. The retentate compartment contained the purified proteins. The permeate obtained was collected and adjusted at pH 2 (in order to avoid phenol oxidation as recommended [28] and stored at -20°C before use. More recently, we proposed an alternative approach in which sunflower proteins are purified from the aqueous extract by tangential filtration [27]. This yielded a highly soluble sunflower protein isolate with high functional properties. Interestingly, phenolic compounds in the extract were only composed of CGA (mainly 5-CQA isomer) [27]. The liquid effluent of this process (the UF permeate) was largely depleted in proteins and probably contained the main part of extracted CGA. Hence, the capture of CGA from this side product would constitute an interesting valorization pathway. The aim of this work was to study CGA adsorption from the UF permeate on five macroporous resins including XAD4, XAD7, XAD16, XAD1180 and HP20 having various properties in term of polarity, pore diameters, specific area). On the resins exhibiting the best adsorption capacities, kinetics and isotherms experimental data were regressed at different temperatures in order to elucidate CGA adsorption mechanisms. Desorption by ethanol/ water mixtures was also investigated and the purity was considered in order to choose the most appropriate resin for CGA capture from this effluent. INTRODUCTION The industrial oil extraction process yields a solid by-product called J Chromatogr Sep Tech., Vol.11 Iss.6 No:435 1 Tuong Thi Le, et al. Tuong Thi Le, et al. OPEN ACCESS Freely available online resins should be chosen based on both adsorption capacities and purity after desorption step, but purity is often not considered. Furthermore, the complex transport phenomenon that are also involved are rarely thoroughly investigated. Sodium chloride (NaCl) and sodium hydroxide (NaOH) pellets were both from VWR (Darmstadt, Germany). Hydrochloric acid (HCl) solution was from Carlo Erba (Milan, Italy). Acetonitrile and formic acid at analytical grade were supplied by Fisher Scientific (Hampton, USA). Sunflower meal is also considered as an promising source of proteins [21,22] for food applications under isolate products. The production of sunflower protein isolates is classically achieved in two main steps (aqueous extraction and purification). The extraction step has to be done under pH 8 and at high NaCl concentration (> 0.2 M NaCl) in order to get satisfying protein extraction yield and color [23,24]. A large part of sunflower phenolic compounds are extracted alongside with proteins in these conditions. Recently, an integrated process was proposed for both protein and phenolic compounds valorization [25,26]. This process integrates a capture of phenolic compounds by adsorption from the aqueous extract prior to protein purification by acid precipitation. The resin had to have low protein binding capacity. XAD16 resin revealed to be the most appropriate resin to do so beyond other apolar and ion exchange resins [26]. However, CGA purity after adsorption step was not considered and adsorption mechanisms and transport were not investigated and remained unclear. Furthermore, in this approach, protein acid precipitation led to an isolate having a poor solubility which is rather detrimental for food applications. Adsorption/ desorption study Before each experiment, resins were washed with methanol for 10 minutes under magnetic stirring (150 rpm) at room temperature and rinsed with ultrapure water as recommended by the manufacturer. For each experiment, 1.5 g resin was mixed with 30 mL of permeate under magnetic stirring (150 rpm) at 25°C. After adsorption, resins were separated from the liquid phase by filtration (using Millex syringe Filter, PVDF, 0.22µm, non-sterile from ThermoFisher Scientific (USA). Liquid phases were analyzed by HPLC for phenolic compounds characterization and/ or CGA quantification. Resin screenings The resins were screened on the basis of massic (Eq. 1) and specific surface (Eq. 2) adsorption capacity : INTRODUCTION Phenolic compounds are a wide group of natural products from plant sources. These molecules are secondary metabolites having one or several benzene ring with hydroxyl groups. They are known to have various bioactivities like antioxidant, anti-inflammatory, anti-cancer, anti-ageing, or anti-osteoporosis activities [1-3]. Among phenolic compounds, chlorogenic acid (CGA) revealed particularly attractive. Chemically, CGA is a polar phenolic compound composed of a quinic acid and caffeic acid linked by an ester bond. The position of the quinic acid on the benzene ring define three CGA isomers including 5-O-caffeoylquinic acid i.e 5-CQA, 4-O-caffeoylquinic acid i.e 4-CQA, and 3-O-caffeoylquinic acid i.e 3-CQA. Interestingly, CGA has demonstrated high anti-oxidant, anti-inflammation and anti-tumor effects [4,5] To date, sunflower meal is mainly used for the feed. For this purpose, phenolic compounds are considered as antinutritionals and need to be eliminated. Many approaches were reported to remove phenolic compounds from the meal. As for other resources, the most classical way is to use organic solvents like acetone, methanol, ethyl acetate or water/ ethanol mixtures [11,12]. Weizs GM et al. 2009 [13] showed that methanol/water mixtures 60% (v/v) was suitable for extraction of sunflower phenols from both kernels and shells. Interestingly, they showed that the obtained extract with a phenolic fraction composed of a particularly high proportion of CGA (around 85%). After solvent extraction, phenolic compounds can be further purified by chromatography [14], membrane processes [15,16] or adsorption on macroporous resins [17]. This last process proved to be particularly efficient for phenolic compounds capture from various plant extracts [6,17,18]. Besides, adsorption processes can easily be scaled up and some resins are food grade [19]. Macroporous resins are divided into polar resins, mild polar resins, and non-polar resins. The mechanism of separation is based on differential affinity between phenolic compounds, impurities, and the adsorbent [20]. Proper Some raw materials like coffee beans, low-grade coffee beans, noxious weeds like Eupatorium adenophorum Spreng (Crofton weed) [6] or Boehmeria nivea L. Gaud (Ramie) leaves [7] revealed to be interesting sources of CGA. Sunflower seed is also rich in phenolic compounds (14.5% on a dry weight basis) and particularly in CGA. It is the second most cultivated oilseed in Europe with an annual production of about 16 million tons (FAO, US Department of Agriculture) in 2018 [8]. It is mainly processed for oil production. Chemical and reagents Chemical and reagents The adsorption capacity (qe, amount of CGA adsorbed per g of resin, Eq. 1) was calculated as: The sunflower meal used was defatted by n-hexane at the industrial scale and provided by Saipol (Bassen, France). Chlorogenic acid (CGA, 5-O-caffeoylquinic acid), 4-O-caffeoylquinic acid (4- CQA), 3-O-caffeoylquinic acid (3-CQA) standards and the five macroporous resins (XAD7, XAD4, XAD16, XAD1180 and HP20) were purchased from Sigma-Aldrich (St. Louis, Missouri, USA). The characteristics of macroporous resins are shown in Table 1. ( ) 0 e i e C C V q W − = where Co and Ce are the initial and equilibrium concentrations of CGA in permeate solution respectively (mg/ mL); Vi is the initial volume of permeate added into the resins (mL); and W is the weight of the dried resin (g). Resins Material Polarity Specific surface (m²/g) Pore (Å) XAD4 SDVB* Non polar 725 50 XAD7 Acrylate Polar 450 90 XAD16 SDVB Non polar 900 100 XAD1180 SDVB Non polar 600 300 HP20 SDVB Non polar 500 260 Table 1: Properties of XAD 7, XAD 16, XAD 4, XAD 1180 and HP 20 The specific surface adsorption capacity (SA, amount of CGA per m² of resin, equation 2) was calculated as: 0 ( ) e i C C V SA SS W − = × 0 ( ) e i C C V SA SS W − = × where SA is the surface adsorption capacity (mg / m2), SS is the resin specific surface (m²/ g). 2 OPEN ACCESS Freely available online Adsorption kinetics Adsorption capacity was monitored after 5, 10, 15, 30, 60, 90, and 120 min. To do so, CGA concentration was measured in the liquid phase by HPLC. From the concentration, qe and/ or SA was deduced. Results were plotted under linearized models (pseudo- first-order, pseudo-second-order, and intra-particle diffusion, Table 2). Kinetic model Linear form Plot Parameter Resin XAD 7 XAD 16 Pseudo-first-order ( ) 1 ln ln e t e q q q k t − = − ( ) ln . e t q q vs t − ( ) ( ) ( ) 1 . . 1/ / ² / ² ² e exp e cal k min q mg m q mg m R -0.082 .053 Pseudo-second-order 0.017 0.016 0.01 0.971 0.982 Pseudo-second-order 2 2 1 1 t e e t q k q q = + 1 . Adsorption kinetics Adsorption capacity was monitored after 5, 10, 15, 30, 60, 90, and 120 min. To do so, CGA concentration was measured in the Table 2: Equations and parameters for the adsorption kinetic models of CGA obtained with the XAD7 and XAD16 resins. solution in the liquid phase at the equilibrium (Ce, mg/ L). For the adsorption study, a duration of 120 min was chosen. Experiments were carried out at 25°C (298.15 K). Langmuir and Freundlich models were used to regress experimental data (Table 3). solution in the liquid phase at the equilibrium (Ce, mg/ L). For the adsorption study, a duration of 120 min was chosen. Experiments were carried out at 25°C (298.15 K). Langmuir and Freundlich models were used to regress experimental data (Table 3). Chemical and reagents t vs t q ( ) ( ) ( ) 2 2 . . ( / . / ² / ² ² e exp e cal k m mg min q mg m q mg m R 0.036 0.057 0.0273 0.017 0.0274 0.017 0.9999 1 Intra-particle diffusion 1 2 tq kt C = + 1 2 . tq vs t ( ) ( ) ( ) 1 1/2 ,1 1 1 1/2 ,2 2 ( ². . / ² ² ². . / ) ² i i K m mg min C mg m R K m mg min C mg m − − − − 0.0024 0.0017 0.014 0.0074 0.9829 0.9848 0 0 0.027 0.016 Table 2: Equations and parameters for the adsorption kinetic models of CGA obtained with the XAD7 and XAD16 resins. Tuong Thi Le, et al. Tuong Thi Le, et al. OPEN ACCESS Freely available online OPEN ACCESS Freely available online liquid phase by HPLC. From the concentration, qe and/ or SA was deduced. Results were plotted under linearized models (pseudo- first-order, pseudo-second-order, and intra-particle diffusion, Table 2). liquid phase by HPLC. From the concentration, qe and/ or SA was deduced. Results were plotted under linearized models (pseudo- first-order, pseudo-second-order, and intra-particle diffusion, Table 2). Table 3: Adsorption isotherm models and parameters for the phenolic compound obtained with the two resins selected XAD7 and XAD16. models and parameters for the phenolic compound obtained with the two resins selected XAD7 and XAD16. Data analysis All experiments and analytical measurements were carried out three times. Data displayed corresponded to the average value with the standard deviation. Statistical analysis was performed by Student’s t-test with Rstudio 3.6.1 (Boston, MA, USA). A p-value inferior to 0.05 was considered a significant difference. All figures were designed by OriginPro 8.5 package (MA, USA). The chemical structures of phenolic compounds and CGA interaction forces with selected resins were illustrated using ChemDrawUltra 8.0 software (Cambridge Soft, MA, USA). ( ) ( ) % 100%(5) d d o e i C V Desorption ratio C C V = − where Cd is the concentration of CGA in desorption solution (mg/ mL), Vd is the volume of the desorption solution (mL). where Cd is the concentration of CGA in desorption solution (mg/ mL), Vd is the volume of the desorption solution (mL). The purity of CGA after desorption step was determined using Eq.6: Nonprotein nitrogen content (4) ∆G = - RT ln Keq Kjeldahl method was used to measure the total nitrogen in the permeate [30]. A nitrogen-to-protein conversion coefficient of 5.6 was applied as recommended in [27] for sunflower source. where ∆G (J/ mol) is the Gibbs energy change. NaCl content NaCl content in the permeate was assessed by conductimetry (Meterlab PHM 210, Radiometer analytical, France). The quantification was done using an NaCl calibration curve ranging from 0.2 to 50 g/L (y = 1.9054x with R² = 0.9908). The effect of the adsorption temperature was investigated by determining the adsorption isotherms at 298.15, 308.15, and 318.15 K. Enthalpy and entropy variations were obtained from the slope and intercept of the linear plot lnKeq vs 1/T according to the linear form of Clausius-Clapeyron Eq. 3: Tuong Thi Le, et al. OPEN ACCESS Freely available online Tuong Thi Le, et al. Dry matter 1 mL of permeate was put in aluminium dish and left in oven during 24 h at 111°C. Afterward, aluminium dish was weighted regularly until reaching a constant weight. Desorption Desorption ratio were determined using different water/ethanol solvents after the adsorption step on XAD7 and XAD16 up to equilibrium. To do so, 40 mL of 30, 50, 70, and 90%, v/v was added to the resins, and shaken at 150 rpm at 25°C for 2 hours (to reach desorption equilibrium). Resins were washed with deionized water twice prior solvent addition. Resins were separated from the liquid filtration using filter paper. CGA concentration in the liquid was quantified by HPLC. Desorption ratio was calculated using Eq.5: Total carbohydrate content H S RT R ln Keq ∆ ∆ = − + H S RT R ln Keq ∆ ∆ = − + Total carbohydrate content in the permeate was measured by the method of anthrone-sulfuric acid presented by Yemm and Willis with some modifications [29]. Briefly, samples were mixed with 2 g/L anthrone in 98% sulfuric acid in boiling water for 10 min. After cooling, the absorbance of the solution was measured at 620 nm in a multiplate (Multiskan GO, Thermo scientific, Japan). Glucose was used as standard with concentration ranging from 0.1 to 1 mg/ mL (y = 1.1845x, R² = 0.9981). where lnKeq is the natural logarithm of the constant of adsorption equilibrium (Keq), ∆H is the enthalpy change (J/ mol), ∆S is entropy change (J/ mol), R is the universal gas constant (8.3144 J/ (molK) and T is the absolute temperature in Kelvin (K). ∆G was determined using Eq. 4: ∆G = - RT ln Keq (4) where ∆G (J/ mol) is the Gibbs energy change. ∆G was determined using Eq. 4: ( ) ( ) ( ) Purity of CGA % 100%(6) Amount of CGA after desorption mg Total total mass mg = ( ) ( ) ( ) Purity of CGA % 100%(6) Amount of CGA after desorption mg Total total mass mg = Characterization of the ultrafiltration permeate from sunflower protein purification Figure 1 shows the SE-HPLC chromatogram at 325 nm and 280 nm of ultrafiltration (UF) permeate obtained from the purification of proteins extracted from sunflower meal at pH 7.5, 0.5 M NaCl. This figure shows the main peak at 32 min of retention time. Marginal signals can be observed at 13, 37 and 39 min of retention time. The first minor peak (13 min) corresponded to the column void volume and showed a higher signal at 280 nm rather than 325 nm. SDS-PAGE analysis of this peak revealed the presence of proteins in a molar weight range of sunflower albumins (SFA, 10- 18 kDa) [27]. Hence, the 3 kDa membrane used for the purification probably allowed the transmission of traces of these proteins. SFA is particularly associated with phenolic compounds by covalent bonds [31]. This would explain the signal observed at 325 nm at this retention time. Peaks at 32, 37, and 39 min revealed the presence of molecules with m/z of 355 in positive mode ESI-MS. This corresponds to chlorogenic acid (CGA) m/z. The figure also shows that standard CGA isomers had the same retention times (5-CQA at 32 min, 3-CQA at 37 min, and 4-CQA at 39 min). This indicated that free polyphenolic fraction in the UF permeate was composed of CGA, mainly in its 5-CQA isomer. The predominance of CGA in sunflower and relative in aqueous extracts of sunflower meal [13,27]. The relative absence of caffeic acid and CGA dimers that account for 15-20% of total phenolic compounds in the meal according to Weisz GM et al. 2009 [13] are probably less extracted in aqueous solvents due to their polarity lower than CGA. In any cases, it makes this side products of sunflower meal protein purification a particularly attractive source of CGA. Indeed, generally recognized Analytical methods OPEN ACCESS Freely available online OPEN ACCESS Freely available online RESULTS AND DISCUSSION ( ) ( ) ( ) Purity of CGA % 100%(6) Amount of CGA after desorption mg Total total mass mg = Adsorption isotherms Adsorption isotherms expressed the relationship between CGA adsorption capacity (qe, mg/g) and the concentration of sample Isotherm model Non-linear form Parameter Constraint Resin XAD 7 XAD 16 Langmuir 1 max L e e L e Q K C q K C = + ( ) ( ) / ² / ² max L Q mg m K L mg R 1 max L e e L e Q K C q K C = + 0.054 0.040 0.026 0.021 0.9899 0.9586 Freundlich n e F e q K C = / ² ² n F mg mg K m L n R          0 < n ≤ 1 0.007 0.003 0.358 0.445 0.8690 0.9119 Table 3: Adsorption isotherm models and parameters for the phenolic compound obtained with the two resins selected XAD7 and XAD16. Table 3: Adsorption isotherm models and parameters for the phenolic compound obtained with the two resins selected XAD7 and XAD16. 3 3 OPEN ACCESS Freely available online OPEN ACCESS Freely available online OPEN ACCESS Freely available online as interesting sources of CGA like low-grade coffee bean, coffee bean methanol extracts or Crafton weed ethanol extracts are composed of many other phenolic compounds [6,32,33]. resins made of styrene divinylbenzene (SDVB). XAD7 was a slightly polar resin made of an acrylic polymer. resins made of styrene divinylbenzene (SDVB). XAD7 was a slightly polar resin made of an acrylic polymer. XAD16 and XAD4 showed the highest adsorption capacity at equilibrium (15.32 ± 0.04 mg/g and 14.07 ± 0.01 mg/g respectively). XAD7, XAD1180 and HP20 had lower values (12.03 ± 0.45, 12.62 ± 0.07 and 11.21 ± 0.02 mg/g dried resin, respectively). The very similar CGA adsorption capacity was observed with XAD7 and HP20 (11 mg/g and 15 mg/g, respectively) from crafton weed extract [6]. Besides, XAD16 was shown to be particularly efficient for CGA adsorption from sunflower meal aqueous extracts [25,26]. The adsorption capacity depends on adsorption kinetics pattern, molecule affinity toward the material and resin specific area [37]. It can also be modulated by the pore diameter. Indeed, if the diameter is low enough (≤ 50 Å) it can induce diffusional limitations by steric hindrance [35,37] or even limit the accessibility of a fraction of the resin area. Component Compound Proportion (%) Liquid by- product(UF permeate) Chlorogenic acid, CGA (5-CQA) % dm 1.13 ± 0.21 NaCl % dm 87.67 ± 11.93 Total carbohydrates % dm 3.53 ± 1.98 Nitrogen contain- ing molecules (N x 5.6) % dm 0.11 ± 0.02 Ash % dm 7.56 ± 0.70 Table 4: Composition of the liquid by-product. Table 1 shows material, pore diameters, and specific area of the resins used in the study. Non-polar resins, Among SDVB resins, XAD16 showed the highest specific area (900 m²/g) followed by XAD4, XAD1180, and HP20. Interestingly, this ranking corresponded to the observed capacity values. XAD4 had the smallest pore diameters (50 Å). This indicated that SDVB resin mass capacities were essentially governed by the contact area. Probably no or few diffusional limitations due to the pore size occurred [38]. This was also suggested by Liu B et al. 2016 [6] with CGA capture from Eupatorium Adenophlorium extracts. Table 4: Composition of the liquid by-product. Table 4 presents the proximate composition of the permeate dry matter. OPEN ACCESS Freely available online The high content in NaCl (87 ± 11.93 %) was consistent with a large amount of NaCl either for the protein extraction step and the protein purification. Ash (7.56 ± 0.70%) were minerals extracted from the meal that also highly crossed the membrane. The permeate also contained carbohydrates (3.53 ± 1.98%). These were probably soluble fibers with molar weights low enough to cross the membrane or simple carbohydrates (both contained in the sunflower meal [34]). The nitrogen-containing biomolecules fraction (0.11 ± 0.02 %) is probably not homogeneous. A part of it might be the albumin traces in the permeate. Besides, about 11% of nitrogen assayed by Kjeldahl analysis in the protein extract remained soluble in 10% (v/v) trichloro acitic acid (TCA) (data not shown). These nitrogen containing molecules are classically referred to as ‘non-protein’ nitrogen and meant to be low molar weight peptides, free amino acids and/ or nucleic acids. These molecules probably constituted the other part of the nitrogen containing molecules observed. In order to investigate the effect of a resin material on CGA adsorption, surface capacity had to be compared. Figure 2B shows that XAD7 had by far the highest surface capacity (27 µg/m² dry resin) whereas its specific area was rather low (450 m²/g). For SDVB resins the value was close to 20 µg/m2). This indicated that acrylic material had a better affinity than SDVB toward CGA. Figure 3: Schemes for mechanism proposed of the adsorption process of the phenolic compound onto the different resins. (A) CGA onto XAD7 resin and (B) CGA onto XAD16 resin. This table also shows that around 25% of the UF permeate organic dry matter is composed of CQA. The rest of the organic matter is composed of polar molecules. Liu B et al. 2016 [6] showed that polar resin exhibited higher CGA adsorption capacity than apolar or mildly polar ones. But they observed poor CGA purity after desorption step. This was due to the competitive adsorption of polar molecules on the polar resin. Or it is likely that polarity and affinity characteristic of the elution solution and the adsorbed components may influence the purity of CGA. Hence, for CGA capture from sunflower meal aqueous extract, apolar or mildly polar resin should be more appropriate. Figure 3: Schemes for mechanism proposed of the adsorption process of the phenolic compound onto the different resins. OPEN ACCESS Freely available online (A) CGA onto XAD7 resin and (B) CGA onto XAD16 resin. Lin L et al. 2012 [39] reported that chemical features of XAD7 and HP20 resins were as important as physical characteristics on phenolic compounds adsorption [39]. The adsorption mechanism remains unclear though. In this study, we propose a possible CGA interaction scheme with acrylic and SDVB resins (Figure 3). CGA would interact with XAD7 through hydrogen bonding implying alcohol functions of the caffeic acid part and the acrylate ester bonds of the resin backbone. Hydrophobic interactions between the caffeic part (benzene ring) and acrylate carbon backbone might also take place (Figure 3B). Water molecules are probably partly excluded from this backbone, otherwise, they would compete with CGA and no CGA adsorption would occur. It is more likely that the highly hydrophilic quinic acid part interacts more favorably with the bulk water molecules. pp p Resins screening Figure 2: Mass (A) and surface (B) adsorption capacity of CGA on XAD4, XAD16, XAD7, XAD1180, and HP20. In a different way, the interaction of CGA – SDVB resins should rather be through - stacking interactions. The highest efficiency of acrylate material to capture CGA could be due to the highest amount of binding sites. But the very high specific area of XAD16 makes it the better resin in terms of massic capacity. Figure 2: Mass (A) and surface (B) adsorption capacity of CGA on XAD4, XAD16, XAD7, XAD1180, and HP20. Figure 2A shows CGA adsorption capacity from XAD4, XAD7, XAD16, XAD1180, and HP20. Those resins were known for their ability to adsorb phenolic compounds from many different plant extract [35,36]. XAD16, XAD4, XAD1189, and HP20 were apolar Figure 2A shows CGA adsorption capacity from XAD4, XAD7, XAD16, XAD1180, and HP20. Those resins were known for their ability to adsorb phenolic compounds from many different plant extract [35,36]. XAD16, XAD4, XAD1189, and HP20 were apolar The adsorption kinetics, isotherms, thermodynamics properties, and CGA desorption were further investigated with XAD7 and XAD16 to better understand the process with these two resins. CGA identification and quantification CGA concentration in the liquid by-product was quantified by HPLC (Shimadzu Corporation, Kyoto, Japan) according to [27] with some modification. The equipment used included a pump with a degasser (LC-20AD), an auto-sampler (SIL-20AC), a column oven (CTO-20A), a diode array detector (CPO-M20A) and a LCSolutions software. The analysis was carried out with a Biosep SEC-s2000 column (300 x 7.8 mm, 5 µm) purchased from Phenomenex (Torrance, CA, USA). The mobile phase consisted of formic acid: ultrapure water: acetonitrile mixture (0.1%: 55%: 45%, v/v). The injection volume was 5 µL. The flow rate was 0.6 mL/ min. The detection wavelength was 325 nm. The oven temperature was kept at 35°C. All solutions and samples were filtered through 0.45 µm membranes (Fisher Scientific, Hampton, USA) before injection. CGA identification was done from standards retention time by on-line electrospray ionization mass spectrometry (ESI-MS). Data acquisition and processing were monitored with LabSolution software (Shimadzu Corporation, Kyoto, Japan). Nitrogen was used as drying gas at a flow rate of 21.5 L/ min. The nebulizer temperature was set at 300°C. CGA was searched in positive mode at m/z = 355.3+. A CGA calibration curve with concentrations ranging from 0.05 to 1.25 mg/mL was used for the quantification (linear regression equation was y = 25057208.61x with R² = 0.9983). 4 Tuong Thi Le, et al. Tuong Thi Le, et al. OPEN ACCESS Freely available online Adsorption kinetics 5 5 Tuong Thi Le, et al. OPEN ACCESS Freely available online Figure 4: Adsorption kinetics of CGA with XAD 7 and XAD 16. (A) surface adsorption kinetic curves, (B) pseudo-first-order model, (C) pseudo-second-order model, and (D) intra-particle diffusion model (in linearized forms). Figure 4: Adsorption kinetics of CGA with XAD 7 and XAD 16. (A) surface adsorption kinetic curves, (B) pseudo-first-order model, (C) pseudo-second-order model, and (D) intra-particle diffusion model (in linearized forms). Adsorption kinetics of CGA (from UF permeate) on XAD7 and XAD16 were presented in Figure 4A. Very similar trends were observed with the two resins. A large part of CGA was quickly adsorbed (up to 80% of the equilibrium capacity in 15 min). Then the adsorption kinetics slowed down and the equilibrium happened between 60 min and 120 min. Thus, the adsorption process should be conducted for 120 min to achieve an equilibrium state. Interestingly, Liu B et al. 2016 [6] observed that CGA adsorption equilibrium with a highly polar resin (NKA – II) was reached at 300 min. This resin had an average pore size similar to XAD7 and XAD16 (around 120 Å). So the adsorption process on apolar or slightly polar resin would be quicker. diffusion. There is also a diffusive transport from the liquid phase to the bead surface (through a limit liquid film) and a diffusive transport inside the particle pores. Adsorption kinetics may be modulated by several diffusional types of transports. The intra- particle diffusion model [42] is commonly used to investigate the diffusive rate-controlling phenomenon [43-45]. Figure 4D shows qt vs t1/2 plots obtained with XAD 7 and XAD 16. These plots correspond to the linear form of the intra-particle diffusion model (Table 2). For both the two resins, a linear evolution with two slopes is observed. The slopes represent the constant rate (ki) of each adsorption step while Ci (intercept at y-axis) is related to the thickness of the limiting layer. R2, ki, and Ci values obtained from linear regressions are displayed in Table 2. In the two cases, k1 (0.0024 and 0.0017 m²/(mgmin0.5) for XAD7 and XAD16, respectively) are by far higher than k2 (approximately 0) for the two resins. It can also be noticed that R² values for XAD7 and XAD16 are 0.9829 and 0.9848, respectively. This indicates that for both the two resins the adsorption process is limited by two diffusional effects. Adsorption kinetics Since the intraparticle diffusion happens inside bead pores, representing the largest part of the resin surface it can be hypothesized that the limitation only concerns the deepest zone of the pores. The diffusion inside pores near the surface is probably only limited by the boundary layer. This would explain why the intraparticle diffusion limitation affects less than 10% of the overall adsorption. The close pore size of the two resins (around 100 Å) explains their close behavior in terms of diffusional limitations. Figure 6: Equilibrium adsorption isotherm using the Langmuir model in linear form. (A) XAD7 and (B) XAD16 at 25°C, 35°C and 45°C; (C) ln KL vs. 1/T plot of adsorption equilibrium constant KL using the Langmuir model. The effect of temperature on both resins adsorption capacity of Adsorption isotherms Figure 5: Adsorption isotherms non-linear fitting of selected XAD 7 and XAD 16 resin. (A) Langmuir model and (B) Freundlich isotherm model. Adsorption isotherms Figure 5: Adsorption isotherms non-linear fitting of selected XAD 7 and XAD 16 resin. (A) Langmuir model and (B) Freundlich isotherm model. Figure 5 shows the adsorption isotherms of CGA on XAD7 and XAD16 at 25°C. Data were regressed with Langmuir (Figure 5A) and Freundlich (Figure 5B) equations as commonly done elsewhere [6,20,26]. Table 3 listed the R² of the regression, equations, and model parameters with XAD7 and XAD16. R2 values indicated that experimental data were better fitted by the Langmuir model (0.9899 and 0.9586 for both XAD7 and XAD16, respectively) than the Freundlich model (0.8690 and 0.9119 for XAD and XAD16, respectively). This indicated that the adsorption mechanism was the same for the two studied resins. It consisted in a monolayer adsorption of phenolic compounds at the surface resin [6,20,26]. These findings also are in agreement with other work on CGA adsorption on XAD16 HP and NKA-II resins from other sources [6,26]. Table 3 indicated that maximum adsorption capacity based on the resin area of XAD7 was higher than XAD16 (0.054 mg/ m² vs. 0.040 mg/m2). However, as expected from screening experiments, XAD16 showed higher maximum capacity when results were based on resin amount (36.59 mg/g for XAD16 vs. 26.21 mg/g for XAD7). Considering kinetics result with the two resins it can be deduced that XAD7 polymer (acrylate resin) had a better adsorption property than XAD16 polymer (styrene divinyl benzene). The far better specific surface area of XAD16 (900 m²/g vs. Adsorption kinetics Very similar results were observed with the adsorption of alfalfa phenolic compounds and on HP20 and AER1 resins [44]. It was interpreted as a two steps adsorption process. The first one is related to the diffusional transport throughout the boundary layer at the liquid/ beads interface. Its high rate constant (Ki,1 was 0.0024 and 0.0017 (mg/m²) for XAD7 and XAD16, respectively) indicates a low diffusional limitation. The second one is due to intraparticle diffusion. The low rate constant (Ki,2 approximately equal to 0 for both resins) associated indicates a strong diffusional limitation. Such observation and explanation were also made by others [36,46,47]. Curiously, it can be noticed that the first step involved the adsorption of the largest part of the phenolic compounds (more than 90% of the adsorption at the equilibrium). This tends to indicate that the intraparticle diffusion limitation Adsorption kinetics were regressed with pseudo-first-order (PFO, [40]) and pseudo-second-order (PSO, [41]) equations in the linearized form (Figure 4B-C). Table 2 shows the corresponding equations, parameter values of the models, and R2 of the linear regressions. The R2 obtained with linearized PFO (ln(qe-qt) vs. t) was less than 0.98. Furthermore, the calculated qe for XAD7 (0.016 mg/m²) and XAD16 (0.010 mg/m²) was found very different from experimental values (0.027 mg/m² for XAD7 and 0.017 mg/m² for XAD16). On the other hand, R2 of the linear regression of the t/ qt vs t plot was very close or equal to 1 (0.999 and 1 for XAD7 and XAD16, respectively). Moreover, the calculated qe (0.0274 mg/ m² for XAD7 and 0.017 mg/m² for XAD16) predicted from PSO model was very near the experimental values of qe (0.0273 mg/m² for XAD7 and 0.017 mg/m² for XAD16). These results indicate that adsorption kinetics followed a PSO model for the two resins. This was also observed with adsorption of CGA from Eupatorium adenophorum Spreng extracts on NKA-II resin [6] and Helianthus tuberosus L. leaves extracts on ADS-21 resin [20]. Solute transport phenomena are complex in adsorption processes. In the liquid phase, solutes are transported by convection and 6 OPEN ACCESS Freely available online Tuong Thi Le, et al. Tuong Thi Le, et al. Tuong Thi Le, et al. only took place for a small part of phenols adsorption. Determination of thermodynamic parameters Adsorption kinetics 450 m²/g) probably made its highest massic adsorption capacity. In any case, it would confirm the hypothesis of a higher adsorption site density on acrylate material. Besides, Qmax value of CGA adsorption from a sunflower (Helianthus annuus L.) aqueous extract on XAD16HP was observed at 42.7 mg/g [25]. This result is close to what is observed here. The discrepancy may be due to the interference of other organic molecules. Indeed, the above-mentioned article dealt with a raw extract containing proteins. It might also be due to the impact of an ionic strength. In this study a UF permeate was used (vs. a whole extract) and the NaCl content was around 0.5 M. Qmax observed here were lower than on the highly polar NKA-II resin (66.863 mg/g) [6]. Authors claimed that this high adsorption capacity was due to the resin polarity, but discrepancies are harder to interpret since the starting material was different (Crofton weed extract). Figure 6: Equilibrium adsorption isotherm using the Langmuir model in linear form. (A) XAD7 and (B) XAD16 at 25°C, 35°C and 45°C; (C) ln KL vs. 1/T plot of adsorption equilibrium constant KL using the Langmuir model. Figure 6: Equilibrium adsorption isotherm using the Langmuir model in linear form. (A) XAD7 and (B) XAD16 at 25°C, 35°C and 45°C; (C) ln KL vs. 1/T plot of adsorption equilibrium constant KL using the Langmuir model. The effect of temperature on both resins adsorption capacity of The effect of temperature on both resins adsorption capacity of Determination of thermodynamic parameters 7 7 OPEN ACCESS Freely available online∆∆∆∆∆∆∆∆ Resin Temperature (0C/ K) Isotherm parameter Thermodynamic parameter Langmuir model KL Qmax(mg/m²) R2 ∆H (kJ/mol) ∆G(kJ/mol K) ∆S (kJ/mol K) XAD7 25/298.15 1.923 0.13 0.9945 -37.75 -2.235 -0.119 35/308.15 1.543 0.12 0.9994 -1.044 45/318.15 0.142 0.09 0.9998 0.147 XAD16 25/298.15 1.923 0.19 0.9994 -22.47 -1.932 -0.068 35/308.15 1.543 0.09 0.9982 -1.243 45/318.15 0.142 0.07 0.9997 Table 5: Isotherm and thermodynamic parameters of phenolic compound adsorption on XAD 7 and XAD 16 at 25°C, 35°C and 45°C. Tuong Thi Le, et al. Tuong Thi Le, et al. Table 5: Isotherm and thermodynamic parameters of phenolic compound adsorption on XAD 7 and XAD 16 at 25°C, 35°C and 45°C. CGA was also investigated in order to get the thermodynamic parameters of adsorption. Figure 6A-B shows adsorption isotherms at 25°C, 35°C and 45°C (linearized Langmuir model). Adsorption kinetics This may indicate stronger interactions throughout hydrogen bonding (occurring with XAD7) rather than  -  interactions (occurring with XAD16). In the past, Liu B et al. 2016 [6] reported that the purity of CGA from the organic extract of Eupatorium adenophorum Spreng was 22.17% using NKA-II macroporous resin (polar). Meanwhile, Sun PC et al. 2015 [20] indicated that CGA separation from Helianthus tuberosus L. leaves extract by ADS-21 (polar resin) was 65.2%. On the contrary, in this study, the purity of CGA from sunflower meal when we used XAD7 and XAD16 were 77.56 ± 0.99 and 74.59 ± 0.12%, respectively. It might be the effect of the polarity of macroporous resin on the adsorption selectivity of CGA to some degree. The results in the current study indicated that selected macroporous resins were efficient for separation of CGA notably with XAD7 resin (moderate polar resin). Adsorption kinetics Langmuir model parameters and R² were listed in Table 5. Obviously, the adsorption of phenolic compounds decreased with increasing temperature (Table 5). XAD7 Qmax ranged from 0.13, to 0.09 mg/m² from 25°C to 45°C while XAD16 Qmax value decreased from 0.19, 0.09 and 0.07 mg/m² in the same temperature range. ∆H and ∆S were determined through the slope and intercept of ln KL against 1/T (Eq. 3) (Figure 6C) according to Van Hoff’s equation. The enthalpy changes (∆H) for the CGA adsorption process on both XAD7 and XAD16 resin were -37.35 and -22.47 kJ/mol, respectively (Table 5). Negative values indicate an exothermic adsorption process. The fact that values were less than 43 kJ/ mol demonstrated that the adsorption process of CGA on the resins was governed by physical rather than chemical interactions [35,48,49]. That demonstrated that XAD7 and XAD16 resins would not undergo structural changes during the CGA adsorption process. Therefore, adsorption of CGA on the resins only takes place through physical mechanism with no chemical reactions. This observation was also reported in the work of Gao ZP et al. 2013 [50] when they studied the adsorption of polyphenols separation from kiwifruit juice using AB-8 resin [50]. Furthermore, it indicated that the adsorption process of CGA for both resins should be conducted at room temperature (around 25°C). In addition, the entropy changes (∆S) values of XAD7 and XAD16 were -0.119 and -0.068 kJ/molK, respectively. These negative values suggested a random adsorption process at the solid-liquid interface [51] which happened owing to the desorption process of water molecules previously adsorbed onto the resins’ surface [50]. The negative free energy change (∆G) deduced from ∆H and ∆S (Table 4). suggested that CGA adsorption onto XAD7 and XAD16 was a spontaneous process. Moreover, the absolute value of ∆G < 20 kJ/mol confirmed physical adsorption of CGA onto both resin [50,52]. These results are also online with the proposed adsorption scheme in Figure 3. showed that the highest desorption ratio (90.9%) was observed when ethanol concentration was 70% (v/v). Such a maximum ratio was also observed with the desorption of flavonoids from G. glabra L. leaf from XAD16 (at ethanol 80%, v/v) [36]. Interestingly, for ethanol concentration below 70% (v/v), the desorption ratio of CGA for XAD16 was higher than for XAD7 (p < 0.05). Relatively large desorption ratio discrepancies were particularly observed at 30% ethanol (48.08% for XAD7 and 56.03% for XAD16). J Chromatogr Sep Tech., Vol.11 Iss.6 No:435 CONCLUSION XAD7 (mildly polar) and XAD16 (apolar) showed high CGA surface and massic adsorption capacities (from an aqueous by-product of sunflower protein purification process). The adsorption kinetics of polyphenols on the two resins followed a pseudo-second-order model with a similar intraparticle diffusion pattern. The Langmuir model described more accurately the adsorption behavior of phenolic compounds on both the two resins indicating a monolayer adsorption behavior. The negative value of enthalpy, entropy, and Gibbs free energy indicated spontaneous and exothermic processes. Hence, the adsorption process was controlled by a physical mechanism and should be favorably performed at low temperatures. We concluded that the high surface adsorption capacity on XAD 7 was probably due to a high frequency of the binding sites on acrylate polymer. The high massic adsorption capacity observed with XAD16 is due to its very high specific area. In addition, the highest desorption ratio was observed at ethanol solution at 70% (v/v) but slight discrepancies were observed between the two resins. This would be due to the different interactions at stake between CGA and the two resins (hydrogen bond with XAD7 and  –  stacking with XAD16). Desorption of CGA on XAD7 and XAD16 Desorption of CGA on XAD7 and XAD16 Ethanol solutions from 30, 50, 70, and 90%, v/v were used to desorb CGA from XAD7 and XAD16. Desorption kinetics showed that 120 min was necessary to reach the equilibrium (data not shown). This observation was in agreement with Xi L et al. 2015 [53] that studied the desorption static of polyphenols from sweet potato Ipomoea batatas L. leaves on AB-8 resin. Figure 7 shows a maximum of CGA desorption ratio from 70% ethanol (v/v) in both the two cases. At this ethanol concentration (and higher), the desorption ratio was around 90%. Xi L et al. 2015 [53] also 8 LIST OF ABBREVIATION/ NOMENCLATURE 2. Ren J, Zheng Y, Lin Z, Han X, Liao W. Macroporous resin purification and characterization of flavonoids from Platycladus orientalis (L.) Franco and their effects on macrophage inflammatory response. Food Funct. 2017;8(1):86-95. LIST OF ABBREVIATION/ NOMENCLATURE SFA Sunflower Albumin DF Diafiltration UF Ultrafiltration TCA Trichloro acetic acid SE-HPLC Size exclusion high-performance liquid chromatography SDS-PAGE Sodium dodecyl sulfate – polyacrylamide gel electrophoresis PVDF Polyvinylidene diflouride CGA chlorogenic acid 3-CQA 3-O-caffeoylquinic acid 4-CQA 4-O-caffeoylquinic acid 5-CQA 5-O-caffeoylquinic acid qe adsorption capacity (mg/g) C0 initial concentration of phenolic compound (mg/mL) Vi volume of the initial sample solution (mL) W weight of the tested dry resin (g) Ce equilibrium concentration of phenolic compound (mg/ mL) SA surface adsorption capacity (mg / m2) SS specific surface (m2/ g) PFO pseudo-first-order PSO pseudo-second-order qt amount of adsorbate uptake per mass of adsorbent at any time t (min) k1 rate constant of the PFO equation (1/min); k2 rate constant of the PSO equation(g/mg×min) ki rate constant of the intra-particle diffusion model (mg/ gmin1/2) C constant associated with the thickness of the boundary layer (mg/g) Qmax maximum saturated monolayer adsorption capacity of polyphenol (mg/g) KL Langmuir constant (L/mg) KF Freundlich constant (mg/g)/(mg/L)n n Freundlich intensity parameter (0 < n ≤1) ∆H enthalpy change (kJ/mol) ∆S entropy change (kJ/mol K) ∆G Gibbs free energy change (kJ/ mol K) R ideal gas constant (8.314 J/ mol K) T temperature (K) 3. Zhang H, Tsao R. Dietary polyphenols, oxidative stress and antioxidant and anti-inflammatory effects. Curr Opin Food Sci. 2016;8:33-42. 4. Tajik N, Tajik M, Mack I, Enck P. The potential effects of chlorogenic acid, the main phenolic components in coffee, on health: a comprehensive review of the literature. Eur J Nutr. 2017;56(7):2215-2244. 5. Li P, Huo L, Su W, Lu R, Deng C, Liu L, et al. Free radical- scavenging capacity, antioxidant activity and phenolic content of Pouzolzia zeylanica. J Serb Chem Soc. 2011;76(5):709-717. 6. Liu B, Dong B, Yuan X, Kuang Q, Zhao Q, Yang M, et al. Enrichment and separation of chlorogenic acid from the extract of Eupatorium adenophorum Spreng by macroporous resin. J Chromatogr B. 2016;1008:58-64. 7. Tan Z, Wang C, Yi Y, Wang H, Li M, Zhou W, et al. Extraction and purification of chlorogenic acid from ramie (Boehmeria nivea L. Gaud) leaf using an ethanol/salt aqueous two-phase system. Sep Purif Technol. 2014;132:396-400. 8. http://www.fao.org/faostat/en/#data/QC/visualize (accessed on 26th May 2020 ), (n.d.). 9. ACKNOWLEDGMENT The authors acknowledge financial support from the “Impact Biomolecules” project of the “Lorraine Université d’Excellence” (Investissements d’avenir-ANR). OPEN ACCESS Freely available online Keq equilibrium distribution coefficient of adsorption isotherm REFERENCES 1. Wong SK, Chin KY, Suhaimi FH, Ahmad F, Ima-Nirwana S. Effects of metabolic syndrome on bone mineral density, histomorphometry and remodelling markers in male rats. PLoS ONE. 2018;13(2):e0192416. J Chromatogr Sep Tech., Vol.11 Iss.6 No:435 J Chromatogr Sep Tech., Vol.11 Iss.6 No:435 Tuong Thi Le, et al. OPEN ACCESS Freely available online CONFLICT OF INTEREST The authors declare that they have no conflict of interest. OPEN ACCESS Freely available online OPEN ACCESS Freely available online Tuong Thi Le, et al. Int J Food Sci Technol. 2011;46(1):109-115. 17. Jia G, Lu X. Enrichment and purification of madecassoside and asiaticoside from Centella asiatica extracts with macroporous resins. J Chromatogr A. 2008;1193:136-141. 34. Bishop CT. Carbohydrates of sunflower heads. Can J Chem. 1955;33:1521-1529. 18. Liu QS, Zheng T, Wang P, Jiang JP, Li N. Adsorption isotherm, kinetic and mechanism studies of some substituted phenols on activated carbon fibers. Chem Eng J. 2010;157(2-3):348-356. 35. Yang Q, Zhao M, Lin L. Adsorption and desorption characteristics of adlay bran free phenolics on macroporous resins. Food Chemistry. 2016;194:900-907. 19. Scordino M, Di Mauro A, Passerini A, Maccarone E. Adsorption of Flavonoids on Resins: Hesperidin. J Agric Food Chem. 2003;51:6998-7004. 36. Dong Y, Zhao M, Sun-Waterhouse D, Zhuang M, Chen H, Feng M, et al. Absorption and desorption behaviour of the flavonoids from Glycyrrhiza glabra L. leaf on macroporous adsorption resins. Food Chemistry. 2015;168:538-545. 20. Sun PC, Liu Y, Yi YT, Li HJ, Fan P, Xia CH. Preliminary enrichment and separation of chlorogenic acid from Helianthus tuberosus L. leaves extract by macroporous resins. Food Chemistry. 2015;168:55-62. 37. Xu Z, Zhang Q, Chen J, Wang L, Anderson GK. Adsorption of naphthalene derivatives on hypercrosslinked polymeric adsorbents. Chemosphere. 1999;38(9):2003-2011. 21. Gonzalez‐Perez S, Vereijken JM. Sunflower proteins: overview of their physicochemical, structural and functional properties. J Sci Food Agric. 2007;87:2173-2191. 38. Sandhu AK, Gu L. Adsorption/Desorption Characteristics and Separation of Anthocyanins from Muscadine (Vitis rotundifolia) Juice Pomace by Use of Macroporous Adsorbent Resins. J Agric Food Chem. 2013;61:1441-1448. 22. Hudson BJF, ed., New and Developing Sources of Food Proteins, Springer US, Boston, MA, 1994. 39. Lin L, Zhao H, Dong Y, Yang B, Zhao M. Macroporous resin purification behavior of phenolics and rosmarinic acid from Rabdosia serra (MAXIM.) HARA leaf. Food Chemistry. 2012;130(2):417-424. 23. Wildermuth SR, Young EE, Were LM. Chlorogenic Acid Oxidation and Its Reaction with Sunflower Proteins to Form Green-Colored Complexes: Chlorogenic acid oxidation. Compr Rev Food Sci Food Saf. 2016;15(5):829-843. 40. Lagergren S. About the Theory of So-Called Adsorption of Soluble Substances. Sven Vetenskapsakad Handingarl. 1898;24:1-39. 24. Bongartz V, Brandt L, Gehrmann ML, Zimmermann BF, Schulze-Kaysers N, Schieber A. Evidence for the Formation of Benzacridine Derivatives in Alkaline-Treated Sunflower Meal and Model Solutions. Molecules. 2016;21(1):91. 41. Ho YS, Mckay G. Kinetic Models for the Sorption of Dye from Aqueous Solution by Wood. Process Saf Environ Prot. 1998;76:183-191. 25. OPEN ACCESS Freely available online Weisz GM, Schneider L, Schweiggert U, Kammerer DR, Carle R. Sustainable sunflower processing-I. Development of a process for the adsorptive decolorization of sunflower [Helianthus annuus L.] protein extracts. Innov Food Sci Emerg Technol. 2010;11(4):733-741. 42. Weber WJ, Morris JC. Kinetics of adsorption on carbon from solution. J Sanit Eng Div. 1963;89:31-60. 43. Annadurai G, Krishnan MRV. Adsorption of acid dye from aqueous solution by chitin: Equilibrium studies. Indian J Chem Technol. 1997;4(5):217-222. 26. Weisz GM, Carle R, Kammerer DR. Sustainable sunflower processing — II. Recovery of phenolic compounds as a by- product of sunflower protein extraction. Innov Food Sci Emerg Technol. 2013;17:169-179. 44. Firdaous L, Fertin B, Khelissa O, Dhainaut M, Nedjar N, Chataigne G, et al. Adsorptive removal of polyphenols from an alfalfa white proteins concentrate: Adsorbent screening, adsorption kinetics and equilibrium study. Sep Pur Technol. 2017;178:29-39. 27. Slabi SA, Mathe C, Framboisier X, Defaix C, Mesieres O, Galet O, et al. A new SE-HPLC method for simultaneous quantification of proteins and main phenolic compounds from sunflower meal aqueous extracts. Anal Bioanal Chem. 2019;411(10):2089-2099. 45. Tran HN, You SJ, Hosseini-Bandegharaei A, Chao HP. Mistakes and inconsistencies regarding adsorption of contaminants from aqueous solutions: A critical review. Water Research. 2017;120:88-116. 28. Sara AS, Mathe C, Basselin M, Fournier F, Aymes A, Bianeis M, et al. Optimization of sunflower albumin extraction from oleaginous meal and characterization of their structure and properties. Food Hydrocoll. 2020;99:105335. 46. Chen Y, Zhang W, Zhao T, Li F, Zhang M, Li J, et al. Adsorption properties of macroporous adsorbent resins for separation of anthocyanins from mulberry. Food Chemistry. 2016;194:712- 722. 29. Yemm EW, Willis A. The estimation of carbohydrates in plant extracts by anthrone. Biochem J. 1954;57(3):508-514. 47. Gok O, Ozcan A, Erdem B, Ozcan AS. Prediction of the kinetics, equilibrium and thermodynamic parameters of adsorption of copper(II) ions onto 8-hydroxy quinoline immobilized bentonite. Colloids Surf A Physicochem Eng Asp. 2008;317:174-185. 30. AOCS International., Official method 991.20. Nitrogen (total) in milk. Official methods of analysis, 19th edition, Gaithersburg: AOCS International. 1995. 31. Slabi SA, Mathe C, Basselin M, Framboisier X, Ndiaye M, Galet O, et al. Multi-objective optimization of solid/liquid extraction of total sunflower proteins from cold press meal. Food Chemistry. 2020;317:126423. 48. Gokmen V, Serpen A. Equilibrium and kinetic studies on the adsorption of dark colored compounds from apple juice using adsorbent resin. J Food Eng. 2002;53:221-227. 32. LIST OF ABBREVIATION/ NOMENCLATURE Geneau-Sbartai C, Leyris J, Silvestre F, Rigal L. Sunflower Cake as a Natural Composite: Composition and Plastic Properties. J Agric Food Chem. 2008;56(23):11198-11208. 10. Pickardt C, Weisz GM, Eisner P, Kammerer DR, Neidhart S, Carle R. Processing of low polyphenol protein isolates from residues of sunflower seed oil production. Procedia Food Sci. 2011;1:1417-1424. 11. Canella M, Sodini G. Extraction of gossypol and oligosaccharides from oilseed meals. J Food Sci. 1977;42(5):1218-1219. 12. Sripad G, Rao MN. Effect of methods to remove polyphenols from sunflower meal on the physicochemical properties of the proteins, J Agric Food Chem. 1987;35(6):962-967. 13. Weisz GM, Kammerer DR, Carle R. Identification and quantification of phenolic compounds from sunflower (Helianthus annuus L.) kernels and shells by HPLC-DAD/ ESI-MSn. Food Chemistry. 2009;115:758-765. 14. Ky CL, Noirot M, Hamon S. Comparison of Five Purification Methods for Chlorogenic Acids in Green Coffee Beans ( Coffea sp.). J Agric Food Chem. 1997;45:786-790. 15. Conidi C, Cassano A, Caiazzo F, Drioli E. Separation and purification of phenolic compounds from pomegranate juice by ultrafiltration and nanofiltration membranes. J Food Eng. 195 (2017) 1–13. 16. Zagklis DP, Vavouraki AI, Kornaros ME, Paraskeva CA. Purification of olive mill wastewater phenols through membrane filtration and resin adsorption/desorption. J Hazard Mater. 2015;285:69-76. 9 9 OPEN ACCESS Freely available online OPEN ACCESS Freely available online Farah A, Monteiro M, Donangelo CM, Lafay S. Chlorogenic Acids from Green Coffee Extract are Highly Bioavailable in Humans. The J Nutr. 200;138(12):2309-2315. 49. Liu T, Li Y, Du Q, Sun J, Jiao Y, Yang G, et al. Adsorption of methylene blue from aqueous solution by graphene. Colloids Surf B. 2012;90:197-203. 33. Ramalakshmi K, Hithamani G, Asha KR, Jagan Mohan Rao L. Separation and characterisation of chlorogenic acid-rich conserves from green coffee beans and their radical scavenging potential: Separation and characterisation of chlorogenic acid. 50. Gao ZP, Yu ZF, Yue TL, Quek SY. Adsorption isotherm, thermodynamics and kinetics studies of polyphenols separation from kiwifruit juice using adsorbent resin. J Food 10 Tuong Thi Le, et al. OPEN ACCESS Freely available online Tuong Thi Le, et al. 51. Gupta VK. Equilibrium Uptake, Sorption Dynamics, Process Development, and Column Operations for the Removal of Copper and Nickel from Aqueous Solution and Wastewater Using Activated Slag, a Low-Cost Adsorbent. Ind Eng Chem Res. 1998;37:192-202. OPEN ACCESS Freely available online OPEN ACCESS Freely available online Eng. 2013;116:195-201. 52. Ding L, Deng H, Wu C, Han X. Affecting factors, equilibrium, kinetics and thermodynamics of bromide removal from aqueous solutions by MIEX resin. Int J Chem Eng. 2012;181:360-370. 51. Gupta VK. Equilibrium Uptake, Sorption Dynamics, Process Development, and Column Operations for the Removal of Copper and Nickel from Aqueous Solution and Wastewater Using Activated Slag, a Low-Cost Adsorbent. Ind Eng Chem Res. 1998;37:192-202. 53. Xi L, Mu T, Sun H. Preparative purification of polyphenols from sweet potato (Ipomoea batatas L.) leaves by AB-8 macroporous resins. Food Chemistry. 2015;172:166-174. 11
https://openalex.org/W2301939152	https://europepmc.org/articles/pmc4808531?pdf=render	English	null	Increased Circulating Th17 Cells, Serum IL-17A, and IL-23 in Takayasu Arteritis	Autoimmune diseases	2,016	cc-by	5,081	Correspondence should be addressed to Ramnath Misra; rnmisra2000@gmail.com Received 14 December 2015; Revised 19 February 2016; Accepted 21 February 2016 Academic Editor: Andras Perl Copyright © 2016 Durga Prasanna Misra et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Introduction. Th17, 𝛾𝛿T, NK, and NKT cells in peripheral blood and serum IL-17 and IL-23 in Takayasu arteritis (TA) were measured and correlated with disease activity. Methods. Th17 (anti-CD3APC, CD4PECy7, and IL-17PE), NKT, NK (anti-CD3APC, CD56FITC), and 𝛾𝛿T (anti-CD3FITC and 𝛾𝛿TCRAPC) cells were enumerated by flow cytometry in peripheral blood of 30 patients with TA (ACR1990 criteria) and 20 healthy controls, serum IL-17 and IL-23 measured by ELISA. Relation with disease activity (NIH criteria, ITAS2010) was analyzed (using nonparametric tests, median with interquartile range). Results. Mean age of patients was 33.47 ± 11.78 years (25 females); mean symptom duration was 7.1 ± 5.3 years. 13 were not on immunosuppressants; 12 were active (ITAS2010 ≥4). The percentage of Th17 cells was significantly expanded in TA (patients 2.1 (1.5–3.2) versus controls 0.75 (0.32–1.2); 𝑝< 0.0001) with no differences in other cell populations. Serum IL-17 and IL-23 (pg/mL) in patients (6.2 (4.6–8.5) and 15 (14.9– 26.5), resp.) were significantly higher (𝑝< 0.001) than controls (3.9 (3.9–7.3) and undetectable median value, resp.). Subgroup analysis revealed no correlation of Th17 cells, serum IL-17, and IL-23 with disease activity or medications, nor any significant difference before and after medication. Conclusions. There is significant expansion of Th17 cells and elevated serum IL-17 and IL-23 levels in TA patients compared to healthy controls. Hindawi Publishing Corporation Autoimmune Diseases Volume 2016, Article ID 7841718, 8 pages http://dx.doi.org/10.1155/2016/7841718 Hindawi Publishing Corporation Autoimmune Diseases Volume 2016, Article ID 7841718, 8 pages http://dx.doi.org/10.1155/2016/7841718 Hindawi Publishing Corporation Autoimmune Diseases Volume 2016, Article ID 7841718, 8 pages http://dx.doi.org/10.1155/2016/7841718 1. Introduction presenting cells [5, 6]. Natural killer (NK) cells, natural killer T (NKT) cells, and 𝛾𝛿T cells also produce IL-17A [5, 7]. Since the role of IL-17A has not yet been clearly defined in TA, we proposed to study cell populations that are producers of IL- 17A in peripheral blood, that is, Th17 cells, NK cells, NKT cells, and 𝛾𝛿T cells, and serum levels of IL-17A and IL-23 in patients with TA, and looked for their relationship with clinical disease activity. Takayasu arteritis (TA) is a rare granulomatous large vessel vasculitis (LVV) of unknown etiology. The pathogenesis of TA has not yet been clearly elucidated; both innate and adaptive immunity are involved. Innate immune cells, gamma delta (𝛾𝛿) T lymphocytes and natural killer cells, are increased in aortic tissue biopsies of TA [1]. Adaptive immunity also plays a role in TA peripheral blood of patients having an increased ratio of CD4+/CD8+ T cells [2]. T-helper cells producing IFN-𝛾(Th1 cells) promote granulomatous inflammation in TA. Studies in Giant Cell Arteritis (GCA), the other variant of LVV, have shown increased Th1 and Th17 (T-helper 17 cells, producing IL-17A) in peripheral blood and also elevated serum IFN-𝛾and IL-17A [3]. Recent studies have shown efficacy of IL-6 receptor blockade (tocilizumab) in refractory TA [4]. IL-6 skews na¨ıve T-helper cells to a Th17 phenotype, maintained by IL-23 secreted by antigen Durga Prasanna Misra, Smriti Chaurasia, and Ramnath Misra Durga Prasanna Misra, Smriti Chaurasia, and Ramnath Misra Department of Clinical Immunology, Sanjay Gandhi Post Graduate Institute of Medical Sciences, Rae Bareily Road, Lucknow, Uttar Pradesh 226 014, India Department of Clinical Immunology, Sanjay Gandhi Post Graduate Institute of Medical Sciences, Rae Bareily Road, Lucknow, Uttar Pradesh 226 014, India Department of Clinical Immunology, Sanjay Gandhi Post Graduate Institute of Medical Sciences, Rae Bareily Roa Lucknow, Uttar Pradesh 226 014, India Correspondence should be addressed to Ramnath Misra; rnmisra2000@gmail.com 2. Materials and Methods The study was approved by the Institute Ethics Committee, SGPGIMS, Lucknow (Ethics committee number 2013-24- DM-Exp) and performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. Thirty consecutive patients ful- filling 1990 American College of Rheumatology criteria for TA [8] were included after seeking informed consent. For 2 Autoimmune Diseases Table 1: Demographic details of patients and controls. Patients Controls 𝑝value Total number 30 20 Mean age 33.5 (±11.8) 34.2 (±11.3) 0.84∗ Sex (female : male) 25 : 5 13 : 7 0.137∗∗ On any immunosuppressant 17 — — Median prednisolone dose (mg/day) 7.5 (6.9–20) (𝑛= 14) — — Median oral methotrexate dose (mg/week) 17.5 (15–20) (𝑛= 11) — — Median azathioprine dose (mg/day) 100 (100–150) (𝑛= 3) — — Angiographic type (Numano’s) V – 66.7% I – 23.3% IIb - 6.7% IIa – 3.3% — — ∗Mann-Whitney 𝑈test. ∗∗𝑧-statistic. Table 1: Demographic details of patients and controls. comparison, 20 healthy controls of similar age and sex were included after obtaining informed consent. Erythrocyte sedimentation rate (ESR) was measured by Westergren’s method for all patients at time of sampling. Clinical activity was assessed using ITAS2010, ITAS-A [9], and NIH criteria [10]. 1.0 mL of heparinized whole blood was collected and flow cytometry performed immediately to enumerate cell populations under study. Serum was stored at −80∘C for IL- 17A and IL-23 estimation. 10 patients presented to us for the first time and were diagnosed with Takayasu arteritis (TA); hence they were not on treatment before (i.e., treatment- na¨ıve). Clinical assessment of active disease merited starting immunosuppressive treatment. They were treated as per the standard practice at our clinic with oral weekly methotrex- ate (15–25 mg/week) and prednisolone (1 mg/kg/day for 6 weeks followed by gradual taper to 7.5 mg/day over 3 to 6 months). For these patients, repeat blood sampling was done after 3 months for comparison with baseline sample after receiving immunosuppressant. All analyses were done using nonparametric tests using Graph Pad Prism software version 6.0e (values represented as median with interquartile range in brackets with Mann-Whitney test used for intergroup comparison). A total of 100,000 events for Th17 cells and 50,000 events each for 𝛾𝛿T cells, NK, and NKT cells were acquired. 2.1. Cell Subsets Analysis by FACS 2.1.1. Th17 Cells. Heparinized whole blood cultured with complete RPMI medium and 10% FCS in the ratio of 1 : 1 was activated with 50 ng/mL phorbol myristate acetate (PMA; Sigma, St. Louis, USA) and 1 𝜇g/mL ionomycin (Sigma, St. Louis, USA) for 6 hours and Golgi Plug 2 𝜇M monensin (Sigma, St. Louis, USA) was added for the last 4 four hours of activation. Cells were surface-stained with PECy7-labeled monoclonal anti-CD4 and APC-labeled monoclonal anti- CD3 antibodies (BD Biosciences, Franklin Lakes, New Jersey, USA) followed by red blood cells (RBC) lysis using FACS lysing solution (BD Biosciences, Franklin Lakes, New Jersey, USA). After washing with phosphate buffered saline (PBS), the cells were fixed using a fixation buffer (Leucoperm, ABD Serotech, San Diego, CA, USA) and washed twice with PBS (pH7.2, 0.15 M). The cells were then permeabilized with per- meabilization buffer (Leucoperm, ABD Serotech, San Diego, CA, USA) and intracellularly stained with phycoerythrin- (PE-) labeled monoclonal antibodies against IL-17A. Cells were washed and resuspended in PBS. 2.1.4. Serum IL-17A and IL-23. IL-17A and IL-23 were mea- sured by ELISA as per manufacturer instructions (eBio- sciences, San-Diego, USA) (sensitivity for serum IL-17A, 4 pg/mL, serum IL-23, 15 pg/mL). 2. Materials and Methods NK cells were identified as CD3−CD56+, NKT cells were identified as CD3+CD56+ (in lymphocyte gate), 𝛾𝛿T cells were 𝛾𝛿-TCR and CD3+ (in CD3 gate), and CD3+CD4+ cells positive for IL-17 were identified as Th17 cells (in CD3 gate). Acquisition of cells was done on Beckman coulter flow cytometer and analyzed by Navios software. 3. Results Demographic characteristics of patients and controls are de- tailed in Table 1. Both were comparable for age and gender distribution.hif There was no significant difference in NK, NKT (Figure 1), or 𝛾𝛿T cells (Figure 2) between patients and control. However, Th17 cells were significantly expanded in patients with TA (3.1-fold higher in patients versus controls (2.69 ± 2.07 versus 0.87 ± 0.61%, resp.), 𝑝 < 0.0001). Similarly serum IL-17A levels showed a significant elevation in patients compared to controls (1.62-fold higher in patients versus controls (8.60 ± 8.07 versus 5.32 ± 1.75 pg/mL, resp.), 𝑝 < 0.0001). Serum IL-23 was detectable in 14 patients and in none of the controls, and the levels in patients were significantly higher than controls (1.75-fold higher in patients versus controls (26.17 ± 29.01 versus 14.9 ± 0.0 pg/mL, resp.), 𝑝= 0.0007) (Figure 3). Th17 cells, serum IL-17 A, or IL-23 did not correlate significantly with ESR (Pearson’s correlation coefficient −0.08, −0.34, and −0.21, resp., 𝑝> 0.05 for all), ITAS2010 (Pearson’s correlation 2.1.2. 𝛾𝛿T Cells. 100 microlitre of heparinized whole blood was surface-stained with FITC-labeled anti-CD3 antibodies and APC-conjugated anti-𝛾𝛿TCR (BD Bioscience, Franklin Lakes, New Jersey, USA) and lysed and washed as mentioned above. 2.1.3. NK and NKT Cells. 100 microlitre of heparinized whole blood was surface-stained with APC-labeled anti-CD3 and FITC-labeled anti-CD56 antibodies (BD Bioscience, Franklin Lakes, New Jersey, USA) and lysed and washed as mentioned above. In each assay, markers were drawn based on data from isotype-specific IgG antibodies. 3 3 Autoimmune Diseases Patients (n = 30) NK cells 0.40 0 10 20 30 (%) 0 5 10 15 20 (%) NKT cells 0.12 Patient Healthy control Gated on PBMC AF1 AF2 AF3 AF4 19.6% 14.6% 28.3% 37.6% 100 101 102 103 CD56-FITC 100 101 102 103 CD56-FITC 103 101 100 102 CD3-APC 103 101 100 102 CD3-APC AB1 AB2 AB3 AB4 8.0% 2.8% 32.2% 57.0% Controls (n = 20) Patients (n = 30) Controls (n = 20) Figure 1: In peripheral blood of 30 patients and 20 controls, natural killer (NK) and natural killer T (NKT) cells were identified by gating on PBMCs and surface staining for CD3 and CD56. NK cells were CD3−CD56+ and NKT cells were CD3+CD56+. Bars represent median with interquartile range. 𝑝value (Mann-Whitney test) is mentioned in each figure. 3. Results Gated on PBMC Healthy control d on PBMC 100 101 102 103 CD56-FITC 103 101 100 102 AB1 AB2 AB3 AB4 8.0% 2.8% 32.2% 57.0% CD56-FITC Controls (n = 20) Patients (n = 30) Controls (n = 20) Patients (n = 30) Patients (n = 30) Figure 1: In peripheral blood of 30 patients and 20 controls, natural killer (NK) and natural killer T (NKT) cells were identified by gating on PBMCs and surface staining for CD3 and CD56. NK cells were CD3−CD56+ and NKT cells were CD3+CD56+. Bars represent median with interquartile range. 𝑝value (Mann-Whitney test) is mentioned in each figure. as well as ITAS2010, ITAS-A, and serum levels of IL-17A and IL-23, in subgroups of patients with Th17 cell expansion (defined as levels of Th17 cells greater than 2 SD of the mean value of controls) versus those who did not and could not demonstrate any differences between these two subgroups (Table 3). coefficient −0.23, +0.05, and −0.25, resp., 𝑝> 0.05 for all), or ITAS-A (Pearson’s correlation coefficient −0.25, −0.02, and −0.26, resp., 𝑝> 0.05 for all). There was no significant correlation demonstrable between Th17 cells and serum IL-17 (Pearson’s correlation coefficient −0.11, 𝑝= 0.58) or IL-23 (Pearson’s correlation coefficient +0.37, 𝑝= 0.052). Thirteen patients were active as assessed by NIH criteria, and 12 were active as assessed by ITAS2010 ≥4. There was no significant difference in populations of Th17 cells or serum levels of IL-17A and IL-23 in between active or inactive patients. Thirteen patients were not on immunosuppressive medications, and comparing Th17 cells and serum IL-17A and IL-23 in them versus patients on immunosuppression revealed no significant difference (Table 2). For the 10 patients for whom we had paired samples before and 3 months after immunosuppression (prednisolone and methotrexate), there was a decline in Th17 cells follow- ing treatment. Serum IL-17A levels remained similar, and serum IL-23 was detectable in 3 patients before and none after immunosuppression. None of these differences reached statistical significance (Figure 4). However, post hoc power analysis revealed that the study was not sufficiently powered to distinguish the effect of medication on the levels of Th17 cells and serum IL-17 and IL-23 levels. 3. Results if Furthermore, we compared proportions of patients active (by NIH criteria) and on immune suppressive medications 4 Autoimmune Diseases 0 5 10 15 20 (%) 0.96 Patients (n = 30) Controls (n = 20) 100 101 102 103 100 101 102 103 100 101 102 103 100 101 102 103 𝛾𝛿-APC 𝛾𝛿-APC CD3-FITC CD3-FITC Patient Healthy control 0.0% 15.8% 0.0% 0.0% 84.2% AE1 AE2 AE3 AE4 0.0% 10.7% 89.3% AF1 AF2 AF3 AF4 𝛾𝛿T cells Gated on CD3+ lymphocytes Figure 2: In peripheral blood of 30 patients and 20 controls, 𝛾𝛿T cells were identified by 𝛾𝛿TCR+ cells on CD3 gate. Bars represent median with interquartile range. 𝑝value (Mann-Whitney test) is mentioned in each figure. 0.96 𝛾𝛿T cells Figure 2: In peripheral blood of 30 patients and 20 controls, 𝛾𝛿T cells were identified by 𝛾𝛿TCR+ cells on CD3 gate. Bars represent median with interquartile range. 𝑝value (Mann-Whitney test) is mentioned in each figure. 4. Discussion in remission, in contrast to our findings. There were no differences in Th17 cells, serum IL-17A, and IL-23 in patients on immunosuppression. Alibaz-Oner et al. explored serum levels of cytokines in patients with TA, and contrary to our findings, they failed to demonstrate an increase in serum IL-23 levels compared to healthy controls [12]. They had not studied serum IL-17A in their patients. The strength of our study was that we assessed levels of Th17 cells, serum IL-17, and IL-23 in a subgroup of patients before and after immunosuppression and could not demonstrate a difference.h We found an increase in serum levels of IL-17A and IL-23 in patients with TA compared with controls. The increased levels neither correlated with overall disease activity state nor became normal with immunosuppressive therapy. We also found increased expansion of Th17 cells, which produce IL- 17 A, but not with NK, NKT, or gamma delta T cells (which are other producers of IL-17A) in the peripheral blood. In a subset of immunosuppressive-na¨ıve patients, there seemed to be no differences before or after immunosuppression in Th17 cells, serum-IL-17A, or IL-23.i f In GCA, expansion of Th17 cells and increased serum IL- 17 compared to controls have been demonstrated, and these levels are suppressed with corticosteroid use [3]. Interestingly, our study confirms contrary findings in TA [11], wherein the circulating Th17 cells are not suppressed with immuno- suppressive medication use, including corticosteroids. This may suggest that Th17 cells in TA and GCA have different responses to corticosteroid therapy, in spite of both being classified as large vessel vasculitides. Our findings corroborate a recent publication by Saadoun et al. studying T cell populations and cytokines which showed increased Th17 cells in TA patients [11], which on in vitro stimulation showed increased IL-17A and IL-23 levels compared to healthy controls. 4. Discussion However, they demonstrated higher in vitro production of IL-17A in patients with active TA (as assessed by NIH criteria) compared with those 5 Autoimmune Diseases Gated on CD3+ lymphocytes Gated on CD3+ lymphocytes Healthy control 100 101 102 103 100 101 102 103 IL-17PE CD4 PEC 7 0.6% 0.8% 38.9% 59.7% AG1 AG2 AG3 AG4 3 lymphocytes Patients (n = 30) Controls (n = 20) Patients (n = 30) Controls (n = 20) Patients (n = 29) Controls (n = 20) 0 2 4 6 8 10 (%) (pg/mL) (pg/mL) 0 10 20 40 50 0 50 100 150 200 Th17 cells <0.0001 <0.0001 Serum IL-17 Serum IL-23 0.007 Patient Healthy control 100 101 102 103 100 101 102 103 100 101 102 103 100 101 102 103 IL-17PE IL-17PE CD4-PECy7 CD4-PECy7 0.2% 2.8% 40.1% 56.9% 0.6% 0.8% 38.9% 59.7% AF1 AF2 AF3 AF4 AG1 AG2 AG3 AG4 Gated on CD3 lymphocytes igure 3: T-helper 17 (Th17) cells were measured in peripheral blood of 29 patients (could not be studied in one patient due to techni ifficulties) and 20 healthy controls, after culture of whole blood for 6 hours followed by gating on CD3+ lymphocytes. Th17 cells w dentified by surface staining for CD4 and intracellular staining for IL-17A. Serum IL-17A and IL-23 were measured by ELISA. Bars represe median with interquartile range. 𝑝value (Mann-Whitney test) is mentioned in each figure. IL-17PE IL-17PE (pg/mL) Patients (n = 29) Controls (n = 20) Serum IL-23 0.007 Figure 3: T-helper 17 (Th17) cells were measured in peripheral blood of 29 patients (could not be studied in one patient due to technical difficulties) and 20 healthy controls, after culture of whole blood for 6 hours followed by gating on CD3+ lymphocytes. Th17 cells were identified by surface staining for CD4 and intracellular staining for IL-17A. Serum IL-17A and IL-23 were measured by ELISA. Bars represent median with interquartile range. 𝑝value (Mann-Whitney test) is mentioned in each figure. Figure 3: T-helper 17 (Th17) cells were measured in peripheral blood of 29 patients (could not be studied in one patient due to technical difficulties) and 20 healthy controls, after culture of whole blood for 6 hours followed by gating on CD3+ lymphocytes. Th17 cells were identified by surface staining for CD4 and intracellular staining for IL-17A. Serum IL-17A and IL-23 were measured by ELISA. Bars represent median with interquartile range. 4. Discussion 𝑝value (Mann-Whitney test) is mentioned in each figure. 6 Autoimmune Diseases Before IS After IS Before IS After IS Before IS After IS Th17 cells 0.27 Serum IL-17 0.83 0.13 Serum IL-23 0 1 2 3 4 5 (%) 0 10 20 30 40 (pg/mL) 0 10 20 30 40 (pg/mL) Figure 4: For the subgroup of 10 patients who were immunosuppressive- (IS-) na¨ıve and were started on treatment (prednisolone and methotrexate), T-helper 17 (Th17) cells and serum IL-17A and IL-23 before and 3 months after therapy were recorded. There were no significant differences. Serum IL-23 was detectable in 3 patients prior to immunosuppression and in none of them after. 𝑝value (Mann-Whitney test) is mentioned in each figure. Before IS After IS Th17 cells 0.27 0 1 2 3 4 5 (%) 0.13 Serum IL-23 After IS Figure 4: For the subgroup of 10 patients who were immunosuppressive- (IS-) na¨ıve and were started on treatment (prednisolone and methotrexate), T-helper 17 (Th17) cells and serum IL-17A and IL-23 before and 3 months after therapy were recorded. There were no significant differences. Serum IL-23 was detectable in 3 patients prior to immunosuppression and in none of them after. 𝑝value (Mann-Whitney test) is mentioned in each figure. Figure 4: For the subgroup of 10 patients who were immunosuppressive- (IS-) na¨ıve and were started on treatment (prednisolone and methotrexate), T-helper 17 (Th17) cells and serum IL-17A and IL-23 before and 3 months after therapy were recorded. There were no significant differences. Serum IL-23 was detectable in 3 patients prior to immunosuppression and in none of them after. 𝑝value (Mann-Whitney test) is mentioned in each figure. Table 2: Th17 cells, serum IL-17, and IL-23 levels with respect to disease activity parameters and immunosuppressants. 4. Discussion Th17 cells have been linked to viral persistence [24]; hence our findings of increased Th17 cells in TA could suggest viral persistence as one of the mechanisms driving TA.fi Recent literature suggests efficacy of IL-6 receptor block- ade (tocilizumab) in patients with difficult-to-treat TA [4]. IL-6, acting on IL-6 receptor, expressed on na¨ıve T-helper cells, skews differentiation towards Th17 subset. Hence our finding of Th17 expansion and increased serum IL-17 in TA may have relevance explaining the efficacy of tocilizumab therapy in TA. In view of emerging Th17-IL-17 targeted therapies in spondyloarthropathies (ustekinumab, antibody to common p40 subunit of IL12 and IL-23; secukinumab, antibody to IL-17A; brodalumab, antibody to IL-17 receptor antagonist) [25], our findings of increased IL-17 and IL-23 in TA offer potential newer therapeutic approaches in refractory TA.h Values represented as median with interquartile range in brackets. Th17 expanders were defined as those patients with TA having percentage of Th17 cells greater than 2 SD of the mean value of controls. 𝑝= ns for all comparisons between the two groups. A recent study showed increased IL-17 mRNA levels in temporal artery biopsy samples compared to controls in spite of peripheral blood IL-17 levels being similar to controls. Levels of IL-17 mRNA when elevated in the temporal artery actually predicted sustained responsiveness to glucocorticoid treatment [13]. IL-17 polymorphisms have been linked to GCA [14], and more recently in TA [15]. Studies on patho- genesis of TA are limited by the availability of aortic tissue samples [16]. However, a recent study shows expression of IL- 17A in aortic tissue biopsies of active TA patients [11]. The limitations of our study were a relatively small sample size. However, this must be considered in view of the fact that TA is a rare large vessel vasculitis. Also, we did not measure serum CRP but instead measured ESR using a standardized technique. Lack of aortic biopsy samples to study Th17 cell and IL-17 and IL-23 mRNA expression is another limitation of our study.ih p p NK cells and gamma delta T cells (which are other producers of IL-17A) have been shown to be increased in vascular tissue biopsies from patients with TA, but we could not demonstrate NK cell or gamma delta T cell expansion in the peripheral blood compared to controls. The disease process in TA originates in the walls of large vessels. 4. Discussion Activity measured by NIH criteria Active (𝑛= 13) Inactive (𝑛= 17) 𝑝value (Mann-Whitney) Th17 cells (%) 2.1 (1.4–3) 2.05 (1.6–4.3) (𝑛= 16∗) 0.55 IL-17A (pg/mL) 7.4 (4.7–9) 5.3 (4.4–8.5) 0.56 IL-23 (pg/mL) 14.9 (14.9–17.7) 15 (14.9–30.5) 0.07 Activity measured by ITAS2010 ≥4 Active (𝑛= 12) Inactive (𝑛= 18) 𝑝value (Mann-Whitney) Th17 cells (%) 2.2 (1.4–3.1) 1.9 (1.6–4.2) (𝑛= 17∗) 0.80 IL-17A (pg/mL) 7.7 (4.7–9.2) 5.4 (4.4–8.6) 0.54 IL-23 (pg/mL) 14.9 (14.9–19) 15 (14.9–28.7) 0.13 Data with respect to immunosuppressants Not on immunosuppressants (𝑛= 13) On immunosuppressants∗∗(𝑛= 17) 𝑝value (Mann-Whitney) Th17 cells (%) 2 (1.3–2.8) 2.3 (1.6–3.8) (𝑛= 16∗) 0.30 IL-17A (pg/mL) 6.5 (4.7–9) 6.1 (4.4–8.6) 0.85 IL-23 (pg/mL) 14.9 (14.9–20.6) 15 (14.9–26.6) 0.36 Values represented as median with interquartile range in brackets. ∗Th17 cell population could not be studied in one patient due to technical difficulties. ∗∗14 patients were on prednisolone, 11 on oral methotrexate, and 3 on azathioprine. e 2: Th17 cells, serum IL-17, and IL-23 levels with respect to disease activity parameters and immunosuppressants Table 2: Th17 cells, serum IL-17, and IL-23 levels with respect to disease activity parameters and 7 Autoimmune Diseases Table 3: Comparison of Th17 expanders versus the rest. Parameter Patients with expansion of Th17 cells (𝑛= 15) Patients without expansion of Th17 cells (𝑛= 14) 𝑝value On immunosuppression 6/15 7/14 0.157 Active by Kerr’s criteria 7/15 6/14 0.157 ITAS2010 1 (0–8) 1 (0–10) 0.80 ITAS-A 3 (2–10) 3.5 (1.75–11.5) 0.73 Serum IL-17A 4.96 (4.72–8.39) 7.28 (4.29–8.78) 0.48 Serum IL-23 14.9 (14.9–26.38) 14.9 (14.9–19.37) 0.52 Values represented as median with interquartile range in brackets. Th17 expanders were defined as those patients with TA having percentage of Th17 cells greater than 2 SD of the mean value of controls. 𝑝= ns for all comparisons between the two groups. Table 3: Comparison of Th17 expanders versus the rest. cells, by virtue of secreting IL-17A, can recruit neutrophils to sites of vascular inflammation, and IL-23 helps sustain the population of Th17 cells. Hence, corroboration with aortic biopsies is needed to further define the exact role of Th17 cells in the pathogenesis of TA. Rare case reports of large vessel arteritis induced by Epstein-Barr virus [22] and coxsackie virus [23] have been described in literature. The etiology of TA is not yet clear, and one of the hypotheses, emerging from such case reports, is a viral infection triggering vasculitis. Disclosure Current affiliation to Durga Prasanna Misra: Department of Clinical Immunology, Jawaharlal Institute of Postgraduate Medical Education and Research (JIPMER), Puducherry – 605006, India. Th17 expansion has also been demonstrated in variable vessel vasculitis, that is, Behcet’s disease [19]. In small vessel vasculitides, Th17 skewing has been demonstrated in patients with inactive granulomatosis with polyangiitis (Wegener’s) and active eosinophilic granulomatosis with polyangiitis [20]. Th17 cells remained elevated in patients with ANCA vasculitis both in active disease as well as in remission, similar to our findings in TA [21].hih 4. Discussion Levels of cytokines and cellular alterations in the peripheral blood are only a surrogate of these changes in the aortic tissue and may not reflect those at tissue level. This might explain why there was no demonstrable correlation between Th17 cells and serum levels of IL-17, or difference with respect to disease activity or medication intake. The clinical assessment of disease activity in TA is imperfect and is a work in progress [17]. Indeed aortic biopsies have shown active inflammation in patients with TA with a normal ESR [18].h To conclude, our finding of increased Th17 cells and serum IL-17A and IL-23 in TA offers newer insights into the pathogenesis of a rare large vessel vasculitis. The lack of suppression of Th17 cells in patients receiving immunosup- pression, unlike in GCA, lends credence to the hypothesis that TA and GCA are pathogenetically different. Tissue level studies looking for both innate and adaptive immune cells and cytokines will provide further support to the role of Th17 cells in this disease. Competing Interests None of the authors have any competing interests to declare. References i The exact pathogenic significance of Th17 cell expansion in TA is unclear. Aortic biopsies from patients with TA have shown neutrophil infiltrates in vascular lesions [1]. Th17 [1] Y. Seko, S. Minota, A. Kawasaki et al., “Perforin-secreting killer cell infiltration and expression of a 65-kD heat-shock protein in 8 Autoimmune Diseases aortic tissue of patients with Takayasu’s arteritis,” The Journal of Clinical Investigation, vol. 93, no. 2, pp. 750–758, 1994. [19] J. Kim, J. A. Park, E. Y. Lee, Y. J. Lee, Y. W. Song, and E. B. Lee, “Imbalance of Th17 to Th1 cells in Behcet’s disease,” Clinical and Experimental Rheumatology, vol. 28, no. 4, supplement 60, pp. S16–S19, 2010. aortic tissue of patients with Takayasu’s arteritis,” The Journal of Clinical Investigation, vol. 93, no. 2, pp. 750–758, 1994. [2] S. Sagar, N. K. Ganguly, M. Koicha, and B. K. Sharma, “Immunopathogenesis of Takayasu arteritis,” Heart and Vessels, vol. 7, no. 7, pp. 85–90, 1992. [20] B. Wilde, M. Thewissen, J. Damoiseaux, P. van Paassen, O. Witzke, and J. W. C. Tervaert, “T cells in ANCA-associated vasculitis: what can we learn from lesional versus circulating T cells?” Arthritis Research and Therapy, vol. 12, article 204, 2010. [3] J. Deng, B. R. Younge, R. A. Olshen, J. J. Goronzy, and C. M. Weyand, “Th17 and th1 T-cell responses in giant cell arteritis,” Circulation, vol. 121, no. 7, pp. 906–915, 2010. [21] E. Nogueira, S. Hamour, D. Sawant et al., “Serum IL-17 and IL-23 levels and autoantigen-specific Th17 cells are elevated in patients with ANCA-associated vasculitis,” Nephrology Dialysis Transplantation, vol. 25, no. 7, pp. 2209–2217, 2010. [4] R. Goel, D. Danda, S. Kumar, and G. Joseph, “Rapid control of disease activity by tocilizumab in 10 ’difficult-to-treat’ cases of Takayasu arteritis,” International Journal of Rheumatic Diseases, vol. 16, no. 6, pp. 754–761, 2013. [22] K. Murakami, M. Ohsawa, S.-X. Hu, H. Kanno, K. Aozasa, and M. Nose, “Large-vessel arteritis associated with chronic active Epstein-Barr virus infection,” Arthritis and Rheumatism, vol. 41, no. 2, pp. 369–373, 1998. [5] W. B. van den Berg and I. B. McInnes, “Th17 cells and IL-17 a—focus on immunopathogenesis and immunotherapeutics,” Seminars in Arthritis and Rheumatism, vol. 43, no. 2, pp. 158– 170, 2013. [23] V. Di Giacomo, F. Meloni, M. G. Transi, G. Mastroberardino, G. Iannucci, and V. Sciacca, “An uncommon systemic arteritis—a case report,” Angiology, vol. 37, no. 1, pp. 63–71, 1986. [6] J. P. References Sherlock, P. C. Taylor, and C. D. Buckley, “The biology of IL-23 and IL-17 and their therapeutic targeting in rheumatic diseases,” Current Opinion in Rheumatology, vol. 27, no. 1, pp. 71–75, 2015. [24] N. E. Martinez, F. Sato, E. Kawai, S. Omura, R. P. Chervenak, and I. Tsunoda, “Regulatory T cells and Th17 cells in viral infections: implications for multiple sclerosis and myocarditis,” Future Virology, vol. 7, no. 6, pp. 593–608, 2012. [7] S. Zhu and Y. Qian, “IL-17/IL-17 receptor system in autoim- mune disease: mechanisms and therapeutic potential,” Clinical Science, vol. 122, no. 11, pp. 487–511, 2012. [25] N. Yeremenko, J. E. Paramarta, and D. Baeten, “The interleukin- 23/interleukin-17 immune axis as a promising new target in the treatment of spondyloarthritis,” Current Opinion in Rheumatol- ogy, vol. 26, no. 4, pp. 361–370, 2014. [8] A. W. S. de Souza and J. F. de Carvalho, “Diagnostic and classi- fication criteria of Takayasu arteritis,” Journal of Autoimmunity, vol. 48-49, pp. 79–83, 2014. [9] R. Misra, D. Danda, S. M. Rajappa et al., “Development and initial validation of the Indian Takayasu Clinical Activity Score (ITAS2010),” Rheumatology, vol. 52, no. 10, pp. 1795–1801, 2013. [10] G. S. Kerr, C. W. Hallahan, J. Giordano et al., “Takayasu arteritis,” Annals of Internal Medicine, vol. 120, no. 11, pp. 919– 929, 1994. [11] D. Saadoun, M. Garrido, C. Comarmond et al., “Th1 and Th17 cytokines drive inflammation in takayasu arteritis,” Arthritis & Rheumatology, vol. 67, no. 5, pp. 1353–1360, 2015. [12] F. Alibaz-Oner, S. P. Yentur, G. Saruhan-Direskeneli, and H. Direskeneli, “Serum cytokine profiles in Takayasu’s arteritis: search for biomarkers,” Clinical and Experimental Rheumatol- ogy, vol. 33, no. 2, supplement 89, pp. S-32–S-35, 2015. [13] G. Esp´ıgol-Frigol´e, M. Corbera-Bellalta, E. Planas-Rigol et al., “Increased IL-17A expression in temporal artery lesions is a predictor of sustained response to glucocorticoid treatment in patients with giant-cell arteritis,” Annals of the Rheumatic Diseases, vol. 72, no. 9, pp. 1481–1487, 2013. [14] A. M´arquez, J. Hern´andez-Rodr´ıguez, M. C. Cid et al., “Influ- ence of the IL17A locus in giant cell arteritis susceptibility,” Annals of the Rheumatic Diseases, vol. 73, no. 9, pp. 1742–1745, 2014. [15] D. Danda, M. Hindhumathi, G. Ruchika et al., “Cytokine gene polymorphisms in Asian Indian patients with Takayasu’s Arteri- tis (TA)—a single center study,” Indian Journal of Rheumatology, vol. 9, article S2, 2014. [16] L. Arnaud, J. Haroche, A. Mathian, G. Gorochov, and Z. References Amoura, “Pathogenesis of Takayasu’s arteritis: a 2011 update,” Autoimmunity Reviews, vol. 11, no. 1, pp. 61–67, 2011. [17] P. Bacon, R. Sivakumar, D. Danda, and R. Misra, “Vasculitis assessment and Takayasu aorto-arteritis,” Indian Journal of Rheumatology, vol. 7, no. 3, pp. 153–158, 2012. [18] G. S. Hoffman, “Takayasu arteritis: lessons from the American National Institutes of Health experience,” International Journal of Cardiology, vol. 54, supplement, pp. S99–102, 1996.
https://openalex.org/W4220692480	https://www.frontiersin.org/articles/10.3389/fcomp.2022.773154/pdf	English	null	An Estimation of Online Video User Engagement From Features of Time- and Value-Continuous, Dimensional Emotions	Frontiers in computer science	2,022	cc-by	13,736	Keywords: user engagement, explainable machine learning, popularity of videos, affective computing, YouTube, continuous emotion annotation Edited by: Zhen Cui, Nanjing University of Science and Technology, China Reviewed by: Robertas Damasevicius, Silesian University of Technology, Poland Tong Zhang, Nanjing University of Science and Technology, China Ming Yin, Jiangsu Police Ofﬁcer College, China Xiaoya Zhang, Nanjing University of Science and Technology, China Edited by: Zhen Cui, Nanjing University of Science and Technology, China Reviewed by: Robertas Damasevicius, Silesian University of Technology, Poland Tong Zhang, Nanjing University of Science and Technology, China Ming Yin, Jiangsu Police Ofﬁcer College, China Xiaoya Zhang, Nanjing University of Science and Technology, China Correspondence: Lukas Stappen stappen@ieee.org Specialty section: This article was submitted to Human-Media Interaction, a section of the journal Frontiers in Computer Science Specialty section: This article was submitted to Human-Media Interaction, a section of the journal Frontiers in Computer Science Received: 09 September 2021 Accepted: 24 February 2022 Published: 23 March 2022 Keywords: user engagement, explainable machine learning, popularity of videos, affective computing, YouTube, continuous emotion annotation An Estimation of Online Video User Engagement From Features of Time- and Value-Continuous, Dimensional Emotions Lukas Stappen 1, Alice Baird 1, Michelle Lienhart 1, Annalena Bätz 1 and Björn Schuller 1,2,3 1 Chair of Embedded Intelligence for Health Care and Wellbeing, University of Augsburg, Augsburg, Germany, 2 audEERING GmbH, Gilching, Germany, 3 GLAM – Group on Language, Audio, & Music, Imperial College London, London, United Kingdom Portraying emotion and trustworthiness is known to increase the appeal of video content. However, the causal relationship between these signals and online user engagement is not well understood. This limited understanding is partly due to a scarcity in emotionally annotated data and the varied modalities which express user engagement online. In this contribution, we utilize a large dataset of YouTube review videos which includes ca. 600 h of dimensional arousal, valence and trustworthiness annotations. We investigate features extracted from these signals against various user engagement indicators including views, like/dislike ratio, as well as the sentiment of comments. In doing so, we identify the positive and negative inﬂuences which single features have, as well as interpretable patterns in each dimension which relate to user engagement. Our results demonstrate that smaller boundary ranges and ﬂuctuations for arousal lead to an increase in user engagement. Furthermore, the extracted time-series features reveal signiﬁcant (p < 0.05) correlations for each dimension, such as, count below signal mean (arousal), number of peaks (valence), and absolute energy (trustworthiness). From this, an effective combination of features is outlined for approaches aiming to automatically predict several user engagement indicators. In a user engagement prediction paradigm we compare all features against semi-automatic (cross-task), and automatic (task-speciﬁc) feature selection methods. These selected feature sets appear to outperform the usage of all features, e.g., using all features achieves 1.55 likes per day (Lp/d) mean absolute error from valence; this improves through semi-automatic and automatic selection to 1.33 and 1.23 Lp/d, respectively (data mean 9.72 Lp/d with a std. 28.75 Lp/d). ORIGINAL RESEARCH published: 23 March 2022 doi: 10.3389/fcomp.2022.773154 1. INTRODUCTION Stappen L, Baird A, Lienhart M, Bätz A and Schuller B (2022) An Estimation of Online Video User Engagement From Features of Time- and Value-Continuous, Dimensional Emotions. Online video content hosted by platforms such as YouTube is now gaining more daily views than traditional television networks (Battaglio, 2016). There are more than 2 billion registered users on YouTube, and a single visitor will remain on the site for at least 10 min (Cooper, 2019). Viewers rate of retention for a single video is between 70–80%, and such retention times may be due to (cross-) social network eﬀects (Roy et al., 2013; Yan et al., 2015; Tan and Zhang, 2019) and Front. Comput. Sci. 4:773154. doi: 10.3389/fcomp.2022.773154 Front. Comput. Sci. 4:773154. doi: 10.3389/fcomp.2022.773154 March 2022 \| Volume 4 \| Article 773154 Frontiers in Computer Science \| www.frontiersin.org User Engagement Estimation From Emotions Stappen et al. the overall improvement in content and connection quality in recent years (Dobrian et al., 2011; Lebreton and Yamagishi, 2020), but arguably caused by intelligent mechanisms (Cheng et al., 2013), e.g., 70% of videos watched on YouTube are recommended from the previous video (Cooper, 2019). To this end, gaining a better understanding of what aspects of a video a user engages with has numerous real-life applications (Dobrian et al., 2011). For example, videos such as misinformation, fake messages, and hate speech are strongly emotionally charged (Knuutila et al., 2020) and detection using conventional methods such as natural language processing is to date a tremendous challenge (Stappen et al., 2020b). Another application is the use by creators who adapt their content to have a greater prospect of the video becoming viral (Trzci´nski and Rokita, 2017) and thus improve advertising opportunities. (Stappen et al., 2021), and derive cross-task features from this initial correlation analysis. Second, we compare these engineered, lean features, to a computationally intensive feature selection approach and to all features when predicting selected engagement indicators (i.e., views, likes, number of comments, likes of the comments). We predict these indicators as a regression task, and train interpretable (linear kernel) Support Vector Regressors (SVR). The main contributions to the research community are two-fold: 1. To the best of the authors’ knowledge, there has been no research which analyses YouTube video user engagement against trustability time-series features. 2. 2. BACKGROUND Within our contribution, the concept of emotions for user- generated content is extended from the conventional Russel concept of emotion dimensions, valence, and arousal (Russell, 1980), to include a continuous measure for trustworthiness. In the following, we introduce these core concepts and related studies. 1. INTRODUCTION Furthermore, we are the ﬁrst to predict cross-modal user popularity indicators as a regression task—purely based on emotional signal features without using typical text, audio, or images/video features as input. Positive emotion (Berger and Milkman, 2012) and trust of the individuals in videos (Nikolinakou and King, 2018) have shown to aﬀect user (i.e., content) engagement (Shehu et al., 2016; Kujur and Singh, 2018). In traditional forms of entertainment (i.e., ﬁlm) portraying emotion captivates the audiences improving their ability to remember details (Subramanian et al., 2014) and similar persuasion appeals are applied within shorter-form YouTube videos (English et al., 2011). When emotion is recognized computationally, research has shown that the emotion (arousal and valence) of a video can be an indicator of popularity, particularly prominent when observing audio features (Sagha et al., 2017). This article is organized as follows; ﬁrstly, in Section 2, we provide a brief background on the core concepts which relate to emotions and user-generated content. We then introduce the data that is used within the experiments in cf. Section 3. This is followed by the experimental methodology, in Section 4, including feature extraction from signals and sentiment extraction from text, and the machine learning pipeline overall. The results are then extensively analyzed and discussed in Section 5, with a mention of study limitations in Section 6. Finally, we oﬀer concluding remarks and future outlook in Section 7. The newly designed and extended datasets, code, and the best models will be made publicly available on in our project repository. The frequency of comments by users is also a strong indicator of how engaged or not users are with a video (Yang et al., 2016). Furthermore, understanding the sentiment of comments (i.e., positive, neutral, or negative) can oﬀer further insights on the type of view engagement, e.g., more positive sentiment correlates to longer audience retention (Yang et al., 2016). In a similar way to the use of emotions, developing trust between the viewer (trustor) and the presenter (trustee) has also shown to improve user engagement. It is a common strategy by content creators to facilitate what is known as a parasocial relationship. A parasocial relationship develops when the viewer begins to consider the presenter as a friend without having ever met them (Chapple and Cownie, 2017). Frontiers in Computer Science \| www.frontiersin.org 2.1. Concepts of Emotion and Trustworthiness With this in mind, we unite multiple emotional signals for an explicit engagement analysis and prediction in this current contribution. Thereby, we focus on the utilization of the emotional dimensions of arousal and valence and extend the typical Russel circumplex model for emotion, by adding trustworthiness as a continuous signal. Hereby, we follow a two- step approach: First, we aim to understand better continuous factors which improve metadata-related (i.e., views, likes, etc.) and comment-related (i.e., sentiment of comments, positive- negative ratios, likes of comments etc.) user engagement across modes (i.e., emotional signals to text-based indicators). To do this, we collect the metadata as well as more than 75 k comments from the videos. We annotate a portion of these comments to be used in combination with other data sets for training a YouTube comment sentiment predictor for the automatic assessment of the unlabeled comments. Furthermore, we utilize a richly annotated data set of ca. 600 h of continuous annotations There are two predominant views in the ﬁeld of aﬀective science: the ﬁrst assumes that emotions are discrete constructs, each acting as an independent emotional system of the human brain, and hence, can be expressed by discrete categories (Ekman, 1992). The second assumes an underlying interconnected dimensional signal system represented by continuous aﬀective states. For emotion recognition using continuous audio-video signals, the circumplex model of emotion developed by Russel is the most prominent (Russell, 1980) and applied (Busso et al., 2008; Kossaiﬁet al., 2019; Stappen et al., 2021) approach of the latter idea. This representation of aﬀect typically consists of continuous valence (the positiveness/ negativity of the emotion) and arousal dimensions (the strength of the activation of the emotion), as well as an optional third focus dimension (Posner et al., 2005). March 2022 \| Volume 4 \| Article 773154 Frontiers in Computer Science \| www.frontiersin.org 2 User Engagement Estimation From Emotions Stappen et al. et al., 2013; Tan and Zhang, 2019), and focusing on both long (Biel and Gatica-Perez, 2013) and short form video sharing (Cheng et al., 2013; Garroppo et al., 2018). In the past, both approaches to classify emotions in user- generated content (Chen et al., 2017) rely on Ekmann’s model to predict six emotional classes in YouTube videos. Similarly, Zadeh et al. (2018) annotated YouTube videos with labels for subjectivity and sentiment intensity (Wöllmer et al., 2013) was the ﬁrst to transfer the dimensional concept to YouTube videos. 1The raw videos and YouTube IDs are available for download: https://zenodo.org/ record/4651164. 2All owners of the data collected for use within the MUSE-CAR data set were contacted in advance for the consent of use for research purposes. 3.1. Video, Meta- and Engagement Data 3.1. Video, Meta- and Engagement Data The dataset contains over 300 user-generated vehicle review videos, equal to almost 40 h of material that cover a broad spectrum of topics within the domain. The videos were collected from YouTube2 and have an average duration of 8 min. The reviews are primarily produced by semi—(“inﬂuencers”) or professional reviewers with an estimated age range of the mid- 20 s until the late-50 s. The speech of the videos is English. We refer the reader to Stappen et al. (2021) for further in- depth explanation about the collection, the annotator training, and the context of the experiments. Utilizing the YouTube ID, we extend the data set by user engagement data. The explicit user engagement indicators are calculated on a per-day basis (p/d) as the videos were uploaded on diﬀerent days resulting in In contrast to the literature, our approach utilizes the predicted sentiment of a ﬁne-tuned Word Embedding Transformer ALBERT (Lan et al., 2020) to automatically classify comments on a large scale to investigate the cross-modal relationship to the continuous emotion and trustworthiness signals. 3. DATA The base for our experimental work is the MUSE-CAR data set1(Stappen et al., 2021). MUSE-CAR is a multi-media dataset originally crafted to improve machine understanding of multi- modal sentiment analysis in real-life media. For the ﬁrst time, it was used for the MUSE 2020 Challenge, which aimed to improve emotion recognition systems, focusing on the prediction of arousal and valence emotion signals (Stappen et al., 2020c). For a detailed description of typical audio-visual feature sets and baseline systems that are not directly related to this work, we point the reader to Stappen et al. (2020a). 2.1. Concepts of Emotion and Trustworthiness Recently, Kollias et al. (2019) annotated 300 videos (ca. 15 h) of “in-the-wild” data, predominantly YouTube videos under the creative commons license. Most previous work analyse view patterns, users’ opinions (comments) and users’ perceptions (likes/dislikes), and their mutual inﬂuence (Bhuiyan et al., 2017). Khan and Vong (2014) correlated these reaction data, while (Rangaswamy et al., 2016) connects them to the popularity of a video. An extended comment analysis has been conducted by Severyn et al. (2016) predicting the type and popularity toward the product and video. The comment ratings, thus the community acceptance, was predicted by Siersdorfer et al. (2010) using the comment language and discrete emotions. Moreover, in Wu and Ito (2014) the authors correlated popularity measures and the sentiment of the comments. Data of other social networking platforms combine sentiment analysis and social media reactions (Ceron et al., 2014; Gilbert and Hutto, 2014), and (Preo¸tiuc-Pietro et al., 2016) attempted to map Facebook posts to the circumplex model to predict the sentiment of new messages. However, none of the mentioned datasets allows the bridging of annotated or predicted emotional signals with user engagement data from videos. We ﬁll this research gap utilizing continuous emotional signals and corresponding data, as well as providing insights into the novel dimension of trustworthiness, entirely without relying on word-based, audio, or video feature extraction. Although general literature lacks in providing a consistent concept of trustworthiness (Horsburgh, 1961; Moturu and Liu, 2011; Cox et al., 2016), in this work, we deﬁne trust as the ability, benevolence, and integrity of a trustee analogous to Colquitt et al. (2007). In the context of user-generated reviews, the viewers assess from their perspective if and to what extent the reviewer communicates unbiased information. In other words, how truthful and knowledgeable does the viewer feel a review is at every moment? As we mentioned, building this trust is part of developing a parasocial relationship with the audience, and in doing so, likely increases repeated viewing (Lim et al., 2020). To the best of our knowledge, no work has so far attempt to investigate the relationship to sophisticated continuous emotional and trustworthiness signals and based on these, predict user engagement as regression tasks. 2.2. Sentiment Analysis of YouTube Comments Sentiment Analysis studies the extraction of opinions, sentiments, and emotions (e.g., “positive,” “negative,” or “neutral”) of user-generated content. The analyzed content usually consists of text (Boiy et al., 2007; Gilbert and Hutto, 2014), such as in movie and product reviews, as well as comments (Singh et al., 2013; Siersdorfer et al., 2014). In recent years, the methods for text classiﬁcation have developed rapidly. Earlier work using rule-based and classical word embedding approaches is now being replaced by what is known as transformer networks, predicting context-based word embeddings (Devlin et al., 2019). State-of-the-art accuracy results on sentiment benchmark datasets using these methods (Cui et al., 2019) range from 77.3 for the 3-classes twitter (Nakov et al., 2013) and between 72.4 and 75.0 on a 2-classes YouTube comment data sets (Uryupina et al., 2014). 3.2. Emotion and Trustworthiness Signals 3.2. Emotion and Trustworthiness Signals As with emotions in general, a certain level of disagreement due to subjectivity can be expected (Russell, 1980). For this reason, nine annotators were trained (Stappen et al., 2021) to have a common understanding of the arousal, valence, and trustworthiness concepts as discussed in Section 2.1. As well established (Busso et al., 2008; Kossaiﬁet al., 2019), the annotator moves the hand up and down using a Logitech Extreme 3D Pro Joystick to annotate one of three dimensions, while watching the videos. The movements are recorded over the entire duration of the video sequence and sampled with a bin size of 0.25 Hz on an axis magnitude between -1 000 and 1 000. Every annotation was checked by an auditor using quantitative and qualitative measures to ensure a high quality (Baird and Schuller, 2020). The time required for annotation alone stands for more than 600 working hours (40 h video * 3 dimensions * 5 annotators per dimension). yEWE n = 1 PA a=1 ra A X a=1 rayn,a, (1) (1) where y is a discrete point of the signal n and ra is the reliability of the ath rater, consequently, A represents the whole population of raters. To use the data for later stages, we z-standardize them. 2.3. Analysis of YouTube Engagement Data and Cross-Modal Studies YouTube meta and engagement data are well researched (Yan et al., 2015) with contributions exploring across domains (Roy March 2022 \| Volume 4 \| Article 773154 Frontiers in Computer Science \| www.frontiersin.org 3 User Engagement Estimation From Emotions Stappen et al. FIGURE 1 \| Example video (video ID 5): Four annotation signals for arousal and valence in addition to the fused EWE “gold-standard” signal (bold purple). The annotator 1 (green) has a negative correlation to the others on valence. The weight (r1) is set to 0, not considered for the fused gold-standard by EWE. FIGURE 1 \| Example video (video ID 5): Four annotation signals for arousal and valence in addition to the fused EWE “gold-standard” signal (bold purple). The annotator 1 (green) has a negative correlation to the others on valence. The weight (r1) is set to 0, not considered for the fused gold-standard by EWE. views (Vp/d), likes (Lp/d), dislikes (Dp/d), comments (Cp/d), and likes of comments (LCp/d). Per video the user engagement criteria is distributed (µ mean, σ standard deviation) as; Vp/d: µ = 863.88, σ = 2048.43; Lp/d: µ = 9.73, σ = 28.75, Dp/d: µ = 0.4125, σ = 1.11; Cp/d: µ = 0.91, σ = 3.00; and LCp/d: µ = 5.28, σ = 16.84. The annotation of ﬁve independent annotators for each video and signal type are fused to obtain a more objective gold- standard signal as depicted in Figure 1. For the fusion of the individual continuous signals, the widely established Evaluator Weighted Estimator (EWE) was computed (Schuller, 2013; Ringeval et al., 2017). It is an estimator of inter-rater agreement, hence, the personal reliability, in which the weighted mean corresponds to the calculated weights for each rater based on the cross-dependency of all other annotators. The EWE can be formulated as Time-series statistics Time series statistics Asymmetry Dynamic sample skewness (skew) Kurtosis (kurt) Energy-related Absolute energy (absE) Sample entropy (SaEn) Change-related Absolute sum of changes (ASOC) Mean absolute change (MACh) Mean change (MCh) Mean value of a central approximation of the second derivatives (MSDC) Strike above the mean (LSAMe) Strike below the mean (LSBMe) Relative points Normalized percentage of reoccurring datapoints (PreDa) First and last location of the minimum and maximum (FLMi, LLMi, FLMa, and FLMa) Number of crossings of a point m (CrM) Peaks of the least support (peaks) 3.3. Video Comments Based on the video IDs of the corpus, we collected more than 79 k YouTube comments and comment-related like counts excluding any other user information, such as the username. We focus exclusively on the parent comments, ignoring reaction from the child comments. The count of comment likes reﬂect the number of people sharing the same opinion and those who “liked” the comment. We randomly select 1 100 comments for March 2022 \| Volume 4 \| Article 773154 Frontiers in Computer Science \| www.frontiersin.org 4 User Engagement Estimation From Emotions Stappen et al. FIGURE 2 \| A comprehensive overview of the approach, as explained in Section 4. Statistics derived from continuous emotion and trustworthiness (purple), automatic sentiment labeling of YouTube comments (blue), and user engagement data (orange) are preprocessed to investigate correlations and predict engagement from features. The development of our signal feature extraction as a cornerstone of our analysis is described in detail in Section 4.1. With its help, we aim to uncover relationships to engagement data (cf. Section 4.3) and the sentiment of the comments predicted by our trained network (cf. Section 4.2). They also serve as input to our regression experiments to predict user engagement (cf. Section 4.5). FIGURE 2 \| A comprehensive overview of the approach, as explained in Section 4. Statistics derived from continuous emotion and trustworthiness (purple), automatic sentiment labeling of YouTube comments (blue), and user engagement data (orange) are preprocessed to investigate correlations and predict engagement from features. The development of our signal feature extraction as a cornerstone of our analysis is described in detail in Section 4.1. With its help, we aim to uncover relationships to engagement data (cf. Section 4.3) and the sentiment of the comments predicted by our trained network (cf. Section 4.2). They also serve as input to our regression experiments to predict user engagement (cf. Section 4.5). TABLE 1 \| List of simple statistics and more complex time-series statistic features extracted by our framework. 4. EXPERIMENTAL METHODOLOGY Figure 2 gives an overview of our approach. As a cornerstone of our analysis (cf. Section 4.1), annotation of arousal, valence, and trustworthiness are annotated by ﬁve independent annotators. These signals are then fused (cf. Section 3.2) to a gold standard label, and meaningful features are extracted (purple). In addition, YouTube user engagement-related data (yellow) and the comments are scraped (blue) from each video. Several sentiment data sets are collected and merged in order to train a robust sentiment classiﬁer using a Transformer network ALBERT to predict unlabeled YouTube comments after ﬁne- tuning on several datasets and our labeled comments. Then, we ﬁrst investigate correlations between the predicted sentiment of the YouTube comments, the YouTube metadata, and the statistics derived from the continuous signals (arousal, valence, and trustworthiness). Additionally, we use derived features to predict user engagement (Vp/d, Lp/d, Cp/d, and CLp/d) directly. 2013). Audio, video, and psychological signals are widely used for computational analysis (Schuller, 2013; Schuller et al., 2020). Simple statistics and advanced feature extraction can be applied in order to condense these signals to meaningful summary 2013). Audio, video, and psychological signals are widely used for computational analysis (Schuller, 2013; Schuller et al., 2020). Simple statistics and advanced feature extraction can be applied in order to condense these signals to meaningful summary 3.3. Video Comments Distribution statistics Standard deviation (std) 5%-, 25%-, 50%-, 75%-, and 95%-quantiles (q5, q25, q50, q75, q95) Time-series statistics Asymmetry Dynamic sample skewness (skew) Kurtosis (kurt) Energy-related Absolute energy (absE) Sample entropy (SaEn) Change-related Absolute sum of changes (ASOC) Mean absolute change (MACh) Mean change (MCh) Mean value of a central approximation of the second derivatives (MSDC) Strike above the mean (LSAMe) Strike below the mean (LSBMe) Relative points Normalized percentage of reoccurring datapoints (PreDa) First and last location of the minimum and maximum (FLMi, LLMi, FLMa, and FLMa) Number of crossings of a point m (CrM) Peaks of the least support (peaks) TABLE 1 \| List of simple statistics and more complex time-series statistic features extracted by our framework. labeling, which is used as a quantitative estimator of how accurate our prediction of the other unlabeled comments are. Three annotators labeled each of them as positive, neutral, negative, and not applicable. The average inter-rater joint probability is 0.47. We use a majority fusion to create a single ground truth, excluding texts where no majority is reached. 4.1. Feature Extraction From Signals Of the energy- related ones, the absolute energy (absE) of a signal can be determined by the sum over the squared values (Christ et al., 2018). 4.2. Sentiment Extraction From Comments Given the vast amount of comments, we decided to carry out the labeling of the sentiment automatically and label only a small share of them by hand to quantify the prediction quality (cf. Section 3.3). For this reason, we built a robust classiﬁer for automatic YouTube sentiment prediction using PyTorch. We opted to use ALBERT as our competitive Transformer architecture (Lan et al., 2020). Compared to other architectures, ALBERT introduces two novel parameter reduction methods: First, the embedding matrix is separated into two more compact matrices, and second, layers are grouped and used repeatedly. Furthermore, it applies a new self-supervised loss function that improves training for downstream ﬁne-tuning tasks. These changes have several advantages, such as reducing the memory footprint, accelerating the converge of the network, and leading to state-of-the-art results in several benchmarks (Devlin et al., 2019). absE = X i=1,...,n x2 i , (2) (2) where x is the signal at point i. Also well known for physiological time-series signals is the sample entropy (SaEn), a variation of the approximate entropy, to measure the complexity independently of the series length (Richman and Moorman, 2000; Yentes et al., 2013). Several change-related features might be valuable to reﬂect the compressed signal (Christ et al., 2018): First, the sum over the absolute value of consecutive changes expresses the absolute sum of changes (ASOC): where x is the signal at point i. Also well known for physiological time-series signals is the sample entropy (SaEn), a variation of the approximate entropy, to measure the complexity independently of the series length (Richman and Moorman, 2000; Yentes et al., 2013). Several change-related features might be valuable to reﬂect the compressed signal (Christ et al., 2018): First, the sum over the absolute value of consecutive changes expresses the absolute sum of changes (ASOC): ASOC = X i=1,...,n−1 \|xi+1 −xi\|. (3) (3) Before training, we remove all words starting with a “#,” “@,” or “http” from all text sources and replace emotions’ unicode by the name. We train ALBERT in a two-step procedure. First, we ﬁne- tune the model for the down-stream task of general sentiment analysis. 4.1. Feature Extraction From Signals 4.1. Feature Extraction From Signals A signal is usually sampled to ﬁne-grained, discrete points of regular intervals, which can be interpreted as a sequential set of successive data points over time (Adhikari and Agrawal, March 2022 \| Volume 4 \| Article 773154 Frontiers in Computer Science \| www.frontiersin.org 5 User Engagement Estimation From Emotions Stappen et al. summarize the distribution similarity, the normalized percentage of reoccurring datapoints (PreDa) of non-unique single points can be calculated by taking the number of data points occurring more than once divided by the number of total points. Also early or late high and low points of the signal are of descriptive value. Four single points describe these: the ﬁrst and last location of the minimum and maximum (FLMi, LLMi, FLMa, and FLMa) relatively to the length of the series. The last two count a) the number of crossings of a point m (here: m=0) (CrM), where for two successive time series steps are ﬁrst lower (or higher) than m followed by two higher (or lower) ones (Christ et al., 2018) and b) the peaks of the least support n. A peak of support n is described as a subsequence of a series where a value occurs, bigger than its n neighbors to the left and the right (Palshikar, 2009; Christ et al., 2018). In total, we extract 24 features from one signal (cf. Table 1). representations and make them more workable (Christ et al., 2018). In this work, we use common statistical measures such as the standard deviation (std), and 5-, 25-, 50-, 75-, and 95%- quantiles (q5, q25, q50, q75, andq95) as they are less complex to interpret, and have been applied in related works (Sagha et al., 2017). Furthermore, to make better use of the changes over time, we manually select and calculate a wide range of time-series statistics following previous work in similar ﬁelds (Geurts, 2001; Schuller et al., 2002). For example, in computational audition (e.g., speech emotion recognition), energy-related features of the audio signals are used to predict emotions (Schuller et al., 2002). We calculate the dynamic sample skewness (skew) of a signal using the adjusted Fisher-Pearson standardized moment coeﬃcient, to have a descriptor for the asymmetry of the series (Ekman, 1992; Doane and Seward, 2011). Similarly, the kurtosis (kurt) measures the “ﬂatness” of the distribution by utilizing the fourth moment (Westfall, 2014). 4.3. Correlation Measure and Signiﬁcance 4.3. Correlation Measure and Signiﬁcance The Pearson correlation (r) explores the relationship between two continuous variables (Ahlgren et al., 2003). Thereby, the relationship has to be linear, meaning that when one variable changes, the other also changes proportionally. r is deﬁned by For our regression experiments, we use a Support Vector Regression (SVR) with a linear kernel as implemented by the open-source machine learning toolkit Scikit-Learn (Pedregosa et al., 2011). The linear kernel allows us to interpret the weights from our various feature selections and has, among other applications, found wide acceptance in the selection of relevant genes from micro-array data (Guyon et al., 2002). Since the coeﬃcients are orthogonal to the hyper-plane, a feature is useful to separate the data when the hyper-plane is orthogonal to the feature axis. The absolute size of the coeﬃcient concerning the other features indicates the importance. rx,z = cov(x, z) σx · σz = Pn i=1(xi −¯x) · (zi −¯z) qPn i=1(xi −¯x)2 · Pn i=1(zi −¯z)2 , (7) (7) where cov(x, z) is the co-variance, a measure of the joint variability, of the variables X, Z, and σx, σz – the standard deviations of both variables (Surhone et al., 2010). The resulting correlation coeﬃcient lies between −1 and +1. If the value is positive, the two variables are positively correlated. A value of ±1 signiﬁes a perfect positive or negative correlation. A coeﬃcient equal to zero implies that there is no linear dependency between the variables. The training is executed on the 60-20-20 training/development/test partition split partitions, pre-deﬁned in the MUSE-CAR emotion recognition sub-task (Stappen et al., 2020c) (cf. Section 3.1). During the training phase, we train a series of models on the training set with diﬀerent parameters C ∈{10−7, 10−6, 10−5, 10−4, 10−3, 10−2, 10−1, 1 } up to 10 000 iterations and validate the performance on the development set. The best performing C value is then used to re-train the model on an enlarged, concatenated training and development set, to estimate the generalization performance on the hold-out test set. This method is repeated for each input signal (combination) on each target (%). Due to the various scales of the input features, we apply standardization to the data but leave the targets, as they allow interpretability of the results. The prediction results are evaluated using the Mean Absolute Error (MAE). 4.1. Feature Extraction From Signals No extensive YouTube comment data set is available, which would span the wide range of writing styles and expressed opinions. Therefore, we aggregate several datasets which aim to classify whether a text is positive, negative, or neutral as our initial training data: all data sets from SemEval (the Semantic Evaluation challenge), a series of challenges for computer-based text classiﬁcation systems with changing domains (Nakov et al., 2013) e.g., Twitter, SMS, sarcasm, from 2013 to 2017 consisting of more than 76 k data points; the popular US Airline Sentiment data set (Air, 2015) (14.5 k tweets), and ﬁnally, 35 k positive and 35 k negative text snippets are selected from Sentiment140 (Go et al., 2009). The 60 k positive, 32 k neutral, and 56 k negative text snippets are equally stratiﬁed and partitioned into 80-10-10 splits for training. We provide this selection for reproducibility in our code. Second, the mean absolute change (MACh) over the absolute diﬀerence between subsequent data points is deﬁned as: MACh = 1 n X i=1,...,n−1 \|xi+1 −xi\|, (4) (4) where n is the number of time-series points. Third, the general diﬀerence between consecutive points over time is called the mean change (MCh): MCh = 1 n −1 X i=1,...,n−1 xi+1 −xi. (5) (5) Fourth, the mean value of a central approximation of the second derivatives (MSDC) is deﬁned as: MSDC = 1 2 ∗(n −1) X i=1,...,n−1 0.5 · (xi+2 −2 · i + 1 + xi). (6) (6) Following the authors’ recommendation, ALBERT is trained using a learning rate of 1e-5, a warmup ratio of 0.06, ǫ set to 1e-8, and gradient clipping set at 1.0. In addition, we use half-precision training and a batch size of 12 to ﬁt the GPU memory restrictions Finally, the length of the normalized consecutive sub-sequence is named strike above (LSAMe) and below (LSBMe) the mean. To Finally, the length of the normalized consecutive sub-sequence is named strike above (LSAMe) and below (LSBMe) the mean. To Frontiers in Computer Science \| www.frontiersin.org March 2022 \| Volume 4 \| Article 773154 6 User Engagement Estimation From Emotions Stappen et al. TABLE 2 \| Example comments and sentiment distribution within the YouTube comments predicted by our developed sentiment model. 4.1. Feature Extraction From Signals Sentiment # Comments Predicted [%] Example Positive 26 032 33 “The metaphors are just ﬂying like the raindrops in this video.” #47620 Neutral 28 518 36 “Are engines for F30 made in Germany?” #4 Negative 24 494 31 “Poor review unfortunately, the microphone quality was very mufﬂed...” #31 TABLE 2 \| Example comments and sentiment distribution within the YouTube comments predicted by our developed sentiment model. all engineered features are equally relevant. Since no previous research can guide us to a reliable selection, we propose two ways for feature selection for our task of predicting user engagement. The ﬁrst is a correlation-based, cross-task semi- automatic selection that uses the correlation between the feature and the target variables. Only those features whose mean value over all prediction tasks is between −0.2 > rmean > +0.2 (minimum low positive/negative correlation) are selected. Sentiment # Comments Predicted [%] Example p g The other concept is a regression-based, task-speciﬁc automatic selection with three steps. First, univariate linear regression (f ) tests act as a scoring function and run successively to measure the individual eﬀect of many regressors: (32 GBs). Counteracting adverse eﬀects of class imbalance, we further inject the class weight to each data point. The model converges after three epochs. Next, we use our own YouTube comment data set to validate the results and further ﬁne-tune the model. This version is then further trained in a second ﬁne- tuning step using the 60% of the YouTube comments and a reduced learning rate of 1e-6 for one epoch. score(f , y) = Xki −¯ Xki · (y −¯y) σXki · σy , (8) (8) where ki is the feature index. The score is converted to an F-test estimate and then to a p-value. Second, the highest k number of features are selected based on the p-value. Finally, this procedure runs brute-force for all number of feature combinations, where 5 < k < kmax−1. Brute-force implies an exhaustive search, which systematically checks all possible combinations until the best one is found based on the provided estimate. The relative distribution of the classiﬁed sentiment of the YouTube comments is given in Table 2. The model achieves an f1 score on the development of 81.13 and 78.09% on the test partitions, as well as 75.41% on the sample of our crawled and manually labeled YouTube test set. 4.3. Correlation Measure and Signiﬁcance For signiﬁcance testing, we ﬁrst compute the t-statistic, and then twice the survival function for the resulting t-value to receive a two-tailed p-value, in which the null hypothesis (two variables are uncorrelated) is rejected at levels of α = {0.01, 0.05, 0.1} (Sham and Purcell, 2014). Since, we intend to give the reader as much transparency as possible with regard to the robustness of the results obtained given the size of the data set, we report the results on three common signiﬁcance levels (see Appendix). Therefore, results signiﬁcant at an alpha level of 0.01 are also signiﬁcant at 0.05 and 0.1. 4.4. Feature Selection Figure 3 depicts the Pearson correlations for the user engagement indicators, and we see that the number of Vp/d, Lp/d, Dp/d, and Cp/d are highly correlated. The To the best of our knowledge, we are the ﬁrst extracting advanced features directly from emotional signals. Usually, not March 2022 \| Volume 4 \| Article 773154 Frontiers in Computer Science \| www.frontiersin.org 7 User Engagement Estimation From Emotions Stappen et al. FIGURE 3 \| Pearson correlation matrix of metadata. All results are considered to be signiﬁcant at a 0.01 level. Com, comment; pos, positive; neg, negative; pd, per day. FIGURE 3 \| Pearson correlation matrix of metadata. All results are considered to be signiﬁcant at a 0.01 level. Com, comment; pos, positive; neg, negative; pd, per day. correlations are based on both, absolute values and ratios. When a correlation to one of the variables occurs, it is likely to be accompanied by correlations to others. We would like to note that all absolute values correlate positively with each other, as all metrics have a positive correlation to the absolute popularity of the video. Therefore, a stronger distribution of the video also increases the absolute number of likes, dislikes, comments, etc., albeit to diﬀerent magnitudes. For example, the average relationship between likes and dislikes in our crawled videos is not as antagonistic as one might expect, which means that as the number of Lp/d increases; so too does the Dp/d. Another example are the number of likes of the comments which increases with the number of dislikes of the video since the number of comments and of dislikes are interdependent. This may relate to the topic of the dataset, being that it is review videos, and the like or dislike may be more objective than other video themes. The correlation in terms of ratios gives a more deﬁnite picture in this context. Arousal: The statistics extracted from the arousal signal indicate several correlations to the engagement data. When the standard deviation or the level of the quantile0.95 increases, the number of Vp/d, Lp/d, Cp/d, and CLp/d slightly decreases (e.g., r(views,std) = −0.293, r(views,q95) = −0.212) with direct eﬀect on the comment-like ratio (clr), e.g., rstd = −0.271. In contrast, the level of the quantile.05 has the opposite eﬀect on all these metrics (e.g., r(views,q0.05) = 0.231, r(clr,q0.05) = −0.248. 4.4. Feature Selection Of the more complex time-series statistics, the peaks as well as the CBM have the strongest correlations across most indicators. These indicates a moderate positive linear relationship, for instance, to Vp/d and Lp/d r(views,peaks) = 0.440, r(likes,CBMe) = 0.456as well as Cp/d rpeaks = 0.409. Further, when these features increase, the share of neutral comments increases much less than the share of positive and negative comments. The next strongest correlated features, CrM, aSoc, abE, and PreDa, also represent upward correlation slopes to the user-engagement criteria. Although these features reﬂect the general change in engagement, no conclusions can be drawn regarding sentiment of the engagement, as there is no signiﬁcant correlation of any feature to the ratios (e.g., like-dislike, and positive-negative comments). correlations are based on both, absolute values and ratios. When a correlation to one of the variables occurs, it is likely to be accompanied by correlations to others. We would like to note that all absolute values correlate positively with each other, as all metrics have a positive correlation to the absolute popularity of the video. Therefore, a stronger distribution of the video also increases the absolute number of likes, dislikes, comments, etc., albeit to diﬀerent magnitudes. For example, the average relationship between likes and dislikes in our crawled videos is not as antagonistic as one might expect, which means that as the number of Lp/d increases; so too does the Dp/d. Another example are the number of likes of the comments which increases with the number of dislikes of the video since the number of comments and of dislikes are interdependent. This may relate to the topic of the dataset, being that it is review videos, and the like or dislike may be more objective than other video themes. The correlation in terms of ratios gives a more deﬁnite picture in this context. 5.1. Relationship Between Features and User Engagement of reoccurring datapoints (PreDa), and sample entropy (SaEn). Features in blue are utilized as cross-task, semi-automatic features for user engagement prediction. seems a stable indicator across all signals. The energy-related features of valence and trustworthiness (valence = r(views,absE) = 0.467, trustworthiness = r(views,absE) = 0.497) seem to have a medium-strong relationship and most likely a valuable predictive feature. proportion of positive comments. Furthermore, higher values around the centre of the distribution (kurtosis – r = −0.313) to more likes per comment. The strongest positively correlated feature is absE e.g., rviews = 0.467, rlikes = 0.422, rdislikes = 0.355, rcomments = 0.350, followed by the peaks, CBME and LSBMe, which suggest the greater the value of these features, the greater the user engagement In contrast, the MaCh and the SaEn have signiﬁcant slight negative correlations, which implies that when the valence signal of a video has a high complexity, the video has a higher tendency to receive fewer user engagement. Regarding the comments, independently of the type of signal and statistic, the negative comments seem to be higher correlated consistently, followed by the number of likes and positive comments. Overall, mostly slight to modest correlations are found. However, signiﬁcant correlations, especially to the more complex time-series features, between valence, arousal, and trustworthiness levels in a video to the user engagement (number of users who watch it, like it, dislike it, or leave a comment) is evident. Trustworthiness: The higher the level of quantile0.05, quantile0.25, median, and quantile0.75 (all slightly positively correlated, with decreasing relevance e.g., r(views,q0.05) = 0.356, r(likes,q0.75) = 0.175), the higher the Vp/d, Lp/d, Dp/d, Cp/d, and CLp/d. Similar to the valence dimension, we see that there is a negative eﬀect on these engagement indicators when the standard deviation in the trustworthiness signal is higher e.g., r(views,std) = −0.304, r(likes,std) = −0.287. As for the other features, the absE and the number of peaks have a moderate positive correlation. The skewness shows a signiﬁcant negative correlation above r < −0.3 for most indicators. In other words, a negative skew of the trustworthiness signal, when the mass of the distribution is concentrated to the right (left- skewed), has a positive inﬂuence on user engagement. Regarding the positiveness/negativeness sentiment ratios (like-dislike, comments positive-negative ratio), none of the features show signiﬁcant associations. 5.1. Relationship Between Features and User Engagement Valence: Most statistics of the signal distribution are below r = 0.2, suggesting that there are only very weak linear dependencies with the engagement indicators. The only exceptions is the positive-negative ratio for the comments (r = −0.276) – a lower standard deviation leads to an increase in the Valence: Most statistics of the signal distribution are below r = 0.2, suggesting that there are only very weak linear dependencies with the engagement indicators. The only exceptions is the positive-negative ratio for the comments (r = −0.276) – a lower standard deviation leads to an increase in the Valence: Most statistics of the signal distribution are below r = 0.2, suggesting that there are only very weak linear dependencies with the engagement indicators. The only exceptions is the positive-negative ratio for the comments (r = −0.276) – a lower standard deviation leads to an increase in the Within this section, we discuss the correlation results for each emotional dimension separately. We report Pearson correlation coeﬃcients, as depicted in Figure 4. Detailed results (r and signiﬁcance level) can be found in Figure 4. March 2022 \| Volume 4 \| Article 773154 Frontiers in Computer Science \| www.frontiersin.org 8 User Engagement Estimation From Emotions Stappen et al. FIGURE 4 \| Pearson correlation matrix of user engagement indicators and the statistics/ features extracted from each dimension. The latter are standard deviation (std), quantile (qx), absolute energy (absE), mean absolute change (MACh), mean change (MCh), mean central approximation of the second derivatives (MSDC), crossings of a point m (CrM), peaks, skewness, kurtosis, strike above the mean (LSAMe), strike below the mean (LSBMe), count below mean (CBMe), absolute sum of changes (ASOC), ﬁrst and last location of the minimum and maximum (FLMi, LLMi, FLMa, FLMa), perc. of reoccurring datapoints (PreDa), and sample entropy (SaEn). Features in blue are utilized as cross-task, semi-automatic features for user engagement prediction. FIGURE 4 \| Pearson correlation matrix of user engagement indicators and the statistics/ features extracted from each dimension. The latter are standard deviation (std), quantile (qx), absolute energy (absE), mean absolute change (MACh), mean change (MCh), mean central approximation of the second derivatives (MSDC), crossings of a point m (CrM), peaks, skewness, kurtosis, strike above the mean (LSAMe), strike below the mean (LSBMe), count below mean (CBMe), absolute sum of changes (ASOC), ﬁrst and last location of the minimum and maximum (FLMi, LLMi, FLMa, FLMa), perc. Frontiers in Computer Science \| www.frontiersin.org C: parameter of the SVR, optimized for from 0.00001 to 1, using the best M: mean absolute error on the dev(elopement) set to deﬁne C for test set prediction. (%) Indicates the relative change of the automatic (auto.) and matically selected (sel.) in % to the unchanged features, “+” indicates an improvement, thus a decrease of the MAE compared to the original feature sets. most weight from the corresponding SVM, indicating that the automatic selection is sensible. In this particular case, the hand selected features have almost identical weights as the automatic ones, whereby the missing features are enough to make the results worse than in the case of the other two (cf. Table 3), indicating a high sensitivity if certain features are left out. TABLE 3 \| Prediction of views, likes, comments, and likes of comments aggregated per day utilizing features extracted and crafted from Arousal (A), Valence (V), and Trustworthiness (T). Type Views Likes Comments Likes of Comments dev test dev test dev test dev test all sel. auto. all sel. auto. all sel. auto. all sel. auto. all sel. auto. all sel. auto. all sel. auto. all sel. auto. C: parameter of the SVR, optimized for from 0.00001 to 1, using the best M: mean absolute error on the dev(elopement) set to deﬁne C for test set prediction. (%) Indicates the relative change of the automatic (auto.) and matically selected (sel.) in % to the unchanged features, “+” indicates an improvement, thus a decrease of the MAE compared to the original feature sets. MAE rel.% rel.% MAE rel.% rel.% MAE rel.% rel.% MAE rel.% rel.% MAE rel.% rel.% MAE rel.% rel.% MAE rel.% rel.% MAE rel.% rel.% A 231.8 +6.8 +5.0 220.3 +9.9 +3.1 2.30 -0.3 +2.6 1.55 +5.9 +3.0 0.288 -0.1 +3.7 0.154 +2.5 +0.6 1.19 +5.7 +5.9 0.50 -19.1 -22.7 V 253.1 +8.7 +7.2 223.8 +17.4 +24.3 2.29 +0.6 +1.0 1.61 +17.6 +24.0 0.288 +3.1 +3.9 0.154 +5.1 +2.4 1.17 -1.4 +3.6 0.51 -2.8 -18.4 T 237.4 +11.8 +16.3 207.9 -5.2 -9.7 2.21 +5.3 +14.4 1.92 +13.3 +3.6 0.262 +5.8 +6.4 0.225 +2.1 -5.3 1.11 -0.1 +9.5 0.75 +8.8 +6.7 A+V 237.6 -1.0 +2.1 210.7 +4.1 +18.3 2.27 -11.4 +0.3 1.79 +24.2 -0.7 0.277 -4.3 +3.4 0.161 +9.9 +0.1 1.16 +0.1 +2.0 0.54 +16.8 -27.3 A+T 240.3 +9.2 +15.7 207.9 -6.7 -3.9 2.26 +4.8 +10.6 2.02 +11.8 +10.3 0.268 +1.6 +7.2 0.182 -34.9 -1.1 1.11 -0.2 +3.7 0.59 -17.9 -14.7 V+T 249.1 +15.5 +20.0 205.8 -3.1 -2.6 2.07 -11.8 -0.2 1.99 +17.2 +0.1 0.262 -2.7 +5.5 0.188 +10.9 -24.7 1.04 -6.2 +0.3 0.78 +11.4 -0.1 A+V+T 228.9 -1.2 +8.7 205.9 -8.4 +0.2 2.06 -12.6 +0.6 2.08 -22.9 +0.3 0.264 -0.0 +2.7 0.192 -7.5 +0.5 1.10 +0.9 +4.3 0.60 -16.6 +8.4 We report C: parameter of the SVR optimized for from 0 00001 to 1 using the best M: mean absolute error on the dev(elopement) set to deﬁne C for test set prediction (%) Indicates the relative change of the automatic (auto ) and Views Per day: When observing the Vp/d prediction from all features, we obtain the best result when performing an early fusion of the valence and trustworthiness signals, and with the addition of arousal, there is a minor decrease (205.8 and 205.8 MAE respectively); this demonstrates the predictive potential of all signals. However, when applying our semi- automatic cross-task feature selection, there is a more substantial improvement particularly for arousal and valence as mono signals, obtaining 198.5, and 184.8 MAE, respectively. This improvement is increased further for valence through automatic feature selection, with our best results for Vp/d of 169.5 MAE. Feature selection appears in all cases to not be beneﬁcial for fused features, with arousal and valence improving slightly but no more than if the signal was alone. Frontiers in Computer Science \| www.frontiersin.org 5.2. Predicting User Engagement From Features of Emotion and Trustworthiness Signals Table 3 shows the results of the prediction tasks Vp/d, Lp/d, Cp/d, and CLp/D. It is worth noting that the scores vary according to the underlying scale of the target variables (cf. Section 3). The features utilized from the cross-task semi-automatic feature selection method are highlighted (in blue) in Figure 4 for each feature type. Across the seven experiments the automatic selection process selected on average the following number of features per each criteria; 7.6 Vp/d, 23.3 Cp/d, 29.3 Lp/d, and 20.1 LCp/d. For each dimension, an average of 9.3 for arousal, 9.5 for valence, and 6.0 for trustworthiness was selected. Figure 5 illustrates an example of both selection methods for predicting CLp/d from a fusion of all three feature types. The p-values of the automatic (univariate) selection and the corresponding weights of all resulting SVMs are shown, indicating the relevance of each feature for the prediction. The interested reader is pointed to Chang and Lin (2008) for an in-depth methodical explanation. The most informative features (largest p-values) also receive Result Discussion: When observing the results from the above sections, we see several patterns between the emotion (including trust) signal statistics and user engagement. While the standard statistics of arousal show that bounded arousal (higher lower quantiles and lower high quantiles) and higher trustworthiness scores (all quantiles are positively correlated, with lower quantiles at a higher level) leads to more user engagement, the sentiment of a video seems less inﬂuential contrary to the ﬁndings of (Sagha et al., 2017). Regarding the time-series features, the number of peaks with support n = 10 March 2022 \| Volume 4 \| Article 773154 9 User Engagement Estimation From Emotions Stappen et al. We report C: parameter of the SVR, optimized for from 0.00001 to 1, using the best M: mean absolute error on the dev(elopement) set to deﬁne C for test set prediction. (%) Indicates the relative change of the automatic (auto.) and semi-automatically selected (sel.) in % to the unchanged features, “+” indicates an improvement, thus a decrease of the MAE compared to the original feature sets. 5.3. General Discussion selection, there is a large decrease. As in other criteria, trustworthiness again performs better than other signals on development, and poorly on test, although the cross-task selection does show improvement for trustworthiness on test, but the absolute value still does not beat that of the arousal and valence. When observing the literature concerning user engagement and the potential advantage of performing this automatically—we see that one essential aspect is the ability for a content creator to develop the parasocial relationship with their viewers (Chapple and Cownie, 2017). In this regard, we see that the features from each emotional dimension (arousal, valence and trustworthiness) can predict core user engagement criteria. Most notably, as we mention previously short-term ﬂuctuations in arousal appear to increase user engagement, and therefore it could be assumed such emotional understanding of video content will lead to higher user-engagement (i.e., an improved parasocial relationship). Result Discussion: When evaluating all results across each user-engagement criteria, it appears that our cross-task feature selection approach obtains the best results more consistently than either automatic selection or all features indicating that a more general selection stabilizes generalization. Through these feature selection approaches valence appears to be a more meaningful signal for most criteria, which can be expected given the positive:negative relationship that is inherent to all the criteria. Furthermore, without any selection, arousal is clearly a strong signal for prediction: with fusion of arousal and valence for Vp/d there is also an improvement. To this end, fusion in general does in no case obtain sustainable better results. With this in mind, further fusion strategies incorporating multiple modes at various stages in the network may be beneﬁcial for further study. Furthermore, the YouTube algorithm itself is known to bias content which has higher user engagement criterion, e.g., comments and likes per day. With this in mind, integration of the emotional features identiﬁed herein (which could be utilized for predicting forthcoming user engagement, cf. Section 6) may result in higher user engagement in other areas, e.g., views per days, resulting in better ﬁnancial outcomes for the creator. The correlations between these aspects, i.e., the increase of comments per day, vs views per day should be further researched concerning these emotional dimensions. Trustworthiness is consistently behind arousal and valence for all criteria. C: parameter of the SVR, optimized for from 0.00001 to 1, using the best M: mean absolute error on the dev(elopement) set to deﬁne C for test set prediction. (%) Indicates the relative change of the automatic (auto.) and matically selected (sel.) in % to the unchanged features, “+” indicates an improvement, thus a decrease of the MAE compared to the original feature sets. Without any feature selection trustworthiness is our strongest signal, for further investigation exploring why trustworthiness does not improve at all with either of the feature selection methods (218.7 and 228.0, for sel. and auto., respectively) would be of interest. Likes per day: As with Vp/d, we see that arousal and valence are strong as singular signals when utilizing all summary features; however, in this case, there is no improvement found through the fusion of multiple feature types. Further to this, the cross- task selection method appears to improve results across all types, aside from the fusion of arousal, valence, and trustworthiness. As with Vp/d, valence again obtains our best result, improved even further by the automatic selection, up to 1.23 MAE Vp/d. Although the automatic selection appears valid for valence, this was not consistent across all the feature type variations. Trustworthiness appears much weaker than all other features types in this case, although when observing scores on the development set; we see that trustworthiness is our strongest singular signal (2.21), even showing promise when fused with the other feature types and from the automatic feature selection. Comments per day: Results obtained from Cp/d continue to show the trend of valence being a meaningful signal. Again for all features, as singular signal both arousal and valence show the best score (0.154 MAE for both). Valence improves by the auto-selection process, and performs better with the cross-task method. Fusion in this case generally does not show much beneﬁt assigned from the combination of arousal and valence, in which our best Cp/d score is obtained from cross-task selection of 0.145 MAE. As previously, trustworthiness is again not the strongest signal on test, however, we see a similar strength on development set. Likes of comments per day: Arousal achieves the strongest result from all features for CLp/d. Unlike the other user engagement criteria, we see a large decrease across most results from the both selection methods. The best improvement comes from the fusion of arousal and valence with the task speciﬁc selection method. However, from automatic March 2022 \| Volume 4 \| Article 773154 10 Stappen et al. User Engagement Estimation From Emotions FIGURE 5 \| SVM weights of arousal, valence, and trustworthiness features predicting the likes of comments comparing the use of all features, manually selected (24), and automatically selected k = 22 features. C: parameter of the SVR, optimized for from 0.00001 to 1, using the best M: mean absolute error on the dev(elopement) set to deﬁne C for test set prediction. (%) Indicates the relative change of the automatic (auto.) and matically selected (sel.) in % to the unchanged features, “+” indicates an improvement, thus a decrease of the MAE compared to the original feature sets. To receive a real output and ﬁt the p-values of the automatic selection in scale, we apply a base 10 logarithm and divide the result by 10 (−Log(pvalue)/10). FIGURE 5 \| SVM weights of arousal, valence, and trustworthiness features predicting the likes of comments comparing the use of all features, manually selected (24), and automatically selected k = 22 features. To receive a real output and ﬁt the p-values of the automatic selection in scale, we apply a base 10 logarithm and divide the result by 10 ( Log(p l )/10) Frontiers in Computer Science \| www.frontiersin.org March 2022 \| Volume 4 \| Article 773154 6. LIMITATIONS AND FUTURE WORK In this section, we would like to point out some aspects of our work that need further exploration, given the novelty of the proposed idea to use continuous emotion signals for modeling explicit user engagement. As with MUSE-CAR , some previously collected datasets harvested YouTube as their primary source (Wöllmer et al., 2013; Zadeh et al., 2018). However, they either do not provide continuous emotion signals or the video metadata (e.g., unique video identiﬁers) of these datasets. Therefore, MUSE-CAR is currently the only dataset that allows studies similar to this, limits extensive exploration in other domains. We want to encourage future dataset creators using social media to provide such identiﬁers. Through a bridge of emotion recognition and user engagement, we see novel applications. The link between emotional and user engagement provides information about what and when (e.g., a part of a video with many arousal peaks) exactly causes a user to feel e.g., aversion, interest or frustration (Picard, 1999). Two parties may particularly beneﬁt from these ﬁndings: (a) Social media network providers: the relationships discovered are directly related to the user retention (e.g., user churn rate) (Lebreton and Yamagishi, 2020) and activity (e.g., recommender systems) (Zhou et al., 2016). These are the most common and important tasks of these platforms and are still extremely diﬃcult to model to this day (Lin et al., 2018; Yang et al., 2018; Liu et al., 2019). Maybe more importantly, critical, emotionally charged videos (e.g., misinformation, fake messages, hate speech) can be recognized and recommendation systems adapted accordingly. (b) Content creators (marketing, advertising): companies act as (video) creators to interact with customers. In our work, we focused to show a connection between generalizable emotional characteristics and user engagement. However, we believe that there are various weaker/stronger inﬂuenced subgroups. A company can identify and target such groups or even explicitly ﬁne-tune their content. When choosing the prediction method, we had to make the diﬃcult choice between interpretability and accuracy. For this study, we opted to use SVMs because we believe that initially, conceivable interactions matter more than a highly optimized outcome. This way, we can reason about relationships between inﬂuencing variables and the output predictions and compare them to ones, extracted from potential other datasets in the future. 5.3. General Discussion A somewhat unexpected result, although this may be caused by generalizability issues on the testing set, further shown by the strong results during development. Interestingly, as a single signal trustworthiness performs better than arousal and valence without feature selection for Vp/d. This result is promising, as it shows a tendency that trust is generally valuable for viewership, a ﬁnding which is supported by the literature in regard to building a parasocial relationship (Lim et al., 2020). We had expected trustworthiness to be useful for predicting user-engagement, given the aforementioned parasocial relationship theory. The results are promising for the prediction of trustworthiness. However, this does not appear to be as successful as the more conventional arousal valence emotional dimensions. The current study implements an arguably conventional method for prediction task and is limited by the data domain. Applying the trustworthiness dimension to other datasets of diﬀerent domains (perhaps more popular topics, such March 2022 \| Volume 4 \| Article 773154 11 User Engagement Estimation From Emotions Stappen et al. contextual factors aﬀecting multi-modal emotion recognition, such as the gender of the speaker and the duration of the emotional episode (Bhattacharya et al., 2021) and the use of non-intrusive on-body electromyography (EMG) sensors as additional input signals (Tamulis et al., 2021). For a broad overview of various (mutlimodal) emotion recognition research, we refer the interested reader to the surveys by Soleymani et al. (2017) and Tian et al. (2022). By using human annotations, we aimed to demonstrate the relationship in a vanilla way (using the targets) to avoid wrong conclusions based on any introduced prediction error bias. We also plan to explore (ii) in-depth in the near future. Another exciting research direction is to incorporate the uncertainty of multi-modal emotion recognition systems (Han et al., 2017), hence, how sure is the system in its prediction based on the availability of (or missing) audio, video, and text data, into the prediction of popularity. Thus, in parallel to the emotion, a measure of uncertainty could be given, which is then factored in the popularity prediction. as comedy or infotainment) where similar metadata is available may show to be more fruitful for exploring the link of trust and improved user engagement. 6. LIMITATIONS AND FUTURE WORK We are fully aware that state-of-the-art black-box methods, e.g., deep learning, may achieve better results but lack in clarity around inner workings and may rely on spurious and non-causal correlations that are less generalisable. However, this does not mean there are no other high non-linearity interactions between inputs, which we want to explore in future work. Another point for future exploration is the emotional spectrum. Although MUSE-CAR provides arousal and valence, which are the most consistently used dimensions in previous research, also other third focus dimensions, for example, dominance (Grimm et al., 2008) and likeability (Kossaiﬁet al., 2019) have previously been annotated. Another interesting aspect might be categorical ratings which summarize an entire video. However, we expect much lower predictive value because of the highly compressed representation of such categories summarizing the emotional content (one value instead of several dynamically extracted features based on a video-length signal). Frontiers in Computer Science \| www.frontiersin.org REFERENCES Chen, Y.-L., Chang, C.-L., and Yeh, C.-S. (2017). Emotion classiﬁcation of youtube videos. Decis. Support Syst. 101, 40–50. doi: 10.1016/j.dss.2017.05.014 (2015). Twitter Us Airline Sentiment. Available online at: https://www.kaggle.com/ crowdﬂower/twitter-airline-sentiment Cheng, X., Liu, J., and Dale, C. (2013). Understanding the characteristics of internet short video sharing: a youtube-based measurement study. IEEE Trans. Multimedia 15, 1184–1194. doi: 10.1109/TMM.2013.2265531 Adhikari, R., and Agrawal, R. K. (2013). An Introductory Study on Time Series Modeling and Forecasting. LAP LAMBERT Academic Publishing. Christ, M., Braun, N., Neuﬀer, J., and Kempa-Liehr, A. W. (2018). Time series feature extraction on basis of scalable hypothesis tests (tsfresh–a python package). Neurocomputing 307, 72–77. doi: 10.1016/j.neucom.2018.03.067 Ahlgren, P., Jarneving, B., and Rousseau, R. (2003). Requirements for a cocitation similarity measure, with special reference to pearson’s correlation coeﬃcient. J. Am. Soc. Inf. Sci. Technol. 54, 550–560. doi: 10.1002/asi.10242 Am. Soc. Inf. Sci. Technol. 54, 550–560. doi: 10.1002/asi.10242 Colquitt, J. A., Scott, B. A., and LePine, J. A. (2007). Trust, trustworthiness, and trust propensity: a meta-analytic test of their unique relationships with risk taking and job performance. J. Appl. Psychol. 92, 909. doi: 10.1037/0021-9010.92.4.909 Baird, A., and Schuller, B. (2020). Considerations for a more ethical approach to data in ai: on data representation and infrastructure. Front. Big Data 3, 25. doi: 10.3389/fdata.2020.00025 Cooper, P. (2019). 23 YouTube Statistics That Matter To Marketers in 2020. Battaglio, S. (2016). Youtube Now Bigger Than TV Among Advertisers’ Target Audience. Available online at: https://www.latimes.com/entertainment/envelop e/cotown/la-et-ct-you-tube-ad-spending-20160506-snap-story.html (October 15, 2020). Cox, J. C., Kerschbamer, R., and Neururer, D. (2016). What is trustworthiness and what drives it? Games Econ. Behav. 98, 197–218. doi: 10.1016/j.geb.2016.05.008 Cui, B., Li, Y., Chen, M., and Zhang, Z. (2019). “Fine-tune BERT with sparse self-attention mechanism,” in Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (Hong Kong: ACL), 3548–3553. Berger, J., and Milkman, K. L. (2012). What makes online content viral? J. Market. Res. 49, 192–205. doi: 10.1509/jmr.10.0353 Bhattacharya, P., Gupta, R. K., and Yang, Y. (2021). Exploring the contextual factors aﬀecting multimodal emotion recognition in videos. IEEE Trans. Aﬀect. Comput.. Devlin, J., Chang, M.-W., Lee, K., and Toutanova, K. (2019). “BERT: pre-training of deep bidirectional transformers for language understanding,” in Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics (ACL), 4171–4186. Bhuiyan, H., Ara, J., Bardhan, R., and Islam, M. R. (2017). 7. CONCLUSION y y g g So far, no link existed between the use of emotional signals and user engagement. That is why, the aim of our paper was to provide a proof of concept that it is valuable to leverage such signals. However, utilizing human annotations can only be the ﬁrst step since they are very limited in scalability. The annotations are usually the prediction target for developing robust emotion recognition models. Our ﬁnal process is intended to be twofold: (i) using audio-visual features to learn to predict the human emotional signals (ii) using the predicted emotional signals on unseen, unlabeled videos to extract our feature set and predict user-engagement. (i) is very well researched in the ﬁeld achieving CCCs of more than 0.7 (high correlation between predicted and human emotional annotations) on similar data sets (Huang et al., 2020). Recent advances aim at understanding For the ﬁrst time, we have empirically (and on a large-scale) presented in this contribution that there are both, intuitive and complex relationships between user engagement indicators and continuously annotated emotion, as well as trustworthiness signals in user-generated data. Of prominence, our contribution ﬁnds that emotion increases engagement when arousal is consistently bounded. In other words, the more consistent the portrayed arousal throughout a video, the better the engagement with it. This ﬁnding contradicted previous emotion literature (Sagha et al., 2017). Arousal shows consistently more robust prediction results, although valence innately (given the link of positive and negative) appears to be more valuable for prediction of video likes. March 2022 \| Volume 4 \| Article 773154 Frontiers in Computer Science \| www.frontiersin.org 12 User Engagement Estimation From Emotions Stappen et al. accession number(s) can be found at: https://doi.org/10.5281/ zenodo.4651164. Further to this, we introduce trustworthiness as a continuous “emotion” dimension for engagement, and ﬁnd when utilizing this for prediction, there is an overall value for monitoring user-engagement in social-media content. However, when fusing the signals, their appears to be little beneﬁt from the current recognition paradigm. Furthermore, we assume that too strict feature selection causes generalization issues since often promising results on the development set seem non-transferable to the test set. AUTHOR CONTRIBUTIONS LS: literature analysis, data acquisition, data preparation, experimental design, computational analysis, and manuscript drafting and preparation. ABa: data acquisition, experimental design, and manuscript drafting and preparation. ML and ABä: data acquisition, data preparation, and computational analysis. BS: technical guidance and manuscript editing. All authors revised, developed, read, and approved the ﬁnal manuscript. From the strong correlation of the results for trustworthiness, we consider that the addition of this dimension is of use for user engagement; however, further investigation in other domains would be valuable. When applying these metrics in a cross-modal sentiment paradigm, there may also be beneﬁts for the prediction of audio-visual hate speech likelihood, as well as fake news. DATA AVAILABILITY STATEMENT The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fcomp. 2022.773154/full#supplementary-material The datasets presented in this study can be found in online repositories. The names of the repository/repositories and REFERENCES Statistical power and signiﬁcance testing in large-scale genetic studies. Nat. Rev. Genet. 15, 335–346. doi: 10.1038/ nrg3706 Shehu, E., Bijmolt, T. H., and Clement, M. (2016). Eﬀects of likeability dynamics on consumers’ intention to share online video advertisements. J. Interact. Market. 35, 27–43. doi: 10.1016/j.intmar.2016.01.001 Lin, Z., Althoﬀ, T., and Leskovec, J. (2018). “I’ll be back: on the multiple lives of users of a mobile activity tracking application,” in Proceedings of the 2018 World Wide Web Conference (WWW) (Geneva), 1501–1511. Liu, Y., Shi, X., Pierce, L., and Ren, X. (2019). “Characterizing and forecasting user engagement with in-app action graph: a case study of snapchat,” in Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (KDD) (Anchorage, AK), 2023–2031. Siersdorfer, S., Chelaru, S., Nejdl, W., and San Pedro, J. (2010). “How useful are your comments? analyzing and predicting youtube comments and comment ratings,” in Proceedings of the 19th International Conference on World Wide Web (WWW) (Raleigh, NC), 891–900. Moturu, S. T., and Liu, H. (2011). Quantifying the trustworthiness of social media content. Distrib. Parallel Databases 29, 239–260. doi: 10.1007/s10619-010-7077-0 Siersdorfer, S., Chelaru, S., Pedro, J. S., Altingovde, I. S., and Nejdl, W. (2014). Analyzing and mining comments and comment ratings on the social web. ACM Trans. Web 8, 1–39. doi: 10.1145/2628441 Nakov, P., Rosenthal, S., Kozareva, Z., Stoyanov, V., Ritter, A., and Wilson, T. (2013). “SemEval-2013 task 2: sentiment analysis in Twitter,” in Second Joint Conference on Lexical and Computational Semantics, Proceedings of the Seventh International Workshop on Semantic Evaluation (Atlanta, GA, ACL), 312–320. Singh, V. K., Piryani, R., Uddin, A., and Waila, P. (2013). “Sentiment analysis of movie reviews: a new feature-based heuristic for aspect- level sentiment classiﬁcation,” in 2013 International Mutli-Conference on Automation, Computing, Communication, Control and Compressed Sensing (Kottayam: IEEE), 712–717. Nikolinakou, A., and King, K. W. (2018). Viral video ads: emotional triggers and social media virality. Psychol. Market. 35, 715–726. doi: 10.1002/mar.21129 Soleymani, M., Garcia, D., Jou, B., Schuller, B., Chang, S.-F., and Pantic, M. (2017). A survey of multimodal sentiment analysis. Image Vis. Comput. 65, 3–14. doi: 10.1016/j.imavis.2017.08.003 Palshikar, G. (2009). “Simple algorithms for peak detection in time-series,” in Proceedings of the 1st International Conference on Advanced Data Analysis, Business Analytics and Intelligence. (Ahmedabad), vol. 122. Stappen, L., Baird, A., Rizos, G., Tzirakis, P., Du, X., Hafner, F., et al. (2020a). REFERENCES D., Mei, T., Zeng, W., and Li, S. (2013). Towards cross-domain learning for social video popularity prediction. IEEE Trans. Multimedia 15, 1255–1267. doi: 10.1109/TMM.2013.2265079 Russell, J. A. (1980). A circumplex model of aﬀect. J. Pers. Soc. Psychol. 39, 1161–1178. Kollias, D., Tzirakis, P., Nicolaou, M. A., Papaioannou, A., Zhao, G., Schuller, B., et al. (2019). Deep aﬀect prediction in-the-wild: aﬀ-wild database and challenge, deep architectures, and beyond. Int. J. Comput. Vis. 127, 1–23. doi: 10.1007/s11263-019-01158-4 Sagha, H., Schmitt, M., Povolny, F., Giefer, A., and Schuller, B. (2017). “Predicting the popularity of a talk-show based on its emotional speech content before publication,” in Proceedings 3rd International Workshop on Aﬀective Social Multimedia Computing, Conference of the International Speech Communication Association (INTERSPEECH) Satellite Workshop (Stockholm, ISCA). Kossaiﬁ, J., Walecki, R., Panagakis, Y., Shen, J., Schmitt, M., Ringeval, F., Han, J., Pandit, V., Toisoul, A., Schuller, B. W., et al. (2019). Sewa db: a rich database for audio-visual emotion and sentiment research in the wild. IEEE Trans. Pattern Anal. Mach. Intell. 43, 1022–1040. doi: 10.1109/TPAMI.2019.2944808 Schuller, B., Lang, M., and Rigoll, G. (2002). “Automatic emotion recognition by the speech signal,” in Proccedings of SCI 2002, 6th World Multiconference on Systemics, Cybernetics and Informatics. (Orlando). Kujur, F., and Singh, S. (2018). Emotions as predictor for consumer engagement in youtube advertisement. J. Adv. Manag. Res. 15, 184–197. doi: 10.1108/JAMR-05-2017-0065 Schuller, B. W. (2013). Intelligent Audio Analysis. (Lyon, Springer). Lan, Z., Chen, M., Goodman, S., Gimpel, K., Sharma, P., and Soricut, R. (2020). “Albert: a lite bert for self-supervised learning of language representations,” in 2020 International Conference on Learning Representations (Addis Ababa). Schuller, B. W., Batliner, A., Bergler, C., Messner, E.-M., Hamilton, A., Amiriparian, S., et al. (2020). “The interspeech 2020 computational paralinguistics challenge: elderly emotion, breathing & masks,” in Proceedings Conference of the International Speech Communication Association (INTERSPEECH) (Shanghai). Lebreton, P., and Yamagishi, K. (2020). Predicting user quitting ratio in adaptive bitrate video streaming. IEEE Trans. Multimedia 23, 4526–4540. doi: 10.1109/TMM.2020.3044452 Severyn, A., Moschitti, A., Uryupina, O., Plank, B., and Filippova, K. (2016). Multi-lingual opinion mining on youtube. Inf. Process. Manag. 52, 46–60. doi: 10.1016/j.ipm.2015.03.002 Lim, J. S., Choe, M.-J., Zhang, J., and Noh, G.-Y. (2020). The role of wishful identiﬁcation, emotional engagement, and parasocial relationships in repeated viewing of live-streaming games: a social cognitive theory perspective. Comput. Hum. Behav. 108, 106327. doi: 10.1016/j.chb.2020.106327 Sham, P. C., and Purcell, S. M. (2014). REFERENCES Go, A., Bhayani, R., and Huang, L. (2009). Twitter sentiment classiﬁcation using distant supervision. CS224N Project Rep. Stanford 1, 2009. Posner, J., Russell, J. A., and Peterson, B. S. (2005). The circumplex model of aﬀect: an integrative approach to aﬀective neuroscience, cognitive development, and psychopathology. Develop. Psychopathol. 17, 715. doi: 10.1017/S0954579405050340 Grimm, M., Kroschel, K., and Narayanan, S. (2008). “The vera am mittag german audio-visual emotional speech database,” in 2008 IEEE International Conference on Multimedia & Expo (ICME) (Hannover: IEEE), 865–868. Preo¸tiuc-Pietro, D., Schwartz, H. A., Park, G., Eichstaedt, J., Kern, M., Ungar, L., et al. (2016). “Modelling valence and arousal in facebook posts,” in Proceedings of the 7th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis co-located to Association for Computer Lingustics (San Diego, CA: ACM), 9–15. Guyon, I., Weston, J., Barnhill, S., and Vapnik, V. (2002). Gene selection for cancer classiﬁcation using support vector machines. J. Mach. Learn. 46, 389–422. doi: 10.1023/A:1012487302797 Han, J., Zhang, Z., Schmitt, M., Pantic, M., and Schuller, B. (2017). “From hard to soft: Towards more human-like emotion recognition by modelling the perception uncertainty,” in Proceedings of the 25th ACM International Conference on Multimedia (New York, NY), 890–897. Rangaswamy, S., Ghosh, S., Jha, S., and Ramalingam, S. (2016). “Metadata extraction and classiﬁcation of youtube videos using sentiment analysis,” in 2016 IEEE International Carnahan Conference on Security Technology (Orlando, FL: IEEE), 1–2. Horsburgh, H. (1961). Trust and social objectives. Ethics 72, 28–40 Richman, J. S., and Moorman, J. R. (2000). Physiological time-series analysis using approximate entropy and sample entropy. Am. J. Physiol. Heart Circul. Physiol. 278, 2039–2049. doi: 10.1152/ajpheart.2000.278.6.H2039 Huang, J., Tao, J., Liu, B., Lian, Z., and Niu, M. (2020). “Multimodal transformer fusion for continuous emotion recognition,” in IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (Barcelona: IEEE), 3507– 3511. Ringeval, F., Schuller, B., Valstar, M., Gratch, J., Cowie, R., Scherer, S., et al. (2017). “Avec 2017: real-life depression, and aﬀect recognition workshop and challenge,” in Proceedings of the 7th Annual Workshop on Audio/Visual Emotion Challenge (Mountain View, CA), 3–9. Khan, G. F., and Vong, S. (2014). Virality over youtube: an empirical analysis. Internet Res. 24, 19. doi: 10.1108/INTR-05-2013-0085 Knuutila, A., Herasimenka, A., Au, H., Bright, J., and Howard, P. N. (2020). Covid-related misinformation on youtube. Oxford Memos: The Spread of Misinformation Videos on Social Media and the Eﬀectiveness of Platform Policies. (Oxford). Roy, S. REFERENCES “Retrieving youtube video by sentiment analysis on user comment,” in 2017 IEEE International Conference on Signal and Image Processing Applications (Kuching: IEEE), 474– 478. Doane, D. P., and Seward, L. E. (2011). Measuring skewness: a forgotten statistic? J. Stat. Educ. 19, 1–18 doi: 10.1080/10691898.2011.11889611 Biel, J., and Gatica-Perez, D. (2013). The youtube lens: Crowdsourced personality impressions and audiovisual analysis of vlogs. IEEE Trans. Multimedia 15, 41–55. doi: 10.1109/TMM.2012.2225032 Dobrian, F., Sekar, V., Awan, A., Stoica, I., Joseph, D., Ganjam, A., et al. (2011). Understanding the impact of video quality on user engagement. ACM SIGCOMM Comput. Commun. Rev. 41, 362–373. doi: 10.1145/2043164.2018478 Boiy, E., Hens, P., Deschacht, K., and Moens, M.-F. (2007). “Automatic sentiment analysis in on-line text,” in Proceedings of the 11th International Conference on Electronic Publishing (Vienna), 349–360. Ekman, P. (1992). An argument for basic emotions. Cogn. Emotion 6, 169–200. Busso, C., Bulut, M., Lee, C.-C., Kazemzadeh, A., Mower, E., Kim, S., et al. (2008). Iemocap: interactive emotional dyadic motion capture database. Lang. Resour. Eval. 42, 335. doi: 10.1007/s10579-008-9076-6 English, K., Sweetser, K. D., and Ancu, M. (2011). Youtube-iﬁcation of political talk: an examination of persuasion appeals in viral video. Am. Behav. Sci. 55, 733–748. doi: 10.1177/0002764211398090 Garroppo, R. G., Ahmed, M., Niccolini, S., and Dusi, M. (2018). A vocabulary for growth: topic modeling of content popularity evolution. IEEE Trans. Multimedia 20, 2683–2692. doi: 10.1109/TMM.2018.2811625 Ceron, A., Curini, L., Iacus, S. M., and Porro, G. (2014). Every tweet counts? how sentiment analysis of social media can improve our knowledge of citizens’ political preferences with an application to italy and france. New Media Soc. 16, 340–358. Geurts, P. (2001). “Pattern extraction for time series classiﬁcation,” in European Conference on Principles of Data Mining and Knowledge Discovery (Freiburg im Breisgau: Springer), 115–127. Chang, Y.-W., and Lin, C.-J. (2008). “Feature ranking using linear svm,” in Causation and Prediction Challenge (PMLR), 53–64. Gilbert, C., and Hutto, E. (2014). “Vader: a parsimonious rule-based model for sentiment analysis of social media text,” in Eighth International Conference on Weblogs and Social Media, vol. 81 (Ann Arbor, MI), 82. Chapple, C., and Cownie, F. (2017). An investigation into viewers’ trust in and response towards disclosed paid-for-endorsements by youtube lifestyle vloggers. J. Promotional Commun. 5, 19–28. Frontiers in Computer Science \| www.frontiersin.org March 2022 \| Volume 4 \| Article 773154 Frontiers in Computer Science \| www.frontiersin.org 13 User Engagement Estimation From Emotions Stappen et al. REFERENCES “Muse 2020 challenge and workshop: Multimodal sentiment analysis, emotion-target engagement and trustworthiness detection in real-life media,” in 1st International Multimodal Sentiment Analysis in Real-life Media Challenge and Workshop, co-located to ACM International Conference on Multimedia. (Seattle, WA: ACM). Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., et al. (2011). Scikit-learn: machine learning in python. J. Mach. Learn. Res. 12, 2825–2830. Picard, R. W. (1999). “Aﬀective computing for hci,” in Human Computer Interaction (Citeseer), 829–833. March 2022 \| Volume 4 \| Article 773154 Frontiers in Computer Science \| www.frontiersin.org 14 User Engagement Estimation From Emotions Stappen et al. Yan, M., Sang, J., Xu, C., and Hossain, M. S. (2015). Youtube video promotion by cross-network association:@ britney to advertise gangnam style. IEEE Trans. Multimedia 17, 1248–1261. doi: 10.1109/TMM.2015.2446949 Stappen, L., Baird, A., Schumann, L., and Schuller, B. (2021). The multimodal sentiment analysis in car reviews (muse-car) dataset: collection, insights and improvements. arXiv preprint arXiv:2101.06053. Stappen, L., Brunn, F., and Schuller, B. (2020b). Cross-lingual zero-and few-shot hate speech detection utilizing frozen transformer language models and axel. arXiv preprint arXiv:2004.13850. Yang, C., Shi, X., Jie, L., and Han, J. (2018). “I know you’ll be back: Interpretable new user clustering and churn prediction on a mobile social application,” in Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (KDD) (London), 914–922. Stappen, L., Schuller, B. W., Lefter, I., Cambria, E., and Kompatsiaris, I. (2020c). “Summary of muse 2020: Multimodal sentiment analysis, emotion-target engagement and trustworthiness detection in real-life media,” in 28th ACM International Conference on Multimedia. (Seattle, WA: ACM). Yang, R., Singh, S., Cao, P., Chi, E., and Fu, B. (2016). “Video watch time and comment sentiment: experiences from youtube,” in 2016 Fourth IEEE Workshop on Hot Topics in Web Systems and Technologies (HotWeb) (Washington, DC: IEEE), 26–28. Subramanian, R., Shankar, D., Sebe, N., and Melcher, D. (2014). Emotion modulates eye movement patterns and subsequent memory for the gist and details of movie scenes. J. Vis. 14, 31–31. doi: 10.1167/14. 3.31 Yentes, J. M., Hunt, N., Schmid, K. K., Kaipust, J. P., McGrath, D., and Stergiou, N. (2013). The appropriate use of approximate entropy and sample entropy with short data sets. Ann. Biomed. Eng. 41, 349–365. doi: 10.1007/s10439-012-0668-3 Zadeh, A., Liang, P. P., Poria, S., Cambria, E., and Morency, L.-P. (2018). REFERENCES “Multimodal language analysis in the wild: Cmu-mosei dataset and interpretable dynamic fusion graph,” in Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Melbourne, VIC), 2236–2246. Surhone, L., Timpledon, M., and Marseken, S. (2010). Spearman’s Rank Correlation Coeﬃcient: Statistics, Non-Parametric Statistics, Raw Score, Null Hypothesis, Fisher Transformation, Statistical Hypothesis Testing, Conﬁdence Interval, Correspondence Analysis. Betascript Publishing. Zhou, P., Zhou, Y., Wu, D., and Jin, H. (2016). Diﬀerentially private online learning for cloud-based video recommendation with multimedia big data in social networks. IEEE Trans. Multimedia 18, 1217–1229. doi: 10.1109/TMM.2016.2537216 Zhou, P., Zhou, Y., Wu, D., and Jin, H. (2016). Diﬀerentially private online learning for cloud-based video recommendation with multimedia big data in social networks. IEEE Trans. Multimedia 18, 1217–1229. doi: 10.1109/TMM.2016.2537216 Tamulis, Ž., Vasiljevas, M., Damaševicius, R., Maskeliunas, R., and Misra, S. (2021). “Aﬀective computing for ehealth using low-cost remote internet of things-based emg platform,” in Intelligent Internet of Things for Healthcare and Industry, vol. 67. (Springer). Tan, Z., and Zhang, Y. (2019). Predicting the top-n popular videos via a cross-domain hybrid model. IEEE Trans. Multimedia 21, 147–156. doi: 10.1109/TMM.2018.2845688 Frontiers in Computer Science \| www.frontiersin.org March 2022 \| Volume 4 \| Article 773154 Conﬂict of Interest: BS was employed by audEERING GmbH. The remaining authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Tian, L., Oviatt, S., Muszynski, M., Chamberlain, B., Healey, J., and Sano, A. (2022). Applied Aﬀective Computing. (Morgan & Claypool). Trzci´nski, T., and Rokita, P. (2017). Predicting popularity of online videos using support vector regression. IEEE Trans. Multimedia 19, 2561–2570. doi: 10.1109/TMM.2017.2695439 Publisher’s Note: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their aﬃliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Uryupina, O., Plank, B., Severyn, A., Rotondi, A., and Moschitti, A. (2014). “SenTube: a corpus for1 sentiment analysis on YouTube social media,” in Proceedings of the Ninth International Conference on Language Resources and Evaluation (Reykjavik, ELRA), 4244–4249. Westfall, P. H. (2014). Kurtosis as peakedness. Am. Stat. 68, 191–195. doi: 10.1080/00031305.2014.917055 Copyright © 2022 Stappen, Baird, Lienhart, Bätz and Schuller. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Wöllmer, M., Weninger, F., Knaup, T., Schuller, B., Sun, C., Sagae, K., et al. (2013). Youtube movie reviews: sentiment analysis in an audio-visual context. IEEE Intell. Syst. 28, 46–53. doi: 10.1109/MIS.2013.34 Wu, Z., and Ito, E. (2014). “Correlation analysis between user’s emotional comments and popularity measures,” in 2014 3rd International Conference on Advanced Applied Informatics (Kokura: IEEE), 280–283. March 2022 \| Volume 4 \| Article 773154 Frontiers in Computer Science \| www.frontiersin.org 15
https://openalex.org/W4285556514	https://rcpjm.cpjm.uerj.br/revista/article/download/31/51	Portuguese	null	Impactos dos programas de compliance na responsabilidade penal individual de dirigentes no âmbito da sociedade empresária	Revista Científica do CPJM	2,021	cc-by	15,407	30 30 Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 1 Graduada em Direito pela Universidade Estácio de Sá do Rio de Janeiro. LL.M. 2014’ pela Boston College Law School com ênfase em Business and Corporate Law. MBA 2017’ em Direito Civil e Processual Civil pela Fundação Getúlio Vargas no Rio de Janeiro. Pós-graduada 2021’ em Direito Público pelo Instituto Superior do Ministério Público do Rio de Janeiro – ISMP. Impactos dos programas de compliance na responsabilidade penal individual de dirigentes no âmbito da sociedade empresária Bruna Perasoli Trade1 Sumário: Introdução. 1. Governança corporativa e Compliance. 1.1 conceito, origem histórica e finalidades dos programas de compliance. 1.2. Origem do fenômeno da autorregulação preventiva e a influência das normas supranacionais. 2. Parâmetros relacionados ao Poder de Punir do Estado. 2.1. Direito Penal Clássico e a Imputação Delitiva. 2.1.1. A (moderna) teoria da imputação objetiva. 2.1.2. A relevância da Teoria do Domínio do Fato. 2.2. Movimentos Teóricos do Direito Penal Moderno. 2.1. Abolicionismo. 2.2. Minimalismo. 2.3. Garantismo. 3. Mecanismos de imputação de delitos cometidos no âmbito empresarial. 3.1. Impactos pelo Descumprimento dos Programas de Compliance. 3.2. Responsabilidade dos dirigentes na qualidade de garante (“compliance officer”). 4. Conclusão. 5. Referências bibliográficas. Resumo: Este artigo aborda os limites do poder de punir do Estado e a influência da autorregulação, de acordo com as modificações político-social na esfera do Direito Penal. Dentre os temas debatidos, temos o surgimento do compliance em termos de sociedade e aspectos legislativos, o desenvolvimento de teorias – clássicas e modernas – que visam consolidar aspectos relevantes dentro do conceito de imputação criminal, seus agentes e ações quando envolvem domínio e controle. Por fim, o tema alude sobre a possibilidade de responsabilização dos agentes quando na qualidade de dirigentes de sociedades empresárias. Palavras-chave: Compliance. Direito Penal Moderno. Responsabilidade da Pessoa Jurídica. Responsabilidade individual. Dirigente. Abstract: This article addresses the limits of the State's power to punish and the influence of self-regulation, according to the political and social changes in the sphere of Criminal Law. Keywords: Compliance. Programs. Business. Criminal Compliance. Keywords: Compliance. Programs. Business. Criminal Compliance. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 Abstract: This article addresses the limits of the State's power to punish and the influence of self-regulation, according to the political and social changes in the sphere of Criminal Law. 1 Graduada em Direito pela Universidade Estácio de Sá do Rio de Janeiro. LL.M. 2014’ pela Boston College Law School com ênfase em Business and Corporate Law. MBA 2017’ em Direito Civil e Processual Civil pela Fundação Getúlio Vargas no Rio de Janeiro. Pós-graduada 2021’ em Direito Público pelo Instituto Superior do Ministério Público do Rio de Janeiro – ISMP. 31 Among the topics discussed, we have the emergence of compliance in terms of society and legislative aspects, the development of theories - classic and modern - that aim to consolidate relevant aspects within the concept of criminal imputation, its agents and actions when they involve dominance and control. Finally, the theme alludes to the possibility of accountability for agents when they are managers of business companies. Keywords: Compliance. Programs. Business. Criminal Compliance. Idem, p. 45. 4 COMPLY. In: MICHAELIS Moderno Dicionário da língua inglesa. São Paulo: Melhoramentos. Disponível em: <https://michaelis.uol.com.br/moderno-ingles/busca/ingles-portugues-moderno/comply/>. Acesso em 18 de dezembro de 2020. 1 RODRIGUES, Anabela Miranda. Direito Penal Econômico: uma política criminal na era compliance. Coimbra: Edições Almedina, 2019, p. 11. 2 Idem, p. 12. 1 RODRIGUES, Anabela Miranda. Direito Penal Econômico: uma política criminal na era com Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 Introdução O trabalho visa, num primeiro momento, situar o leitor quanto aos reflexos dos regramentos adotados nos ambientes empresariais – os chamados programas de compliance, o qual possui o intuito de consolidar a cultura da empresa e definir os limites tolerados dentro daquele grupo profissional. No segundo momento, iremos analisar como estes programas são estruturados, quais as ferramentas são utilizadas para obter eficácia da aplicação destes conjuntos de regulação e quais seus efeitos internos e externos, tanto em relação às próprias pessoas jurídicas envolvidas quanto aos dirigentes envolvidos nas práticas consideradas contrárias aos programas. Posteriormente, necessária a análise destes tão inovadores programas sob a perspectiva do Direito Penal Moderno, uma vez que estes regramentos fazem previsão e aplicação de sanções como métodos coercitivos face ao descumprimento das regras dos referidos programas, bem como os efeitos externos que recaem, inclusive, sobre a figura individual dos dirigentes. Na sequência, serão analisados os mecanismos que envolvem as empresas e como configura a responsabilização dos agentes à frente da gestão da pessoa jurídica, a configuração de seus atos de gestão, e principalmente suas falhas – sejam estas intencionais, ou não – e quais as consequências destes atos para o ordenamento jurídico vigente, em especial no que tange a esfera do Direito Penal. Lembrando que todo o estudo apresentado foi regido pela investigação dos institutos relativos ao conteúdo, o que foi realizado através da metodologia de pesquisa bibliográfica com a utilização de artigos, livros de doutrina e jurisprudência, com intuito de embasar toda a fundamentação teórica, consubstanciando o raciocínio apresentado neste trabalho. 32 Governança corporativa e compliance. Governança corporativa e compliance. 3 Idem, p. 45. 2 Idem, p. 12. IGUES, Anabela Miranda. Direito Penal Econômico: uma política criminal na era compliance. Edi õ Al di 2019 11 IGUES, Anabela Miranda. Direito Penal Econômico: uma política criminal na era compliance. a: Edições Almedina, 2019, p. 11. 2.1 Conceito geral do termo compliance Discussões sobre a legitimidade e a necessidade de intervenção penal do domínio econômico encontram-se num momento particularmente sensível. A referência à crise financeira de 2008 tem suscitado respostas absolutamente díspares e contraditórias1, que vão do discurso de resistência, contrário à intervenção penal econômica, ao moderno conceito de crime econômico de caráter político, que afirma haver uma delinquência que não apenas se conecta com o poder econômico, mas que é verdadeiramente tolerada pelo Estado pelas conexões políticas que possui.2 Nessa quadra a empresa se torna o pano de fundo da afetação de bens jurídicos transindividuais e os proprietários, muitas das vezes, não tem qualquer controle sobre os comportamentos delituosos que no seu âmbito se desenvolvem, que se consolidam por meio dos resultados indesejados das políticas econômicas neoliberais e da desregulação das quais decorreram os escândalos financeiros que a supervisão não conseguiu evitar. Nesta perspectiva, surge o modelo que se convencionou denominar de autorregulação regulada consubstanciada nas orientações de corporate governance e de compliance3, onde se buscou instituir uma governança corporativa de sorte a proteger sócios e a própria coletividade frente a possibilidade de desvios na condução da empresa. A escolha da terminologia compliance, a qual deu origem aos ditos programas provém do verbo em inglês “to comply” que significa “condescender, aquiescer, aceder, concordar, consentir, ceder; cumprir, satisfazer, corresponder a obedecer, estar de acordo”4, seja agir de acordo com um comando, seja este estabelecido por lei, instrução interna ou conduta ética. Contudo, não se pode resumir o sentido do que vem a ser esta prática apenas ao seu sentido literal, considerando sua amplitude como método sistêmico que funciona como instrumento de mitigação de riscos, preservação de valores éticos e de sustentabilidade 33 corporativa, assegurando a continuidade do negócio e os interesses das pessoas coligadas ao sistema corporativo. ( ) p p 6 A Lei nº. 12.846/2013 preferiu o termo “integridade”, mas já a Lei nº. 13.303/2016, que dispõe sobre o estatuto jurídico da empresa pública, da sociedade de economia mista e de suas subsidiarias utilizou o termo compliance mais utilizado pelos profissionais que atual na área. 5 CARVALHO, André Castro; BERTOCCELLI, Rodrigo de Pinho; ALVIM, Tiago Cripa; VENTURINI, Otavio (Coord.). Manual de Compliance. 2. ed. Rio de Janeiro: Forense, 2020. p. 39-55. p p q 7 HEGENBERG, Leonidas. Definições – termos teóricos e significados. São Paulo: Cultrix, 1973. p. 42. In: Op. cit., item 6. p. 62. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 2.1 Conceito geral do termo compliance Nesse sentido, ilustra de Rodrigo de Pinho Bertoccelli que se trata de sistema complexo e organizado de procedimentos de controle de riscos e preservação de valores intangíveis que deve ser coerente com a estrutura societária, o compromisso efetivo da sua liderança e a estratégia da empresa, como elemento, cuja adoção resulta na criação de um ambiente de segurança jurídica e confiança indispensável para a boa tomada de decisão.5 Consoante aos aspectos acima mencionados, esta ferramenta também é entendida como um sistema interno muito chamado por “programa de integridade”6 (sendo a referência “programa de “compliance” mais utilizada no âmbito empresarial), tem por finalidade prevenir, detectar e corrigir atos não condizentes com os princípios e valores da empresa, assim como perante o ordenamento jurídico vigente. Contudo, apesar desta ser uma conceituação objetiva, o dinamismo desta temática perfaz a dificuldade de compreender o real significado desta prática. Isto ocorre por se tratar de um conceito relativamente novo no Brasil que gerou largas discussões e respectivas adaptações aos padrões sociais, jurídicos e culturais. Logo, visando determinar quais seriam as propriedades suficientes e necessárias7 deste termo, o sistema empresarial brasileiro adota como referência apenas a compreensão de “estar de acordo com”. E, apesar de se notar a dupla aplicação do termo compliance (seja como processo, em que se compreende por atividades, modelos, caminhos a serem adotados pela pessoa jurídica; e como produto, a palavra se refere aos resultados das tais atividades), aqui iremos concentrar o estudo nos aspectos desta prática como processo, por notoriamente ser o que mais implica dentro do cenário empresarial, bem como abordar seus respectivos reflexos no âmbito jurídico. 2.2 Origem do fenômeno da autorregulação preventiva e a influência das normas supranacionais 2.2 Origem do fenômeno da autorregulação preventiva e a influência das normas supranacionais Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 8 GOMES, Abel Fernandes. Responsabilidade penal pela omissão de compliance. Centro de Pesquisa em Crimes Empresariais e Compliance Prof. João Marcello de Araújo Jr. (CPJM), Rio de Janeiro, 2020. Disponível em: <http://www.cpjm.uerj.br/wp-content/uploads/2020/06/RESPONSABILIDADE-PENAL-PELA-OMISSÃO- DE-COMPLIANCE.pdf>. Acesso em 21 de dezembro de 2020. 9 Idem 2.2 Origem do fenômeno da autorregulação preventiva e a influência das normas supranacionais Numa visão ampliada, uma série de iniciativas de governos e organismos internacionais, com vistas ao estabelecimento de um conjunto de ações a serem adotadas na linha da Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 34 constituição de um alicerce de princípios e regras às entidades públicas e privadas, visando o controle preventivo de atos ilícitos (basicamente: a criminalidade organizada, a lavagem de dinheiro, as fraudes financeiras, o financiamento ao terrorismo e a corrupção, as quais geram um nível de interferência altíssimo no contexto socioeconômico e financeiro), praticados no fluxo de algumas atividades econômicas e financeiras nacionais e internacionais8. Diante do alastrar destes fenômenos ilícitos e seus rebotes negativos em diversas esperas sociais, podemos indicar como marco temporal mais distante o período que sucedeu a dois acontecimentos mundiais críticos: a crise econômica de 1929 e a Segunda Guerra Mundial.9 Retroagindo à década de 1950, se faz possível notar oficialmente a preocupação com atividades desempenhadas à margem dos controles oficias e contrárias às leis vigentes, que uma vez criminalizadas ensejaram práticas clandestinas e mais organizadas, sendo o exemplo mais marcante a repressão ao comércio de bebidas alcoólicas nos Estados Unidos da América, logo revelando o quanto tal criminalidade organizada era capaz de desenvolver uma atividade econômica tão rentável quanto marginal, retroalimentando o poderio econômico e até político daquilo que se denominou uma “delinquência organizada”, se valendo exatamente da corrupção, da sonegação fiscal e da lavagem de dinheiro.10 Como tentativa de coerção, a Assembleia Geral da Organização das Nações Unidas (ONU) preconizou a necessidade de da realização periódica (a cada cinco anos) de um Congresso Internacional de Prevenção ao Crime e Punição de Criminosos, mas que acabou esvaziada na sua efetivação prática, haja vista os obstáculos necessários para atingir a cooperação internacional entre Estados, neste ato impossibilitada em razão do fenômeno da denominada Guerra Fria que cindiu tantas nações. Neste viés, surge a primitiva ideia de compliance, por intermédio da legislação norte- americana, com a criação da Prudential Securities (1950), e por meio da regulação da Securities Exchange Commission – SEC (1960), onde se enfatizou a necessidade de institucionalizar os programas com foco de governança preventiva, intentando o controle e monitoramento interno de diversas operações. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. 11 GLYN, Patrick; KOBRIN, Stephen J.; NAIM, Moisés. Globalização da corrupção, a economia política da corrupção. Brasília: UNB, 2002. p. 74. 12 Diretriz do Estatuto em sua íntegra acessível por meio do site do Departamento de Justiça dos Estados Unidos. Disponível em: <https://www.justice.gov/sites/default/files/criminal-fraud/legacy/2012/11/14/fcpa- english.pdf>. Acesso em 09 de dezembro de 2020. 13 KLITGAARD, Robert. A Corrupção sob Controle. Rio de Janeiro: Jorge Zahar Editor, 1994. p. 83. 14 Op cit item 6 g p 13 KLITGAARD, Robert. A Corrupção sob Controle. Rio de Janeiro: Jorge Zahar Editor, 1994. p. 83. 14 Op. cit., item 6. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 11 GLYN, Patrick; KOBRIN, Stephen J.; NAIM, Moisés. Globalização da corrupção, a economia política da corrupção. Brasília: UNB, 2002. p. 74. 12 Diretriz do Estatuto em sua íntegra acessível por meio do site do Departamento de Justiça dos Estados Unidos. 2.2 Origem do fenômeno da autorregulação preventiva e a influência das normas supranacionais Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 35 Apesar destes primeiros diplomas regulamentares, nos Estados Unidos acontece o “Caso Watergate”, escândalo político que ocorreu em 1974, o qual ficou conhecido pela apuração de violação e espionagem no comitê do Partido Democrata, desmascarando a existência de um financiamento ilegal para a campanha do presidente Nixon, envolvendo um esquema de propinas pagas por empresas americanas a governos estrangeiros11, as autoridades americanas e empresas privadas. Diante do alvoroço político, as autoridades legislativas engrenaram no movimento para determinar limites à corrupção nas relações entre empresas nacionais e funcionários de governos estrangeiros com os quais eram mantidas relações empresariais. Isto posto, em 1977 ainda nos Estados Unidos, criou-se a Lei Contra Práticas Corruptas Internacionais (Foreign Corrupt Practices Act - FCPA)12 a qual tem por base estrutural a interferência preventiva e repressiva na fonte do pagamento, o denominado “cliente”, levando em conta o modelo proposto por Klitgaar13 o qual entende como figuras de atuação (i) o outorgante, (ii) o agente e (iii) o cliente, sendo o primeiro o agente oficial da Administração Pública (aquele que é o destinatário do suborno: o corrupto); o segundo o intermediário (que faz a ponte: lobista, advogado, despachante etc.); e o terceiro é o sujeito do setor privado que paga o suborno (corruptor). No mesmo momento, exatamente em 09 de dezembro de 1977, se registra na Europa a Convenção Relativa à Obrigação de Diligência dos Bancos no Marco da Associação dos Bancos Suíços, instituindo como base o sistema de autorregulação de conduta, vinculando as instituições cujo descumprimento resultaria na aplicação de sanções – como por exemplo, multas e outras penalidades.14 Em consonância, ainda durante da década de 1970, nos Estados Unidos advém o fenômeno da lavagem de dinheiro derivado do tráfico ilícito de drogas, resultando na elaboração da Convenção das Nações Unidas Contra o Tráfico Ilícito de Entorpecentes e Substâncias Psicotrópicas, conhecida como Convenção de Viena, realizada em dezembro de 1988 na Áustria, que juntamente com o Money Laudering Control Act, legislação elaborada em 1986 Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 17 No Brasil, no cumprimento das obrigações assumidas nessa Convenção, alterou o Código Penal, por meio da Lei nº. 10.476/01, criando tipos incriminadores e tráfico de influência em transação comercial internacional (arts. 337-B e 337-C). 15 Op. cit., item 9. 15 Op. cit., item 9. 16 CUNHA, Rogério Sanches. SOUZA, Renée do Ó. Lei Anticorrupção Empresarial. 3. ed., rev. atual. e ampl. Salvador: Ed. Juspodivm, 2020. p. 22. 17 No Brasil, no cumprimento das obrigações assumidas nessa Convenção, alterou o Código Penal, por meio da Lei nº. 10.476/01, criando tipos incriminadores e tráfico de influência em transação comercial internacional (arts. 337-B e 337-C). 15 Op. cit., item 9. 16 CUNHA, Rogério Sanches. SOUZA, Renée do Ó. Lei Anticorrupção Empresarial. 3. ed., rev. atual. e ampl. Salvador: Ed. Juspodivm, 2020. p. 22. 17 No Brasil, no cumprimento das obrigações assumidas nessa Convenção, alterou o Código Penal, por meio da Lei nº 10 476/01 criando tipos incriminadores e tráfico de influência em transação comercial internacional (arts 2.2 Origem do fenômeno da autorregulação preventiva e a influência das normas supranacionais 348/05 e promulgada pelo Decreto Presidencial nº. 5.687/06. Posteriormente, com a vinda da edição da Lei nº. 12.846 em 2013 (Lei Anticorrupção), o compliance foi novamente invocado, agora como fator capaz de influir positivamente na aplicação das sanções às pessoas jurídicas que tenham estabelecido mecanismos para tais procedimentos internos de conformidade e integridade (art. 7 ̊, inciso VIII), como auditorias, incentivos a denúncias de fatos ilícitos (hot line) e códigos de ética, cabendo ao Poder Executivo editar o regulamento próprio para avaliação de tais mecanismos (art. 7 ̊, parágrafo único). Pois bem, considerando que o compliance se constitui por meio de políticas, ações e procedimentos de controle aptos a fazer cumprir normas legais e regulamentares a uma determinada área de atividade econômica ou financeira, devemos explorar de quais formas estes programas funcionalizam as máquinas empresariais. Os métodos que viabilizaram a adesão e execução dos referidos programas foram motivados pelo gerenciamento de sistemas internos, tomando por base a ideia de ser mais fácil prevenir a corrupção partindo do controle no setor privado, do que modificar a malha legislativa de alguns países, motivando assim o estabelecimento de normas rígidas de cumprimento de certas obrigações por parte das empresas, as quais preconizam a análise cuidadosa de riscos operacionais, prevenção de ilícitos e outros atos que possam resvalar na responsabilidade de pessoas jurídicas, incluindo as pessoas pertencentes a cada estrutura empresarial que se submeteu à normativa de um programas de integridade. 2.2 Origem do fenômeno da autorregulação preventiva e a influência das normas supranacionais DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 36 nos Estados Unidos, a qual se tornou o principal instrumento originário e disciplinador das regras de controle dessa atividade15 Muitos outros atos foram deliberados até o final do século XX e início do século XXI, momento pelo qual finalmente são fixadas regulamentações preventivas no âmbito internacional, quais sejam: Convenção Interamericana contra a Corrupção (de Caracas de 1996), a Convenção sobre o Combate a Corrupção de Funcionários Públicos Estrangeiros em Transações Comerciais Internacionais (de Paris de 1997), a Convenção de Palermo Contra o Crime Organizado (2000/2004) e a Convenção de Mérida Contra a Corrupção (2003/2006), instrumentos nos quais se passou a prever uma série de regras de conformidade ou compliance. No Brasil o implemento se dá por meio do advento da Lei nº. 9.613/98 (Lei de Lavagem de Dinheiro), que traz o conceito prático de política de prevenção e controle de atos ilícitos envolvendo a organização criminosa, que dispõe sobre os crimes de “lavagem” ou “ocultação” de bens, direitos e valores provenientes direta ou indiretamente da prática de infrações penais. Face à intensificação da globalização na economia e nas relações sociais no século passado, implicações transacionais da corrupção e constatação de leis internas tradicionais quase sempre fracassaram como resposta às práticas corruptas que ocorreram no mundo, obrigando o tema corrupção a ser prioridade na agenda de discussão da comunidade internacional, demandando no desenvolvimento dos mecanismos de prevenção e punição dos atos lesivos à administração pública.16 Assim, o tema corrupção movimentou ferozmente o cenário nacional e internacional, sendo possível elencar como importantes diplomas no Brasil as três Convenções abaixo: a) Convenção sobre o Combate à Corrupção de Funcionários Públicos Estrangeiros em Transações Comerciais Internacionais da Organização para Cooperação e Desenvolvimento Econômico (OCDE), firmada pelo Brasil e, Paris no dia 17 de dezembro de 1997; e ratificada por meio do Decreto Legislativo nº. 125/00, e promulgada pelo Decreto nº. 3.678/00;17 b) Convenção Interamericana Contra a Corrupção da Organização dos Estados Americanos (OEA), aprovada por meio do Decreto Legislativo nº. 152/02 e promulgada pelo Decreto nº. 4.410/02; 37 c) Convenção das Nações Unidas contra a Corrupção da Organização da Nações Unidas (ONU), assinada no México e ratificada no Brasil pelo Decreto Legislativo nº. 348/05 e promulgada pelo Decreto Presidencial nº. 5.687/06. c) Convenção das Nações Unidas contra a Corrupção da Organização da Nações Unidas (ONU), assinada no México e ratificada no Brasil pelo Decreto Legislativo nº. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 Bruna Perasoli Trade 3.1 Direito penal clássico e a imputação delitiva 3.1 Direito penal clássico e a imputação delitiva p , , p 24 SOUZA, Artur de Brito Gueiros; JAPIASSÚ, Carlos Eduardo Adriano. Curso De Direito Penal: Parte Geral. ed. 2a. Rio de Janeiro: Editora Forense, 2015. p. 214. ROUSSEAU, Jean-Jacques. O contrato social. Trad. de Paulo Neves. Porto Alegre: L&PM, 2013. p. 60. 19 SANTIN, Miguel Ângelo. O Estado e o direito de punir. In: MARIN, Jeferson Dytz (Coord.). Jur processo II: reformas processuais, ordinarização e racionalismo. Curitiba: Juruá, 2009. p. 52. Parâmetros relacionados ao poder de punir do estado Após consolidados os principais referenciais dos programas de compliance, respeitável se faz compreender a esfera do direito penal, explorando um pouco da evolução do conceito dentro do direito penal e seu poder punitivo perante a sociedade, o qual por muitas décadas foi regido apenas pelas escolas clássicas e, atualmente, se percebe um movimento conjugado com o direito penal moderno. E é por meio desta união de conceitos que encontramos a coerência dentro das regências que entrelaçam o direito penal e o ambiente empresarial em suas respectivas dinâmicas de funcionamento. Partiremos pela elucidação de duas teorias que desenvolvem a ocorrência de condutas ilícitas e a atuação de seus respectivos agentes relacionadas aos conceitos clássicos: (i) a teoria da imputação objetiva, e (ii) a teoria do domínio do fato. 38 Em seguida, o debate versa sobre a concepção trazida pelo direito penal moderno no que tange à atribuição de responsabilidade individual e coletiva, em especial, dentro do ramo do direito empresarial para melhor compreender as diretrizes que norteiam o exercício dos programas de integridade nos âmbitos empresariais. 21 GRECO, Luís. A teoria da imputação objetiva: uma introdução. In: Funcionalismo e imputação objetiva no direito penal. ed. 3a. Rio de Janeiro: Renovar, 2002. p. 10. 22 Op cit item 21 p 15 20 GRECO, Luís. Um panorama da teoria da imputação objetiva. 4. ed. São Paulo: Editora Revista dos Tribunais, 2014, p. 122-129. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade cit., item 22, p. 15-16. UZA A d B i G i JAPIASSÚ C l Ed d Ad i C D Di it P l P t G Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI 10 55689/ j 2021 02 003 \| ISSN 2764 1899 A (moderna) teoria da imputação objetiva Em tempos atuais, sabe-se que ao Estado é rogada a posição de punir, assim como dizia o filósofo Rousseau que, quando firmado um contrato social, “todo malfeitor que ataca o direito social torna-se por seus crimes rebelde e traidor da pátria”18; já Santin que “todo o sistema montado pelo Estado, no sentido de punir seus componentes, deve observar a busca do equilíbrio social entre os mesmos”19. A temática é enfatizada pela tese desenvolvida por Roxin na década de 197020 chamada de “teoria moderna da imputação objetiva”, sendo atribuída a qualidade de “moderna” por ter sido antecedida por outras teses, as quais foram apresentadas na primeira metade do século XX, cuja finalidade também se firmava em delimitar critérios valorativos para determinar se um evento causal deve ser imputado ao agente.21 Como exemplos de teorias precedentes, desenvolvidas no início do século XX, com intuito de conceituar a imputação no âmbito do direito, destaca-se a tese formulada por Karl Larenz em 1927, denominada “A teoria da imputação de Hegel e o conceito de imputação objetiva”22, a qual tinha o critério de imputação desenvolvido por meio do conceito hegeliano23 (considerando o verdadeiro antecessor da teoria da imputação objetiva atual24) definido como o juízo pelo qual um determinado acontecimento pode ser descrito como uma ação, uma vez DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 39 que é compreendida não só como uma soma de causas e efeitos, mas um todo dirigido pela vontade.25 Em semelhante caminho, o polonês Richard Honig, em artigo publicado em 1930 transportou uma ideia diferenciada para o Direito Penal daquela trabalhada por Larenz. Assim, segundo o sustentado por Honig, para que determinado nexo causal seja relevante para o direito, faz-se necessário não apenas uma mera relação fortuita entre a ação e o resultado, mas uma relação jurídica especial26 , um nexo normativo, somente interessando ao direito penal as ações típicas derivadas da vontade humana.27 Logo, a ação humana consiste em uma exteriorização final de sua vontade, pela qual o ser humano intervém na natureza28. Portanto, o juízo de imputação depende da antecedente atuação teleológica da vontade humana29 sendo a possibilidade objetiva de pretender um curso causal danoso o critério de distinção entre os acontecimentos imprevisíveis daqueles decorrentes da direção objetiva da vontade30. 25 LARRAURI, Elena. Notas preliminares para una discusión sobre la imputación objetiva. In: BUSTOS/LARRAURI. La imputación objetiva. Santa Fé de Bogotá: Temis, 1998. p. 736. 26 Idem, p. 738. 30 ROXIN, Claus. Problemas basicos del derecho penal. Madrid: Editorial Reus S.A., 1976. p. 130-131. 31 A primeira menção de Claus Roxin a respeito da Teoria da Imputação Objetiva foi em um artigo de 1970, intitulado “Gedanken zur Problematik der Zurechnung im Strafrecht”, o qual fazia parte de um livro dedicado a HONIG, “Festschrift für Richard Honig”, publicado em Göttingen. A tradução foi feita por Ana Paula dos Santos Luís Natscheradet na obra “Problemas Fundamentais de Direito Penal”, publicada em Lisboa em 1993. 32 ROXIN, Claus. Estudos de direito penal. Trad. de Luís Greco. 2. ed. Rio de Janeiro: Renovar, 2008. p. 125- 126. 27 SANCINETTI, Marcelo A. Observaciones sobre la teoría de la imputación objetiva. In: Teorias Actuales en el derecho penal. Buenos Aires: Ad-Hoc, 1998, p. 186. 28 O it it 22 21 Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 A (moderna) teoria da imputação objetiva As formulações teóricas reconhecidamente preocupadas com o problema da imputação voltaram a tomar força na dogmática penal com a publicação do artigo “Reflexões sobre a problemática da imputação objetiva em direito penal” de Roxin, em 1970.31 A “moderna” teoria desenvolvida por Roxin (ou a evolução da anteriormente encontrada em Honig e Larenz) neste momento se ocupa, é verdade, de excluir os acontecimentos fortuitos do tipo (...). Mas os resultados que ocorrem por ocasião de uma diminuição do risco ou de um risco permitido, bem como aqueles que se encontram fora do fim de proteção da norma de cuidado ou fora do alcance do tipo, não são fortuitos, e ainda assim não são imputados.32 25 LARRAURI, Elena. Notas preliminares para una discusión sobre la imputación objetiva. In: BUSTOS/LARRAURI. La imputación objetiva. Santa Fé de Bogotá: Temis, 1998. p. 736. 26 Id 738 40 Neste ato, Roxin priorizou em seu estudo uma face da teoria geral da imputação, completamente desprovida do dogma causal33. A essência de sua construção teve como base a doutrina elaborada por Honig para o qual o centro da teoria da ação é transportado de uma esfera ontológica para a esfera normativa, na medida em que a questão jurídica fundamental deixa de ser a verificação a respeito de se determinada conduta pode ser qualificada como ação humana, estruturando tal teoria por meio de critérios normativos baseados no princípio do risco.34 Entrelaçando a concepção e Roxin com os tempos atuais, entendemos que apesar de nem todas as condutas serem passíveis de proteção por não estarem elencadas dentro do rol do direito penal, àquelas as quais se compreendem como nocivas à dinâmica social devem ser observadas e reprimidas de alguma forma. A (moderna) teoria da imputação objetiva Esta concepção é mais trabalhada por meio dos conceitos que permeiam o entendimento do direito penal econômico, que na visão de João Marcello de Araújo Junior é apresentado tanto a regular o comportamento daqueles que participam do mercado, quanto a proteger a estrutura e o funcionamento do próprio mercado, como também, a política econômica estatal, sob o manto garantista, sem preocupação de segurança.35 tanto a regular o comportamento daqueles que participam do mercado, quanto a proteger a estrutura e o funcionamento do próprio mercado, como também, a política econômica estatal, sob o manto garantista, sem preocupação de segurança.35 Portanto, primordial o protecionismo econômico, o qual apesar de não se ater a condutas específicas, foca no combate à atos nocivos à saúde empresarial, que detém a capacidade de desordenar a atuação tanto no âmbito público quanto privado. E, é por meio desta construção ideológica que se começa a engajar a necessidade da criação de programas de integridade, como ferramenta de prevenção aos crimes econômicos. Op. cit., item 38, p. 148. 35 ARAUJO JUNIOR, João Marcello. O direito penal econômico. Revista Brasileira de Ciências Criminais nº. 25, ano 07. São Paulo: Revista dos Tribunais, jan./mar. 1999. p. 150. 33 ROXIN, Claus. Reflexões sobre a problemática da imputação em direito penal. In: Problemas fundamentais de direito penal. ed. 2a. Lisboa: Vega, 1993. p. 145-168. 34 Op cit item 38 p 148 Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. B P li T d Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 36 WELZEL, Hans. Derecho Penal Aleman. 11ª ed. Trad. por Bustos Ramírez e Yáñez Pérez, Santiago: Ed. Jurídica de Chile, 1997, p. 3-5. In: GUEIROS, Arthur. Valores ético-sociais e os programas de compliance. Coordenador Acadêmico do Centro de Pesquisas em Crimes Empresariais e Compliance (CPJM) de 10/09/2020. Disponível em: <http://www.cpjm.uerj.br/texto-blog-valores-etico-sociais-e-os-programas-de-compliance/>. Acesso em 15 de janeiro de 2021. A relevância da teoria do domínio do fato Além dos critérios de imputação de ato ilícito, importante destacar a íntima ligação destas teses à doutrina de Hans Welzel quando combinada com os programas de compliance, na busca de definir a figura de autor e partícipe durante a prática delituosa. Conforme dita o pensamento penal do séc. XX, a missão primária do Direito Penal não é a proteção atual de bens jurídicos, pois quando entra efetivamente em ação, em geral chega Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 41 demasiadamente tarde. Em verdade, a missão mais profunda do Direito Penal é de natureza ético-social e de caráter positivo: ao proibir e punir a inobservância efetiva dos valores fundamentais da consciência jurídica, revela-se a disposição da vigência inquebrantável daqueles valores, dando forma ao juízo ético-social dos cidadãos e fortalecendo a consciência de permanente fidelidade jurídica36. Em 1939, Hans Welzel emana a ideia da teoria do domínio do fato, sendo a mesma aprimorada por Roxin em 1963, fundamentando-se na teoria restritiva da realização da conduta típica, assenta em princípios relacionados à conduta e não ao resultado. O ponto inicial de Roxin na elaboração de sua teoria foi a peculiar abstração e a impalpabilidade da concepção welzeliana, que o levaram a repulsar a ideia de domínio do fato finalista. Inclusive, Roxin foi categórico ao esclarecer que (i) Welzel introduziu o conceito “de forma absolutamente repentina e sem explicação, como se seu significado fosse compreensível por si mesmo”; e (ii) que a “unilateralidade dos critérios compreendidos de forma lógica e exata” e a “sua incapacidade de satisfazer as diversas formas de manifestação da vida em suas expressões individuais” não servem como critérios para definir a ideia de domínio do fato.37 Neste diapasão, esta teoria se consolida por meio de uma base teórica restritiva, que conceitua o autor e o partícipe no concurso de pessoas: determina que o autor é quem realiza a conduta típica expressa no verbo (atuação direta), sendo ele o executor material do fato; e o partícipe é quem participa de forma acessória (“auxiliar”) na realização do fato. Por este motivo, a figura do partícipe invariavelmente não se ajusta ao verbo do tipo e não tem poder de decisão sobre a execução ou consumação do crime, é o mero colaborador. j 37 ROXIN, Claus. Täterscha und Tatherrscha. 6. Aufl, Berlin: Walter de Gruyter, 1994, p. 112. In: ALFLEN, Pablo Rodrigo. Teoria do domínio do fato na doutrina e na jurisprudência brasileiras. Universitas JUS, v.25, n.2, 2014. p. 15-33. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 A relevância da teoria do domínio do fato Para tanto, a teoria restritiva subdivide-se em: (i) teoria objetivo-formal e (ii) teoria objetivo-material, para decifrar a comportamentos dos sujeitos, que de forma diversa, são considerados autores. Sendo assim, brevemente temos que: Para tanto, a teoria restritiva subdivide-se em: (i) teoria objetivo-formal e (ii) teoria objetivo-material, para decifrar a comportamentos dos sujeitos, que de forma diversa, são considerados autores. Sendo assim, brevemente temos que: (i). Na teoria objetivo-formal diferencia autoria da participação na hipótese em que o sujeito realizar uma ação ou omissão que não se enquadrar no verbo central, concorrendo para o crime mediante induzimento, instigação ou auxílio; DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 42 (ii). Já na teoria objetivo-material, prefere por distinguir o autor de participe pela maior contribuição do primeiro na causa do resultado; Por melhor que sejam estas classificações, estas teorias não resolvem problemas como o da autoria mediata – figura principal na caracterização do agente no descumprimento dos programas de compliance. Aqui, o autor mediato é o sujeito que utiliza de outrem para cometer o delito, prática muito comum nos casos em que se apresentam autores materiais e intelectuais de crimes, como os chefes de organizações criminosas que diretamente não praticam o verbo do tipo, mas detém o controle das operações tendo em vista suas posições administrativas dentro dos ambientes empresariais. Diante destas qualidades, o estudo da teoria Roxiniana do domínio do fato traz preciosa contribuição para a compreensão e sistematização das diversas formas de cometimento de delitos em que estejam envolvidas uma ou mais pessoas, delimitando, principalmente, o contexto do domínio da vontade por meio denominados “aparatos organizados de poder” (ou “domínio por organização”). Incialmente esta categoria foi desenvolvida como uma forma independente de autoria mediata, ao lado do domínio da vontade por erro e por coação, e posteriormente adicionada à peculiaridade do domínio por organização, onde o executor direto é considerado plenamente responsável. Considerando este desenvolvimento, Roxin demonstra que há três maneiras de dominar o fato típico, quais sejam: (i) domínio do fato pelo domínio da ação; (ii) domínio do fato pelo domínio da vontade; e (iii) domínio funcional do fato. Essas três maneiras de dominar o acontecer típico instituem as categorias da autoria direta/imediata, da autoria indireta/mediata e da coautoria, respectivamente. Contudo, para o estudo em questão, nos interessa enfatizar a forma de autoria mediata, pertinente ao comando da vontade em virtude do domínio de um aparato organizado de poder. 38 ROXIN, Claus. Problemas de autoría y participación em la criminalidad organizada. Revista Penal, n. 2, p. 61-65, jul. 1998. p. 63. A relevância da teoria do domínio do fato Apesar das inúmeras divergências doutrinarias sobre a autoria mediata, na tentativa de clarear a questão, aborda Roxin que a solução da coautoria que JAKOBS propugna, descansa em uma consideração mais normativa do domínio do fato. O entende como responsabilidade jurídica, não como domínio real. Para ele, a autoria mediata pressupõe que quem atua diretamente, haja de acordo com o Direito, a saber, que juridicamente não se o/a responsável, ou não completamente responsável. Pois se era completamente responsável, então, segundo este entendimento, não poderia ser um instrumento.38 38 ROXIN, Claus. Problemas de autoría y participación em la criminalidad organizada. Revista Penal, n. 2, p. 61-65, jul. 1998. p. 63. N, Claus. Problemas de autoría y participación em la criminalidad organizada. Revista Penal, n. 2, p. ul. 1998. p. 63. 43 Assim sendo, fica nítido que Roxin enfrenta a questão trazendo como solução a ideia de que na coautoria há uma tomada de decisão conjunta para o fato, ou seja, um acordo entre coautores que não possuem hierarquia entre si; ao passo que na autoria mediata o homem de trás serve-se de uma estrutura vertical hierarquizada, onde apenas a concretização da ordem fica a cargo do executor direto que é fungível e não pode evitar que o resultado aconteça, pois existem muitos outros executores a serviço da organização. Este acerto quanto ao posicionamento de Roxin parece-nos irrefutável, na medida em que dentro de um aparato organizado de poder, e os casos históricos em que a teoria foi aplicada, ficou comprovado que o “homem de trás” (ou seja, o manipulador da ordem) possui elevado controle sobre o resultado do acontecimento típico. Além disso, o fato de existirem inúmeros executores diretos a serviço do aparato faz com que o resultado típico independa da sua vontade. Trazendo para os dias atuais, podemos notar que esta teoria se encontra firmada pelo Código Penal Brasileiro, em seu artigo 62, inciso I, onde é agravada a pena de quem “promove, ou organiza a cooperação no crime ou dirige a atividade dos demais agentes”39, consolidando o conceito extensivo de autor, como corolário da teoria dos equivalentes causais. Todavia, no advento da reforma de 1984 que surgiram indagações sobre qual a real concepção adotada pelo nosso atual Código Penal. g g 40 SANTOS, Juarez Cirino dos. Direito Penal - Parte Geral. 4. ed. rev., ampl. Florianópolis: Conceito Editorial, 2010. p. 247. 39 Artigo 62, inciso I do Código Penal brasileiro (Decreto-Lei nº. 2.848/1940). Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. 39 Artigo 62, inciso I do Código Penal brasileiro (Decreto-Lei nº. 2.848/1940). A relevância da teoria do domínio do fato Por isso, é importante observar a ressalva trazida por Juarez Cirino dos Santos: A lei penal brasileira adota, a princípio, a teoria unitária de autor, mas a introdução legal de critérios de distinção entre autor e partícipe transforma, na prática judicial, o paradigma monístico da teoria unitária em paradigma diferenciador, admitindo o emprego de teorias modernas sobre autoria e participação, como, por exemplo, a teoria do domínio do fato, cujos postulados são inteiramente compatíveis com a disciplina legal de autoria e participação no Código Penal [...]40 Ao passo que parte da doutrina aponta que a reforma adotou a teoria restritiva de autor em face da diferenciação entre autor e participe, inclusive diante de institutos como a participação de menor importância (vide art. 29, §1º, do CP), outra parte assegura que o legislador passou a adotar a teoria restritiva complementada pela teoria do domínio do fato, haja Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 44 vista a aceitação da teoria finalista da ação, conforme se observa pela redação do artigo 29, caput, §§1º e 2º, e no referido artigo 62, I, ambos do Código Penal. À vista disso, compreende-se que a teoria do domínio do fato se aplica face à necessidade de alcançar a figura do agente mediato na prática delituosa, que, embora não tenha cometido o ato propriamente, manipula terceiro, com o propósito de que este pratique a ação típica. Nesse sentido, afirma Fernando Capez que “o executor atua sem vontade ou consciência, considerando-se, por essa razão, que a conduta principal foi realizada pelo autor mediato”.41 Contudo, graças ao grau de complexidade dos casos concretos, surge a dificuldade de se identificar com clareza o controle completo do agente ante ao fato típico, diminuindo assim a incidência direta da teoria do domínio do fato. Neste viés, rebatem André Estefam e Victor Eduardo Rios Gonçalves afirmando: Por outro lado, a teoria do domínio do fato não pode ser aceita em sua integralidade porque não é possível identificar com clareza, em grande número dos casos, quando uma pessoa tem ou não o controle completo da situação. Quando o mandante, por exemplo, contrata uma pessoa para matar a vítima, o executor contratado pode fugir com o dinheiro, ser preso antes de cometer o crime, ou, por outro lado, cometer delito mais grave do que o combinado. 41 CAPEZ, Fernando. Curso de direito penal: parte geral. 17. ed. vol 1. São Paulo: Saraiva, 2011, p. 370. 42 ESTEFAM, André. GONÇALVES, Victor Eduardo Rios. Direito penal esquematizado: Parte Geral. 3. ed. São Paulo: Saraiva, 2014. p. 444. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 CAPEZ, Fernando. Curso de direito penal: parte geral. 17. ed. vol 1. São Paulo: Saraiva, 2011, p. 370. p p g p ESTEFAM, André. GONÇALVES, Victor Eduardo Rios. Direito penal esquematizado: Parte Geral. 3. ed Paulo: Saraiva, 2014. p. 444. 3.2 Movimentos teóricos do direito penal moderno Apesar dos ditames das teorias clássicas, dentro da teoria moderna se questiona o poder punitivo exercido pelo Estado de modo geral, considerando estarmos numa esfera onde a autonomia das organizações sociais deva prevalecer. Elucidando pelas palavras de Heleno Cláudio Fragoso, “uma política criminal moderna orienta-se no sentido da descriminalização e da desjudicialização, ou seja, no sentido de contrair ao máximo o sistema punitivo do Estado, dele retirando todas as condutas antissociais que podem ser reprimidas e controladas sem o emprego de sanções criminais.”43 Perfazendo esta onda ideológica, surgem propostas sobre os novos ditames de política criminal que deve ser seguida, com intuito de extinguir certas condutas ilícitas. Nesse sentido, doutrinadores criminalistas recorrem ao desenvolvimento de teorias que servem de elementais na difusão de resultados positivos dentro do sistema jurídico penal, sendo algumas destas: i) abolicionismo; (ii) minimalismo; e (iii) garantista, trabalhadas a seguir. A relevância da teoria do domínio do fato Em nenhum desses casos, o mandante tinha pleno controle da situação.42 Portanto, vale observar que é preciso entender que não se pode exigir uma aplicação automática da categoria do domínio da organização como forma independente de autoria mediata, sem a comprovação segura da existência, no caso concreto, dos seus pressupostos fundamentais. Por este motivo, quando se trata de delito ocorrido dentro do âmbito empresarial, envolvendo uma série de agentes, sendo cada um destes ocupante de uma posição abarcada de um volume específico de poder, se faz necessário verificar o fluxo de ordens, desde sua origem, para então conseguir lastrear os comandos e seus respectivos ordenadores – ou, pela lógica de Roxin, aquele que se encontra no domínio do fato como autor mediato da ação. 45 43 FRAGOSO, Heleno Cláudio. Lições de direito penal: a nova parte geral. 2. Ed. Rio de Janeiro: Forense, 1991. p. 17. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Abolicionismo O surgimento da ideia de abolicionismo do Direito Penal acontece como rebote ao período do final da Segunda Guerra Mundial, onde se buscava um direito mais humanistas. No entanto, o movimento se consolidou nas décadas de 1960-70 em razão das teorias sociológicas que se floresciam na época. Assim, o nome deste instituto se efetivou na segunda metade do século XX na Escandinávia, abarcado anteriormente nos movimentos que pediram o fim da escravidão e da pena de morte em outras épocas, sendo seu objetivo buscar soluções para a violência desmedida, sem, contudo, utilizar do mesmo meio como forma de coerção, desviando-se das formas de correção por meio de disciplina por meio da valorização do indivíduo. Além disso, se refere tanto a um movimento social quanto a um conjunto de teorias na academia, que nega a legitimidade do sistema penal contemporâneo, e se contrapõe às formas de castigo usadas hoje pela justiça criminal na resolução de conflitos, especialmente as prisões, abrindo janelas à criação de microrganismos sociais baseados na solidariedade e fraternidade, 46 buscando assim à reapropriação social dos conflitos entre agressores e ofendidos e a criação espontânea de métodos ou formas de composição.44 buscando assim à reapropriação social dos conflitos entre agressores e ofendidos e a criação espontânea de métodos ou formas de composição.44 Tomando pela ótica teórica, encontramos três diferentes tipos de abolicionismos, consubstanciados conforme a concepção metodológica de seus protagonistas, sendo estas: a) fenomenológica de Louk HULSMAN; a) fenomenológica de Louk HULSMAN; b) materialista com fundo marxista de Thomas MATHIESEN; c) estrutural de Michael FOUCAULT; Resumidamente, podemos dizer que enquanto Hulsman coloca a abolição de todo o sistema de justiça penal, Mathiesen tenta abolir os absorventes sistemas sociais repressivos da última etapa do capitalismo de Estado, e Mathiesen tenta chegar à “(...) transcendência da estrutura repressiva de nossa sociedade, na última instância do modelo básico de produção dessa sociedade.”, sendo que seu abolicionismo tem (ao menos em princípio) uma maior extensão que o abolicionismo de Hulsman. Nos fatos, porém, seu abolicionismo baseia-se na ideia de que a política penal norueguesa “envolve em grande medida um tratamento irracional e injusto dos grupos marginais da sociedade — tratamento que deveria ser abolido em uma medida considerável. 44 OLIVEIRA, Mara Elisa de. Breve análise sobre o abolicionismo e o minimalismo. Artigo disponível em: <https://jus.com.br/artigos/22596/breve-analise-sobre-o-abolicionismo-e-o-minimalismo>. Acesso em 18 de janeiro de 2021. 45 Thomas Mathiesen. The Politics of Abolition. Oslo, Robertson, 1974, p. 3. In: FOLTER, Rolf S. de. Sobre a fundamentação metodológica do enfoque abolicionista do sistema de justiça penal — uma comparação das idéias de Hulsman, Mathiesen e Foucault. Trad. do espanhol por Natalia Montebello. Disponível em: <https://cpisp.org.br/wp-content/uploads/2018/05/5140-12199-1-SM.pdf>. Acesso em 18 de janeiro de 2021. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 44 OLIVEIRA, Mara Elisa de. Breve análise sobre o abolicionismo e o minimalismo. Artigo disponível em: <https://jus.com.br/artigos/22596/breve-analise-sobre-o-abolicionismo-e-o-minimalismo>. Acesso em 18 de janeiro de 2021. janeiro de 2021. 45 Thomas Mathiesen. The Politics of Abolition. Oslo, Robertson, 1974, p. 3. In: FOLTER, Rolf S. de. Sobre a fundamentação metodológica do enfoque abolicionista do sistema de justiça penal — uma comparação das idéias de Hulsman, Mathiesen e Foucault. Trad. do espanhol por Natalia Montebello. Disponível em: <https://cpisp.org.br/wp-content/uploads/2018/05/5140-12199-1-SM.pdf>. Acesso em 18 de janeiro de 2021. 44 OLIVEIRA, Mara Elisa de. Breve análise sobre o abolicionismo e o minimalismo. Artigo disponível em: <https://jus.com.br/artigos/22596/breve-analise-sobre-o-abolicionismo-e-o-minimalismo>. Acesso em 18 de janeiro de 2021. 45 Thomas Mathiesen. The Politics of Abolition. Oslo, Robertson, 1974, p. 3. In: FOLTER, Rolf S. de. Sobre a fundamentação metodológica do enfoque abolicionista do sistema de justiça penal uma comparação das idéias Seguindo o mesmo entendimento, discorre Bittencourt que A abolição da prisão supõe o desenvolvimento de formas alternativas de autogestão da sociedade no campo de controle da delinquência. Tais formas auto gestionárias de controle da delinquência exigiriam a colaboração das entidades locais e das associações obreiras, a fim de evitar o isolamento social que sofre o infrator quando é recolhido a uma instituição penitenciária.47 Em apertada síntese, os adeptos à visão abolicionista enxergam um direito penal deslegitimado, a julgar por ser um sistema que não pune de forma igualitária todas os ilícitos, levando à subsunção de punição quase sempre aos mais desfavorecidos, atestando assim, um sistema de elevada desvantagem. Consequentemente, pleiteia a eliminação total desse modelo de controle estatal. Abolicionismo E Foucault, em sua percepção geral consoante aos pensamentos de Nietzsche e Bataille, entende que se deve abolir todos esses limites que fixam relações assimétricas, oposicionais, como as relações entre inocência e culpabilidade, razão e loucura, o bom e o mau, o normal e o patológico pois percebia o problema das prisões como “local e marginal”.45 Na América Latina, destacamos Eugenio Raúl Zaffaroni como defensor do realismo criminológico, sustentando que como em todos os tipos de teorias, existem os radicais e eles seriam a corrente abolicionista. Logo, pela visão de Zaffaroni o abolicionismo nega a legitimidade do sistema penal tal como atua na realidade social contemporânea e, como princípio geral, nega a legitimação de qualquer outro sistema penal que se possa imaginar no futuro como alternativa a modelos formais e abstratos 47 de solução de conflitos, postulando a abolição radical dos sistemas penais e a solução dos conflitos por instâncias ou mecanismos informais.46 46 ZAFFARONI, Eugênio Raúl. Em busca das Penas Perdidas: a perda da legitimidade do sistema penal. 5ª Ed Ri d J i R 1991 2014 89 , , p 47 BITTENCOURT, Cezar Roberto. Tratado de Direito Penal: Parte Geral, 1. São Paulo: Saraiva, 2013. p. 593. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade 46 ZAFFARONI, Eugênio Raúl. Em busca das Penas Perdidas: a perda da legitimidade do sistema penal. 5ª Ed. Rio de Janeiro: Revan, 1991, 2014. p. 89. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 ZAFFARONI, Eugênio Raúl. Em busca das Penas Perdidas: a perda da legitimidade do sistema Ed. Rio de Janeiro: Revan, 1991, 2014. p. 89. 46 ZAFFARONI, Eugênio Raúl. Em busca das Penas Perdidas: a perda da legitimidade do sistema penal. 5ª Ed. Rio de Janeiro: Revan, 1991, 2014. p. 89. 47 BITTENCOURT, Cezar Roberto. Tratado de Direito Penal: Parte Geral, 1. São Paulo: Saraiva, 2013. p. 593. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade 46 ZAFFARONI, Eugênio Raúl. Em busca das Penas Perdidas: a perda da legitimidade do sistema penal. 5ª Ed. Rio de Janeiro: Revan, 1991, 2014. p. 89. 47 BITTENCOURT, Cezar Roberto. Tratado de Direito Penal: Parte Geral, 1. São Paulo: Saraiva, 2013. p. 593. Minimalismo Por mais falho que possa ser o sistema punitivo, uma sociedade sem uma estrutura mínima punitiva certamente recairia no abismo da ordenação social. Sendo assim, temos como vertente os surgimento do minimalismo, o qual se dedica a criticar o sistema penal afirmando a necessidade do direito penal reduzir sua incidência a um mínimo necessário, sendo sua aplicação estabelecida apenas quando essencial sobre a conduta danosa que deseja cercear. Dentre os argumentos elaborados, verifica-se um consenso de que apenas bens de elevada valia devam ser tutelados pelo Direito penal, tendo em vista que a utilização de recurso tão danoso à liberdade individual somente se justifica em face do grau de importância que o bem tutelado assume. Neste ponto, surge a preocupação com a dignidade do bem jurídico, dado que o Direito penal só deve atuar na defesa dos bens jurídicos imprescindíveis à coexistência pacífica dos homens. Ademais, não bastaria apenas a comprovação do grau de importância do bem jurídico a ser tutelado pelo direito penal, mas sim, verificar qual o tamanho do impacto social e suas consequências, e somente depois disto, determinar quando à aplicação a intervenção estatal. 48 Estes pensamentos foram impulsionados por propostas elaboradas principalmente na década passada, em especial pelos pensadores Luigi Ferrajoli e Alessandro Baratta, ambos italianos, sendo que o primeiro tem sua base liberal iluminista e o segundo de base interacionista-materialista. E palavras de Ferrajoli, a justiça penal, com o caráter inevitavelmente desonroso de suas intervenções, não pode ser incomodada e, sobretudo, não pode incomodar os cidadãos por fatos de escasso relevo, como o são a maior parte dos atualmente castigados com simples multas.48 a justiça penal, com o caráter inevitavelmente desonroso de suas intervenções, não pode ser incomodada e, sobretudo, não pode incomodar os cidadãos por fatos de escasso relevo, como o são a maior parte dos atualmente castigados com simples multas.48 Considerando as alegações minimalistas a partir da década 90 do século XX no Brasil, em especial com o advento da reforma penal em 1984, os diplomas advindos nesta época causaram a inserção das penas alternativas por meio das Leis nº 7.209 e 7.210, ambas de 1984, e, atualmente, a Lei das penas alternativas Lei nº 9.714, de 1998, sendo mais notória a implantação dos juizados especiais criminais estaduais, que buscou tratar de crimes de menor potencial, pela Lei nº 9.099, de 1995. Minimalismo Logo, à luz dos simpatizantes desta teoria, estas reformas proveram, de certa forma, o controle da sociedade sobre os fatos que ocorrem atualmente. Contudo, seria abusivo dizer que o sistema penal seja minimalista, tendo em vista a ampla quantidade de crimes e de condutas que ainda são dosadas pela legislação vigente no País. 48 FERRAJOLI, Luigi. Direito e razão: teoria do garantismo penal. 2. Ed. São Paulo: Revista dos Tribunais, 2006. p. 383-384. 49 Op. cit., item 49. Garantismo Por fim, este último aspecto teórico surgiu pela primeira vez por meio de Luigi Ferrajoli em sua obra “Direito e Razão”, considerada por muitos o princípio e o mais importante de todos os livros referentes a esta teoria. Segundo sua obra, Ferrajoli afirma que o garantismo penal deve que procurar estabelecer um conjunto de conceitos, princípios e normas capazes de fundamentar a legitimação do poder punitivo do Estado buscando garantir uma posição de relevo para o indivíduo.49 Garantismo, com efeito, significa [..] precisamente a tutela daqueles valores ou direitos fundamentais, cuja satisfação, mesmo contra os interesses da maioria, constitui o objetivo justificante do direito penal, vale dizer, a imunidade dos cidadãos contra arbitrariedade das proibições e das punições, a defesa dos fracos mediante Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 49 regras do jogo iguais para todos, a dignidade da pessoa do imputado, e, consequentemente, a garantia de sua liberdade, inclusive por meio do respeito à sua verdade. É precisamente a garantia desses direitos fundamentais que torna aceitável por todos, inclusive pela minoria formada pelos réus e pelos imputados, o direito penal50 Enquanto as demais teorias fixavam suas vertentes na tentativa de retirar do Estado o poder de punir, sendo por meio da redução do mesmo ou até sua abolição, o garantismo prega por uma necessidade de fortalecimento do Estado de Direito, onde não somente a lei se incumbirá a determinar os limites do poder punitivo estatal, como determina que o próprio legislador deverá seguir os limites predispostos na Constituição Federal da respectiva sociedade a qual irá regulamentar. Considerando este conceito, Ferrajoli definiu em sua teoria três diferentes significados, sendo estes: (i) o modelo normativo de direito; (ii) a teoria jurídica, separando a vigência da validade das normas, determinando categorias distintas; e (iii) a filosofia política, onde o exige a justificação moderna do Direito e do Estado. Para ilustrar tais argumentos, temos o que segue: “Segundo um primeiro significado, ‘garantismo’ designa um modelo normativo de direito: precisamente, no que diz respeito ao direito penal, o modelo de “estrita legalidade”, próprio do Estado de direito, (…) se caracteriza como uma técnica de tutela idônea (…) em garantia dos direitos. Em um segundo significado, ‘garantismo’ designa uma teoria jurídica da “validade” e da “efetividade” como categorias distintas não só entre si mas, também, pela “existência” ou “vigor” das normas. p p 51 Op. Cit., item 49, p. 785-787. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. 50 Op. cit., item 49, p. 271. Bruna Perasoli Trade Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 52 SHECAIRA, Sérgio Salomão. Criminologia. São Paulo: RT. 2014. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 Garantismo (…) exprime uma aproximação teórica que mantém separados o “ser” e o “dever ser” no direito; (…) Num terceiro significado, por fim, ‘garantismo’ designa uma filosofia política que requer do direito e do Estado o ônus da justificação externa com base nos bens e nos interesses dos quais a tutela ou a garantia constituem a finalidade.”51 Em suma, o garantismo penal tem por finalidade garantir ao indivíduo uma maior segurança nos procedimentos adotados pelo sistema penal, pregando o método mais justo possível, tomando por base os princípios fundamentais e processuais a fim de que o Estado atenda o que for necessário em defesa da população, seja cerceando o indivíduo que comete a conduta ilícita quanto resguardando àquele que foi lesado. 50 Mecanismos de imputação de delitos cometidos no âmbito empresarial 53 PAGOTTO, Leopoldo; NAKAHARA, Eric Felipe Sabadini. O programa de compliance como mecanismo de prevenção de responsabilidade penal no ambiente corporativo. In: SOUZA, Luciano Anderson de. (coord.). Compliance no Direito Penal. vol. 5. São Paulo: Thomson Reuters RT, 2021. p. 235. Mecanismos de imputação de delitos cometidos no âmbito empresarial Face aos benefícios trazidos pelos movimentos do direito penal moderno, estes incentivaram a mudança na forma de ser dos tipos penais, os quais se adaptaram à contemporaneidade, culminando na exigência de um aparato estatal de maior intervenção para a proteção dos bens jurídicos em geral. Por exemplo: enquanto nas décadas de 1950-80 o Direito Penal era consolidado pelos crimes de sangue e nas camadas menos favorecidas de cada sociedade, o Direito Penal Moderno contemplou condutas que envolvem a sociedade como um todo, em atenção àquelas praticadas em meio de estruturas complexas – ambientes chamados de “in company”. Inclusive, a temática foi abordada primitivamente por Edwin Sutherland52, por meio de seu estudo após a quebra da Bolsa de Nova Iorque em 1929, onde enfatizou seu estudo nos crimes empresariais (chamados de “white-collar crimes”). No Brasil, movimentos políticos contribuíram para a flexibilização dos princípios penais e das garantias fundamentais, que haviam sido conquistadas através da Revolução Francesa no século XVIII, permitindo que diversas novas condutas ilícitas emergissem, em especial no âmbito empresarial, como por exemplo os crimes “de colarinho branco”. Por conseguinte, a atenção foi voltada para a busca de meios alternativos de repressão à estas condutas delitivas, mas que não necessariamente impactassem de modo brusco a liberdade das atividades empresariais. E é por meio deste raciocínio que se compreende a importância e adequação dos programas de criminal compliance, como forma de autorregulação, a qual tem por finalidade conter delitos econômicos, através da fiscalização e gestão de riscos e a prevenção de ilícitos em geral. Por isso, para que esta máquina de gestão exerça sua eficácia, se faz necessária a concentração do controle da empresa, mediante a constituição de uma posição de garante, que no caso é a figura do compliance officer. E é pelo intermédio da figura do garantidor na execução dos atos empresariais que são levantadas diversas discussões acerca da responsabilização dentro do ambiente empresarial. 51 Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 Compliance no Direito Penal. vol. 5. São Paulo: Thomson Reuters RT, 2021. p. 235. 54 SILVEIRA, Renato de Mello Jorge; SAAD-DINIZ, Eduardo. Compliance, Direito penal e a lei anticorrupção. 1. ed. São Paulo: Saraiva, 2015. p.258. , p IRA, Renato de Mello Jorge; SAAD-DINIZ, Eduardo. Compliance, Direito penal e a lei anticorrupção. ão Paulo: Saraiva, 2015. p.258. 4.1 Impactos pelo descumprimento dos programas de compliance e o "criminal compliance" 4.1 Impactos pelo descumprimento dos programas de compliance e o "criminal compliance" Resumidamente, programas de compliance podem ser compreendidos como instrumentos normativos que utilizam procedimentos empresariais internos para a gestão e prevenção de riscos.53 Nesse sentido, entendemos que os programas de compliance defendem valores que vão muito além dos elencados no rol de bens jurídicos contidos no Código Penal vigente. Neste ramo, as medidas se destinam desde a evitar delitos até serem responsáveis pela prevenção de tantos outros ilícitos de ordem ética e normativas empresariais. Por isso a abrangente capacidade de implementar mecanismos de controle internos para cada cenário societário, em conformidade com o tipo específico de negócio ao qual se pretende proteger, sempre buscando estar de acordo com a legislação vigente. Destarte, pelas palavras de Renato de Mello Jorge Silveira, o criminal compliance se reconhece no mundo jurídico como o comportamento decisório na Economia é marcado pela ideia de oportunidade, a qual advém do juízo de cálculo de probabilidade sobre o potencial e a disposição de recursos em níveis mais elevados de produtividade. É nesse contexto de agitações do sistema econômico que se inserem as teses de compliance penal, na tentativa de se alcançarem, desde as representações um tanto mais sofisticadas da responsabilidade penal, a gestão e o controle deste novo cenário de risco, já de alguma forma vinculada às estruturas de governança.54 o comportamento decisório na Economia é marcado pela ideia de oportunidade, a qual advém do juízo de cálculo de probabilidade sobre o potencial e a disposição de recursos em níveis mais elevados de produtividade. É nesse contexto de agitações do sistema econômico que se inserem as teses de compliance penal, na tentativa de se alcançarem, desde as representações um tanto mais sofisticadas da responsabilidade penal, a gestão e o controle deste novo cenário de risco, já de alguma forma vinculada às estruturas de governança.54 Neste ponto vale esclarecer um detalhe importante: apesar da similaridade do criminal compliance com o Direito Penal Moderno/Econômico, estes indicadores atuam de modo diverso, uma vez que o primeiro atua na antecipação da responsabilidade penal em razão da sua característica preventiva; e o segundo – que à despeito de ter como viés o reconhecimento de condutas que antes eram descriminalizadas e agora são alvo de cerceamento – atua nos moldes clássicos, pois incide sobre condutas que já afetam algum bem jurídico tutelado pelo ordenamento jurídico vigente. 55 JOBIM, Rosana Kim. Compliance e Trabalho: entre o poder diretivo do empregador e os direitos inespecíficos do empregado. Florianópolis: Tirant Lo Blanch, 2018. p. 27. p g p p 56 MAEDA, Bruno Carneiro; AYRES, Carlos Henrique da Silva (Coord.). Temas de anticorrupção e Compliance. Rio de Janeiro: Elsevier, 2013. p. 167-201. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 55 JOBIM, Rosana Kim. Compliance e Trabalho: entre o poder diretivo do empregador e os direitos inespecíficos do empregado. Florianópolis: Tirant Lo Blanch, 2018. p. 27. 56 MAEDA, Bruno Carneiro; AYRES, Carlos Henrique da Silva (Coord.). Temas de anticorrupção e Compliance. Rio de Janeiro: Elsevier, 2013. p. 167-201. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. 55 JOBIM, Rosana Kim. Compliance e Trabalho: entre o poder diretivo do empregador e os direitos ines 55 JOBIM, Rosana Kim. Compliance e Trabalho: entre o poder diretivo do empregador e os direitos ines do empregado. Florianópolis: Tirant Lo Blanch, 2018. p. 27. 4.1 Impactos pelo descumprimento dos programas de compliance e o "criminal compliance" 52 No cenário brasileiro, onde o compliance se firmou com o advento da Lei de Lavagem de Dinheiro – Lei nº. 9.613/98, que apesar de na época de sua entrada em vigor parecer não ter aplicabilidade prática, conquistou sua viabilidade por meio das alterações promovidas com a chegada da Lei nº. 12.683/2012, dispondo explicitamente sobre os crimes de “lavagem” ou ocultação de bens, direitos e valores, sob o pretexto de tornar mais eficiente a persecução penal dos crimes de “lavagem de dinheiro”. E aproveitando o embalo desta onda de medidas preventivas, logo no ano seguinte temos a promulgação da Lei Anticorrupção – Lei nº. 12.846/13, a qual deu forças ao criminal compliance, uma vez que prevê a redução das sanções para as empresas que cooperam com as autoridades na apuração das infrações e que estabelecem procedimentos internos de auditoria e fiscalização, viabilizando a denúncia de irregularidades. Esta Lei determina que a responsabilidade de pessoas jurídicas ocorre na forma objetiva, ou seja, não há discussão de culpa. Por isso, ainda que os atos de corrupção tenham sido cometidos exclusivamente por empregado ou preposto, sem anuência ou participação direta da empresa, mas em seu benefício, a pessoa jurídica será responsabilizada. Como reforço da perspectiva preventiva da chegada destas Lei, em especial da Lei Anticorrupção, Rosana Jobim afirma que Há o entendimento que a Lei Anticorrupção contribuiu para o fortalecimento da implementação de controles internos e de programas de compliance, já que prevê a responsabilização objetiva das pessoas jurídicas envolvidas, incentivando uma atuação empresarial preventiva, ética, e combativa, a qual reforça a confiança dos investidores no âmbito nacional e internacional, trazendo benefícios a toda sociedade brasileira.55 Sendo assim, sabemos que o descumprimento da legislação nacional e internacional pelas empresas pode trazer efeitos prejudiciais à reputação da empresa, especialmente quando a conduta violar padrões socialmente aceitos. 4.1 Impactos pelo descumprimento dos programas de compliance e o "criminal compliance" Nesse sentido, opina Bruno Maeda que Além de incentivar condutas socialmente desejáveis, o tratamento diferenciado para empresas que investem em medidas de prevenção e de promoção de integridade corporativa serve para minimizar desvantagens competitivas e reduzir distorções de mercado que beneficiaram aquelas que nada fazem para evitar práticas ilícitas.56 DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 53 Isto posto, com base na influência das mudanças de ordem jurídica previamente descritas, inferimos que temas relativos à moral e à ética estão em grande destaque, tanto no cenário internacional como no Brasil, principalmente em razão do aumento nos números de casos de corrupção que vieram ao conhecimento público nos últimos tempos. E, por este motivo, os programas de prevenção se adequam como uma importante ferramenta para mitigar os riscos empresariais e para garantir o atendimento às normas legais e ao comportamento ético, tendo como resultado da dinâmica da execução destes programas um trinômio: prevenir, detectar e responder.57 4.2 Responsabilidade dos dirigentes na qualidade de garante ("compliance officer") 4.2 Responsabilidade dos dirigentes na qualidade de garante ("compliance officer") Em regra, todo programa de integridade inicia sua ação pela avaliação de riscos, tomando por base o cenário interno da empresa a qual se destina e o mercado ao qual se vincula. Entretanto, para que este se efetive, necessária se faz a adesão total deste regramento pela alta direção do referido segmento empresarial, com intuito de atribuir credibilidade e pautar a tomada de decisões. Considerando que estes programas têm por ponto principal o combate à corrupção e outros atos antijurídicos, se percebe que cada vez mais o criminal compliance vêm sendo utilizado e até entendido como requisito essencial para atuação dentro do mercado financeiro no combate de delitos econômicos. Desta forma, a fiscalização e a gestão de riscos para prevenção de ilícitos em geral têm seu papel crucial. E, no intuito de viabilizar esta fiscalização, se faz necessária a concentração de controle por meio de um representante que detém o poder na tomada de decisões da empresa (ou seja, figura na qualidade de “garante”), chamado de compliance officer, o qual fica responsável pela coordenação de ordem objetiva e/ou subjetiva na execução dos programas. Ressalta-se que a referida expressão veio à tona pela primeira vez na legislação brasileira pelo Decreto nº. 8.420/2015, em seu artigo 42, inciso IX, que elenca a figura do responsável pela aplicação do programa de integridade. 57 Op. cit., item 6, p. 519. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. li d Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 Bruna Perasoli Trade Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 60 ESTELLITA, Heloisa. Responsabilidade penal dos dirigentes de empresas por omissão: estudo sobre a responsabilidade omissiva impropria de dirigentes de sociedades anônimas, limitadas e encarregados pelo cumprimento por crimes praticados por membros da empresa. São Paulo: Marcial Pons, 2017. p.41. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 BETTE, Eduardo Luiz Santos; NAHUR, Marcius Tadeu Maciel. Criminal Compliance e ética empresaria os desafios do direito penal econômico. Porto Alegre: Núria Fabris, 2013. p. 22. In: Op. cit., item 6. NEDETTI C l R h l C i i l C li Sã P l Q ti L ti 2014 59 os desafios do direito penal econômico. Porto Alegre: Núria Fabris, 2013. p. 22. In: Op. cit., item 6. NEDETTI, Carla Rahal. Criminal Compliance. São Paulo: Quartier Latin, 2014. p. 59. 58 CABETTE, Eduardo Luiz Santos; NAHUR, Marcius Tadeu Maciel. Criminal Compliance e ética empresarial: novos desafios do direito penal econômico. Porto Alegre: Núria Fabris, 2013. p. 22. In: Op. cit., item 6. 58 CABETTE, Eduardo Luiz Santos; NAHUR, Marcius Tadeu Maciel. Criminal Compliance e ética empresarial: novos desafios do direito penal econômico. Porto Alegre: Núria Fabris, 2013. p. 22. In: Op. cit., item 6. 59 BENEDETTI, Carla Rahal. Criminal Compliance. São Paulo: Quartier Latin, 2014. p. 59. 60 ESTELLITA, Heloisa. Responsabilidade penal dos dirigentes de empresas por omissão: estudo sobre a responsabilidade omissiva impropria de dirigentes de sociedades anônimas, limitadas e encarregados pelo cumprimento por crimes praticados por membros da empresa. São Paulo: Marcial Pons, 2017. p.41. 58 CABETTE, Eduardo Luiz Santos; NAHUR, Marcius Tadeu Maciel. Criminal Compliance e ética emp novos desafios do direito penal econômico. Porto Alegre: Núria Fabris, 2013. p. 22. In: Op. cit., item BETTE, Eduardo Luiz Santos; NAHUR, Marcius Tadeu Maciel. Criminal Compliance e ética empresaria Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. 4.1 Impactos pelo descumprimento dos programas de compliance e o "criminal compliance" Em complemento, vale enfatizar que são pessoas dotadas de conhecimento técnico de gestão, com capacidade para avaliar os riscos internos e externos da empresa, com o fulcro de prevenir e/ou minimizar os riscos de responsabilidade legal no âmbito judicial. Portanto, trata- 54 se de um cargo de alta confiança, de natureza pública a qual resguarda o interesse da coletividade, cuja finalidade é o desempenho com probidade da atividade empresarial. Diante disso, podemos classificar esta prática em duas espécies: (i) do interesse da empresa em si, quando realiza fiscalização interna como um auditor, na busca de verificar se há ou não práticas ilícitas dentro daquele ambiente; e (ii) do equilíbrio social da empresa com o setor público, onde o compliance officer tenta evitar qualquer tipo de infração.58 Como podemos notar, há uma vasta lista de atividades que dependem da figura do compliance officer, o que pode culminar em sua responsabilização penal, caso atue de forma precária quanto à prevenção criminal da empresa. Por isso, há discussão de que tal função de redução de risco pressupõe um comportamento que acaba por antecipar sua própria responsabilidade de suas atribuições. Dentro do âmbito nacional, a figura do compliance officer sempre existiu, sendo que ocorreu uma transferência de posição de garantia, que antigamente era refletida apenas sob o empresário de alto cargo empresarial e hoje recai sob àquele que está na posição no ato da ocorrência da conduta ilícita. Com efeito, os questionamentos acerca da responsabilidade criminal em relação ao compliance officer surgem a partir da adoção da teoria formal do dever jurídico, prevista no artigo 13 §2º do Código Penal, concomitantemente com a edição da Lei de Lavagem de Dinheiro (Lei nº. 9.613/98, posteriormente alterada pela Lei nº. 12.638/2012) e pela Lei de Defesa da Concorrência (Lei nº. 12.529/2011). 4.1 Impactos pelo descumprimento dos programas de compliance e o "criminal compliance" Por conseguinte, a responsabilidade aumenta de acordo com a complexidade das sociedades, onde os garantidores de gestão se tornam réus de condutas delitivas que não foram diretamente por eles praticadas, mas que vem sendo categorizadas em razão do expansionismo do Direito Penal como um meio de retorno dos anseios sociais.59 Isto porque a tutela de bens jurídicos em crimes econômicos geralmente é coletiva, apresentando distanciamento da vida do cidadão comum, consequentemente, afastando a relação direta entre a conduta praticada e o ato lesivo.60 DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 55 Por este motivo, o que mais se nota é a incidência da responsabilidade objetiva recaindo sob tais garantidores, independente da comprovação de dolo ou culpa para a atribuição de responsabilidade. Em complemento, este ponto que pode ser alvo de questionamento sobre a questão da autoria mediata – aceita pela doutrina brasileira – a qual o autor realiza a ação por meio de outra pessoa, na qualidade de mandante imbuído de erro ou coação.61 No entanto, conforme explica Roxin, essa qualificação seria problemática no caso em que houvesse múltiplos garantes, pois haveria grande dificuldade de determinar como cada um agiria, gerando insegurança acerca da ação de cada indivíduo.62 Diante desta inoperância, nos restaria então classificar a atuação do compliance officer nos crimes de omissão imprópria63 (também chamados de comissivos por omissão), eis que não há causalidade fática na ação, mas sim jurídica. Logo, o agente responderia não pelo que causou, mas sim pelo fato de não ter evitado a ocorrência do ato lesivo.64 Seguindo esta lógica, temos no artigo 13, §2º do Código Penal a previsão do indivíduo assumindo o posicionamento como garante da situação, onde sua obrigação é de dever e poder agir para evitar o resultado, dividindo então entre a possibilidade de atuação e o dever de atuar, considerando o rol taxativo para caracterizar tal questão. Emparelhando ao âmbito empresarial, poderíamos elencar a atuação do compliance officer na alínea “b” do artigo supramencionado, uma vez que tal posição hierárquica se procedeu mediante delegação de poderes de vigilância e supervisão do programa de integridade, implicando em seu dever cumprir as normas legais. Vale lembrar que aqui não se trata de um dever contratual, mas sim, de qualquer imputação de responsabilidade, o que decorre da sua posição empresarial no ato da ocorrência do ato ilícito. 61 BITTENCOURT, Cezar Roberto. Tratado de Direito Penal: Parte Geral, 1. São Paulo: Saraiva, 2014. p. 560. 62 ROXIN, Claus. Novos estudos do direito penal. São Paulo: Marcial Pons, 2014. p. 169. , , , p 62 ROXIN, Claus. Novos estudos do direito penal. São Paulo: Marcial Pons, 2014. p. 169. 63 Diferentemente dos crimes comissivos, os omissivos ocorrem quando o agente não faz ou não pode ou não deve fazer tal ato. Os crimes omissivos se dividem em dois tipos: (i) próprios – previstos no Código Penal e preveem uma conduta específica; e (ii) impróprios – sem previsão legal, como por exemplo, o crime de homicídio por omissão (art. 121 c/c art. 13, ambos do Código Penal). 64 Op. Cit. Item 63, p. 308. 4.1 Impactos pelo descumprimento dos programas de compliance e o "criminal compliance" Apesar desta última hipótese de enquadramento se consolidar em uma saída viável para a imputação da responsabilidade ao compliance officer, este raciocínio é questionável, pois a lei brasileira não institui a posição de garante de forma solida, e ainda deixa de modo vago a NCOURT, Cezar Roberto. Tratado de Direito Penal: Parte Geral, 1. São Paulo: Saraiva, 2014. p. 560. N, Claus. Novos estudos do direito penal. São Paulo: Marcial Pons, 2014. p. 169. p p 63 Diferentemente dos crimes comissivos, os omissivos ocorrem quando o agente não faz ou não pode ou não deve fazer tal ato. Os crimes omissivos se dividem em dois tipos: (i) próprios – previstos no Código Penal e preveem uma conduta específica; e (ii) impróprios – sem previsão legal, como por exemplo, o crime de homicídio por omissão (art. 121 c/c art. 13, ambos do Código Penal). 64 Op. Cit. Item 63, p. 308. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 56 delimitação das condutas omissivas de resultado de dano. Por isso, as ações deste agente não configuram tipos penais abarcados de modo inequívoco na legislação brasileira. Ademais, destaca-se que na jurisprudência brasileira há um movimento de distanciamento das funções intrínsecas de garante da responsabilidade criminal. Como exemplo, podemos citar um caso advindo do Supremo Tribunal de Justiça de 2009 com manifestação contraria à responsabilização objetiva de profissional que exaure parecer técnico, de acordo com sua atividade, e venha a ser incriminado.65 Não obstante a discussão e o posicionamento acima, em 2013 o Supremo Tribunal Federal, no julgamento da Ação nº. 470 (caso conhecido como “Mensalão”), decide por responsabilizar criminalmente os agentes financeiros apontados no caso, por sua atuação na execução de forma indireta do crime, a qualidade de autores mediatos.66 De forma recorrente aos fundamentos do caso, foi citada a teoria do domínio do fato (exposta em capítulo anterior), consubstanciando a possibilidade de incriminar aqueles que não tivessem participado diretamente dos atos executórios, mas que de alguma forma tinham o domínio do fato, por coordenarem o comando no controle da execução do crime. j g 66 BOTTINI, Pierpaolo Cruz. Aplicação da teoria do domínio do fato na AP 470. Conjur, São Paulo, 13 ago. 2013. Disponível em: <http://www.conjur.com.br/2013-ago-13/direito-defesa-aplicacao-teoria-dominio-fatos- ap-470>. Acesso em: 18 de janeiro 2021. p j 67 STJ, RHC nº. 45.872/MG, Rel. Min. Ribeiro Dantas, Quinta Turma, julgado em 17.08.2017, DJe 28.08.2017. 65 STJ, HC nº. 108.985/DF, Rel. Min. Laurita Vaz, Quinta Turma, julgado e, 26.05.2009, DJe 15.06.200 Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 4.1 Impactos pelo descumprimento dos programas de compliance e o "criminal compliance" Neste caso, o tema compliance foi especificamente trabalhado, em especial quanto à condenação de Vinicius Samarane, considerado compliance officer do setor de controle compliance do Banco Rural, o qual restou comprovado que obtinha vinculação direta a fraudes no sistema de prevenção de lavagem de dinheiro, por meio de sua gestão fraudulenta e de lavagem de dinheiro, uma vez que determinava a alteração de relatórios internos do bando relativos ao programa de compliance vigente. Por fim, mais recentemente, o Superior Tribunal de Justiça, em Recurso Ordinário em Habeas Corpus (agosto/2017), também a Quinta Turma, entendeu que nos crimes societários cometidos por representantes da empresa deve haver correlação entre a concretude dos fatos delituosos com a atividade do acusado, não sendo suficiente a condição de socio da sociedade, sob pena de responsabilização objetiva.67 Em complemento, vale lembrar que não há tipo penal específico que categorize a figura do compliance officer, contudo na doutrina, existem autores que argumentam que haverá Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 57 responsabilização em casos de omissão impropria, quando esta omissão for inicial e dolosa do agente, dando causa ao resultado posterior, eis que tinha o dever de evitar tal resultado.68 Em suma, o STJ entendeu que não basta ocupar a posição de garante ao cumprimento dos programas de integridade, atuando assim como barreira à possibilidade de prática de ilícitos penais. Há, na visão desta Corte, a necessidade de comprovação de nexo entre o agente e a conduta, para que então possa haver imputação de responsabilidade do representante da pessoa jurídica. Salienta-se que, apesar das decisões conflitantes dentro das Cortes brasileiras, na Alemanha, Estados Unidos e países da Europa Ocidental o compliance officer tem uma função abrangente e predefinida, havendo o chamado “deve de compliance”, que é exigido para todas as empresas constituídas como sociedades anônimas ou microempresas. Em vista disso, o cenário ideal seria a delimitação, tanto pela legislação brasileira quanto pelos programas de integridade, das funções e poderes que serão desenvolvidos pelo agente garantidor, bem como suas respectivas responsabilizações quanto ao descumprimento destas, de forma a deixar evidente as atribuições ao assumir o cargo de compliance officer efetivamente. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 68 NUCCI, Guilherme de Souza. Manual de Direito Penal. 2. Ed. São Paulo: RT, 2006. Conclusão Diante de todo o exposto, conclui-se a importância da figura do compliance officer vem aumentando ao longo do tempo, devido à complexidade no desempenho de suas tarefas perante o sistema econômico e legislativo, o que explica o aumento na adesão aos programas de compliance dentro do cenário empresarial brasileiro. Todavia, há um rol ínfimo de leis que abordam o assunto no ordenamento jurídico brasileiro, o que dificulta (e muito) a definição da responsabilização juridicamente, visto que não se trata de mera individualização de imputação criminal, em que pese o comportamento falho do indivíduo como representante da empresa como figura de garante da situação, mas sim, sobre os reflexos envolvendo comportamentos coletivos e sua respectiva individualização, correlacionados à tutela coletiva, neste caso, a defesa da sociedade contra os atos corruptivos que influencial a harmonia social e econômica. 68 NUCCI, Guilherme de Souza. Manual de Direito Penal. 2. Ed. São Paulo: RT, 2006. 58 Por conseguinte, entende-se que para que haja uma responsabilização ponderada, necessário definir as dinâmicas de cada tipo societário, divisões de trabalho e funções dos agentes ocupantes dos cargos de controle, a fim de que não haja margem para desoneração de reponsabilidade dos agentes de alto escalão societário. Além disso, apesar das duras críticas de que não se pode condenar executivos pela inoperâncias de seus atos em seus respectivos cargos de alta gestão, também não se pode eximir completamente de toda e qualquer responsabilidade criminal, dado que esta máxima incitaria ainda mais o caos e a falta de comprometimento destes agente para com suas funções. Logo, compreensível que para que haja responsabilização subjetiva do compliance officer, por meio de comprovação de nexo causal (conexão da conduta com o resultado), inclusive através de tipificação como crime omissivo improprio, se comprovada, além da omissão inicial, o dolo nas ações praticadas em razão da posição de garantidor da operação empresarial. Caso contrário, estaríamos invariavelmente recaindo sobre a regra da responsabilização objetiva, que já não se adequa mais aos ditames do Direito Penal Moderno. Outrossim, entendível o distanciamento da aplicação da teoria do domínio do fato, já que não se trata de agente que serve de outrem para a prática do delito. No caso, é o próprio agente quem determina a distorção da conduta preventiva, tornando-a deturpada em relação ao seu respectivo programa de compliance. 69 RAWLS, John. Uma teoria da justiça. Trad. de Jussara Simões. São Paulo: M. Fontes, 2008. p. 11. 70 Id 4 Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 69 RAWLS, John. Uma teoria da justiça. Trad. de Jussara Simões. São Paulo: M. Fontes, 2008. p. 11. 70 Idem, p. 4. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. BITTENCOURT, Cezar Roberto. Tratado de Direito Penal: Parte Geral, 1. São Paulo: Saraiva, 2013 e 2014. BOTTINI, Pierpaolo Cruz. Aplicação da teoria do domínio do fato na AP 470. Conjur, São Paulo, 13 ago. 2013. Disponível em: <http://www.conjur.com.br/2013-ago-13/direito-defesa- aplicacao-teoria-dominio-fatos-ap-470>. Acesso em: 18 de janeiro de 2021. BRASIL. Decreto-Lei 2.848, de 07 de dezembro de 1940. Código Penal. Diário Oficial da União, Rio de Janeiro, 31 dez. 1940. CAPEZ, Fernando. Curso de direito penal: parte geral. 17. ed. vol 1. São Paulo: Saraiva, 2011. BENEDETTI, Carla Rahal. Criminal Compliance. São Paulo: Quartier Latin, 2014. BENEDETTI, Carla Rahal. Criminal Compliance. São Paulo: Quartier Latin, 2014. BITTENCOURT, Cezar Roberto. Tratado de Direito Penal: Parte Geral, 1. São Paulo: Saraiva, 2013 e 2014. Referências bibliográficas BENEDETTI, Carla Rahal. Criminal Compliance. São Paulo: Quartier Latin, 2014. Conclusão Assim, até seria admissível a adoção da conduta omissiva impropria se esta estivesse definida pela esfera criminal, o que não ocorre atualmente. Portanto, para que não haja o uso indiscriminado desta possibilidade, e consequentemente resulte na violação do princípio da legalidade, o melhor seria a delimitação da atuação da figura do compliance officer, através da elaboração de uma lei especifica para esclarecer seus deveres jurídicos. Por fim, como reflexão, servem as palavras do filósofo político Rawls, aludindo que “as diversas concepções de justiça provêm das distintas noções de sociedade, contra um pano de fundo de visões conflitantes acerca das necessidades naturais e das oportunidades da vida humana”69 E continua: “a única coisa que nos permite aquiescer a uma teoria errônea é a falta de uma melhor; de maneira análoga, a injustiça só é tolerável quando é necessária para evitar uma injustiça ainda maior”.70 59 Por isso, torna-se necessário, o aprimoramento das teorias penais, as quais são essenciais para organizar e reger o fluxo de contatos sociais na sociedade contemporânea. Referências bibliográficas ARAUJO JUNIOR, João Marcello. O direito penal econômico. Revista Brasileira de Ciências Criminais, nº 25, ano 07. São Paulo: Revista dos Tribunais, jan./mar. 1999. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 FRAGOSO, Heleno Cláudio. Lições de direito penal: a nova parte geral. 2. Ed. Rio de Janeiro: Forense, 1991. FRAGOSO, Heleno Cláudio. Lições de direito penal: a nova parte geral. 2. Ed. Rio de Janeiro: Forense, 1991. GLYN, Patrick; KOBRIN, Stephen J.; NAIM, Moisés. Globalização da corrupção, a economia política da corrupção. Brasília: UNB, 2002. GRECO, Luís. A Teoria da Imputação Objetiva: uma introdução. In: Funcionalismo e imputação objetiva no direito penal. ed. 3a. Rio de Janeiro: Renovar, 2002. GRECO, Luís. A Teoria da Imputação Objetiva: uma introdução. In: Funcionalismo e imputação objetiva no direito penal. ed. 3a. Rio de Janeiro: Renovar, 2002. ESTATUTO sobre fraudes criminais do Departamento de Justiça dos Estados Unidos. Disponível em: <https://www.justice.gov/sites/default/files/criminal- fraud/legacy/2012/11/14/fcpa-english.pdf>. Acesso em 09 de dezembro de 2020. ESTATUTO sobre fraudes criminais do Departamento de Justiça dos Estados Unidos. Disponível em: <https://www.justice.gov/sites/default/files/criminal- fraud/legacy/2012/11/14/fcpa-english.pdf>. Acesso em 09 de dezembro de 2020. ESTEFAM, André. GONÇALVES, Victor Eduardo Rios. Direito penal esquematizado: Parte Geral. 3. ed. São Paulo: Saraiva, 2014. FERRAJOLI, Luigi. Direito e razão: teoria do garantismo penal. 2. Ed. São Paulo: Revista dos Tribunais, 2006. FERRAJOLI, Luigi. Direito e razão: teoria do garantismo penal. 2. Ed. São Paulo: Revista dos Tribunais, 2006. ESTEFAM, André. GONÇALVES, Victor Eduardo Rios. Direito penal esquematizado: Parte Geral. 3. ed. São Paulo: Saraiva, 2014. ESTELLITA, Heloisa. Responsabilidade penal dos dirigentes de empresas por omissão: estudo sobre a responsabilidade omissiva impropria de dirigentes de sociedades anônimas, limitadas e encarregados pelo cumprimento por crimes praticados por membros da empresa. São Paulo: Marcial Pons, 2017. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 CAPEZ, Fernando. Curso de direito penal: parte geral. 17. ed. vol 1. São Paulo: Saraiva, 2011. CARVALHO, André Castro; BERTOCCELLI, Rodrigo de Pinho; ALVIM, Tiago Cripa; VENTURINI, Otavio (Coord.). Manual de Compliance. 2. ed. Rio de Janeiro: Forense, 2020. COMPLY. In: MICHAELIS Moderno Dicionário da língua inglesa. São Paulo: Melhoramentos. Disponível em: <https://michaelis.uol.com.br/moderno-ingles/busca/ingles- portugues-moderno/comply/>. Acesso em 18 de dezembro de 2020. COMPLY. In: MICHAELIS Moderno Dicionário da língua inglesa. São Paulo: Melhoramentos. Disponível em: <https://michaelis.uol.com.br/moderno-ingles/busca/ingles- portugues-moderno/comply/>. Acesso em 18 de dezembro de 2020. CUNHA, Rogério Sanches. SOUZA, Renee do Ó. Lei Anticorrupção Empresarial. 3. ed., rev. atual. e ampl. Salvador: Ed. Juspodivm, 2020. CUNHA, Rogério Sanches. SOUZA, Renee do Ó. Lei Anticorrupção Empresarial. 3. ed., rev. atual. e ampl. Salvador: Ed. Juspodivm, 2020. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 60 LARRAURI, Elena. Notas preliminares para una discusión sobre la imputación objetiva. In: BUSTOS/LARRAURI. La imputación objetiva. Santa Fé de Bogotá: Temis, 1998. MAEDA, Bruno Carneiro; AYRES, Carlos Henrique da Silva (Coord.). Temas de anticorrupção e Compliance. Rio de Janeiro: Elsevier, 2013. MATHIESEN, Thomas. The Politics of Abolition. Oslo, Robertson, 1974, p. 3. In: FOLTER, Rolf S. de. Sobre a fundamentação metodológica do enfoque abolicionista do sistema de justiça penal — uma comparação das ideias de Hulsman, Mathiesen e Foucault. Trad. do espanhol por Natalia Montebello. Disponível em: <https://cpisp.org.br/wp- content/uploads/2018/05/5140-12199-1-SM.pdf>. Acesso em 18 de janeiro de 2021. NUCCI, Guilherme de Souza. Manual de Direito Penal. 2. Ed. São Paulo: RT, 2006. _____. Um panorama da teoria da imputação objetiva. ed. 4a. São Paulo: Editora Revista dos Tribunais, 2014. GOMES, Abel Fernandes. Responsabilidade penal pela omissão de compliance. Centro de Pesquisa em Crimes Empresariais e Compliance Prof. João Marcello de Araújo Jr. (CPJM), Rio de Janeiro, 2020. Disponível em: <http://www.cpjm.uerj.br/wp- content/uploads/2020/06/RESPONSABILIDADE-PENAL-PELA-OMISSÃO-DE- COMPLIANCE.pdf>. Acesso em 21 de dezembro de 2020. 61 JOBIM, Rosana Kim. Compliance e Trabalho: entre o poder diretivo do empregador e os direitos inespecíficos do empregado. Florianópolis: Tirant Lo Blanch, 2018. JOBIM, Rosana Kim. Compliance e Trabalho: entre o poder diretivo do empregador e os direitos inespecíficos do empregado. Florianópolis: Tirant Lo Blanch, 2018. JOBIM, Rosana Kim. Compliance e Trabalho: entre o poder diretivo do empregador e os direitos inespecíficos do empregado. Florianópolis: Tirant Lo Blanch, 2018. KLITGAARD, Robert. A Corrupção sob Controle. Rio de Janeiro: Jorge Zahar Editor, 1994. KLITGAARD, Robert. A Corrupção sob Controle. Rio de Janeiro: Jorge Zahar Editor, 1994. LARRAURI, Elena. Notas preliminares para una discusión sobre la imputación objetiva. In: BUSTOS/LARRAURI. La imputación objetiva. Santa Fé de Bogotá: Temis, 1998. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 NUCCI, Guilherme de Souza. Manual de Direito Penal. 2. Ed. São Paulo: RT, 2006. OLIVEIRA, Mara Elisa de. Breve análise sobre o abolicionismo e o minimalismo. Artigo disponível em: <https://jus.com.br/artigos/22596/breve-analise-sobre-o-abolicionismo-e-o- minimalismo>. Acesso em 18 de janeiro de 2021. RAWLS, John. Uma teoria da justiça. Trad. de Jussara Simões. São Paulo: M. Fontes, 2008. ROXIN, Claus. Problemas basicos del derecho penal. Madrid: Editorial Reus S.A., 1976. ROXIN, Claus. Problemas basicos del derecho penal. Madrid: Editorial Reus S.A., 1976. _____. Reflexões sobre a problemática da imputação em direito penal. In: Problemas fundamentais de direito penal. ed. 2a. Lisboa: Vega, 1993. _____. Problemas de autoría y participación em la criminalidad organizada. Revista Penal, n. 2, jul. 1998. _____. Estudos de Direito Penal. Trad. de Luís Greco. 2. ed. Rio de Janeiro: Renovar, 2008. ____. Estudos de Direito Penal. Trad. de Luís Greco. 2. ed. Rio de Janeiro: Renovar, 2008. _____. Täterscha und Tatherrscha. 6. Aufl, Berlin: Walter de Gruyter, 1994, p. 112. In ALFLEN, Pablo Rodrigo. Teoria do domínio do fato na doutrina e na jurisprudência brasileiras. Universitas JUS, v.25, n.2, 2014. _____. Täterscha und Tatherrscha. 6. Aufl, Berlin: Walter de Gruyter, 1994, p. 112. In ALFLEN, Pablo Rodrigo. Teoria do domínio do fato na doutrina e na jurisprudência brasileiras. Universitas JUS, v.25, n.2, 2014. SANTIN, Miguel Ângelo. O Estado e o direito de punir. In: MARIN, Jeferson Dytz (Coord.). Jurisdição e processo II: reformas processuais ordinarização e racionalismo Curitiba: Juruá Jurisdição e processo II: reformas processuais, ordinarização e racionalismo. Curitiba: Juruá, 2009. SANTOS, Juarez Cirino dos. Direito Penal - Parte Geral. 4. ed. rev., ampl. Florianópolis: Conceito Editorial, 2010. SHECAIRA, Sérgio Salomão. Criminologia. São Paulo: RT, 2014. RAWLS, John. Uma teoria da justiça. Trad. de Jussara Simões. São Paulo: M. Fontes, 2008. RODRIGUES, Anabela Miranda. Direito Penal Econômico: uma política criminal na era compliance. Coimbra: Edições Almedina, 2019. ROUSSEAU, Jean-Jacques. O contrato social. Trad. de Paulo Neves. Porto Alegre: L&PM, 2013. 62 SANCINETTI, Marcelo A. Observaciones sobre la teoría de la imputación objetiva. In: Teorias Actuales en el derecho penal. Buenos Aires: Ad-Hoc, 1998. SANCINETTI, Marcelo A. Observaciones sobre la teoría de la imputación objetiva. In: Teorias Actuales en el derecho penal. Buenos Aires: Ad-Hoc, 1998. TIN, Miguel Ângelo. O Estado e o direito de punir. In: MARIN, Jeferson Dytz (Coord SANTIN, Miguel Ângelo. O Estado e o direito de punir. In: MARIN, Jeferson Dytz (Coord.). Jurisdição e processo II: reformas processuais, ordinarização e racionalismo. Curitiba: Juruá, 2009. Revista Científica do CPJM, Rio de Janeiro, Vol.1, N.02, 2021. Bruna Perasoli Trade DOI: 10.55689/rcpjm.2021.02.003 \| ISSN: 2764-1899 SHECAIRA, Sérgio Salomão. Criminologia. São Paulo: RT, 2014. SILVEIRA, Renato de Mello Jorge; SAAD-DINIZ, Eduardo. Compliance, Direito penal e a lei anticorrupção. 1. ed. São Paulo: Saraiva, 2015. SOUZA, Luciano Anderson de. (coord.) Compliance no Direito Penal. vol. 5. São Paulo: Thomson Reuters RT, 2021. 63 SOUZA, Artur de Brito Gueiros; JAPIASSÚ, Carlos Eduardo Adriano. Curso De Direito Penal: Parte Geral. ed. 2. Rio de Janeiro: Editora Forense, 2015. STJ, HC nº. 108.985/DF, Rel. Min. Laurita Vaz, Quinta Turma, julgado e, 26.05.2009, DJe 15.06.2009. STJ, HC nº. 108.985/DF, Rel. Min. Laurita Vaz, Quinta Turma, julgado e, 26.05.2009, DJe 15.06.2009. STJ, RHC nº. 45.872/MG, Rel. Min. Ribeiro Dantas, Quinta Turma, julgado em 17.08.2017, DJe 28.08.2017. STJ, RHC nº. 45.872/MG, Rel. Min. Ribeiro Dantas, Quinta Turma, julgado em 17.08.2017, DJe 28.08.2017. WELZEL, Hans. Derecho Penal Aleman. 11ª ed. Trad. por Bustos Ramírez e Yáñez Pérez, Santiago: Ed. Jurídica de Chile, 1997, p. 3-5. In: GUEIROS, Arthur. Valores ético-sociais e os programas de compliance. Coordenador Acadêmico do Centro de Pesquisas em Crimes Empresariais e Compliance (CPJM) de 10/09/2020. Disponível em: <http://www.cpjm.uerj.br/texto-blog-valores-etico-sociais-e-os-programas-de-compliance/>. Acesso em 15 de janeiro de 2021. ZAFFARONI, Eugênio Raúl. Em busca das Penas Perdidas: a perda da legitimidade do sistema penal. 5. Ed. Rio de Janeiro: Revan, 1991, 2014. ZAFFARONI, Eugênio Raúl. Em busca das Penas Perdidas: a perda da legitimidade do sistema penal. 5. Ed. Rio de Janeiro: Revan, 1991, 2014.
https://openalex.org/W2969540450	https://research-information.bris.ac.uk/ws/files/233390191/untitled.pdf	English	null	Implementing a Digital Tool to Support Shared Care Planning in Community-Based Mental Health Services: Qualitative Evaluation	JMIR. Journal of medical internet research/Journal of medical internet research	2,020	cc-by	9,969	Pithara, C., Farr, M. C., Sullivan, S. A., Edwards, H. B., Hall, W., Gadd, C., Walker, J., Hebden, N., & Horwood, J. P. (2020). Implementing a Digital Tool to Support Shared Care Planning in Community-Based Mental Health Services: Qualitative Evaluation. Journal of Medical Internet Research, 22(3), Article e14868. https://doi.org/10.2196/14868 Publisher's PDF, also known as Version of record License (if available): CC BY Link to published version (if available): 10.2196/14868 Publisher's PDF, also known as Version of record License (if available): CC BY Link to published version (if available): 10.2196/14868 Link to publication record on the Bristol Research Portal PDF-document This is the final published version of the article (version of record). It first appeared online via JMIR at https://www.jmir.org/2020/3/e14868/. Please refer to any applicable terms of use of the publisher. Pithara, C., Farr, M. C., Sullivan, S. A., Edwards, H. B., Hall, W., Gadd, C., Walker, J., Hebden, N., & Horwood, J. P. (2020). Implementing a Digital Tool to Support Shared Care Planning in Community-Based Mental Health Services: Qualitative Evaluation. Journal of Medical Internet Research, 22(3), Article e14868. https://doi.org/10.2196/14868 Pithara, C., Farr, M. C., Sullivan, S. A., Edwards, H. B., Hall, W., Gadd, C., Walker, J., Hebden, N., & Horwood, J. P. (2020). Implementing a Digital Tool to Support Shared Care Planning in Community-Based Mental Health Services: Qualitative Evaluation. Journal of Medical Internet Research, 22(3), Article e14868. https://doi.org/10.2196/14868 Corresponding Author: Corresponding Author: Christalla Pithara, PhD Population Health Sciences, Bristol Medical School University of Bristol 9th Floor Whitefriars, Lewins Mead Bristol, BS1 2NT United Kingdom Phone: 44 1173421272 ext 21272 Email: christalla pithara@bristol ac uk Email: christalla.pithara@bristol.ac.uk (J Med Internet Res 2020;22(3):e14868) doi: 10.2196/14868 (J Med Internet Res 2020;22(3):e14868) doi: 10.2196/14868 University of Bristol – Bristol Research Portal General rights This document is made available in accordance with publisher policies. Please cite only the published version using the reference above. Full terms of use are available: http://www.bristol.ac.uk/red/research-policy/pure/user-guides/brp-terms/ JOURNAL OF MEDICAL INTERNET RESEARCH Pithara et al Original Paper Implementing a Digital Tool to Support Shared Care Planning in Community-Based Mental Health Services: Qualitative Evaluation Abstract Background: Mental health services aim to provide recovery-focused care and facilitate coproduced care planning. In practice, mental health providers can find supporting individualized coproduced care with service users difficult while balancing administrative and performance demands. To help meet this aim and using principles of coproduction, an innovative mobile digital care pathway tool (CPT) was developed to be used on a tablet computer and piloted in the West of England. Objective: The aim of this study was to examine mental health care providers’ views of and experiences with the CPT during the pilot implementation phase and identify factors influencing its implementation. Methods: A total of 20 in-depth telephone interviews were conducted with providers participating in the pilot and managers in the host organization. Interviews were audio recorded, transcribed, anonymized, and thematically analyzed guided by the Consolidated Framework for Implementation Research. Results: The tool was thought to facilitate coproduced recovery-focused care planning, a policy and organizational as well as professional priority. Internet connectivity issues, system interoperability, and access to service users’ health records affected use of the tool during mobile working. The organization’s resources, such as information technology (IT) infrastructure and staff time and IT culture, influenced implementation. Participants’ levels of use of the tool were dependent on knowledge of the tool and self-efficacy; perceived service-user needs and characteristics; and perceptions of impact on the therapeutic relationship. Training and preparation time influenced participants’ confidence in using the tool. Conclusions: Findings highlight the importance of congruence between staff, organization, and external policy priorities and digital technologies in aiding intervention engagement, and the need for ongoing training and support of those intended to use the technology during and after the end of implementation interventions. (J Med Internet Res 2020;22(3):e14868) doi: 10.2196/14868 J Med Internet Res 2020 \| vol. 22 \| iss. 3 \| e14868 \| p. 1 (page number not for citation purposes) 1Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, United Kingdom 2National Institute for Health Research Applied Research Collaboration West at University Hospitals Bristol Foundation Trust, Bristol, United Kingdom 3Centre for Academic Mental Health, Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, United Kingdom 4Avon and Wiltshire Mental Health Partnership NHS Trust, Bristol, United Kingdom 5Ot k H lth S l ti Sl h U it d Ki d Implementing a Digital Tool to Support Shared Care Planning in Community-Based Mental Health Services: Qualitative Evaluation Christalla Pithara1,2, PhD; Michelle Farr1,2, PhD; Sarah A Sullivan1,2,3, PhD; Hannah B Edwards1,2, MSc; William Hall4, MBChB, MRCPsych; Caroline Gadd5, RN; Julian Walker4, PhD; Nick Hebden5, BSc; Jeremy Horwood1,2*, PhD Background Mental health digital technologies provide opportunities to improve care [1,2], service efficiency, and health outcomes [3-7]. Previous studies have explored mental health service users’ [5-10] and providers’ [11,12] experiences with health technology and technological innovations. There remains a need for coproduced real-world evaluation research [13], increasing understanding of contextual and organizational factors involved in successful implementation [14], particularly use of digital interventions within a therapeutic context [13,15-18]. The CPT was developed using the coproduction principles [20]: an iterative, collaborative approach involving service users and providers through (1) consultations that identified gaps and care needs, (2) a Joint Project Board with representation from service users, practitioners, managers, and software developers, and (3) feedback from mental health trust staff and service users via detailed observations, interviews, and focus groups on their experiences of consecutive versions of the CPT. As part of this collaboration, a film reporting on the coproduction process from the service users’ perspective was put together by Rethink Mental Illness, an organization facilitating service user involvement [28]. Recovery-focused care, embracing principles of shared decision making and coproduction of care plans, is recommended to improve mental health care delivery [19]. Coproduction requires services to be delivered “in an equal and reciprocal relationship between professionals, people using services, [and] their families” [20]. For decision making, both care providers and service users should possess skills and ability to access, share, and use information to meet service users’ often complex, individual needs [21]. Care providers, however, may find it challenging to ensure individualized, person-centered care, while also balancing administrative and performance demands [22]. Coproduction in care can be compromised by individual and organizational factors [22-25], including health information technology (IT) systems [25,26]. Here lies the need for innovations supporting coproduced care, while addressing performance and efficiency concerns. On the basis of this work, the CPT was designed to be used on a mobile tablet computer and incorporated 4 different components (Table 1). Screenshots of the CPT are included in Multimedia Appendix 1. The pilot implementation of the CPT took place between March and December 2016. A total of 30 providers involved in care planning and recovery support for Community Mental Health Teams were recruited through engagement events or word-of-mouth to pilot the CPT in routine practice. KEYWORDS health care technology; mental health; community health care; patient-centered care; patient care planning; implementation science J Med Internet Res 2020 \| vol. 22 \| iss. 3 \| e14868 \| p. 1 (page number not for citation purposes) https://www.jmir.org/2020/3/e14868 XSL•FO RenderX JOURNAL OF MEDICAL INTERNET RESEARCH Pithara et al through patient involvement in care planning, and introducing specific exercises to encourage new ways of interacting. The experiences of staff using the CPT to coproduce recovery-oriented care planning are reported here [27]. Background Face-to-face training was provided for all staff from an experienced mental health worker on how the tool worked and how it could be used to facilitate coproduction in care planning. Any issues arising from the CPT during these meetings were raised with the software developers. Help information was also included as part of the tool itself and as part of a service user information leaflet. This information was specific to navigating the CPT and using the CPT components. J Med Internet Res 2020 \| vol. 22 \| iss. 3 \| e14868 \| p. 2 (page number not for citation purposes) Quick notes The study was reviewed by the NHS Health Research Authority (ID: 199385) and ethically reviewed by the University of Bristol, Faculty of Health Sciences, Research Ethics Committee (Application: 29045). The study was reviewed by the NHS Health Research Authority (ID: 199385) and ethically reviewed by the University of Bristol, Faculty of Health Sciences, Research Ethics Committee (Application: 29045). Analysis Data collection and analysis occurred in parallel. Sample size was driven by the concept of information power, with information within our sample continuously assessed in relation to our study objective as data collection progressed [31]. Interviews were transcribed, anonymized, and thematically analyzed [32,33] in NVivo 10 (QSR). An initial inductive coding scheme was developed and refined as new data were analyzed to understand the main themes emerging from the data. Data were coded into thematic categories representing participants’ attitudes toward the CPT, positive and negative aspects of the CPT, and barriers and facilitators to implementation. MF and CP double coded a subset of transcripts, and any discrepancies were discussed and resolved. Analytical uncertainties or disagreements were discussed by the multidisciplinary research team to ensure credibility and confirmability. The CFIR was then used as a framework to order codes [30,34], in line with the CFIR Qualitative Codebook Guidelines [35] to deepen our analysis, rather than impose deductive codes on the data. The CFIR incorporates a repository of 39 standardized implementation-related constructs organized across 5 domains, which interact to influence implementation effectiveness (Multimedia Appendix 3) [34]. The 5 domains are as follows: (1) intervention characteristics: includes 8 constructs related to characteristics of the intervention being implemented; (2) outer setting: includes 4 constructs related to external factors such as the economic, political, and social context within which an organization is situated; (3) inner setting: includes 12 constructs related to features such as the structural, political, and cultural The aim of this paper was to present findings arising from using the Consolidated Framework for Implementation Research (CFIR) [30] to identify factors influencing adoption and use of the CPT in care delivery routine interactions. Evaluation of the Care Pathway Tool Pilot Implementation As part of the CPT pilot implementation, an independent qualitative evaluation was commissioned and undertaken by the National Institute for Health Research Collaboration for Leadership in Applied Health Research and Care West (NIHR CLAHRC West) to investigate CPT acceptability by care providers and effectiveness in facilitating coproduced care plans and recording information more efficiently. Details about how the interactive features of the CPT supported coproduction in care planning are reported separately [27]. J Med Internet Res 2020 \| vol. 22 \| iss. 3 \| e14868 \| p. 3 (page number not for citation purposes) Development and Pilot Implementation of the Care Pathway Tool To support coproduced mental health care, a care pathway tool (CPT) was developed through a collaborative effort between a mental health care service provider (Avon and Wiltshire Mental Health Partnership NHS Trust [AWP], the lead regional provider for community mental health services), users of community mental health services, and technology developers (Otsuka Health Solutions [OHS]), as part of a project piloted in the West of England (Joining the Dots). Staff volunteers were asked to use the CPT with up to 5 mental health service users whose clinical risk assessment for a mental health crisis (such as an exacerbation in their clinical condition, which would require urgent attention) was set at low or medium risk. This decision was informed by a cautious approach to testing new mental health electronic tools in the National Health Service (NHS). The project aimed to use computer tools for better use of data and information to improve care delivery and facilitate collaborative working in care planning. The CPT aimed to (1) enable providers’ and service users’ direct access to electronic care plans to support efficient working and (2) enable coproduced, recovery-focused care during community visits, Mental health providers introduced the tool to service users during routine meetings and integrated it into practice if service users agreed. https://www.jmir.org/2020/3/e14868 XSL•FO RenderX Pithara et al JOURNAL OF MEDICAL INTERNET RESEARCH Table 1. Components and features of the care pathway tool. Planning for my future • Focuses on identifying goals, split into specific actions, to be pursued by the service user • It visually illustrates the goal and its action pathway in the form of an infographic • Focuses on identifying goals, split into specific actions, to be pursued by the service user • It visually illustrates the goal and its action pathway in the form of an infographic • It visually illustrates the goal and its action pathway in the form of an infographic • Enables providers to use the tablet computer for making notes during the meeting with the service users, including service user–written progress notes • Enables providers to use the tablet computer for making notes during the meeting with the service users, including service user–written progress notes Development and Pilot Implementation of the Care Pathway Tool Description Care pathway tool component and feature My life My journey • An interactive timeline illustrating service users’ health care system journey • Combines data extracted from clinical records, for example, referrals and admissions, alongside care experience–specific information inputted by service users People in my life • Enables service users to graphically present key people in their life, including social networks, providers, or services they are engaged with • Enables service users to visualize their social networks and explore their relationships My plan to stay well Managing my warning signs • An electronic version of Max Birchwood’s early signs of psychosis approach [29] (with permission from authors) • Allows the individual to identify early warning signs and psychotic symptoms specific to their experiences • Enables discussions about warning signs of relapse and identify ways of managing these Planning for my future Goals and actions • Focuses on identifying goals, split into specific actions, to be pursued by the service user • It visually illustrates the goal and its action pathway in the form of an infographic Quick notes • Enables providers to use the tablet computer for making notes during the meeting with the service users, including service user–written progress notes E l i f h C P h T l Pil The study was reviewed by the NHS Health ResearchAuthority Table 1. Components and features of the care pathway tool. • An electronic version of Max Birchwood’s early signs of psychosis approach [29] (with permission from authors) • Allows the individual to identify early warning signs and psychotic symptoms specific to their experiences • Enables discussions about warning signs of relapse and identify ways of managing these Planning for my future Goals and actions Intervention Source Staff’s involvement in tool development through coproduction activities influenced engagement with the pilot. Managers were aware of the need for staff to feel involved in new interventions: External Policy and Incentives Another management idea coming in isn’t necessarily something they (staff) are going to embrace. [Manager 01] A recent Care Quality Commission (CQC) report of the mental health organization highlighted the need to improve inclusion of service users’ views in care plans [33]. The CPT could be a key mechanism to improve practices in response to the CQC report’s comments, and Joint Project Board members saw this as an important facilitator to encourage its use within the organization. Outer Setting There was [...] a big meeting with our team [...] to see what kind of solutions they could come up with to improve co-production and help our sort of work. And [...] I put my name down. [Practitioner 03] The CFIR conceptualizes the outer setting as factors external to the organization [34]. Design Quality, Packaging, and Complexity characteristics of the organization implementing the intervention; (4) characteristics of individuals: includes 5 constructs related to the individuals involved with the intervention and implementation process; and (5) process: includes 8 constructs related to essential activities of the implementation process. Some participants distinguished between the CPT’s features, which they thought were well designed, useful, and easy to use, and limitations of the tablet computer on which the CPT was hosted: Part of it is to separate the tool from the piece of equipment it's on. The piece of equipment, there've been lots of problems, but the actual tool itself [CPT], the different bits of it have been really good, really easy to use. [Practitioner 12] Sampling and Interviews In-depth interviews were conducted with (1) providers piloting the CPT to explore their views and experiences, and (2) staff with performance and management roles to explore views from a strategic level on the implementation and use of new digital technologies. The interview guide is included in Multimedia Appendix 2. Purposeful sampling ensured that a range of roles were recruited. All interviews were conducted over the telephone and, with oral consent, audio recorded. Interviews took place between October and November 2016 and lasted between 13 and 60 min (mean 32 min). https://www.jmir.org/2020/3/e14868 XSL•FO RenderX JOURNAL OF MEDICAL INTERNET RESEARCH Pithara et al Relative Advantage The CPT was thought to enable more efficient information recording and facilitate coproduced care. Creating notes using the CPT alongside service users facilitated transparency and involvement of service users, and it was quicker than traditional ways of working: Participants In total, 20 providers were interviewed (11 female). Participants included 15 (out of the 30) practitioners who piloted the CPT with service users (mental health support workers, peer support workers, psychiatrists, occupational therapists, community psychiatric nurses, and social workers) and 5 managers; 6 practitioners piloted the CPT for 6 months or more, 4 between 3 to 6 months and 5 between 6 weeks and 3 months. These practitioners provided community-based care, and contact with service users was described to be needs driven, ranging from weekly to monthly. Practitioners discussed how service users involved in piloting the tool had a range of mental health diagnoses, including psychosis, anxiety, depression, and previous experiences of trauma. Issues with internet connectivity and tablet computer log-in problems were a common barrier to using the CPT with service users: I don't have confidence yet [...] that it'll work first time [...] I feel positive about the software itself but not about being able to use it when I need to in remote locations. [Practitioner 03] Other barriers included security limitations, and lack of live cross-system communication between the CPT and the host organization’s main electronic patient record (EPR) system. Until secure platforms for information exchange were developed, all data were manually transferred from the CPT system to service users’ EPRs by administrators. The delay in transfer (up to 48 hours) impacted on information available during meetings, sharing of information between providers involved in care, and ultimately how and how often the tool was used: A total of 13 CFIR constructs were seen to influence the processes of CPT implementation in all 5 framework domains. The factors identified and their relationship to these constructs and domains are outlined in Multimedia Appendix 3. The 5 CFIR domains are used to structure presentation of findings with illustrative verbatim participant quotes. (If) the service user is a bit unstable with their mental health and you need to update a lot of other information related to the meetings [...] If the service user is at risk, immediate risk, then we might need to go on the actual system [ERP] and record it to avoid any other issues. [Practitioner 005] Patient Needs and Resources In this construct, quotes relating to awareness of service user–specific factors influencing implementation were included. Some participants were guided in their use of the tool by perceived needs and characteristics of individual service users. For example, English language literacy was taken into consideration when deciding to use the tool: It saved me a lot of time. You don’t have to go back to the main [IT] system. [Practitioner 05] Implementation Climate and Culture The organization already had a focus on recovery-oriented care, including engaging service users in care planning. Participants agreed that the CPT facilitated coproduced recovery-focused care planning by supporting novel and more user-centered conversations with a psychosocial focus: I thought it would be really useful, [...] we quite often do WRAP (wellness recovery action plans) plans with people [...] But having the ability to actually sit down and work with somebody and do a process holistically together rather than it being me-led was quite nice. [Practitioner 09] I believe in a real connection between people and a connection in the room and that, to me, comes from face-to-face and eye contact and us sitting opposite each other almost and me being really attentive to the other person. So I think any device is going to take away from that. [Practitioner 13] However, some medically trained staff saw the CPT’s focus on psychosocial aspects of recovery as contradictory to their professional roles, for example, discussing medications: Inner Setting This construct relates to characteristics of the organization implementing the intervention [34]. It saved me a lot of time. You don’t have to go back to the main [IT] system. [Practitioner 05] Declining to use the tool once, however, did not always exclude use of the tool in subsequent sessions: It’s very hard to have that time before sessions to thoroughly think it through and plan it. You kind of like rushing to an appointment. [Practitioner 06] It’s very hard to have that time before sessions to thoroughly think it through and plan it. You kind of like rushing to an appointment. [Practitioner 06] we’re short of money and we’re short of time, therefore we’re short of people. I spend much less time face-to-face supporting people than I would like to and I think [...] that will impact on the use of the tool. [Practitioner 07] we’re short of money and we’re short of time, therefore we’re short of people. I spend much less time face-to-face supporting people than I would like to and I think [...] that will impact on the use of the tool. [Practitioner 07] We did a first session and then he was like oh God I couldn't concentrate, [...] I don't really want to do it [...] And then actually recently he said, “Oh why don't I do that tool with you anymore?” [Practitioner 12] Characteristics of Individuals This CFIR construct includes features of individuals involved in the intervention [34]. Self-Efficacy (If service users are) expecting to talk to me about their medication [...] and so I have to make sure that it (using the tool) wouldn’t be [...] something that would leave them feeling dissatisfied. [Practitioner 04] Issues of confidence and perceived ability to use the CPT were raised by both experienced users of the tool, that is, those trained 6 to 8 months before the interview, as well as inexperienced users, that is, participants trained around 2 months prior. Familiarity with the tool was needed when deciding who to use the CPT with, which components to use, how, and when. It also related to how comfortable staff were in changing practice and introducing new ways of working, potentially influencing the therapeutic relationship: Knowledge and Beliefs Most participants were positive about the CPT’s ability to aid in coproducing care plans, stating that they “liked the idea of doing a support plan on a device (that was) client facing” [Practitioner 02]. Some participants used most or all of the tool features with different service users, whereas others believed the tool could only be used in certain contexts, for example, guided by service users’ needs. There was reticence among some providers to introduce a tablet computer to the therapeutic relationship: It saved me a lot of time. You don’t have to go back to the main [IT] system. [Practitioner 05] I’ve had feedback that service users have felt in the centre of the process. [...] more in control of their support [...] by being part of that process and by having the opportunity to use the tool. [Manager 04] I support quite a lot of people whose first language isn’t English [...] so it’s not so useful in that sense. [Practitioner 02] https://www.jmir.org/2020/3/e14868 J Med Internet Res 2020 \| vol. 22 \| iss. 3 \| e14868 \| p. 4 (page number not for citation purposes) https://www.jmir.org/2020/3/e14868 XSL•FO RenderX XSL•FO RenderX JOURNAL OF MEDICAL INTERNET RESEARCH Pithara et al The organization and staff were under pressure from increasing numbers of referrals. This provided an incentive for innovations that would compensate for the lack of capacity: Service users’ levels of self-awareness and stage in their illness influenced CPT use, for example, when service users were not thought to be able or ready to engage in recovery care. Another influencing factor was service users’ attitudes toward technology, with age being a related factor: Services got very, very swamped with huge numbers of referrals coming in [...] there not being sufficient capacity to manage that. [Manager 01] It depends on their level of awareness, where they are at in their recovery as well. That's quite key and to a certain extent age but not exclusively. [Practitioner 03] It depends on their level of awareness, where they are at in their recovery as well. That's quite key and to a certain extent age but not exclusively. [Practitioner 03] Pressures on the organization impacted on the time and resources staff had available to learn how to use and implement the CPT. Carrying a heavy caseload also shaped capacity for recovery-focused work, as staff had to address the service user’s immediate needs. This influenced perceptions of how the tool could support meetings with service users: (I would want to use it with) people who are fairly articulate and in touch with how they’re feeling and wanting to engage with services. [Practitioner 04] When introducing the tool to service users, reasons for declining its use included wanting to “talk to a human being” [Practitioner 03]; seeing it as a wall between themselves and providers; distrust toward technology; and thinking it did not enhance their care experience. J Med Internet Res 2020 \| vol. 22 \| iss. 3 \| e14868 \| p. 5 (page number not for citation purposes) Engaging Training provided during implementation helped support participants using the tool. Training included group or one-to-one sessions followed by feedback meetings, but some participants recruited later in the pilot did not always receive similar training: I didn’t have the training really I had, like half an hour. [Practitioner 13] One participant thought training on the interactive element of the tool was needed to guide integration into practice in a way that does not compromise the therapeutic relationship. Such aspects of using the tool were seen as important because of its objective to support collaborative service provision, coproduced with service users: Importance attached by providers to the therapeutic relationship when adopting digital health technologies highlights the need to better understand interpersonal aspects of health technologies in clinical contexts [13,18,27,37]. Perceptions of impact on, and concerns about, the therapeutic relationship influenced whether and how the CPT was used, and so did perceptions of relevance to role priorities. Staff assessments of service user needs and characteristics, for example, crisis management, literacy or attitudes toward technology, and uncertainty as to how to most successfully integrate the tool in practice, influenced providers’ choice of who to use the tool with, and how often. (the training was) very functional, what the functions are, how you log in, so it wasn’t at all about the human element or the relation element. [Practitioners 13] (the training was) very functional, what the functions are, how you log in, so it wasn’t at all about the human element or the relation element. [Practitioners 13] Information leaflets explaining what the tool is about and how to introduce the project to service users were thought to be particularly useful. We had leaflets given to us to introduce the tool, which I have to say were really good, because it gave it, it had a bit of a talking point with the service users. [Practitioners 02] We had leaflets given to us to introduce the tool, which I have to say were really good, because it gave it, it had a bit of a talking point with the service users. [Practitioners 02] We had leaflets given to us to introduce the tool, which I have to say were really good, because it gave it, it had a bit of a talking point with the service users. JOURNAL OF MEDICAL INTERNET RESEARCH JOURNAL OF MEDICAL INTERNET RESEARCH Pithara et al Factors Impacting on Implementation It has been quite difficult to use the tool because of the level of risks I’m working with. We are only supposed to use the tool with people whose recorded risk level is medium or low. And the nature of my job means that most people would have a recorded risk level of high. [Practitioner 07] Aref-Adib et al [14] state the need for better understanding of the contextual and organizational determinants of successful implementation. Our findings highlight the importance of ensuring alignment between external policy, organization and staff priorities, and CPT features to aid intervention engagement. Externally, the CQC inspection findings and increased demand for mental health services provided an incentive for organizational change in ways of working. The CPT facilitated changing practice in ways that met these pressures while aligning with organizational and professional values. Some participants thought risk should not be a limitation and the CPT could potentially be used with high-risk individuals if enough consideration was given to which aspects of the tool were used, and how it could be used during interactions: I don't think risk per se would stop me using it with someone because that's the point of it, to help people who potentially are struggling. [Practitioner 12] Other highlighted factors include tool adaptability to existing ways of working and attitudes and beliefs toward the digital innovation; stakeholder involvement in the development process is recommended to address these factors and facilitate implementation [14]. In this study, coproduction principles [20] in tool design and development supported engagement in implementation, and the tool was thought to facilitate service user–centered, recovery-focused coproduced care [27], a professional and organizational priority. Uptake was impacted by available organization resources, including IT infrastructure, staff caseloads, and time pressures; staff self-efficacy and knowledge of using the tool; service user attributes; and mobile working–related factors, including internet connectivity and IT system compatibility with the CPT. Findings reiterate the importance of considering such issues early on in digital innovation design [14,36]. Readiness for Implementation Organizational IT factors were also raised. The organization was thought to lack adequate IT infrastructure, to be paper heavy, and not incorporating technology within current practice: I’ve got a way of (working with service users) and a process that I go through probably sort of subconsciously or not really thinking about it, it’s just kind of what you do. So changing that is always a bit challenging. [Practitioner 08] It's still a very paper heavy mindset [...] there's a cultural shift that needs to happen for them to fully get on board with another bit of IT equipment. [Practitioner 03] It's still a very paper heavy mindset [...] there's a cultural shift that needs to happen for them to fully get on board with another bit of IT equipment. [Practitioner 03] J Med Internet Res 2020 \| vol. 22 \| iss. 3 \| e14868 \| p. 5 (page number not for citation purposes) https://www.jmir.org/2020/3/e14868 XSL•FO RenderX Planning This study used the CFIR to evaluate the implementation of a digital CPT in a mental health care community setting. Findings contribute to the evidence base by first adding to our understanding of organizational and contextual factors, as well as individual ones, involved when implementing digital health technologies in mental health settings [14]; and their use within the therapeutic relationship [13]. Second, it provides evidence on the experience of using the CFIR to explore implementation barriers and facilitators, adding to ongoing discussions on its use in health technology implementation research [34,35]. Staff recruitment to the pilot was led by practitioners from inside the organization, which facilitated getting staff on board. Having continuing user feedback allowed the software developers to make improvements to the CPT software. This meant that some tablet computers needed to be replaced to support updated versions of the CPT, but it also meant that some participants received new tablet computers toward the end of the pilot phase, not allowing enough time to use the new tool in their practice. Restrictions placed by the organization on which service users could be involved in the pilot, that is, individuals who were assessed to be at low/medium risk for experiencing a mental health crisis, also acted as a barrier: Engaging [Practitioners 02] Mental health service users are open to using digital health technologies [5,38], but barriers, such as, intervention complexity [14], can prevent widespread access and use among individuals with increased needs, for example, those [Practitioners 02] https://www.jmir.org/2020/3/e14868 https://www.jmir.org/2020/3/e14868 J Med Internet Res 2020 \| vol. 22 \| iss. 3 \| e14868 \| p. 6 (page number not for citation purposes) XSL•FO RenderX XSL•FO RenderX Pithara et al JOURNAL OF MEDICAL INTERNET RESEARCH quality when categorizing data items. With more treatment interventions provided in an interactive way through mobile technology, it is essential for frameworks such as the CFIR to capture the dynamic and multidimensional nature of technological interventions [10]. The complex nature of such interventions, its impact on implementation, and the CFIR’s ability to adapt to and capture this complexity should be further explored in future evaluations of such mobile mental health care interventions. experiencing psychotic symptoms or learning difficulties. At the same time, low IT and health literacy and digital inequalities among individuals with mental illness also impact on innovation uptake [14,39-41]. Our findings suggest in some cases, staff’s perceptions of service user characteristics, for example, literacy skills, may result in some service users being excluded from interactive health technologies in a care setting, but more research is needed to better understand staff decision making on this aspect, including differences in perspectives between providers from different professional backgrounds. Views of service users from underrepresented groups at risk of digital inequalities should also be explored [41]. Varsi et al [46] discussed the broadness of the CFIR that can restrict one study’s ability to capture the big picture represented in the framework, without explicitly addressing all dimensions during data collection [34]. This can be because of time limitations restricting researchers’ ability to explore all constructs within a single interview, but also recruitment limitations, when stakeholders that might represent different views are not included in the sample [46]. One way our study tackled this was to interview both practitioners using the CPT and senior managers who had a broader perspective on implementing technological innovations. Training and access to continuing support on technical and interactive aspects of the intervention during and after implementation may enhance efforts to integrate technology into routine practice [14,42,43]. Strengths, Limitations, and Conclusions One limitation of the study is that participants were sampled from those that had volunteered to take part in the pilot; these may have been providers who were more enthusiastic about using technology. Strengths include recruitment of a diverse participant sample in terms of professional roles to enable a comprehensive insight into the CPT implementation. There was often consensus in the views expressed across professional roles providing confidence of the attainment of information power. Consolidated Framework for Implementation Research Methodological Considerations Consolidated Framework for Implementation Research Methodological Considerations Theoretical grounding of implementation research allows for conclusions to be drawn as to the relevance of findings to other settings and contexts, allows comparisons, and guides further research [34]. In our study, the CFIR informed data analysis and identified factors shaping intervention implementation, following examples supporting its use in qualitative research evaluations [30,34,35]. The CFIR was useful in guiding categorization of factors and capturing overarching implementation factors involved in the CPT’s uptake. Using the framework in the analysis stage presented challenges in assigning data items to individual dimensions or constructs, and identifying which ones were the most salient in our data, an issue already raised in the literature [10]. In our case, challenges reflected the unique nature of the CPT as a tool used by both providers and service users simultaneously in the context of a care meeting. Findings highlight the value of congruence between staff, organization, and external policy priorities and digital technologies to aid intervention engagement. Only a handful of health technologies have addressed mental health recovery in community settings [10,17,18], although there is a need for health technology design interventions that follow principles of coproduction to address needs and capabilities of both staff and service users [14,16,20]. Integrating training alongside and after health technology implementation might be a way to address some of the challenges identified. Integrating an action research component within health technology implementation efforts could help to identify early training needs to address uncertainty and lack of confidence in adopting innovations, support reflexive practice, and promote effective practice change. The crucial role played by perceived impact of the technology on the therapeutic relationship highlights the importance of better understanding the ways digital health technologies impact on the therapeutic process as well as on outcomes [13]. Analytical ambiguity existed between the categories Patient needs and resources and Individual characteristics when coding service user–related factors, as both were end users of the tool in the same context and setting; the tool’s dynamic and interactive nature also made difficult distinctions between, for example, the intervention’s adaptability, complexity, and design Engaging Training that approaches use of digital innovations in a more reflexive and critical way [26] might address skepticism toward the innovation and concerns over its impact on the therapeutic relationship. When planning such activities, workload pressures and time available for staff to attend training need to be considered [14]. Training informed by action research can result in changing practice [44], and its usefulness in promoting acceptance of health technologies should be explored [45], especially because of its philosophical similarities with principles of coproduction [20]. J Med Internet Res 2020 \| vol. 22 \| iss. 3 \| e14868 \| p. 7 (page number not for citation purposes) References 1. Limb M. Digital technologies offer new ways to tackle mental health problems. Br Med J 2012 Jul 30;345:e5163. [doi: 10.1136/bmj.e5163] [Medline: 22846418] 2. Hollis C, Morriss R, Martin J, Amani S, Cotton R, Denis M, et al. Technological innovations in mental healthcare: harnessing the digital revolution. Br J Psychiatry 2015 Apr;206(4):263-265. [doi: 10.1192/bjp.bp.113.142612] [Medline: 25833865] 3. Palmier-Claus JE, Rogers A, Ainsworth J, Machin M, Barrowclough C, Laverty L, et al. Integrating mobile-phone based assessment for psychosis into people's everyday lives and clinical care: a qualitative study. BMC Psychiatry 2013 Jan 23;13:34 [FREE Full text] [doi: 10.1186/1471-244X-13-34] [Medline: 23343329] 4. Alvarez-Jimenez M, Alcazar-Corcoles MA, González-Blanch C, Bendall S, McGorry PD, Gleeson JF. Online, social media and mobile technologies for psychosis treatment: a systematic review on novel user-led interventions. Schizophr Res 2014 Jun;156(1):96-106. [doi: 10.1016/j.schres.2014.03.021] [Medline: 24746468] 5. Ennis L, Robotham D, Denis M, Pandit N, Newton D, Rose D, et al. Collaborative development of an electronic Personal Health Record for people with severe and enduring mental health problems. BMC Psychiatry 2014 Nov 18;14:305 [FREE Full text] [doi: 10.1186/s12888-014-0305-9] [Medline: 25403285] 6. Goodwin J, Cummins J, Behan L, O'Brien SM. Development of a mental health smartphone app: perspectives of mental health service users. J Ment Health 2016 Oct;25(5):434-440. [doi: 10.3109/09638237.2015.1124392] [Medline: 26732242] 6. Goodwin J, Cummins J, Behan L, O'Brien SM. Development of a mental health smartphone app: perspectives of mental health service users. J Ment Health 2016 Oct;25(5):434-440. [doi: 10.3109/09638237.2015.1124392] [Medline: 26732242] 7. Ben-Zeev D, Davis KE, Kaiser S, Krzsos I, Drake RE. Mobile technologies among people with serious mental illness: i i f f i Ad P li M H l h 2013 J l 40(4) 340 343 [FREE F ll ] [d i 7. Ben-Zeev D, Davis KE, Kaiser S, Krzsos I, Drake RE. Mobile technologies among people with serious mental illness: opportunities for future services. Adm Policy Ment Health 2013 Jul;40(4):340-343 [FREE Full text] [doi: 10.1007/s10488-012-0424-x] [Medline: 22648635] 8. Nazi KM. The personal health record paradox: health care professionals' perspectives and the information ecology of personal health record systems in organizational and clinical settings. J Med Internet Res 2013 Apr 4;15(4):e70 [FREE Full text] [doi: 10.2196/jmir.2443] [Medline: 23557596] 9. Robotham D, Mayhew M, Rose D, Wykes T. Electronic personal health records for people with severe mental illness; a feasibility study. BMC Psychiatry 2015 Aug 6;15:192 [FREE Full text] [doi: 10.1186/s12888-015-0558-y] [Medline: 26245805] 9. Acknowledgments The study was jointly funded by OHS and the National Institute for Health Research Collaboration for Leadership in Applied Health Research and Care West (NIHR CLAHRC West) at University Hospitals Bristol NHS Foundation Trust and conducted in collaboration with AWP. The views expressed are those of the authors and not necessarily those of the NHS, the NIHR, and the Department of Health and Social Care. The authors thank Sam Kozak for her co-ordination and communication role within J Med Internet Res 2020 \| vol. 22 \| iss. 3 \| e14868 \| p. 7 (page number not for citation purposes) J Med Internet Res 2020 \| vol. 22 \| iss. 3 \| e14868 \| p. 7 (page number not for citation purposes) https://www.jmir.org/2020/3/e14868 XSL•FO RenderX JOURNAL OF MEDICAL INTERNET RESEARCH Pithara et al the research. They also thank the service users involved in developing the interview topic guide and all interviewees taking part in the study. Multimedia Appendix 2 Interview topic guide. p g [DOCX File , 74 KB-Multimedia Appendix 2] p g [DOCX File , 74 KB-Multimedia Appendix 2] Multimedia Appendix 1 Care Pathway Tool Screenshots. (a) Managing my warning signs Step 2: Choose cards representing servi Managing my warning signs Step 3: Sorting cards into a timeline. (c) Goals and Actions. [PPTX File 384 KB Multimedia Appendix 1] Care Pathway Tool Screenshots. (a) Managing my warning signs Step 2: Choose cards representing service user experiences. (b) Managing my warning signs Step 3: Sorting cards into a timeline. (c) Goals and Actions. [PPTX File , 384 KB-Multimedia Appendix 1] Conflicts of Interest OHS were paid under contract by AWP to design and develop digital solutions including the CPT described in the publication from April 2015 to December 2016. OHS paid the NIHR CLAHRC West a fee for the evaluation project which represented 50% of the costs of the project. OHS own the software referred to in this publication as the Care Pathway Tool. Authors' Contributions JH, SS, MF, CP, and HE were responsible for study design and collection of data. JH, SS, WH, CG, NH, and JW were responsible for study management and co-ordination. MF and CP analyzed the data. CP drafted the paper. All authors read, commented on, and approved the final manuscript. Multimedia Appendix 3 Consolidated Framework for Implementation Research–linked factors impacting on implementation. [DOCX File , 27 KB-Multimedia Appendix 3] JOURNAL OF MEDICAL INTERNET RESEARCH Exploring working relationships in mental health care via an e-recovery portal: qualitative study on the experiences of service users and health providers. JMIR Ment Health 2017 Nov 14;4(4):e54 [FREE Full text] [doi: 10.2196/mental.8491] [Medline: 29138127] 18. Gammon D, Strand M, Eng LS, Børøsund E, Varsi C, Ruland C. Shifting practices toward recovery-oriented care through an e-recovery portal in community mental health care: a mixed-methods exploratory study. J Med Internet Res 2017 May 2;19(5):e145 [FREE Full text] [doi: 10.2196/jmir.7524] [Medline: 28465277] 19. Clark M. Co-production in mental health care. Ment Health Rev J 2015;20(4):213-219. [doi: 10.1108/MHRJ-10-2015-0030] 19. Clark M. Co-production in mental health care. Ment Health Rev J 2015;20(4):213-219. [doi: 10.1108/MHRJ-10-2015-0030] 20. Penny J, Slay J, Stephens L. Nesta \| The Innovation Foundation. London, UK: Nesta and New Economics Foundation; 2012. People Powered Health Co-Production Catalogue URL: https://media.nesta.org.uk/documents/co-production catalogue. p 20. Penny J, Slay J, Stephens L. Nesta \| The Innovation Foundation. London, UK: Nesta and New Economics Foundation; 2012. People Powered Health Co-Production Catalogue URL: https://media.nesta.org.uk/documents/co-production_catalogue. pdf [accessed 2019-07-25] p 21. Realpe A, Wallace LM. Warwick WRAP. London, United Kingdom: The Health Foundation; 2010. What is Co-Production? URL: http://wrap.warwick.ac.uk/61814/ [accessed 2020-01-02] 22. Simpson A, Hannigan B, Coffey M, Barlow S, Cohen R, Jones A, et al. Recovery-focused care planning and coordination in England and Wales: a cross-national mixed methods comparative case study. BMC Psychiatry 2016 May 16;16:147 [FREE Full text] [doi: 10.1186/s12888-016-0858-x] [Medline: 27184888] 23. Bee P, Price O, Baker J, Lovell K. Systematic synthesis of barriers and facilitators to service user-led care planning. Br J Psychiatry 2015 Aug;207(2):104-114 [FREE Full text] [doi: 10.1192/bjp.bp.114.152447] [Medline: 26243762] 23. Bee P, Price O, Baker J, Lovell K. Systematic synthesis of barriers and facilitators to service user-led care planning. Br J Psychiatry 2015 Aug;207(2):104-114 [FREE Full text] [doi: 10.1192/bjp.bp.114.152447] [Medline: 26243762] 24. Grundy AC, Bee P, Meade O, Callaghan P, Beatty S, Olleveant N, et al. Bringing meaning to user involvement in mental 23. Bee P, Price O, Baker J, Lovell K. Systematic synthesis of barriers and facilitators to service user-led care planning. Br J Psychiatry 2015 Aug;207(2):104-114 [FREE Full text] [doi: 10.1192/bjp.bp.114.152447] [Medline: 26243762] 24. Grundy AC, Bee P, Meade O, Callaghan P, Beatty S, Olleveant N, et al. Bringing meaning to user involvement in mental health care planning: a qualitative exploration of service user perspectives. J Psychiatr Ment Health Nurs 2016 Feb;23(1):12-21. [doi: 10.1111/jpm.12275] [Medline: 26634415] 24. JOURNAL OF MEDICAL INTERNET RESEARCH Pithara et al 11. Doran D, Paterson J, Clark C, Srivastava R, Goering PN, Kushniruk AW, et al. A pilot study of an electronic interprofessional evidence-based care planning tool for clients with mental health problems and addictions. Worldviews Evid Based Nurs 2010 Sep;7(3):174-184. [doi: 10.1111/j.1741-6787.2010.00191.x] [Medline: 20367805] p j 12. Clarke A, Adamson J, Sheard L, Cairns P, Watt I, Wright J. Implementing electronic patient record systems (EPRs) into England's acute, mental health and community care trusts: a mixed methods study. BMC Med Inform Decis Mak 2015 Oct 14;15:85 [FREE Full text] [doi: 10.1186/s12911-015-0204-0] [Medline: 26466894] 14;15:85 [FREE Full text] [doi: 10.1186/s12911 015 0204 0] [Medline: 26466894] 13. Hollis C, Sampson S, Simons L, Davies EB, Churchill R, Betton V, et al. Identifying research priorities for digital technology in mental health care: results of the James Lind Alliance Priority Setting Partnership. Lancet Psychiatry 2018 Oct;5(10):845-854 [doi: 10 1016/S2215-0366(18)30296-7] [Medline: 30170964] 13. Hollis C, Sampson S, Simons L, Davies EB, Churchill R, Betton V, et al. Identifying research priorities for digital technology in mental health care: results of the James Lind Alliance Priority Setting Partnership. Lancet Psychiatry 2018 Oct;5(10):845-854. [doi: 10.1016/S2215-0366(18)30296-7] [Medline: 30170964] ; ( ) [ ( ) ] [ ] 14. Aref-Adib G, McCloud T, Ross J, O'Hanlon P, Appleton V, Rowe S, et al. Factors affecting implementation of digital health interventions for people with psychosis or bipolar disorder, and their family and friends: a systematic review. Lancet Psychiatry 2019 Mar;6(3):257-266. [doi: 10.1016/S2215-0366(18)30302-X] [Medline: 30522979] 14. Aref-Adib G, McCloud T, Ross J, O'Hanlon P, Appleton V, Rowe S, et al. Factors affecting implementation of digital health interventions for people with psychosis or bipolar disorder, and their family and friends: a systematic review. Lancet Psychiatry 2019 Mar;6(3):257-266. [doi: 10.1016/S2215-0366(18)30302-X] [Medline: 30522979] 15. Maguire D, Evans H, Honeyman M, Omojomolo D. Digital Change in Health and Social Care. Fund; 2018. 16. Strand M, Gammon D, Ruland CM. Transitions from biomedical to recovery-oriented practices in mental health: a scoping review to explore the role of Internet-based interventions. BMC Health Serv Res 2017 Apr 7;17(1):257 [FREE Full text] [doi: 10.1186/s12913-017-2176-5] [Medline: 28388907] 16. Strand M, Gammon D, Ruland CM. Transitions from biomedical to recovery-oriented practices in mental health: a scoping review to explore the role of Internet-based interventions. BMC Health Serv Res 2017 Apr 7;17(1):257 [FREE Full text] [doi: 10.1186/s12913-017-2176-5] [Medline: 28388907] 17. Strand M, Gammon D, Eng LS, Ruland C. References Robotham D, Mayhew M, Rose D, Wykes T. Electronic personal health records for people with severe mental illness; a feasibility study. BMC Psychiatry 2015 Aug 6;15:192 [FREE Full text] [doi: 10.1186/s12888-015-0558-y] [Medline: 26245805] ] 10. Lord S, Moore SK, Ramsey A, Dinauer S, Johnson K. Implementation of a substance use recovery support mobile phone app in community settings: qualitative study of clinician and staff perspectives of facilitators and barriers. JMIR Ment Health 2016 Jun 28;3(2):e24 [FREE Full text] [doi: 10.2196/mental.4927] [Medline: 27352884] 10. Lord S, Moore SK, Ramsey A, Dinauer S, Johnson K. Implementation of a substance use recovery support mobile phone app in community settings: qualitative study of clinician and staff perspectives of facilitators and barriers. JMIR Ment Health 2016 Jun 28;3(2):e24 [FREE Full text] [doi: 10.2196/mental.4927] [Medline: 27352884] 10. Lord S, Moore SK, Ramsey A, Dinauer S, Johnson K. Implementation of a substance use recovery support mobile phone app in community settings: qualitative study of clinician and staff perspectives of facilitators and barriers. JMIR Ment Health 2016 Jun 28;3(2):e24 [FREE Full text] [doi: 10.2196/mental.4927] [Medline: 27352884] J Med Internet Res 2020 \| vol. 22 \| iss. 3 \| e14868 \| p. 8 (page number not for citation purposes) https://www.jmir.org/2020/3/e14868 XSL•FO RenderX XSL•FO RenderX JOURNAL OF MEDICAL INTERNET RESEARCH JOURNAL OF MEDICAL INTERNET RESEARCH Grundy AC, Bee P, Meade O, Callaghan P, Beatty S, Olleveant N, et al. Bringing meaning to user involvement in mental health care planning: a qualitative exploration of service user perspectives. J Psychiatr Ment Health Nurs 2016 Feb;23(1):12-21. [doi: 10.1111/jpm.12275] [Medline: 26634415] 25. Simpson A, Hannigan B, Coffey M, Jones A, Barlow S, Cohen R, et al. Cross-national comparative mixed-methods case study of recovery-focused mental health care planning and co-ordination: Collaborative Care Planning Project (COCAPP). Health Serv Delivery Res 2016;4(5):1-190 [FREE Full text] [doi: 10.3310/hsdr04050] [Medline: 26866206] 26. Grundy AC, Walker L, Meade O, Fraser C, Cree L, Bee P, et al. Evaluation of a co-delivered training package for community mental health professionals on service user- and carer-involved care planning. J Psychiatr Ment Health Nurs 2017 Aug;24(6):358-366 [FREE Full text] [doi: 10.1111/jpm.12378] [Medline: 28218977] g jp 27. Farr M, Pithara C, Sullivan S, Edwards H, Hall W, Gadd C, et al. Pilot implementation of co-designed software for co-production in mental health care planning: a qualitative evaluation of staff perspectives. J Ment Health 2019 Oct;28(5):495-504. [doi: 10.1080/09638237.2019.1608925] [Medline: 31240971] 27. Farr M, Pithara C, Sullivan S, Edwards H, Hall W, Gadd C, et al. Pilot implementation of co-designed software for co-production in mental health care planning: a qualitative evaluation of staff perspectives. J Ment Health 2019 Oct;28(5):495-504. [doi: 10.1080/09638237.2019.1608925] [Medline: 31240971] ( ) 28. Rethink Mental Illness. YouTube. UK: YouTube; 2017 Aug 24. Co-Production in Mental Health URL: https://www. youtube.com/watch?v=rg0lDuoUonY&feature=youtu.be [accessed 2020-01-02] ( ) 28. Rethink Mental Illness. YouTube. UK: YouTube; 2017 Aug 24. Co-Production in M youtube.com/watch?v=rg0lDuoUonY&feature=youtu.be [accessed 2020-01-02] 28. Rethink Mental Illness. YouTube. UK: YouTube; 2017 Aug 24. Co-Production in Mental Health URL: https://www. youtube.com/watch?v=rg0lDuoUonY&feature=youtu.be [accessed 2020-01-02] y g y 29. Birchwood M, Spencer E, McGovern D. Schizophrenia: early warning signs. Adv Psychiatr Treat 2000 Mar;6(2):93-101. [doi: 10.1192/apt.6.2.93] 29. Birchwood M, Spencer E, McGovern D. Schizophrenia: early warning signs. Adv Psychiatr Treat 2000 Mar;6(2):93-101. [doi: 10.1192/apt.6.2.93] 30. Damschroder LJ, Aron DC, Keith RE, Kirsh SR, Alexander JA, Lowery JC. Fostering implementation of health services research findings into practice: a consolidated framework for advancing implementation science. Implement Sci 2009 Aug 7;4:50 [FREE Full text] [doi: 10.1186/1748-5908-4-50] [Medline: 19664226] 30. Damschroder LJ, Aron DC, Keith RE, Kirsh SR, Alexander JA, Lowery JC. Fostering implementation of health services research findings into practice: a consolidated framework for advancing implementation science. Implement Sci 2009 Aug 7;4:50 [FREE Full text] [doi: 10.1186/1748-5908-4-50] [Medline: 19664226] 31. https://www.jmir.org/2020/3/e14868 JOURNAL OF MEDICAL INTERNET RESEARCH Pithara et al 34. Kirk MA, Kelley C, Yankey N, Birken SA, Abadie B, Damschroder L. A systematic review of the use of the consolidated framework for implementation research. Implement Sci 2016 May 17;11:72 [FREE Full text] [doi: 10.1186/s13012-016-0437-z] [Medline: 27189233] 35. The Consolidated Framework for Implementation Research. Michigan: CFIR Research Team; 2019. URL: https://cfirguide. org/ [accessed 2019-03-08] [WebCite Cache ID 76hTKZAAm] 36. Gammon D, Strand M, Eng LS. Service users' perspectives in the design of an online tool for assisted self-help in mental health: a case study of implications. Int J Ment Health Syst 2014 Jan 9;8(1):2 [FREE Full text] [doi: 10.1186/1752-4458-8-2] [Medline: 24405765] 37. Farr M, Banks J, Edwards HB, Northstone K, Bernard E, Salisbury C, et al. Implementing online consultations in primary care: a mixed-method evaluation extending normalisation process theory through service co-production. BMJ Open 2018 37. Farr M, Banks J, Edwards HB, Northstone K, Bernard E, Salisbury C, et al. Implementing online consultations in primary care: a mixed-method evaluation extending normalisation process theory through service co-production. BMJ Open 2018 Mar 19;8(3):e019966 [FREE Full text] [doi: 10.1136/bmjopen-2017-019966] [Medline: 29555817] 38. Ennis L, Rose D, Denis M, Pandit N, Wykes T. Can't surf, won't surf: the digital divide in mental health. J Ment Health 2012 Aug;21(4):395-403 [FREE Full text] [doi: 10.3109/09638237.2012.689437] [Medline: 22712756] 39. Clausen W, Watanabe-Galloway S, Baerentzen MB, Britigan DH. Health literacy among people with serious mental illness. Community Ment Health J 2016 May;52(4):399-405. [doi: 10.1007/s10597-015-9951-8] [Medline: 26443671] 39. Clausen W, Watanabe-Galloway S, Baerentzen MB, Britigan DH. Health literacy among people with serious mental illness. Community Ment Health J 2016 May;52(4):399-405. [doi: 10.1007/s10597-015-9951-8] [Medline: 26443671] 40 Latulippe K Hamel C Giroux D Social health inequalities and eHealth: a literature review with qualitative synthesis of 39. Clausen W, Watanabe-Galloway S, Baerentzen MB, Britigan DH. Health literacy among people with serious mental illness. Community Ment Health J 2016 May;52(4):399-405. [doi: 10.1007/s10597-015-9951-8] [Medline: 26443671] 40. Latulippe K, Hamel C, Giroux D. Social health inequalities and eHealth: a literature review with qualitative synthesis of theoretical and empirical studies. J Med Internet Res 2017 Apr 27;19(4):e136 [FREE Full text] [doi: 10.2196/jmir.6731] 40. Latulippe K, Hamel C, Giroux D. Social health inequalities and eHealth: a literature review with qualitative synthesis of theoretical and empirical studies. J Med Internet Res 2017 Apr 27;19(4):e136 [FREE Full text] [doi: 10.2196/jmir.6731] [Medline: 28450271] 41. Greer B, Robotham D, Simblett S, Curtis H, Griffiths H, Wykes T. JOURNAL OF MEDICAL INTERNET RESEARCH J Med Internet Res 2015 Nov 18;17(11):e262 [FREE Full text] [doi: 10.2196/jmir.5091] [Medline: 26582138] Abbreviations AWP: Avon and Wiltshire Mental Health Partnership NHS Trust CFIR: Consolidated Framework for Implementation Research CPT: Care Pathway Tool CQC: Care Quality Commission EPR: electronic patient record IT: information technology NHS: National Health Service NIHR CLAHRC West: National Institute for Health Research, Collaborations for Leadership in Applied Health Research and Care West OHS: Otsuka Health Solutions Edited by M Focsa; submitted 05.06.19; peer-reviewed by C Varsi, K Machin; comments to author 24.06.19; revised version received 13.08.19; accepted 21.08.19; published 19.03.20 Please cite as: Pithara C, Farr M, Sullivan SA, Edwards HB, Hall W, Gadd C, Walker J, Hebden N, Horwood J Implementing a Digital Tool to Support Shared Care Planning in Community-Based Mental Health Services: Qualitative Evaluation J Med Internet Res 2020;22(3):e14868 URL: https://www.jmir.org/2020/3/e14868 doi: 10.2196/14868 PMID: J Med Internet Res 2020 \| vol. 22 \| iss. 3 \| e14868 \| https://www.jmir.org/2020/3/e14868 (page number not for citation purp FO JOURNAL OF MEDICAL INTERNET RESEARCH Malterud K, Siersma VD, Guassora AD. Sample size in qualitative interview studies: guided by information power. Qual Health Res 2016 Nov;26(13):1753-1760. [doi: 10.1177/1049732315617444] [Medline: 26613970] 31. Malterud K, Siersma VD, Guassora AD. Sample size in qualitative interview studies: guided by information power. Qual Health Res 2016 Nov;26(13):1753-1760. [doi: 10.1177/1049732315617444] [Medline: 26613970] 32. Boyatzis RE. Transforming Qualitative Information: Thematic Analysis and Code Development. California, USA: Sage Publications, Inc; 1998. 32. Boyatzis RE. Transforming Qualitative Information: Thematic Analysis and Code Development. California, USA: Sage Publications, Inc; 1998. 33. Braun V, Clarke V. Using thematic analysis in psychology. Qual Res Psychol 2006;3(2):77-101. [doi: 10.1191/1478088706qp063oa] 33. Braun V, Clarke V. Using thematic analysis in psychology. Qual Res Psychol 2006;3(2):77-101. [doi: 10.1191/1478088706qp063oa] J Med Internet Res 2020 \| vol. 22 \| iss. 3 \| e14868 \| p. 9 (page number not for citation purposes) J Med Internet Res 2020 \| vol. 22 \| iss. 3 \| e14868 \| p. 9 (page number not for citation purposes) https://www.jmir.org/2020/3/e14868 JOURNAL OF MEDICAL INTERNET RESEARCH JOURNAL OF MEDICAL INTERNET RESEARCH Digital exclusion among mental health service users: qualitative investigation. J Med Internet Res 2019 Jan 9;21(1):e11696 [FREE Full text] [doi: 10.2196/11696] [Medline: 30626564] 42. Bredfeldt CE, Awad EB, Joseph K, Snyder MH. Training providers: beyond the basics of electronic health records. BMC Health Serv Res 2013 Dec 2;13:503 [FREE Full text] [doi: 10.1186/1472-6963-13-503] [Medline: 24295150] 43. Ross J, Stevenson F, Lau R, Murray E. Factors that influence the implementation of e-health: a systematic review of systematic reviews (an update). Implement Sci 2016 Oct 26;11(1):146 [FREE Full text] [doi: 10.1186/s13012-016-0510-7] [Medline: 27782832] 43. Ross J, Stevenson F, Lau R, Murray E. Factors that influence the implementation of e-health: a systematic review of systematic reviews (an update). Implement Sci 2016 Oct 26;11(1):146 [FREE Full text] [doi: 10.1186/s13012-016-0510-7] [Medline: 27782832] 44. Lindeman MA, Black K, Smith R, Gough J, Bryce A, Gilsenan B, et al. Changing practice in residential aged care using participatory methods. Educ Health (Abingdon) 2003 Mar;16(1):22-31. [doi: 10.1080/1357628031000066651] [Medline: 14741920] 44. Lindeman MA, Black K, Smith R, Gough J, Bryce A, Gilsenan B, et al. Changing practice in residential aged care using participatory methods. Educ Health (Abingdon) 2003 Mar;16(1):22-31. [doi: 10.1080/1357628031000066651] [Medline: 14741920] 45. Vabo G, Slettebø Å, Fossum M. Participants' perceptions of an intervention implemented in an Action Research Nursing Documentation Project. J Clin Nurs 2017 Apr;26(7-8):983-993. [doi: 10.1111/jocn.13389] [Medline: 27192412] 45. Vabo G, Slettebø Å, Fossum M. Participants' perceptions of an intervention implemented in an Action Research Nursing Documentation Project. J Clin Nurs 2017 Apr;26(7-8):983-993. [doi: 10.1111/jocn.13389] [Medline: 27192412] 46. Varsi C, Ekstedt M, Gammon D, Ruland CM. Using the consolidated framework for implementation research to identify barriers and facilitators for the implementation of an internet-based patient-provider communication service in five settings: a qualitative study. J Med Internet Res 2015 Nov 18;17(11):e262 [FREE Full text] [doi: 10.2196/jmir.5091] [Medline: 26582138] 46. Varsi C, Ekstedt M, Gammon D, Ruland CM. Using the consolidated framework for implementation research to identify barriers and facilitators for the implementation of an internet-based patient-provider communication service in five settings: a qualitative study. J Med Internet Res 2015 Nov 18;17(11):e262 [FREE Full text] [doi: 10.2196/jmir.5091] [Medline: 26582138] 46. Varsi C, Ekstedt M, Gammon D, Ruland CM. Using the consolidated framework for implementation research to identify barriers and facilitators for the implementation of an internet-based patient-provider communication service in five settings: a qualitative study. Abbreviations Edited by M Focsa; submitted 05.06.19; peer-reviewed by C Varsi, K Machin; comments to author 24.06.19; revised version received 13.08.19; accepted 21.08.19; published 19.03.20 J Med Internet Res 2020 \| vol. 22 \| iss. 3 \| e14868 \| p. 10 (page number not for citation purposes) https://www.jmir.org/2020/3/e14868 J Med Internet Res 2020 \| vol. 22 \| iss. 3 \| e14868 \| p. 11 (page number not for citation purposes) ©Christalla Pithara, Michelle Farr, Sarah A Sullivan, Hannah B Edwards, William Hall, Caroline Gadd, Julian Walker, Nick Hebden, Jeremy Horwood. Originally published in the Journal of Medical Internet Research (http://www.jmir.org), 19.03.2020. This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included. JOURNAL OF MEDICAL INTERNET RESEARCH Pithara et al ©Christalla Pithara, Michelle Farr, Sarah A Sullivan, Hannah B Edwards, William Hall, Caroline Gadd, Julian Walker, Nick Hebden, Jeremy Horwood. Originally published in the Journal of Medical Internet Research (http://www.jmir.org), 19.03.2020. This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included. https://www.jmir.org/2020/3/e14868 https://www.jmir.org/2020/3/e14868 XSL•FO RenderX XSL•FO RenderX
https://openalex.org/W2949687088	https://eprints.bbk.ac.uk/id/eprint/21420/1/21420.pdf	English	null	Extensive tonotopic mapping across auditory cortex is recapitulated by spectrally-directed attention, and systematically related to cortical myeloarchitecture	bioRxiv (Cold Spring Harbor Laboratory)	2,017	cc-by	23,597	Research Articles: Systems/Circuits Extensive tonotopic mapping across auditory cortex is recapitulated by spectrally-directed attention, and systematically related to cortical myeloarchitecture Frederic K. Dick1,2,3, Matt I. Lehet4,5, Martina F. Callaghan7, Tim A. Keller4,6, Martin I. Sereno8 and Lori L. Holt4,5 1Department of Psychological Sciences, Birkbeck College, University of London, London, WC1E 7HX 2Birkbeck-UCL Centre for Neuroimaging, London, WC1H 0AP 3Department of Experimental Psychology, University College London, London, WC1H 0AP 4Department of Psychology, Carnegie Mellon University, Pittsburgh, PA 15213 5Center for the Neural Basis of Cognition, Carnegie Mellon University, Pittsburgh, PA 15213 6Scientific Imaging & Brain Research Center, Carnegie Mellon University, Pittsburgh, PA 15213 7Wellcome Trust Center for Neuroimaging, Institute of Neurology, University College London, London, WC1N 3BG 8D t t f P h l S Di St t U i it S Di CA S Di CA 92182 4611 8Department of Psychology, San Diego State University, San Diego, CA San Diego, CA 92182-4611 BIROn - Birkbeck Institutional Research Online Dick, Frederic and Lehet, M.I. and Callaghan, M.F. and Keller, T.A. and Sereno, M.I. and Holt, L.L. (2017) Extensive Tonotopic Mapping across Auditory Cortex Is recapitulated by spectrally directed attention and systematically related to Cortical Myeloarchitecture. The Journal of Neuroscience 37 (50), pp. 12187-12201. ISSN 0270-6474. Downloaded from: https://eprints.bbk.ac.uk/id/eprint/21420/ Usage Guidelines: Please refer to usage guidelines at https://eprints.bbk.ac.uk/policies.html contact lib-eprints@bbk.ac.uk. his Accepted Manuscript has not been copyedited and formatted. The final version may differ from this version. This Accepted Manuscript has not been copyedited and formatted. The final version may differ from this version. Research Articles: Systems/Circuits DOI: 10.1523/JNEUROSCI.1436-17.2017 Received: 24 May 2017 Revised: 4 October 2017 Accepted: 6 October 2017 Published: 6 November 2017 Received: 24 May 2017 Revised: 4 October 2017 Accepted: 6 October 2017 Published: 6 November 2017 Author contributions: F.D. and L.L.H. designed research; F.D., M.I.L., T.K., and L.L.H. performed research; F.D., M.I.L., and L.L.H. analyzed data; F.D., M.F.C., M.S., and L.L.H. wrote the paper; M.F.C. and M.S. contributed unpublished reagents/analytic tools. Author contributions: F.D. and L.L.H. designed research; F.D., M.I.L., T.K., and L.L.H. performed research; F.D., M.I.L., and L.L.H. analyzed data; F.D., M.F.C., M.S., and L.L.H. wrote the paper; M.F.C. and M.S. contributed unpublished reagents/analytic tools. Conflict of Interest: The authors declare no competing financial interests. Conflict of Interest: The authors declare no competing financial interests. Research supported by Rothberg Research Award in Human Brain Imaging of Carnegie Mellon University. M.L. was supported by NIH T90DA022761 and T32GM081760. Thanks to Scott Kurdilla and Debbie Viszlay (SIBR) for imaging support, Antoine Lutti and Nikolaus Weiskopf for physics expertise and generosity with porting the MPM protocol to the SIBR Verio, and the developers of FSL, AFNI, and FreeSurfer for their scientific work and software. Finally, we are grateful to Marlene Behrmann, Jenny Bizley, Maria Chait, Tim Griffiths, Jen Linden, Catherine Perrodin, Chris Petkov, Lars Riecke, Sam Schwarzkopf, Jeremy Skipper, Ediz Sohoglu, Adam Tierney, and three anonymous reviewers of a previous version of the manuscript for extremely useful suggestions and feedback. The Wellcome Centre for Human Neuroimaging is supported by core funding from Wellcome [203147/Z/16/Z]. Corresponding Authors: Frederic K. Dick, Birkbeck/UCL Centre for NeuroImaging, 26 Bedford Way, London, WC1H 0AP, UK +44 20 7079 0815, f.dick@bbk.ac.uk; Lori L. Holt, Department of Psychology, Baker Hall, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA 15213, +1 (412) 268-4964, loriholt@cmu.edu Cite as: J. Neurosci ; 10.1523/JNEUROSCI.1436-17.2017 Alerts: Sign up at www.jneurosci.org/cgi/alerts to receive customized email alerts when the fully formatted version of this article is published. Accepted manuscripts are peer-reviewed but have not been through the copyediting, formatting, or proofreading process. p Copyright © 2017 Dick et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International license, which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed. Copyright © 2017 Dick et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International license, which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed. Extensive tonotopic mapping across auditory cortex is recapitulated by 1 spectrally-directed attention, and systematically related to cortical 2 myeloarchitecture 3 4 Frederic K. Dick 1,2,3 (ORCID ID 0000-0002-2933-3912) 5 Matt I. Lehet 4.5 (ORCID 0000-0002-7601-9360) 6 Martina F. Callaghan 7 (ORCID ID 0000-0003-0374-1659) 7 Tim A. Keller 4,6 (ORCID ID 0000-0002-2345-8626) 8 Martin I. Sereno 8 (ORCID ID 0000-0002-7598-7829) 9 Lori L. Conflict of Interest: The authors declare no competing financial interests. Holt 4,5 (ORCID ID 0000-0002-8732-4977) 10 11 1 Department of Psychological Sciences, Birkbeck College, University of London, London, 12 WC1E 7HX 13 2 Birkbeck-UCL Centre for Neuroimaging, London, WC1H 0AP 14 3 Department of Experimental Psychology, University College London, London, WC1H 0AP 15 4 Department of Psychology, Carnegie Mellon University, Pittsburgh, PA 15213 16 5 Center for the Neural Basis of Cognition, Carnegie Mellon University, Pittsburgh, PA 15213 17 6 Scientific Imaging & Brain Research Center, Carnegie Mellon University, Pittsburgh, PA 15213 18 7 Wellcome Trust Center for Neuroimaging, Institute of Neurology, University College London, 19 London, WC1N 3BG 20 8 Department of Psychology, San Diego State University, San Diego, CA San Diego, CA 92182- 21 4611 22 23 Corresponding Authors: 24 25 Frederic K. Dick, Birkbeck/UCL Centre for NeuroImaging, 26 Bedford Way, London, WC1H 26 0AP, UK +44 20 7079 0815, f.dick@bbk.ac.uk 27 28 Lori L. Holt, Department of Psychology, Baker Hall, Carnegie Mellon University, 5000 Forbes 29 Avenue, Pittsburgh, PA 15213, +1 (412) 268-4964, loriholt@cmu.edu 30 31 Abbreviated Title: Spectrally-directed attention maps in auditory cortex 32 Number of Pages: 49 33 Number of Figures: 8 34 Number of Words for Abstract: 228 35 Number of Words for Introduction: 655 36 Number of Words for Discussion: 1402 37 Conflict of Interest: The authors declare no competing financial interests. 38 39 Acknowledgements: Research supported by Rothberg Research Award in Human Brain 40 Imaging of Carnegie Mellon University. M.L. was supported by NIH T90DA022761 and 41 T32GM081760. Thanks to Scott Kurdilla and Debbie Viszlay (SIBR) for imaging support, 42 Antoine Lutti and Nikolaus Weiskopf for physics expertise and generosity with porting the MPM 43 protocol to the SIBR Verio, and the developers of FSL, AFNI, and FreeSurfer for their scientific 44 work and software. Finally, we are grateful to Marlene Behrmann, Jenny Bizley, Maria Chait, 45 Tim Griffiths, Jen Linden, Catherine Perrodin, Chris Petkov, Lars Riecke, Sam Schwarzkopf, 46 Jeremy Skipper, Ediz Sohoglu, Adam Tierney, and three anonymous reviewers of a previous 47 version of the manuscript for extremely useful suggestions and feedback. The Wellcome Centre 48 for Human Neuroimaging is supported by core funding from Wellcome [203147/Z/16/Z]. 49 50 3 Department of Experimental Psychology, University College London, London, W 15 4 Department of Psychology Carnegie Mellon University Pittsburgh PA 15213 16 5 Center for the Neural Basis of Cognition, Carnegie Mellon University, Pittsburgh, PA 15213 17 6 Scientific Imaging & Brain Research Center, Carnegie Mellon University, Pittsburgh, PA 15213 18 7 Wellcome Trust Center for Neuroimaging, Institute of Neurology, University College London, 19 London, WC1N 3BG 20 8 Department of Psychology, San Diego State University, San Diego, CA San Diego, CA 92182- 21 4611 22 23 8 Department of Psychology, San Diego State University, San Diego, CA San Diego, CA 92182- 21 4611 22 23 Frederic K. Dick, Birkbeck/UCL Centre for NeuroImaging, 26 Bedford Way, London, WC1H 26 0AP, UK +44 20 7079 0815, f.dick@bbk.ac.uk 27 28 28 Lori L. Holt, Department of Psychology, Baker Hall, Carnegie Mellon University, 5000 Forbes 29 Avenue, Pittsburgh, PA 15213, +1 (412) 268-4964, loriholt@cmu.edu 30 31 Conflict of Interest: The authors declare no competing financial interests. 38 Conflict of Interest: The authors declare no competing financial interests. 38 Acknowledgements: Research supported by Rothberg Research Award in Human Brain 40 Imaging of Carnegie Mellon University. M.L. was supported by NIH T90DA022761 and 41 T32GM081760. Thanks to Scott Kurdilla and Debbie Viszlay (SIBR) for imaging support, 42 Antoine Lutti and Nikolaus Weiskopf for physics expertise and generosity with porting the MPM 43 protocol to the SIBR Verio, and the developers of FSL, AFNI, and FreeSurfer for their scientific 44 work and software. Finally, we are grateful to Marlene Behrmann, Jenny Bizley, Maria Chait, 45 Tim Griffiths, Jen Linden, Catherine Perrodin, Chris Petkov, Lars Riecke, Sam Schwarzkopf, 46 Jeremy Skipper, Ediz Sohoglu, Adam Tierney, and three anonymous reviewers of a previous 47 version of the manuscript for extremely useful suggestions and feedback. The Wellcome Centre 48 for Human Neuroimaging is supported by core funding from Wellcome [203147/Z/16/Z]. 49 50 1 Abstract 51 Abstract 51 52 Auditory selective attention is vital in natural soundscapes. But, it is unclear how attentional 53 focus on the primary dimension of auditory representation - acoustic frequency - might modulate 54 basic auditory functional topography during active listening. In contrast to visual selective 55 attention, which is supported by motor-mediated optimization of input across saccades and pupil 56 dilation, the primate auditory system has fewer means of differentially sampling the world. This 57 makes spectrally-directed endogenous attention a particularly crucial aspect of auditory 58 attention. Using a novel functional paradigm combined with quantitative MRI, we establish in 59 male and female listeners that human frequency-band-selective attention drives activation in 60 both myeloarchitectonically-estimated auditory core, and across the majority of tonotopically- 61 mapped non-primary auditory cortex. The attentionally-driven best-frequency maps show strong 62 concordance with sensory-driven maps in the same subjects across much of the temporal 63 plane, with poor concordance in areas outside traditional auditory cortex. There is significantly 64 greater activation across most of auditory cortex when best frequency is attended, versus 65 ignored; the same regions do not show this enhancement when attending to the least-preferred 66 frequency band. Finally, the results demonstrate that there is spatial correspondence between 67 the degree of myelination and the strength of the tonotopic signal across a number of regions in 68 auditory cortex. Strong frequency preferences across tonotopically-mapped auditory cortex 69 spatially correlate with R1-estimated myeloarchitecture, indicating shared functional and 70 anatomical organization that may underlie intrinsic auditory regionalization. 71 72 52 Auditory selective attention is vital in natural soundscapes. But, it is unclear how attentional 53 focus on the primary dimension of auditory representation - acoustic frequency - might modulate 54 basic auditory functional topography during active listening. In contrast to visual selective 55 attention, which is supported by motor-mediated optimization of input across saccades and pupil 56 dilation, the primate auditory system has fewer means of differentially sampling the world. This 57 makes spectrally-directed endogenous attention a particularly crucial aspect of auditory 58 attention. Using a novel functional paradigm combined with quantitative MRI, we establish in 59 male and female listeners that human frequency-band-selective attention drives activation in 60 both myeloarchitectonically-estimated auditory core, and across the majority of tonotopically- 61 mapped non-primary auditory cortex. Abstract 51 The attentionally-driven best-frequency maps show strong 62 concordance with sensory-driven maps in the same subjects across much of the temporal 63 plane, with poor concordance in areas outside traditional auditory cortex. There is significantly 64 greater activation across most of auditory cortex when best frequency is attended, versus 65 ignored; the same regions do not show this enhancement when attending to the least-preferred 66 frequency band. Finally, the results demonstrate that there is spatial correspondence between 67 the degree of myelination and the strength of the tonotopic signal across a number of regions in 68 auditory cortex. Strong frequency preferences across tonotopically-mapped auditory cortex 69 spatially correlate with R1-estimated myeloarchitecture, indicating shared functional and 70 anatomical organization that may underlie intrinsic auditory regionalization. 71 76 2 2 Significance 77 78 Perception is an active process especially sensitive to attentional state. Listeners direct auditory 79 attention to track a violin’s melody within an ensemble performance, or to follow a voice in a 80 crowded cafe. Although diverse pathologies reduce quality of life by impacting such spectrally- 81 directed auditory attention, its neurobiological bases are unclear. We demonstrate that human 82 primary and non-primary auditory cortical activation is modulated by spectrally-directed attention 83 in a manner that recapitulates its tonotopic sensory organization. Further, the graded activation 84 profiles evoked by single frequency bands are correlated with attentionally-driven activation 85 when these bands are presented in complex soundscapes. Finally, we observe a strong 86 concordance in the degree of cortical myelination and the strength of tonotopic activation across 87 several auditory cortical regions. 88 89 3 3 Introduction 90 91 Listeners shift attention across multiple simultaneously-present acoustic dimensions to home in 92 on those that are diagnostic in guiding behavior (Idemaru and Holt, 2011; Herrmann and Henry, 93 2013; Shamma and Fritz, 2014). In nonhuman animal studies task-based spectral attention 94 adaptively modulates auditory neurons’ spectrotemporal response fields (Fritz et al., 2010). 95 Human neuroimaging results reveal that attention to streams of high- versus low-frequency 96 acoustic input can modulate activity in tonotopically-defined regions (Paltoglou et al., 2009), as 97 can imagery of higher versus lower frequencies (Oh et al., 2013). In and directly around 98 Heschl’s gyrus, there are strong frequency-band-specific attentional effects to high and low 99 pure-tone streams presented to opposite ears (Da Costa et al., 2013) and a shared topography 100 of sensory and attentionally-driven responses (Riecke et al., 2016). These results establish that 101 endogenous attention directed across acoustic frequency, the primary axis of auditory 102 representation, can modulate human cortical activity in a tonotopic manner around Heschl’s 103 gyrus. But, there remain important unanswered questions about the neurobiological basis of 104 human spectrally-directed attention. 105 Introduction 90 t oduct o 90 91 Listeners shift attention across multiple simultaneously-present acoustic dimensions to home in 92 on those that are diagnostic in guiding behavior (Idemaru and Holt, 2011; Herrmann and Henry, 93 2013; Shamma and Fritz, 2014). In nonhuman animal studies task-based spectral attention 94 adaptively modulates auditory neurons’ spectrotemporal response fields (Fritz et al., 2010). 95 Human neuroimaging results reveal that attention to streams of high- versus low-frequency 96 acoustic input can modulate activity in tonotopically-defined regions (Paltoglou et al., 2009), as 97 can imagery of higher versus lower frequencies (Oh et al., 2013). In and directly around 98 Heschl’s gyrus, there are strong frequency-band-specific attentional effects to high and low 99 pure-tone streams presented to opposite ears (Da Costa et al., 2013) and a shared topography 100 of sensory and attentionally-driven responses (Riecke et al., 2016). These results establish that 101 endogenous attention directed across acoustic frequency, the primary axis of auditory 102 representation, can modulate human cortical activity in a tonotopic manner around Heschl’s 103 gyrus. But, there remain important unanswered questions about the neurobiological basis of 104 human spectrally-directed attention. 105 First, does the topography of attention to different frequency bands recapitulate tonotopic 106 organization in human primary auditory cortex? Non-human animal physiology establishes 107 spectrally-directed attention in myelo- and cyto-architectonically-defined primary areas in 108 ‘auditory core’ (Fritz et al., 2007b; Shamma and Fritz, 2014). However, although two recent 109 neuroimaging studies have shown strong similarities between stimulus- and attentionally-driven 110 tonotopic organization in and directly around Heschl’s gyrus (Da Costa et al., 2013; Riecke et 111 al., 2016), it has not yet been possible to unambiguously localize this effect to human auditory 112 core. Here, we use high-resolution quantitative MRI (Pierpaoli, 2010) to estimate myelo- 113 4 architectonically-defined auditory core, and demonstrate that spectrally-directed attention 114 modulates its activation in a tonotopically-organized manner. 115 Second, is attentionally-driven tonotopic organization present outside of auditory core? In 116 humans, Riecke et al. (2016) found no significant evidence for tonotopically organized effects of 117 spectral attention outside of early auditory areas, but did show that the information content of 118 non-primary cortical frequency representations was sufficient for above-chance decoding of 119 listeners’ frequency-selective attentional focus. The lack of attentionally-driven tonotopic maps 120 contrasts with the finding that most non-primary cortical visual areas exhibit strong 121 retinotopically-specific attentional effects (Saygin and Sereno, 2008). Using intensive data 122 collection (>7000 functional volumes per subject) we present evidence for widespread, 123 tonotopically-organized modulation by spectral attention across much of auditory cortex, with 124 individual differences in individual participants' tonotopic maps reproduced in attentionally- 125 driven maps. 126 Third, what is the effect of frequency-selective attention being directed to a voxel's non- 127 preferred frequency band? Detailed fMRI studies of stimulus-driven frequency response 128 functions (Schönwiesner and Zatorre, 2009; Moerel et al., 2013) have shown graded and multi- 129 peaked frequency responses across human auditory cortex. However, it is unclear whether 130 these more complex patterns are recapitulated by attention to a given frequency band. In the 131 context of three distinct frequency bands, (Riecke et al., 2016) found that attentional filters 132 appeared to be bandpass in and around Heschl’s gyrus. Here, using a five-frequency-band 133 paradigm, we establish that graded response profiles evoked by single frequency bands are 134 strongly associated with attentionally-driven response profiles to those frequencies across much 135 of auditory cortex. We also show that a systematic topography of ‘dis-preferred’ frequency can 136 be driven by attention, and establish the regionalization of spectral attentional effects relative to 137 prior studies of crossmodal auditory attention (Petkov et al., 2004). 138 5 5 Finally, is there spatial correspondence between auditory cortical anatomy, as measured by the 139 local change in R1-estimated myelination, and fMRI-assessed strength of relative frequency 140 selectivity? Post-mortem Gallyas staining to establish human myeloarchitecture reveals 141 considerable variability in auditory cortical myelination that is associated with MRI signal change 142 in the same brain (Wallace et al., 2016). Likewise, variation in cortical myelination estimated 143 using T1-weighted/T2-weighted ratio approaches also appears to correspond spatially with 144 some functional variation in the superior temporal lobe (Glasser et al., 2016). Here, we 145 demonstrate that there is spatial concordance between the degree of myelination and the 146 amplitude of the frequency-selective tonotopic signal across several regions in auditory cortex. 147 Finally, is there spatial correspondence between auditory cortical anatomy, as measured by the 139 local change in R1-estimated myelination, and fMRI-assessed strength of relative frequency 140 selectivity? Post-mortem Gallyas staining to establish human myeloarchitecture reveals 141 considerable variability in auditory cortical myelination that is associated with MRI signal change 142 in the same brain (Wallace et al., 2016). Likewise, variation in cortical myelination estimated 143 using T1-weighted/T2-weighted ratio approaches also appears to correspond spatially with 144 some functional variation in the superior temporal lobe (Glasser et al., 2016). Here, we 145 demonstrate that there is spatial concordance between the degree of myelination and the 146 amplitude of the frequency-selective tonotopic signal across several regions in auditory cortex. 147 Methods 148 149 Experiment Overview. We used a novel paradigm in which listeners direct attention to a series 150 of four-tone ‘mini-sequences’ that fall within one of five possible spectral bands, without any 151 spatial cues. The task is to monitor for temporally-adjacent mini-sequence repeats within the 152 attended band. Inasmuch as this places a very high demand on encoding and integrating 153 spectral sequences within a delimited frequency range, we expect it to be especially effective in 154 evoking strong responses in non-primary auditory cortical areas. The goal is to address where, 155 specifically, in the auditory system spectral gain from attention is evident, and akin to long- 156 standing work in vision (Kastner and Ungerleider, 2000), to delineate the topographic maps 157 across which attentional modulation is apparent. 158 159 The target mini-sequences were embedded in either an informationally-sparse or 160 informationally-dense acoustic scene (Figure (Fig.) 1). Streams of four-tone mini-sequences 161 were presented in either a single band (‘tonotopy’, Fig. 1a), or accompanied by mini-sequences 162 in a ‘distractor’ frequency band, the center frequency of which varied in the frequency distance 163 Methods The tone stimuli from this condition 176 were identical to the ‘stepped’ attn-tono condition, but the order of the verbal cues directing 177 listeners’ attention to a specific frequency band was scrambled in their assignment to blocks. 178 This preserved the acoustics across conditions, but eliminated the consistent ‘stepping’ of 179 attention through the frequency spectrum in the randomized condition thereby destroying the 180 consistent phase-lag associated with specific frequency bands that support Fourier analyses 181 [schematized in Fig. 1e, and see below]. Thus, to the extent that there are attentionally-driven 182 frequency-selective maps in auditory cortex we expect tonotopically-organized attentional maps 183 to be apparent in the 'stepped,' but not the 'randomized' attn-tono conditions under Fourier 184 analyses. In contrast, regression analyses include a model of attention, allowing ‘stepped’ and 185 ‘randomized’ attn-tono conditions to be pooled to investigate the impact of attention on cortical 186 activation. Across both Fourier and regression analyses, the ‘stepped’ attn-tono conditions were 187 collapsed across runs for which the cued frequency band stepped up in frequency and those 188 from the attended band across blocks (‘attention-tonotopy’, abbreviated 'attn-tono', Fig. 1b). A 164 verbal cue directed listeners’ attention to a specific frequency band, within which listeners 165 monitored four-tone mini-sequences for repeats; the ‘distractor’ band in attn-tono blocks also 166 contained repeats. Using a discretized version of a phase-encoded fMRI design (Sereno et al., 167 1995; Schwarzkopf et al., 2011; Rao and Talavage, 2005; Herdener et al., 2013; Langers et al., 168 2014), the cued frequency band stepped up or down in orderly steps across the acoustic 169 spectrum across a 64-sec cycle [Fig. 1c]. Phase-encoded tonotopic designs benefit from the 170 power and robustness of the 'travelling wave' method for topographic cortical mapping of 171 smoothly varying representations (Engel, 2012); the discretized (blocked) version we use here 172 allows use of the verbal cue, and has the advantage of being able to be analyzed using both 173 Fourier and regression approaches. This allowed us to include an additional, ‘randomized’ attn- 174 tono condition that contributed both as a control condition in Fourier analyses, and also as an 175 additional attn-tono run in regression analyses (see Fig. 1d). Methods Experiment Overview. We used a novel paradigm in which listeners direct attention to a series 150 of four-tone ‘mini-sequences’ that fall within one of five possible spectral bands, without any 151 spatial cues. The task is to monitor for temporally-adjacent mini-sequence repeats within the 152 attended band. Inasmuch as this places a very high demand on encoding and integrating 153 spectral sequences within a delimited frequency range, we expect it to be especially effective in 154 evoking strong responses in non-primary auditory cortical areas. The goal is to address where, 155 specifically, in the auditory system spectral gain from attention is evident, and akin to long- 156 standing work in vision (Kastner and Ungerleider, 2000), to delineate the topographic maps 157 across which attentional modulation is apparent. 158 The target mini-sequences were embedded in either an informationally-sparse or 160 informationally-dense acoustic scene (Figure (Fig.) 1). Streams of four-tone mini-sequences 161 were presented in either a single band (‘tonotopy’, Fig. 1a), or accompanied by mini-sequences 162 in a ‘distractor’ frequency band, the center frequency of which varied in the frequency distance 163 6 from the attended band across blocks (‘attention-tonotopy’, abbreviated 'attn-tono', Fig. 1b). A 164 verbal cue directed listeners’ attention to a specific frequency band, within which listeners 165 monitored four-tone mini-sequences for repeats; the ‘distractor’ band in attn-tono blocks also 166 contained repeats. Using a discretized version of a phase-encoded fMRI design (Sereno et al., 167 1995; Schwarzkopf et al., 2011; Rao and Talavage, 2005; Herdener et al., 2013; Langers et al., 168 2014), the cued frequency band stepped up or down in orderly steps across the acoustic 169 spectrum across a 64-sec cycle [Fig. 1c]. Phase-encoded tonotopic designs benefit from the 170 power and robustness of the 'travelling wave' method for topographic cortical mapping of 171 smoothly varying representations (Engel, 2012); the discretized (blocked) version we use here 172 allows use of the verbal cue, and has the advantage of being able to be analyzed using both 173 Fourier and regression approaches. This allowed us to include an additional, ‘randomized’ attn- 174 tono condition that contributed both as a control condition in Fourier analyses, and also as an 175 additional attn-tono run in regression analyses (see Fig. 1d). Methods The tone stimuli from this condition 176 were identical to the ‘stepped’ attn-tono condition, but the order of the verbal cues directing 177 listeners’ attention to a specific frequency band was scrambled in their assignment to blocks. 178 This preserved the acoustics across conditions, but eliminated the consistent ‘stepping’ of 179 attention through the frequency spectrum in the randomized condition thereby destroying the 180 consistent phase-lag associated with specific frequency bands that support Fourier analyses 181 [schematized in Fig. 1e, and see below]. Thus, to the extent that there are attentionally-driven 182 frequency-selective maps in auditory cortex we expect tonotopically-organized attentional maps 183 to be apparent in the 'stepped,' but not the 'randomized' attn-tono conditions under Fourier 184 analyses. In contrast, regression analyses include a model of attention, allowing ‘stepped’ and 185 ‘randomized’ attn-tono conditions to be pooled to investigate the impact of attention on cortical 186 activation. Across both Fourier and regression analyses, the ‘stepped’ attn-tono conditions were 187 collapsed across runs for which the cued frequency band stepped up in frequency and those 188 7 7 that stepped down; inclusion of each simply balanced the directional movement of attention 189 through the acoustic spectrum across the experiment. 190 191 In summary, in the attn-tono conditions, attention alone was available to differentially drive 192 responses to an approximately constant acoustic input, whereas in the tonotopy condition, 193 responses were driven by spectrally-selective stimuli as well as by attention. 194 195 We analyzed mapping data using Fourier methods with individual and surface-based group 196 analysis methods, as described previously (Sereno et al., 1995; Hagler et al., 2006; 2007). With 197 this approach, voxels preferentially responding to a certain phase in a ‘stepped’ stimulus cycle 198 are defined as those that have a significantly higher signal amplitude at this stimulus temporal 199 frequency (meaning the slow frequency of the repeat of the spectral ramp) than at the average 200 of other 'noise' frequencies (see Fig. 1e). Significant signal phases (a particular position in the 201 cycle) are then mapped to a color scale to indicate the voxel’s ‘best frequency’ and signal 202 amplitude is mapped to the voxel’s color saturation (Fig. 1f). We time-reversed runs stepping 203 down in frequency and averaged them with runs stepping up in frequency (Sereno et al., 1995; 204 Talavage et al., 2004; Dick et al., 2012; Ahveninen et al., 2016). Cross-subject averaging of 205 phase-encoded mapping data was performed using a method described previously (Hagler et 206 al., 2007) in which the real and imaginary components of the signal with respect to the stepped 207 cycle were sampled to the cortical surface and then averaged across subjects, preserving any 208 phase information that was coherent over subjects. 209 210 Using previously established methods (Dick et al., 2012; Sereno et al., 2013; Lutti et al., 2014) 211 that stepped down; inclusion of each simply balanced the directional movement of attention 189 through the acoustic spectrum across the experiment. 190 that stepped down; inclusion of each simply balanced the directional movement of attention 189 through the acoustic spectrum across the experiment. 190 that stepped down; inclusion of each simply balanced the directional movement of attention 189 through the acoustic spectrum across the experiment. 190 In summary, in the attn-tono conditions, attention alone was available to differentially drive 192 responses to an approximately constant acoustic input, whereas in the tonotopy condition, 193 responses were driven by spectrally-selective stimuli as well as by attention. 194 We analyzed mapping data using Fourier methods with individual and surface-based group 196 analysis methods, as described previously (Sereno et al., 1995; Hagler et al., 2006; 2007). With 197 this approach, voxels preferentially responding to a certain phase in a ‘stepped’ stimulus cycle 198 are defined as those that have a significantly higher signal amplitude at this stimulus temporal 199 frequency (meaning the slow frequency of the repeat of the spectral ramp) than at the average 200 of other 'noise' frequencies (see Fig. 1e). Significant signal phases (a particular position in the 201 cycle) are then mapped to a color scale to indicate the voxel’s ‘best frequency’ and signal 202 amplitude is mapped to the voxel’s color saturation (Fig. 1f). We time-reversed runs stepping 203 down in frequency and averaged them with runs stepping up in frequency (Sereno et al., 1995; 204 Talavage et al., 2004; Dick et al., 2012; Ahveninen et al., 2016). Cross-subject averaging of 205 phase-encoded mapping data was performed using a method described previously (Hagler et 206 al., 2007) in which the real and imaginary components of the signal with respect to the stepped 207 cycle were sampled to the cortical surface and then averaged across subjects, preserving any 208 phase information that was coherent over subjects. 209 Using previously established methods (Dick et al., 2012; Sereno et al., 2013; Lutti et al., 2014) 211 see also (Glasser et al., 2016), we used high-resolution quantitative multi-parameter mapping to 212 generate maps of estimated cortical myelination based on longitudinal relaxation times 213 (quantitative T1). Recent work by multiple labs supports the hypothesis that T1 relaxation is 214 Using previously established methods (Dick et al., 2012; Sereno et al., 2013; Lutti et al., 2014) 211 see also (Glasser et al., 2016), we used high-resolution quantitative multi-parameter mapping to 212 generate maps of estimated cortical myelination based on longitudinal relaxation times 213 (quantitative T1). Recent work by multiple labs supports the hypothesis that T1 relaxation is 214 8 8 reliably associated with quantitative differences in myelination in white matter and cortex 215 (Sereno et al., 2013; Callaghan et al., 2014; Stüber et al., 2014; Dinse et al., 2015; Tardif et al., 216 2015; Turner, 2015; Tardif et al., 2016). Here, we calculated each subject’s R1 (1/T1) values, 217 where the greater the R1, the higher the inferred myelin content. These R1 values were 218 resampled onto his or her surface at a cortical depth fraction of 0.5, and also averaged across 219 individuals using sulcus-aligned cortical-surface-based procedures (see below for further 220 details). 221 222 Participants. Eight adults (aged 23-45 y, mean 28 y; 6 female) participated; none reported a 223 history of neurological disease or communication disorders. All had some childhood and/or adult 224 musical training (one had a music degree), and had previous experience with longer scanning 225 sessions. While musical training seemed to facilitate learning the experimental task, subsequent 226 behavioral studies in the lab have shown that musically-naive subjects can also attain excellent 227 performance with similar levels of training on this and even more demanding related tasks. 228 229 Stimuli and Design. Stimuli were created using custom code in Matlab version 2015a 230 (Mathworks, Inc.) and SoX version 14.4.2 (sourceforge.net). The basic stimulus unit was a four- 231 tone mini-sequence (140 ms sine-wave tones including 10 ms linear amplitude ramp), with each 232 tone drawn with replacement from a seven-semitone, band-delimited pool centered around one 233 of five frequencies (300, 566, 1068, 2016, 3805 Hz; Fig. 1a). Fourteen mini-sequences formed a 234 block (mean inter-sequence silent interval 240 ms, SD 10 ms). Each block contained one to 235 three mini-sequence repeats (1:2:1 ratio of 1, 2, and 3 repeats). When there was more than one 236 repeat per block, mini-sequence repeat pairs were separated by at least one intervening mini- 237 sequence. Each block began with a verbal prompt: ‘hear’, ‘high’, or ‘low’; loudness-equalized 238 Mac ‘Victoria’ voice, mean duration 506 ms (SD 36 ms), padded with silence to 800 ms total 239 reliably associated with quantitative differences in myelination in white matter and cortex 215 (Sereno et al., 2013; Callaghan et al., 2014; Stüber et al., 2014; Dinse et al., 2015; Tardif et al., 216 2015; Turner, 2015; Tardif et al., 2016). Here, we calculated each subject’s R1 (1/T1) values, 217 where the greater the R1, the higher the inferred myelin content. These R1 values were 218 resampled onto his or her surface at a cortical depth fraction of 0.5, and also averaged across 219 individuals using sulcus-aligned cortical-surface-based procedures (see below for further 220 details). 221 reliably associated with quantitative differences in myelination in white matter and cortex 215 (Sereno et al., 2013; Callaghan et al., 2014; Stüber et al., 2014; Dinse et al., 2015; Tardif et al., 216 2015; Turner, 2015; Tardif et al., 2016). Here, we calculated each subject’s R1 (1/T1) values, 217 where the greater the R1, the higher the inferred myelin content. These R1 values were 218 resampled onto his or her surface at a cortical depth fraction of 0.5, and also averaged across 219 individuals using sulcus-aligned cortical-surface-based procedures (see below for further 220 details). 221 222 Participants. Eight adults (aged 23-45 y, mean 28 y; 6 female) participated; none reported a 223 history of neurological disease or communication disorders. All had some childhood and/or adult 224 musical training (one had a music degree), and had previous experience with longer scanning 225 sessions. While musical training seemed to facilitate learning the experimental task, subsequent 226 behavioral studies in the lab have shown that musically-naive subjects can also attain excellent 227 performance with similar levels of training on this and even more demanding related tasks. 228 229 Participants. Eight adults (aged 23-45 y, mean 28 y; 6 female) participated; none reported a 223 history of neurological disease or communication disorders. All had some childhood and/or adult 224 musical training (one had a music degree), and had previous experience with longer scanning 225 sessions. While musical training seemed to facilitate learning the experimental task, subsequent 226 behavioral studies in the lab have shown that musically-naive subjects can also attain excellent 227 performance with similar levels of training on this and even more demanding related tasks. 228 Stimuli and Design. Stimuli were created using custom code in Matlab version 2015a 230 (Mathworks, Inc.) and SoX version 14.4.2 (sourceforge.net). The basic stimulus unit was a four- 231 tone mini-sequence (140 ms sine-wave tones including 10 ms linear amplitude ramp), with each 232 tone drawn with replacement from a seven-semitone, band-delimited pool centered around one 233 of five frequencies (300, 566, 1068, 2016, 3805 Hz; Fig. 1a). Fourteen mini-sequences formed a 234 block (mean inter-sequence silent interval 240 ms, SD 10 ms). Each block contained one to 235 three mini-sequence repeats (1:2:1 ratio of 1, 2, and 3 repeats). When there was more than one 236 repeat per block, mini-sequence repeat pairs were separated by at least one intervening mini- 237 sequence. Each block began with a verbal prompt: ‘hear’, ‘high’, or ‘low’; loudness-equalized 238 Mac ‘Victoria’ voice, mean duration 506 ms (SD 36 ms), padded with silence to 800 ms total 239 9 9 duration. This prompt was followed by 800 ms silent gap (tonotopy) or tone-cue (attn-tono), then 240 the 14 mini-sequences (11.2s in total), for a total block duration of 12.8s. 241 The task was to detect mini-sequence repeats in the attended frequency band (i.e., a 1-back 243 task). In the tonotopy condition, mini-sequences were confined to a single frequency band 244 preceded by the neutral verbal prompt ‘hear’ (Fig. 1a). In two of the four single-band tonotopy 245 runs, block center frequency was stepped from low-to-high over a 64-second cycle with 8 246 cycles/run; step direction was reversed (high-to-low) for the other two runs. This is a ‘discrete’ 247 version of phase-encoded designs commonly used in visual, somatosensory, and auditory 248 mapping studies (Engel et al., 1994; Sereno et al., 1995; Da Costa et al., 2011; Dick et al., 249 2012; Langers and van Dijk, 2012; Langers et al., 2014; Saenz and Langers, 2014). 250 251 The attn-tono condition had the exact mini-sequence patterns from the tonotopy blocks, but 252 there also were simultaneous, competing mini-sequences in a distinct frequency band with a 253 center frequency at least 14 semitones apart (Fig. 1b; 300 vs. 1068 Hz; 300 vs. 2016 Hz; 300 254 vs. 3805 Hz; 566 vs. 2016 Hz; 566 vs. 3805 Hz; 1068 vs. 3805 Hz; not all center frequencies 255 were paired due to the 14-semitone constraint). The verbal prompt (high or low) initiating each 256 attn-tono block signaled participants to perform the 1-back task on either the higher or lower 257 frequency band. Immediately after the verbal prompt, a randomly-ordered pair of sine-wave 258 tones cued the center frequencies of the upcoming block (140 ms tones including 8 ms linear 259 on/off amplitude ramp; 26 ms inter-tone silence, tone pair followed by 494 ms silence, total 260 duration 800 ms). Crucially, there were mini-sequence repeats even in the unattended band to 261 assure that attention directed to the task was endogenously driven rather than being attracted 262 by stimulus repetition effects (Barascud et al., 2016). 263 10 10 There were two attn-tono conditions: ‘stepped’ and ‘randomized’. Analogous to single-band 265 tonotopy runs, in stepped attn-tono runs the verbally-cued frequency band implicitly stepped up 266 (2 runs) or down (2 runs) in frequency over a 64-sec cycle (Fig. 1c). This cued iterative stepping 267 through the frequency spectrum facilitates transfer of attention to each frequency band (as in 268 traditional phase-encoded designs) and supports Fourier approaches to analysis (Fig. 1e). Each 269 randomized attn-tono run was acoustically identical to a stepped run, but the verbal prompt was 270 manipulated so that there was no systematic, stepped organization of mini-sequence center 271 frequencies through the spectrum (Fig. 1d). For this condition, frequency bands were cued at 272 inconsistent phase lags within the 8 cycles/run, thereby phase-canceling any periodic attentional 273 response; this is schematized in Fig 1e. This randomized-order control is important, as there is 274 a small (~1 octave) overall shift in spectral mean over the course of an attn-tono stimulus cycle 275 that is unavoidable due to the constraints on the pairing of frequency bands. 276 11 There were two attn-tono conditions: ‘stepped’ and ‘randomized’. Analogous to single-band 265 tonotopy runs, in stepped attn-tono runs the verbally-cued frequency band implicitly stepped up 266 (2 runs) or down (2 runs) in frequency over a 64-sec cycle (Fig. 1c). This cued iterative stepping 267 through the frequency spectrum facilitates transfer of attention to each frequency band (as in 268 traditional phase-encoded designs) and supports Fourier approaches to analysis (Fig. 1e). Each 269 randomized attn-tono run was acoustically identical to a stepped run, but the verbal prompt was 270 manipulated so that there was no systematic, stepped organization of mini-sequence center 271 frequencies through the spectrum (Fig. 1d). For this condition, frequency bands were cued at 272 inconsistent phase lags within the 8 cycles/run, thereby phase-canceling any periodic attentional 273 response; this is schematized in Fig 1e. This randomized-order control is important, as there is 274 a small (~1 octave) overall shift in spectral mean over the course of an attn-tono stimulus cycle 275 that is unavoidable due to the constraints on the pairing of frequency bands. 276 277 Each of the twelve 9.6-minute-long runs was composed of eight 64s cycles plus 32-sec silent 278 periods at the beginning and end of each run to allow for calculation of baseline auditory 279 activation (Klein et al., 2014). 280 281 Behavioral Thresholds and Training. Participants first underwent behavioral tests of monaural 282 pure-tone thresholds and binaural thresholds for detecting mini-sequences in quiet and in 283 acoustic noise generated by the MRI scanner running the multiband EPI sequence. This 284 provided a basis for adjusting center frequency amplitudes to approximate equal loudness in 285 scanner noise. Participants also trained on the mini-sequence detection task in quiet and in 286 acoustic scanner noise across two sessions. 287 288 Imaging Data Acquisition. Structural and functional images were acquired on a 3-Tesla 289 Siemens Verio wide-bore MRI scanner at the Scientific Imaging and Brain Research (SIBR) 290 There were two attn-tono conditions: ‘stepped’ and ‘randomized’. Analogous to single-band 265 tonotopy runs, in stepped attn-tono runs the verbally-cued frequency band implicitly stepped up 266 (2 runs) or down (2 runs) in frequency over a 64-sec cycle (Fig. 1c). This cued iterative stepping 267 through the frequency spectrum facilitates transfer of attention to each frequency band (as in 268 traditional phase-encoded designs) and supports Fourier approaches to analysis (Fig. 1e). Each 269 randomized attn-tono run was acoustically identical to a stepped run, but the verbal prompt was 270 manipulated so that there was no systematic, stepped organization of mini-sequence center 271 frequencies through the spectrum (Fig. 1d). For this condition, frequency bands were cued at 272 inconsistent phase lags within the 8 cycles/run, thereby phase-canceling any periodic attentional 273 response; this is schematized in Fig 1e. This randomized-order control is important, as there is 274 a small (~1 octave) overall shift in spectral mean over the course of an attn-tono stimulus cycle 275 that is unavoidable due to the constraints on the pairing of frequency bands. 276 277 Each of the twelve 9.6-minute-long runs was composed of eight 64s cycles plus 32-sec silent 278 periods at the beginning and end of each run to allow for calculation of baseline auditory 279 activation (Klein et al., 2014). 280 Behavioral Thresholds and Training. Participants first underwent behavioral tests of monaural 282 pure-tone thresholds and binaural thresholds for detecting mini-sequences in quiet and in 283 acoustic noise generated by the MRI scanner running the multiband EPI sequence. This 284 provided a basis for adjusting center frequency amplitudes to approximate equal loudness in 285 scanner noise. Participants also trained on the mini-sequence detection task in quiet and in 286 acoustic scanner noise across two sessions. 287 Behavioral Thresholds and Training. Participants first underwent behavioral tests of monaural 282 pure-tone thresholds and binaural thresholds for detecting mini-sequences in quiet and in 283 acoustic noise generated by the MRI scanner running the multiband EPI sequence. This 284 provided a basis for adjusting center frequency amplitudes to approximate equal loudness in 285 scanner noise. Participants also trained on the mini-sequence detection task in quiet and in 286 acoustic scanner noise across two sessions. 287 Imaging Data Acquisition. Structural and functional images were acquired on a 3-Tesla 289 Siemens Verio wide-bore MRI scanner at the Scientific Imaging and Brain Research (SIBR) 290 11 Center at Carnegie Mellon University using a phased array 32-channel head coil across three 291 scan sessions on separate days. Stimulus presentation was under the control of a MacPro 292 running PsychToolbox 3.0.12 in Matlab (The Mathworks, Inc.), with audio output to an external 293 AD/DA converter (Babyface, RME) connected to an amplifier (Pylepro) that delivered stimuli to 294 participants in the scanner diotically over MRI-compatible earbuds (Sensimetrics S14). All 295 stimuli were pre-filtered to equalize sound stimuli according to the earbuds’ frequency response. 296 After participants were settled into the bore, sound volume was adjusted so that participants 297 could comfortably hear all frequencies through scanner noise. Participants wore a fiber optic 298 response glove (Current Designs) that communicated with a Brain Logics I/O device 299 (Psychology Software Tools, Inc); participants used the glove to respond to mini-sequence 300 repeats using the right index finger. During all functional scans, subjects closed their eyes to 301 reduce the potential for stimulus-correlated eye movements. 302 In the initial scanning session (~50 min), we acquired multi-parameter mapping (MPM) images 304 for quantitative myelin mapping and structural identification of primary auditory cortex on an 305 individual basis while participants watched a film. Proton density-weighted (PDw), T1-weighted 306 (T1w), and magnetization transfer (MTw) images were acquired using an in-house 3D FLASH 307 pulse sequence (voxel size: 0.8 x 0.8 x 0.8 mm3, matrix = 320 x 280 x 208, TR = 25.0 ms, 308 bandwidth 488 Hz/px, excitation flip angle: 6o (PDw/MTw) or 21o (T1w), slab rotation 30o). To 309 accelerate this high resolution acquisition, a partial Fourier acquisition (6/8 coverage) was used 310 in the inner phase-encoded direction (RL) and parallel imaging was used along the outer phase 311 encoding direction (AP), reconstructed using the GRAPPA algorithm (acceleration factor 2, 18 312 integrated auto-calibration lines) as implemented on the scanner platform. Four gradient echoes 313 were acquired for each contrast (TE=2.5, 4.74, 6.98, 9.22 ms) after each excitation pulse and 314 averaged to improve SNR (Helms et al., 2009). Each FLASH acquisition lasted 9 minutes 45 315 seconds. Quantitative R1 (=1/T1) maps were estimated from the PDw and T1w images 316 12 according to the model developed by Helms et al. (Helms et al., 2008) including a correction for 317 RF transmit field inhomogeneities (Lutti et al., 2010) and imperfect spoiling (Preibisch and 318 Deichmann, 2009). The transmit field map was calculated using a 3D EPI spin-echo 319 (SE)/stimulated echo (STE) method (Lutti et al., 2010; 2012); FOV = 256 x 192 x 192 mm, 320 matrix = 64 x 64 x 48, TE = 53.14 ms, TM = 47.60 ms, TR = 500 ms, bandwidth = 2298, nominal 321 α varying from 135o to 65o in steps of 5o, acquisition time 6 minutes) and was corrected for off- 322 resonance effects using a standard B0 field map (double gradient echo FLASH, 3x3x2 mm 323 isotropic resolution, whole-brain coverage). 324 Image Preprocessing. Cortical surface creation, and mapping of R1 values. Each subject’s 345 cortical surface was reconstructed from a contrast-optimized synthetic FLASH volume, created 346 with mri_synthesize in Freesurfer from scaled and truncated versions of the T1 and proton- 347 density volumes; another MPRAGE-like synthetic image was created for use with the automated 348 Freesurfer Talairach procedure. Both volumes were conformed to 1mm isotropic resolution and 349 used in a customized reconstruction pipeline version. (In particular, the subject’s PD volume 350 was used to deskull the synthetic FLASH image using a ‘shrink-wrap’ technique (Dale and 351 Sereno, 1993). After inspection for reconstruction quality, R1 values were resampled from 50% 352 cortical depth fraction to the subject’s surface, and also morphed to the unit icosahedron for 353 cross-subject curvature-aligned cortical surface based averaging (Fischl et al., 1999). 354 355 Image Preprocessing. Cortical surface creation, and mapping of R1 values. Each subject’s 345 cortical surface was reconstructed from a contrast-optimized synthetic FLASH volume, created 346 with mri_synthesize in Freesurfer from scaled and truncated versions of the T1 and proton- 347 density volumes; another MPRAGE-like synthetic image was created for use with the automated 348 Freesurfer Talairach procedure. Both volumes were conformed to 1mm isotropic resolution and 349 used in a customized reconstruction pipeline version. (In particular, the subject’s PD volume 350 was used to deskull the synthetic FLASH image using a ‘shrink-wrap’ technique (Dale and 351 Sereno, 1993). After inspection for reconstruction quality, R1 values were resampled from 50% 352 cortical depth fraction to the subject’s surface, and also morphed to the unit icosahedron for 353 cross-subject curvature-aligned cortical surface based averaging (Fischl et al., 1999). 354 355 EPI processing. Each functional image from both sessions was aligned to a reference volume 356 from the middle of the first run using AFNI’s 3dvolreg; registration and motion correction 357 goodness were hand-checked for each run. The reference volume was aligned to the subject’s 358 cortical surface using boundary-based registration in Freesurfer (Greve and Fischl, 2009), 359 verified using manual blink comparison, and applied to the volume-aligned EPI data for 360 resampling. EPI data were analyzed in native space without any spatial smoothing using both 361 Fourier and general linear methods. 362 EPI processing. The final two scanning sessions acquired functional data for four runs each of the tonotopy, 326 ‘stepped’ attn-tono, and ‘randomized’ attn-tono conditions. The runs were interleaved across 327 conditions and designed to assess phase-encoded functional influences of selective attention 328 across frequency (‘stepped’ attn-tono), the functional response to identical acoustics without 329 systematic phase encoded shifts of attention (‘randomized’ attn-tono), and functional responses 330 to single frequency bands identical to the attended bands in attn-tono, with phase-encoded 331 steps through frequency and no distractor frequency bands (tonotopy). Across all functional 332 runs, participants engaged in detecting repeats (1-back) of the four-tone mini-sequences. Run 333 order was counterbalanced according to condition and whether the cycle involved steps up or 334 down in frequency. 335 Functional images were acquired using a T2*-weighted echo-planar imaging (EPI) pulse 337 sequence (44 oblique axial slices, in plane resolution 3 mm x 3 mm, 3 mm slice thickness, no 338 gap, repetition time TR = 1000 ms, echo time TE = 41 ms, flip angle = 61°, matrix size = 64 x 339 64, field of view FOV = 192 mm). All EPI functional scans were performed using 4x multi-band 340 acceleration (Feinberg et al., 2010; Feinberg and Setsompop, 2013). There were 584 repetitions 341 13 acquired per run, with the first 8 images discarded to allow for longitudinal magnetization to 342 arrive at equilibrium. Runs were pseudo-randomly ordered across participants. 343 Each functional image from both sessions was aligned to a reference volume 356 from the middle of the first run using AFNI’s 3dvolreg; registration and motion correction 357 goodness were hand-checked for each run. The reference volume was aligned to the subject’s 358 cortical surface using boundary-based registration in Freesurfer (Greve and Fischl, 2009), 359 verified using manual blink comparison, and applied to the volume-aligned EPI data for 360 resampling. EPI data were analyzed in native space without any spatial smoothing using both 361 Fourier and general linear methods. 362 EPI processing. Each functional image from both sessions was aligned to a reference volume 356 from the middle of the first run using AFNI’s 3dvolreg; registration and motion correction 357 goodness were hand-checked for each run. The reference volume was aligned to the subject’s 358 cortical surface using boundary-based registration in Freesurfer (Greve and Fischl, 2009), 359 verified using manual blink comparison, and applied to the volume-aligned EPI data for 360 resampling. EPI data were analyzed in native space without any spatial smoothing using both 361 Fourier and general linear methods. 362 Experimental Design and Statistical Analysis. As noted above, the fMRI experiment used a 364 'discrete' version of a traditional phase-encoded design, such that both Fourier-based and 365 general linear model approaches could be used. Fourier analyses were carried out in csurf 366 (http://www.cogsci.ucsd.edu/~sereno/.tmp/dist/csurf) with individual and group analysis methods 367 Experimental Design and Statistical Analysis. As noted above, the fMRI experiment used a 364 'discrete' version of a traditional phase-encoded design, such that both Fourier-based and 365 general linear model approaches could be used. Fourier analyses were carried out in csurf 366 (http://www.cogsci.ucsd.edu/~sereno/.tmp/dist/csurf) with individual and group analysis methods 367 Experimental Design and Statistical Analysis. As noted above, the fMRI experiment used a 364 'discrete' version of a traditional phase-encoded design, such that both Fourier-based and 365 general linear model approaches could be used. Fourier analyses were carried out in csurf 366 (http://www.cogsci.ucsd.edu/~sereno/.tmp/dist/csurf) with individual and group analysis methods 367 14 employed as previously described (Sereno et al., 1995; Sereno and Huang, 2006; Hagler et al., 368 2007). Functional activation amplitude was estimated as the Fourier amplitude of the periodic 369 BOLD signal (proportional to percent response) at the frequency of the stimulus cycle (8 370 repetitions per run). An F statistic was calculated by comparing that amplitude to the average 371 amplitude of other ‘noise’ frequencies (Hagler et al., 2007). Periodic signal components with 372 very low frequencies (due to slow head motion) and the second and third harmonic of the 373 stimulus were excluded as neither signal nor noise (this is mathematically equivalent to first 374 linearly regressing out these frequencies as nuisance variables before calculating significance). 375 The phase of the signal, which corresponds to a particular point of the stimulus cycle, was then 376 mapped to a color scale and the amplitude of the signal at each vertex was mapped to color 377 saturation (Gouraud sharing within each face). Runs with downward frequency steps were time- 378 reversed and averaged with upward-stepped scans in order to cancel fixed voxel-specific delays 379 in the BOLD response. 380 employed as previously described (Sereno et al., 1995; Sereno and Huang, 2006; Hagler et al., 368 2007). Functional activation amplitude was estimated as the Fourier amplitude of the periodic 369 BOLD signal (proportional to percent response) at the frequency of the stimulus cycle (8 370 repetitions per run). An F statistic was calculated by comparing that amplitude to the average 371 amplitude of other ‘noise’ frequencies (Hagler et al., 2007). Periodic signal components with 372 very low frequencies (due to slow head motion) and the second and third harmonic of the 373 stimulus were excluded as neither signal nor noise (this is mathematically equivalent to first 374 linearly regressing out these frequencies as nuisance variables before calculating significance). 375 The phase of the signal, which corresponds to a particular point of the stimulus cycle, was then 376 mapped to a color scale and the amplitude of the signal at each vertex was mapped to color 377 saturation (Gouraud sharing within each face). Runs with downward frequency steps were time- 378 reversed and averaged with upward-stepped scans in order to cancel fixed voxel-specific delays 379 in the BOLD response. 380 employed as previously described (Sereno et al., 1995; Sereno and Huang, 2006; Hagler et al., 368 2007). Functional activation amplitude was estimated as the Fourier amplitude of the periodic 369 BOLD signal (proportional to percent response) at the frequency of the stimulus cycle (8 370 repetitions per run). An F statistic was calculated by comparing that amplitude to the average 371 amplitude of other ‘noise’ frequencies (Hagler et al., 2007). Periodic signal components with 372 very low frequencies (due to slow head motion) and the second and third harmonic of the 373 stimulus were excluded as neither signal nor noise (this is mathematically equivalent to first 374 linearly regressing out these frequencies as nuisance variables before calculating significance). 375 The phase of the signal, which corresponds to a particular point of the stimulus cycle, was then 376 mapped to a color scale and the amplitude of the signal at each vertex was mapped to color 377 saturation (Gouraud sharing within each face). Runs with downward frequency steps were time- 378 reversed and averaged with upward-stepped scans in order to cancel fixed voxel-specific delays 379 in the BOLD response. 380 381 Linear modeling was carried out in FSL (Smith et al., 2004). For all runs, the motion-registered 382 data were high-pass-filtered (100 sec) and prewhitened; a hemodynamic model corresponding 383 to each stimulated and attended (tonotopy condition) or attended (stepped, randomized attn- 384 tono conditions) frequency band was created by convolving the 12.8-sec block with a gamma 385 function (lag 6s, SD sc). In a separate multiple regression, the unattended (ignored) frequency 386 band was modeled for both stepped and randomized attn-tono conditions. The verbal cue was 387 also modeled; all models were temporally filtered before multiple regression. Coefficients from 388 the first-level contrasts for each of the four runs were combined in a fixed-effects analysis for 389 each condition; data from the stepped and random block conditions were also combined in an 390 eight-run average. 391 392 Surface-based cluster exclusion was used to correct for multiple comparisons in the group-wise 403 averages (surfclust and randsurfclust from (Hagler et al., 2006)). The exclusion criterion (only 404 surface clusters > 78 mm2 unless otherwise noted) was determined based on the minimum 405 estimated cortical area from iterative random sampling of cluster sizes (N=10000 iterations per 406 hemisphere in randsurfclust) required to achieve a corrected alpha of p < 0.001 for each 407 hemisphere, based on an initial uncorrected alpha of vertex-wise p < 0.01. 408 Surface-based cluster exclusion was used to correct for multiple comparisons in the group-wise 403 averages (surfclust and randsurfclust from (Hagler et al., 2006)). The exclusion criterion (only 404 surface clusters > 78 mm2 unless otherwise noted) was determined based on the minimum 405 estimated cortical area from iterative random sampling of cluster sizes (N=10000 iterations per 406 hemisphere in randsurfclust) required to achieve a corrected alpha of p < 0.001 for each 407 hemisphere, based on an initial uncorrected alpha of vertex-wise p < 0.01. 408 409 As an alternative means of defining primary auditory cortex, we projected the Morosan et al. 410 (2001) 3D raw probability maps provided in the AFNI (version 16.3.13, Cox, 2012) to a 411 FreeSurfer “fsaverage” brain registered to the Talairach target brain, resampled the data onto 412 the cortical surface, and thresholded at p > 0.30 to create ROI labels. The labels were ~2 mm 413 FWHM (five steps) surface-smoothed with manual removal of isolated marked vertices (due to 414 “spillover” from the 3D to 2D projection within the lateral fissure), then spherically morphed to 415 each subject. The labels were individually inspected (and filled if there were small holes in the 416 label), then a boundary was delineated around each label on each subject's flattened auditory 417 cortical patch. 418 As an alternative means of defining primary auditory cortex, we projected the Morosan et al. 410 (2001) 3D raw probability maps provided in the AFNI (version 16.3.13, Cox, 2012) to a 411 FreeSurfer “fsaverage” brain registered to the Talairach target brain, resampled the data onto 412 the cortical surface, and thresholded at p > 0.30 to create ROI labels. The labels were ~2 mm 413 FWHM (five steps) surface-smoothed with manual removal of isolated marked vertices (due to 414 “spillover” from the 3D to 2D projection within the lateral fissure), then spherically morphed to 415 each subject. 381 Linear modeling was carried out in FSL (Smith et al., 2004). For all runs, the motion-registered 382 data were high-pass-filtered (100 sec) and prewhitened; a hemodynamic model corresponding 383 to each stimulated and attended (tonotopy condition) or attended (stepped, randomized attn- 384 tono conditions) frequency band was created by convolving the 12.8-sec block with a gamma 385 function (lag 6s, SD sc). In a separate multiple regression, the unattended (ignored) frequency 386 band was modeled for both stepped and randomized attn-tono conditions. The verbal cue was 387 also modeled; all models were temporally filtered before multiple regression. Coefficients from 388 the first-level contrasts for each of the four runs were combined in a fixed-effects analysis for 389 each condition; data from the stepped and random block conditions were also combined in an 390 eight-run average. 391 15 15 Cross-subject averaging of phase-encoded mapping data was performed using the 393 methodology developed by Hagler and Sereno (2006) in which the real and imaginary 394 components of the signal with respect to the stimulus ramp are averaged across subjects, 395 preserving any phase information consistent between subjects. This was performed by 396 projecting each participant’s phase-encoded map to the FreeSurfer spherical atlas using 397 mri_surf2surf, performing 1 step of surface-based smoothing (< 1 mm FWHM in 2D), averaging 398 across subjects at each vertex, then painting back onto a single subject’s surface for viewing. 399 For the multiple regression analyses, the same sampling process was used to sample each 400 subject’s contrast parameter estimates for cross-subject averaging and t-tests. 401 Cross-subject averaging of phase-encoded mapping data was performed using the 393 methodology developed by Hagler and Sereno (2006) in which the real and imaginary 394 components of the signal with respect to the stimulus ramp are averaged across subjects, 395 preserving any phase information consistent between subjects. This was performed by 396 projecting each participant’s phase-encoded map to the FreeSurfer spherical atlas using 397 mri_surf2surf, performing 1 step of surface-based smoothing (< 1 mm FWHM in 2D), averaging 398 across subjects at each vertex, then painting back onto a single subject’s surface for viewing. 399 For the multiple regression analyses, the same sampling process was used to sample each 400 subject’s contrast parameter estimates for cross-subject averaging and t-tests. 401 402 set of ROIs that surpass this z threshold. 437 438 Results 439 440 Fourier-based Analyses 441 442 Stimulus- and attentionally-driven tonotopic organization in human auditory core. As a 443 necessary first step, we characterized basic tonotopic (stimulus-driven) organization in and 444 The labels were individually inspected (and filled if there were small holes in the 416 label), then a boundary was delineated around each label on each subject's flattened auditory 417 cortical patch. 418 16 419 ROI analyses. We quantified the similarity between frequency band response profiles driven by 420 stimulus+attention (tonotopy) versus attention alone (attn-tono) in a 'quilt' of small cortical- 421 surface-based ROIs that tiled the temporal plane. ROIs (as seen in Fig. 5b) were created on a 422 single subject's right and left hemisphere flattened patches by flooding all vertices within a 4mm 423 radius around a central selected vertex. Each of the ROIs (57 in the right hemisphere patch, and 424 68 in the slightly larger left patch) were then spherically morphed to the other 7 subjects' 425 flattened patches. Spurious ROI sampling on the edges of the patches was manually corrected 426 on the original subject's inflated surface and re-morphed to all other subjects. Each ROI was 427 then projected into the registered native-space EPI volume using Freesurfer's mri_label2vol 428 (sampled from the grey-white boundary to 0.8 of the calculated cortical depth, with fillthresh set 429 to 0.5). For each subject, within each ROI, we calculated the average parameter estimate for 430 each frequency band for tonotopy, and combined stepped and randomized attn-tono conditions. 431 For each ROI, we then ran a linear model with average tonotopy parameter estimates for the 5 432 frequency bands predicting average attn-tono parameter estimates for the same bands, 433 including subjects as a random factor. The resulting partial t-statistic for each ROI was z- 434 transformed and color-rendered in Fig. 5b, with p-value thresholds Bonferroni-corrected to p < 435 0.05 for the number of ROIs per hemisphere, and indicated by the white outline surrounding the 436 set of ROIs that surpass this z-threshold. 437 Results 17 immediately around myelin-estimated auditory core. The group-average R1-based estimates of 445 myelination (inflated hemispheres in left-most panel of Fig. 2) show that the highest R1 values 446 occur within primary somatomotor areas along the central sulcus, and in the typically keyhole- 447 shaped presumptive ‘auditory core’ lying along and immediately surrounding Heschl’s gyrus. It 448 is important to note that myelination varies within auditory core and that the lateral and medial 449 borders are less sharply demarcated (reviewed in Hackett et al., 2001, and Dick et al., 2012). To 450 show this variation, we plot isointensity R1 contours in the cortical flat patches in Figs 2a-c (with 451 the curvature-based boundaries of Heschl’s gyrus overlaid in dotted lines). To help identify the 452 discontinuities in R1 that would correspond to the putative borders of auditory core, we 453 calculated the R1 gradient along the surface (Glasser et al., 2016). Lines drawn along the peak 454 gradient amplitude (not shown) corresponded well with the outermost R1 isointensity contour in 455 Fig. 2 (0.66 s-1). (It is important to note the gradient at the lateral edge of presumptive core is 456 quite shallow, and - as in postmortem myelin stains - it is therefore more difficult to establish an 457 unambiguous lateral border, as could be surmised from the greater lateral spread of the 458 isointensity contours. The shape and size (~1.2cm x 2.4cm) of presumptive auditory core in this 459 sample also agreed with the results from Dick et al. (2012) at the same R1 threshold (with the 460 latter average core slightly narrower, not shown here). 461 The group-averaged topography of preferred frequency around auditory core has a typical 463 arrangement (Dick et al., 2012; De Martino et al., 2015b), with the core surrounded by a high- 464 frequency ‘V’. Preferred frequency descends into the center of core (where R1 values are 465 highest) before reversing and slowly ascending to mid-frequency preferred frequencies 466 anterolaterally (and to some extent posterolaterally). Fig. 3 shows tonotopic maps for each 467 individual listener. In general, the relationship between auditory core and tonotopy group is 468 conserved across listeners, but with some variability in the shape and extent of the isointensity 469 R1 contours. In particular, S2, S6 (right hemisphere), S7, and S8 (right hemisphere) had 470 18 irregularly shaped and 'blotchy' isointensity contours. While there is a fair degree of individual 471 variability in results from human postmortem cyto- and myelo-architectonic studies of auditory 472 core and surrounding areas (Hackett et al., 2001; Sweet et al., 2005), this was somewhat 473 greater than expected variation given other work in our laboratory (Dick et al., 2012; Lutti et al., 474 2014; Carey et al., 2017). As an independent estimate of primary auditory areas, we also 475 morphed the Morosan et al. (2001) 3D probabilistic map of primary auditory areas (TE1.0) using 476 previously established methods (see Methods); the outlines of the morphed labels 477 corresponding to p > 0.30 of being within TE1.0 are overlaid in black dotted lines in Figure 3. 478 479 We then asked whether ‘attention-tonotopic’ mapping resembled the tonotopic case in and 480 around auditory core. Here, the group-level spatial distribution of tonotopy is closely 481 recapitulated when spectrally-directed attention (‘stepped’ attn-tono condition) alone modulates 482 activation (Fig. 2b). This holds true in and around the keyhole-shaped hyperintensity defining 483 core, with a slight exception in the transition from higher to lower frequency preference in mid 484 core. In contrast (and as expected) the group-level attn-tono response for the randomized 485 control condition is much weaker (Fig. 2c), with almost no correspondence with the tonotopic 486 map, despite being acoustically identical to ‘stepped’ attn-tono but for the shuffling of the verbal 487 prompt ordering that destroyed the consistent phase-lag associated with specific frequency 488 bands, and supporting these Fourier-based analyses. The one potential exception is in and 489 around posterolateral core, where there is a low-to-mid frequency progression that is similar in 490 attn-tono and tonotopic maps, particularly in the left hemisphere. (This may be due to the small 491 (~1 octave) overall shift in spectral mean over the course of a stimulus cycle noted in Methods). 492 493 Stimulus- and attentionally-driven tonotopic organization outside of auditory core. In line 494 with results from previous fMRI studies (Talavage et al., 2004; Woods et al., 2009; Humphries et 495 al., 2010; Barton et al., 2012; Dick et al., 2012; Moerel et al., 2012; Saenz and Langers, 2014; 496 19 Thomas et al., 2015; De Martino et al., 2015b; Ahveninen et al., 2016; Leaver and Rauschecker, 497 2016; Riecke et al., 2016), there is stimulus-driven tonotopic mapping extending well beyond 498 auditory core, spanning the temporal plane and continuing into the superior temporal sulcus 499 (STS). As shown in Fig. 2a, the overall arrangement is characterized by two pairs of three 500 interlacing best-frequency ‘fingers,’ with the high-frequency fingers (red/orange colormap) 501 predominating medially and extending laterally, where they meet interdigitated lower-frequency 502 fingers (green/yellow colormap) extending lateral to medial, with the longest ‘middle’ lower- 503 frequency finger extending about halfway into auditory core. Similar to tonotopy within auditory 504 core, the overall pattern of group activation can be observed in the majority of individual 505 subjects (Fig. 3), but there is also considerable individual variability in the complexity, 506 topography, and extent of tonotopic and attn-tono mapping, similar to that observed in the fMRI 507 studies cited above (as well as electrophysiological studies in a number of studies in macaque 508 and owl monkey, e.g., Merzenich and Brugge, 1973; Morel et al., 1993). 509 As can be seen in the maps in Fig. 2b, the tonotopically-aligned maps evoked by spectrally- 511 directed attention are also present in the majority of auditory cortex outside of auditory core. 512 Again, the structure of the tonotopic map (as revealed by Fourier analysis) is abolished when 513 the attentional cue is randomized, thereby eliminating any consistent relationship between 514 attended frequency band and phase lag (Fig. 2c). 515 The similarity between the maps evoked by presentation of a single frequency band (tonotopy) 517 versus attention to one of two simultaneously-presented frequency bands (‘stepped’ attn-tono) 518 can also be seen in each individual subject (Fig. 3). As with the group-averaged data, there is a 519 close correspondence in the progression of preferred frequencies across auditory cortex in 520 individual subjects. The similarity between the tonotopic and attn-tono maps is particularly 521 striking in subjects 1, 2, 5, 6, and 7. The tonotopic organization of individual subjects 522 The similarity between the maps evoked by presentation of a single frequency band (tonotopy) 517 versus attention to one of two simultaneously-presented frequency bands (‘stepped’ attn-tono) 518 can also be seen in each individual subject (Fig. 3). As with the group-averaged data, there is a 519 close correspondence in the progression of preferred frequencies across auditory cortex in 520 individual subjects. The similarity between the tonotopic and attn-tono maps is particularly 521 striking in subjects 1, 2, 5, 6, and 7. The tonotopic organization of individual subjects 522 20 demonstrated overall commonalities, but with notable differences, even between individual 523 subjects’ right and left hemispheres, particularly outside of auditory core (Humphries et al., 524 2010; Moerel et al., 2014; Saenz and Langers, 2014). However, individual peculiarities were 525 replicated across tonotopic and attn-tono conditions. In some subjects, there was a surprising 526 lack of strong tonotopic mapping (subject 3 for which poor tonotopy may be due to greater EPI 527 warping, and also subject 4 for which low frequencies dominated the tonotopic maps). In 528 summary, there was a strong correspondence between tonotopic and attn-tono maps at both 529 the group and individual levels. 530 531 Multiple regression analyses: 532 533 Winner-Takes-All: Maps of ‘stepped’ versus ‘randomized’ attention conditions, and 534 quantitative concordance of tonotopic and attn-tono maps. In a complementary analysis, 535 we used standard multiple regression techniques (see Methods) to estimate the BOLD 536 response to each center-frequency band when it was presented in isolation (tonotopy), versus 537 when it was attended in the presence of a distractor band (attn-tono). This allowed us to make 538 use of the attentionally-driven signal in the randomized attn-tono condition and to combine these 539 data with the results from the stepped attn-tono condition to increase statistical power. It also 540 allowed us to verify that the attention effects generalize when listeners direct attention without 541 the ‘crutch’ of consistent stepping up or down across attended frequency bands. 542 543 The auditory cortical patches in Fig. 4 show the cross-subject average ‘Winner-Takes-All’ (WTA) 544 best frequency band (most positive-going BOLD response relative to resting baseline) maps for 545 tonotopy and attn-tono conditions (with no shading for response amplitude). These are overlaid 546 with the outermost R1 isocontour (dashed yellow) corresponding to auditory core. As should be 547 demonstrated overall commonalities, but with notable differences, even between individual 523 subjects’ right and left hemispheres, particularly outside of auditory core (Humphries et al., 524 2010; Moerel et al., 2014; Saenz and Langers, 2014). However, individual peculiarities were 525 replicated across tonotopic and attn-tono conditions. In some subjects, there was a surprising 526 lack of strong tonotopic mapping (subject 3 for which poor tonotopy may be due to greater EPI 527 warping, and also subject 4 for which low frequencies dominated the tonotopic maps). In 528 summary, there was a strong correspondence between tonotopic and attn-tono maps at both 529 the group and individual levels. 530 Winner-Takes-All: Maps of ‘stepped’ versus ‘randomized’ attention conditions, and 534 quantitative concordance of tonotopic and attn-tono maps. In a complementary analysis, 535 we used standard multiple regression techniques (see Methods) to estimate the BOLD 536 response to each center-frequency band when it was presented in isolation (tonotopy), versus 537 when it was attended in the presence of a distractor band (attn-tono). This allowed us to make 538 use of the attentionally-driven signal in the randomized attn-tono condition and to combine these 539 data with the results from the stepped attn-tono condition to increase statistical power. It also 540 allowed us to verify that the attention effects generalize when listeners direct attention without 541 the ‘crutch’ of consistent stepping up or down across attended frequency bands. 542 543 Winner-Takes-All: Maps of ‘stepped’ versus ‘randomized’ attention conditions, and 534 quantitative concordance of tonotopic and attn-tono maps. In a complementary analysis, 535 we used standard multiple regression techniques (see Methods) to estimate the BOLD 536 response to each center-frequency band when it was presented in isolation (tonotopy), versus 537 when it was attended in the presence of a distractor band (attn-tono). This allowed us to make 538 use of the attentionally-driven signal in the randomized attn-tono condition and to combine these 539 data with the results from the stepped attn-tono condition to increase statistical power. It also 540 allowed us to verify that the attention effects generalize when listeners direct attention without 541 the ‘crutch’ of consistent stepping up or down across attended frequency bands. 542 543 The auditory cortical patches in Fig. 4 show the cross-subject average ‘Winner-Takes-All’ (WTA) 544 best frequency band (most positive-going BOLD response relative to resting baseline) maps for 545 tonotopy and attn-tono conditions (with no shading for response amplitude). These are overlaid 546 with the outermost R1 isocontour (dashed yellow) corresponding to auditory core. As should be 547 expected, the topography of the WTA maps essentially recapitulates the topography revealed 548 The auditory cortical patches in Fig. 4 show the cross-subject average ‘Winner-Takes-All’ (WTA) 544 best frequency band (most positive-going BOLD response relative to resting baseline) maps for 545 tonotopy and attn-tono conditions (with no shading for response amplitude). These are overlaid 546 with the outermost R1 isocontour (dashed yellow) corresponding to auditory core. As should be 547 expected, the topography of the WTA maps essentially recapitulates the topography revealed 548 21 21 by the phase-encoded analyses. The same holds true of the attn-tono WTA maps from both the 549 ‘stepped’ and importantly, the ‘randomized’ block conditions (Fig. 4); this result confirms that 550 even without the 'crutch' of the stepping frequency band, listeners can direct their attention to 551 specific frequency bands. 552 by the phase-encoded analyses. The same holds true of the attn-tono WTA maps from both the 549 ‘stepped’ and importantly, the ‘randomized’ block conditions (Fig. 4); this result confirms that 550 even without the 'crutch' of the stepping frequency band, listeners can direct their attention to 551 specific frequency bands. 552 by the phase-encoded analyses. The same holds true of the attn-tono WTA maps from both the 549 ‘stepped’ and importantly, the ‘randomized’ block conditions (Fig. 4); this result confirms that 550 even without the 'crutch' of the stepping frequency band, listeners can direct their attention to 551 specific frequency bands. 552 The WTA approach also allowed us to straightforwardly quantify the within-subject 554 correspondence between voxel-wise best frequency, as estimated by tonotopy and by attn-tono. 555 Here, we coded each voxel in native space as a ‘1’ when best frequency was identical in both 556 conditions, and a ‘0’ otherwise. We then resampled each subject’s binary maps to their cortical 557 surface, and then averaged across subjects to create a ‘concordance’ map (Fig. 5a). These 558 maps (statistically thresholded at vertex-wise p< 0.01, with surface-cluster-corrected alpha of p 559 < 0.001) show that across subjects there was high concordance across best frequency maps 560 evoked by stimulus and by attention across much of the temporal plane in both hemispheres, 561 with little concordance in non-auditory areas. The extent of attentionally-driven tonotopic 562 mapping relative to overall tonotopicity is shown in the cortical patches below each concordance 563 map in Fig. 5a. Here, the outer contour of the significant (p<0.001 clusterwise corrected) 564 concordance map is overlaid on the phase-averaged group tonotopy map (same as Fig. 2a). 565 Averaging over all subjects, the majority of consistently tonotopically mapped cortex medial to 566 the crown of the superior temporal gyrus also shows preferred-frequency-aligned attn-tono 567 maps, as does a small posterior cluster. In the left hemisphere, almost all consistently mapped 568 tonotopic cortex also shows aligned attn-tono maps. However, it is important to note that there 569 are considerable individual differences in regional best-frequency alignment across the tonotopy 570 and attn-tono maps (as can be seen in Figure 3). 571 22 frequency in auditory cortex are not necessarily bandpass, but can be more complex and 575 multipeaked. Therefore, we also examined whether attention to a given frequency band in the 576 presence of a distractor band recapitulates the more graded response to non-preferred 577 frequencies observed when that frequency band is presented in isolation. To do this, we created 578 and surface-morphed a set of small cortical ROIs to each subject (see Fig. 5b and Methods), 579 and quantified the similarity between the tonotopy and attn-tono response profiles in each ROI 580 in each hemisphere by regressing the mean tonotopic parameter estimate for each frequency 581 band against the attn-tono parameter estimate (with subjects as a random factor). We used this 582 'ROI quilt' analysis (as opposed to a vertex-wise one) to capture regional variation in cross- 583 condition response profile similarity across subjects, which might be obscured by individual 584 differences in tonotopic map topography and surface-based registration errors. This also 585 reduced the number of statistical comparisons that must be corrected for, thus increasing power 586 to detect differences. 587 frequency in auditory cortex are not necessarily bandpass, but can be more complex and 575 multipeaked. Therefore, we also examined whether attention to a given frequency band in the 576 presence of a distractor band recapitulates the more graded response to non-preferred 577 frequencies observed when that frequency band is presented in isolation. To do this, we created 578 and surface-morphed a set of small cortical ROIs to each subject (see Fig. 5b and Methods), 579 and quantified the similarity between the tonotopy and attn-tono response profiles in each ROI 580 in each hemisphere by regressing the mean tonotopic parameter estimate for each frequency 581 band against the attn-tono parameter estimate (with subjects as a random factor). We used this 582 'ROI quilt' analysis (as opposed to a vertex-wise one) to capture regional variation in cross- 583 condition response profile similarity across subjects, which might be obscured by individual 584 differences in tonotopic map topography and surface-based registration errors. This also 585 reduced the number of statistical comparisons that must be corrected for, thus increasing power 586 to detect differences. 587 588 The region of interest (ROI) analyses (Fig. 5) further support the results from the 'concordance' 589 maps from the Winner-Takes-All analyses (Fig. 4). The ROI analyses (Fig. 5b) show that 590 individual subjects' tonotopy and attn-tono responses profiles are significantly associated across 591 most of auditory cortex (all ROIs within the white border), with the exception of the most lateral 592 aspects of the STG and upper bank of the STS. (Note that while there is a strong relationship 593 between tonotopy and attn-tono response profiles of each subject within a given ROI, there is 594 cross-subject variability in the particular shape of those response profiles, as suggested by the 595 individual maps in Fig. 3). There is a broad tendency for tonotopy/attn-tono profile similarity to 596 be strongest posterior-medially in both hemispheres, and no clear indication that profile 597 similarity is higher in auditory core (indeed, this is not the case in the left hemisphere). As 598 shown by the white line on the tonotopic flat maps in Fig. 5a, the area showing significant 599 response profile similarity extends over the majority of cortex showing strong tonotopic mapping 600 The region of interest (ROI) analyses (Fig. 5) further support the results from the 'concordance' 589 maps from the Winner-Takes-All analyses (Fig. 4). The ROI analyses (Fig. 5b) show that 590 individual subjects' tonotopy and attn-tono responses profiles are significantly associated across 591 most of auditory cortex (all ROIs within the white border), with the exception of the most lateral 592 aspects of the STG and upper bank of the STS. (Note that while there is a strong relationship 593 between tonotopy and attn-tono response profiles of each subject within a given ROI, there is 594 cross-subject variability in the particular shape of those response profiles, as suggested by the 595 individual maps in Fig. 3). There is a broad tendency for tonotopy/attn-tono profile similarity to 596 be strongest posterior-medially in both hemispheres, and no clear indication that profile 597 similarity is higher in auditory core (indeed, this is not the case in the left hemisphere). As 598 shown by the white line on the tonotopic flat maps in Fig. 5a, the area showing significant 599 response profile similarity extends over the majority of cortex showing strong tonotopic mapping 600 23 with topologic similarity across subjects. The response profile similarity extends into less 601 tonotopically consistent regions medially and posteriorly, but does not include the more postero- 602 lateral tonotopically mapped regions along the crown of the superior temporal gyrus. 603 604 with topologic similarity across subjects. The response profile similarity extends into less 601 tonotopically consistent regions medially and posteriorly, but does not include the more postero- 602 lateral tonotopically mapped regions along the crown of the superior temporal gyrus. 603 604 634 We then resampled each subject’s native-space ‘attention maps’ to her/his cortical surface to 635 allow for surface-based cross-subject averaging and statistical testing (all again with a vertex- 636 wise p< 0.01 threshold and surface-cluster-corrected alpha of p < 0.001). The top row of Fig. 7 637 shows that across subjects there was significantly greater activation across most of auditory 638 cortex when best frequency was attended versus ignored. The widespread attention effect 639 included all of R1-estimated auditory core (outlined in green), extending from the inferior circular 640 sulcus laterally to the upper bank of the STS, and antero-posteriorly from the temporal pole onto 641 the planum temporale. By contrast, there were relatively few regions where attention to a 642 voxel's least-preferred frequency band evoked greater activation than when the same 643 frequency band was the distractor. Attending to a voxel’s least-preferred frequency 644 band only significantly increased activation along the posterior lateral STG in both 645 hemispheres, extending more medially in the left, and more anteriorly in the right (Fig. 7, 646 middle row). A direct comparison between these maps (cross-subject t-test on the difference 647 of differences, Fig. 7, bottom row) showed that there were considerable regional differences in 648 activation between attending to a voxel's preferred versus dis-preferred frequency band. In both 649 hemispheres, there was greater activation across most of the anterior temporal plane when 650 attention was directed to the preferred versus dis-preferred frequency; in the right hemisphere, 651 selective attention across subjects, and how the effect of attention varied with preferred or dis- 627 preferred frequency. We first used a subject’s native-space WTA map to establish each voxel’s 628 best frequency. Then, we assigned each voxel the parameter estimate for the difference in 629 activation between attending to its best frequency in the presence of a distractor, versus 630 attending to the distractor and ignoring its best frequency. (In other words, the value at each 631 voxel was the estimated difference in activation between attending to, versus ignoring, its best 632 frequency in the presence of other frequency bands). We repeated this process to estimate the 633 parallel attention effect for each voxel's 'worst' frequency (using the corresponding LTA map). 634 We then resampled each subject’s native-space ‘attention maps’ to her/his cortical surface to 635 allow for surface-based cross-subject averaging and statistical testing (all again with a vertex- 636 wise p< 0.01 threshold and surface-cluster-corrected alpha of p < 0.001). The top row of Fig. 7 637 shows that across subjects there was significantly greater activation across most of auditory 638 cortex when best frequency was attended versus ignored. The widespread attention effect 639 included all of R1-estimated auditory core (outlined in green), extending from the inferior circular 640 sulcus laterally to the upper bank of the STS, and antero-posteriorly from the temporal pole onto 641 the planum temporale. By contrast, there were relatively few regions where attention to a 642 voxel's least-preferred frequency band evoked greater activation than when the same 643 frequency band was the distractor. Attending to a voxel’s least-preferred frequency 644 band only significantly increased activation along the posterior lateral STG in both 645 hemispheres, extending more medially in the left, and more anteriorly in the right (Fig. 7, 646 middle row). A direct comparison between these maps (cross-subject t-test on the difference 647 of differences, Fig. 7, bottom row) showed that there were considerable regional differences in 648 activation between attending to a voxel's preferred versus dis-preferred frequency band. In both 649 hemispheres, there was greater activation across most of the anterior temporal plane when 650 attention was directed to the preferred versus dis-preferred frequency; in the right hemisphere, 651 Loser-Takes-All: Maps of ‘dis-preferred’ frequency. Given the graded nature of frequency 605 response preferences we observed, we suspected that there would be a large-scale topography 606 associated with the minimum BOLD response across frequency, and that this topography would 607 also be recapitulated by attention. Thus, we also performed a parallel ‘Loser-Takes-All’ (LTA) 608 analysis, in which we coded voxels by the frequency band driving the minimum BOLD response 609 (again relative to resting baseline) and analyzed as above. The average descriptive LTA maps 610 show roughly opposite frequency responses compared to the WTA tonotopic maps, with higher- 611 frequency-band-preferring regions in the tonotopic map being least driven by lower-frequency- 612 bands, and vice versa (Fig. 6a). There is also some overlap in the ‘mid-frequency-preferring’ 613 regions, likely due to blurring of values when averaging subjects’ integer-based maps. There is 614 also quite close correspondence between the frequency band evoking the least response in the 615 tonotopy (stimulus) condition and the smallest BOLD response evoked by attending to a given 616 frequency band. The LTA concordance maps (Fig. 6b, statistical thresholding as in Fig. 5a) 617 show that in the right hemisphere, the alignment of tonotopic and attn-tono maps is greatest in 618 more lateral and anterior auditory cortex, with qualitatively somewhat greater concordance more 619 medially in the left hemisphere. (Note that the hemispheric difference and also the apparent 620 qualitative contrast with the WTA concordance maps seen in Figures 4 and 6 are exaggerated 621 by the cluster-wise statistical thresholding combined with the overall slightly lower concordance 622 in the LTA maps). 623 24 selective attention across subjects, and how the effect of attention varied with preferred or dis- 627 preferred frequency. We first used a subject’s native-space WTA map to establish each voxel’s 628 best frequency. Then, we assigned each voxel the parameter estimate for the difference in 629 activation between attending to its best frequency in the presence of a distractor, versus 630 attending to the distractor and ignoring its best frequency. (In other words, the value at each 631 voxel was the estimated difference in activation between attending to, versus ignoring, its best 632 frequency in the presence of other frequency bands). We repeated this process to estimate the 633 parallel attention effect for each voxel's 'worst' frequency (using the corresponding LTA map). Relationship of tonotopic and attn-tono map strength to MR-estimated 659 myeloarchitecture. Typically, assays of cortical myelination are used to differentiate the most 660 highly-myelinated cortical regions (like auditory core, MT/V5, or V1) from adjacent regions. This 661 is true whether cortical myelination is assessed using ex-vivo ‘gold-standard’ approaches such 662 as Gallyas staining, or estimated through in-vivo MRI T1-weighted/T2-weighted ratio, 663 quantitative R1, or magnetization transfer measures. However, more subtle myelination changes 664 that occur throughout cortex may spatially correspond with changes in functional characteristics 665 (Glasser et al., 2016; Wallace et al., 2016). For instance, recent combined fMRI and high- 666 resolution quantitative MR show that slight reductions in cortical myelination in primary 667 somatomotor cortex reliably occur at the border between face and hand areas (Kuehn et al., 668 2017). 669 670 Here, we asked whether the change in the degree to which cortex showed a strong frequency- 671 Relationship of tonotopic and attn-tono map strength to MR-estimated 659 myeloarchitecture. Typically, assays of cortical myelination are used to differentiate the most 660 highly-myelinated cortical regions (like auditory core, MT/V5, or V1) from adjacent regions. This 661 is true whether cortical myelination is assessed using ex-vivo ‘gold-standard’ approaches such 662 as Gallyas staining, or estimated through in-vivo MRI T1-weighted/T2-weighted ratio, 663 quantitative R1, or magnetization transfer measures. However, more subtle myelination changes 664 that occur throughout cortex may spatially correspond with changes in functional characteristics 665 (Glasser et al., 2016; Wallace et al., 2016). For instance, recent combined fMRI and high- 666 resolution quantitative MR show that slight reductions in cortical myelination in primary 667 somatomotor cortex reliably occur at the border between face and hand areas (Kuehn et al., 668 2017). 669 670 Here, we asked whether the change in the degree to which cortex showed a strong frequency- 671 band preference (namely, the amplitude of the phase-encoded tonotopic or attn-tono signal) 672 spatially corresponded with changes in myelination as assessed by quantitative R1 (within a 4 673 mm radius disk that roved across the entire cortical surface). The cross-subject-average 674 normalized covariance map in Fig. 8a shows that there is a shared local gradient in tonotopic 675 amplitude and R1 along the entire inferior circular sulcus and the anterior part of the superior 676 temporal gyrus, where tonotopic amplitude and R1 drop in tandem over a narrow band of cortex. 25 this effect extended throughout the temporal plane, as well as including a patch in the posterior 652 STG. There were no regions in which the converse effect was observed (greater attend > 653 distract activation for dis-preferred versus preferred frequency band). This shows that the 654 frequency-selective attention-related BOLD gain is strongly modulated by frequency-preference 655 - and provides some evidence for models of multiplicative, and not additive, attentional gain (but 656 see Discussion). 657 677 Here, we asked whether the change in the degree to which cortex showed a strong frequency- 671 band preference (namely, the amplitude of the phase-encoded tonotopic or attn-tono signal) 672 spatially corresponded with changes in myelination as assessed by quantitative R1 (within a 4 673 mm radius disk that roved across the entire cortical surface). The cross-subject-average 674 normalized covariance map in Fig. 8a shows that there is a shared local gradient in tonotopic 675 amplitude and R1 along the entire inferior circular sulcus and the anterior part of the superior 676 temporal gyrus, where tonotopic amplitude and R1 drop in tandem over a narrow band of cortex. 677 26 There is also negative local spatial covariance between tonotopic amplitude and R1 within the 678 center of auditory cortex, where tonotopic amplitude increases but R1 remains relatively stable. 679 (There is also some tonotopic/R1 spatial covariance within and around the central sulcus; these 680 regions showed considerably less overall amplitude in tonotopic response, but one that spatially 681 covaries with changes in R1). 682 There is also negative local spatial covariance between tonotopic amplitude and R1 within the 678 center of auditory cortex, where tonotopic amplitude increases but R1 remains relatively stable. 679 (There is also some tonotopic/R1 spatial covariance within and around the central sulcus; these 680 regions showed considerably less overall amplitude in tonotopic response, but one that spatially 681 covaries with changes in R1). 682 To test the replicability of this novel tonotopy-versus-R1 searchlight cross-correlation, we 684 reanalyzed R1 and tonotopy data from a previous study (Dick et al., 2012) that used a different 685 tonotopic stimulus (bandpass-filter-swept non-linguistic vocalizations) and a slightly different 686 multiparameter mapping protocol. Despite these methodological differences, we found a very 687 similar pattern of tonotopic/R1 positive local spatial covariance within the circular sulcus and 688 along the lateral STG, with negative spatial covariance again in the center of auditory cortex 689 (Fig. 8b). The shared and relatively steep anterolateral and medial gradients in putative 690 myelination and degree of frequency specificity - observed in two independently-acquired 691 datasets - suggests a shared functional and myeloarchitectonic border, possibly similar in 692 character to those reported recently relating resting state, standard task activation, and T1- 693 weighted/T2-weighted derived myelination estimates across cortex (Glasser et al., 2016; Kuehn 694 et al., 2017). 695 As seen in Fig. 8c, the spatial relationship between local R1 and attn-tono amplitude changes is 697 much less clear. Here, there is a weak relationship within and around auditory cortex that is only 698 observed within the circular sulcus (particularly in the right hemisphere). There are also stripes 699 of spatial covariation along the banks of the central sulcus, although not closely aligned with the 700 pattern observed with the tonotopy versus R1 covariance maps. Although very preliminary, 701 these results suggest that changes in the degree of spectral attentional modulation in auditory 702 cortex are not strongly linked to the underlying myeloarchitecture, and stands in contrast to the 703 As seen in Fig. 8c, the spatial relationship between local R1 and attn-tono amplitude changes is 697 much less clear. Here, there is a weak relationship within and around auditory cortex that is only 698 observed within the circular sulcus (particularly in the right hemisphere). There are also stripes 699 of spatial covariation along the banks of the central sulcus, although not closely aligned with the 700 pattern observed with the tonotopy versus R1 covariance maps. Although very preliminary, 701 these results suggest that changes in the degree of spectral attentional modulation in auditory 702 cortex are not strongly linked to the underlying myeloarchitecture, and stands in contrast to the 703 27 consistent spatial association in lateral and medial auditory cortex between local changes in R1 704 and the strength of stimulus-driven frequency response preference. 705 706 Summary. Everyday listening ordinarily takes place in rich soundscapes within multiple, 707 simultaneous sound sources contributing to the overlapping mix of sound waves that arrives at 708 the ears. Auditory attention is crucial to sorting out the mix. Listeners direct attentional focus to a 709 sound source, or even to specific acoustic dimensions within a single sound source, to zero in 710 on auditory information that is diagnostic in guiding behavior. 711 712 We asked how endogenous attention directed to specific acoustic frequency bands modulates 713 human auditory cortical activity. Using high-resolution quantitative MRI and a novel fMRI 714 paradigm for driving sustained selective attention within specific frequency bands, we 715 established effects of spectrally-specific attention in myeloarchitectonically-estimated human 716 auditory core. These effects extend across the majority of tonotopically-mapped auditory cortex, 717 and are apparent in individual listeners. Sensory-driven best-frequency tonotopic maps align 718 with attentionally-driven maps across much of the temporal plane, with poor concordance in 719 non-auditory areas. Individual tonotopic and attn-tono maps show correlated idiosyncracies. The 720 frequency bands that evoke the least BOLD response from input and from attention also exhibit 721 close spatial correspondence. There is greater activation across most of auditory cortex when 722 best frequency is attended, versus ignored. Finally, there is local spatial correspondence in 723 multiple auditory regions between the degree of R1-estimated myelination, and the strength of 724 the frequency-band-selective fMRI response for tonotopic stimuli. 725 726 consistent spatial association in lateral and medial auditory cortex between local changes in R1 704 and the strength of stimulus-driven frequency response preference. 705 consistent spatial association in lateral and medial auditory cortex between local changes in R1 704 and the strength of stimulus-driven frequency response preference. 705 Summary. Everyday listening ordinarily takes place in rich soundscapes within multiple, 707 simultaneous sound sources contributing to the overlapping mix of sound waves that arrives at 708 the ears. Auditory attention is crucial to sorting out the mix. Listeners direct attentional focus to a 709 sound source, or even to specific acoustic dimensions within a single sound source, to zero in 710 on auditory information that is diagnostic in guiding behavior. 711 We asked how endogenous attention directed to specific acoustic frequency bands modulates 713 human auditory cortical activity. Using high-resolution quantitative MRI and a novel fMRI 714 paradigm for driving sustained selective attention within specific frequency bands, we 715 established effects of spectrally-specific attention in myeloarchitectonically-estimated human 716 auditory core. These effects extend across the majority of tonotopically-mapped auditory cortex, 717 and are apparent in individual listeners. Sensory-driven best-frequency tonotopic maps align 718 with attentionally-driven maps across much of the temporal plane, with poor concordance in 719 non-auditory areas. Individual tonotopic and attn-tono maps show correlated idiosyncracies. The 720 frequency bands that evoke the least BOLD response from input and from attention also exhibit 721 close spatial correspondence. There is greater activation across most of auditory cortex when 722 best frequency is attended, versus ignored. Finally, there is local spatial correspondence in 723 multiple auditory regions between the degree of R1-estimated myelination, and the strength of 724 the frequency-band-selective fMRI response for tonotopic stimuli. 725 Attentionally-driven tonotopic organization extends across much of auditory cortex. We 743 also find strong evidence for tonotopically-mapped spectrally-directed attention in much of 744 auditory cortex, particularly along the lateral superior temporal gyrus (potentially analogous to 745 lateral auditory belt and parabelt regions in macaque (Hackett, 2007)). In addition to the 746 concordance in and around auditory core, the most consistent group-level alignment of these 747 maps lies lateral to auditory core, with each map characterized by three higher-to-lower best- 748 frequency-band traversals, moving from posterior to anterior roughly along the superior 749 temporal gyrus. 750 Discussion Discussion 28 Human auditory core exhibits attentionally-driven tonotopic organization. Previous 729 findings showed similar stimulus- and attentionally-driven frequency preference in and around 730 Heschl’s gyrus, a macroanatomical landmark associated with primary auditory areas (Da Costa 731 et al., 2013; Riecke et al., 2016). Here, we demonstrate that, within quantitative-R1-defined 732 primary auditory areas, the attentionally-driven maps in each hemisphere are very similar to the 733 detailed tonotopic maps in the same subjects. As shown by comparison maps across the 734 acoustically-identical stepped and randomized attn-tono conditions (Figs. 2 and 4), the 735 alignment between tonotopic and attention maps depends on allocation of attention to the cued 736 frequency band, not perceptual interference or other stimulus-driven effects. The fact that there 737 is considerable, high-level attentional modulation within primary auditory areas is interesting 738 given previous results suggesting more limited attentional topographic modulation in primary 739 auditory (Atiani et al., 2014) and visual (Saygin and Sereno, 2008) cortex, compared to more 740 robust attentional modulation in areas immediately adjacent to primary ones. 741 742 Attentionally-driven tonotopic organization extends across much of auditory cortex. We This pattern suggests a cross-species parallel to results reported in ferret (Atiani et al., 2014), 752 where task-evoked attentional modulation of frequency-tuned neurons is particularly strong in 753 non-primary (dPEG) tonotopically mapped auditory areas in ferret. In this regard, the stimulus 754 29 complexity, variability, and memory demands of the current task may have helped to drive 755 attentional response in these more lateral and anterior areas. Our results are consistent with a 756 human fMRI comparison of cross-modal attentional effects (Petkov et al., 2004), which showed 757 greater activation in lateral auditory regions when attention was directed to a demanding 758 auditory repetition detection task than when the same sounds were played as subjects 759 performed a demanding visual detection task. However, our results differ from these studies to 760 some degree in that attentionally-driven tonotopic modulation in auditory core was also robust 761 (similar to cross-modal attention studies in macaque A1 (O'Connell et al., 2014) and primary 762 auditory areas (De Martino et al., 2015a)), and did not differ significantly from that in lateral belt. 763 764 complexity, variability, and memory demands of the current task may have helped to drive 755 attentional response in these more lateral and anterior areas. Our results are consistent with a 756 human fMRI comparison of cross-modal attentional effects (Petkov et al., 2004), which showed 757 greater activation in lateral auditory regions when attention was directed to a demanding 758 auditory repetition detection task than when the same sounds were played as subjects 759 performed a demanding visual detection task. However, our results differ from these studies to 760 some degree in that attentionally-driven tonotopic modulation in auditory core was also robust 761 (similar to cross-modal attention studies in macaque A1 (O'Connell et al., 2014) and primary 762 auditory areas (De Martino et al., 2015a)), and did not differ significantly from that in lateral belt. 763 764 There was good correspondence between the voxel-wise best frequency-band for tonotopy and 765 attention-tonotopy in individual listeners. Like several prior studies (data and review in 766 Humphries et al., 2010; Moerel et al., 2014; Saenz and Langers, 2014; Brewer and Barton, 767 2016; Leaver and Rauschecker, 2016), we observed quite substantial variation in the detailed 768 topography of tonotopy across individuals (but cf. Ahveninen et al., 2016). It is especially 769 noteworthy that attention-tonotopy recapitulated these topographic idiosyncrasies (as observed 770 in the concordance analyses, Fig. 4b and 5b). 771 There was good correspondence between the voxel-wise best frequency-band for tonotopy and 765 attention-tonotopy in individual listeners. Like several prior studies (data and review in 766 Humphries et al., 2010; Moerel et al., 2014; Saenz and Langers, 2014; Brewer and Barton, 767 2016; Leaver and Rauschecker, 2016), we observed quite substantial variation in the detailed 768 topography of tonotopy across individuals (but cf. Ahveninen et al., 2016). It is especially 769 noteworthy that attention-tonotopy recapitulated these topographic idiosyncrasies (as observed 770 in the concordance analyses, Fig. 4b and 5b). 771 It is intriguing that there was a systematic frequency-band-associated topography not only of 773 best frequency but also of dis-preferred frequency and, also, that the frequency-selective 774 attenuation of BOLD gain relative to other frequencies can be recapitulated by selective 775 attention to that frequency band in the presence of other spectral information. One could 776 speculate that this map structure might be a population-level reflection of an ‘inhibitory surround’ 777 structure observed in some electrophysiology studies (Calford and Semple, 1995; Sutter et al., 778 1999; but cf Wehr and Zador, 2003), with the frequency band driving the least BOLD response 779 corresponding to the deepest trough in an asymmetric surround -- an effect that could drive the 780 30 very similar tonotopic and attn-tono graded frequency response preferences revealed in the 781 multiple ROI analysis (Fig. 5b). 782 very similar tonotopic and attn-tono graded frequency response preferences revealed in the 781 multiple ROI analysis (Fig. 5b). 782 very similar tonotopic and attn-tono graded frequency response preferences revealed in the 781 multiple ROI analysis (Fig. 5b). 782 Here, the average frequency response profile evoked by the single-band tonotopic stimuli was 784 recapitulated by attention to the same frequency bands in the context of distractors. Prior 785 human neuroimaging research has been consistent with the possibility that the shape of the 786 frequency response in and around Heschl’s gyrus is attentionally-modulated in a bandpass 787 manner that relies on amplification rather than attenuation (Riecke et al., 2016). Based on 788 results from a larger number of spectral bands, the current findings suggest that, at least at a 789 more macroscopic scale, spectrally-directed attention modulates cortical activity in a more 790 graded fashion, with the shape of the attentional response to both preferred and less-preferred 791 frequency bands similar to that evoked by stimulus alone - a contention supported by the 792 alignment of the ‘Loser-Takes-All’ tonotopic and attn-tono maps (Fig. 6). That is, the frequency 793 band that drives the smallest fMRI response when presented alone is also the frequency band 794 that elicits the least activation when attended in the presence of a distractor. A better 795 understanding of the mechanisms underlying these maps will require more fine-grained 796 characterization of frequency-directed attentional modulation, preferably at very high spectral 797 and temporal resolution (Moerel et al., 2013; Lutti et al., 2014; Moerel et al., 2014; Ahveninen et 798 al., 2016) that might also help to unveil cortical-depth-specific attentional effects (De Martino et 799 al., 2015a). In particular, it will be important to see whether different fMRI tasks - using more 800 complex naturalistic sounds, or more or less abstract cues to frequency - can mimic the task-, 801 valence- and context-dependent effects observed in non-human animal cortical auditory 802 receptive fields, where the character of the ‘contrast-enhancing’ modulations differs markedly 803 with experimental manipulation (Fritz et al., 2005; 2007a; 2007b; David et al., 2012; Atiani et al., 804 2014; Kuchibhotla et al., 2017). (It is worth noting that task-related modulation of frequency- 805 selective attentional effects has long been of interest in human auditory psychophysics 806 31 (Greenberg, 1968; Scharf et al., 1987; Scharf, 1989; Moore et al., 1996; Green and McKeown, 807 2001)). 808 809 There is correspondence between local change in R1-estimated myelination and the 810 strength of fMRI-assessed relative frequency selectivity. We found that the change in the 811 degree to which a small (4 mm radius) patch of cortex shows strong frequency preferences in 812 tonotopy was positively spatially correlated with its degree of myelination as estimated by R1. 813 The strength of the correlation was anatomically specific, marking the medial border of auditory 814 cortex (within the circular sulcus) and revealing a potential anatomical index of ‘processing style’ 815 (from more to less tonotopically mapped) along anterolateral superior temporal gyrus. We found 816 this pattern to hold true in the data from the current study as well as in an independent cohort 817 scanned with quite different tonotopic stimuli and with multiparameter maps acquired on a 818 different scanner model, with different sequence settings (Fig. 8c). Although there was a 819 relatively reliable pattern of R1-tonotopy correspondence at a group level, there was some 820 notable individual variation in local shared R1/tonotopy gradients relative to gyral anatomy. 821 Thus, these patterns may be more useful than curvature for establishing areal borders on an 822 individual subject basis, particularly when there is no obvious sharp change in a single measure 823 (for discussion see also Glasser et al., 2016). Such work holds promise for generating novel 824 hypotheses for more intensively characterized species like mouse, ferret, or marmoset, 825 particularly in tandem with imaging techniques that that can cover multiple cortical areas 826 simultaneously. 827 809 There is correspondence between local change in R1-estimated myelination and the 810 strength of fMRI-assessed relative frequency selectivity. We found that the change in the 811 degree to which a small (4 mm radius) patch of cortex shows strong frequency preferences in 812 tonotopy was positively spatially correlated with its degree of myelination as estimated by R1. 813 The strength of the correlation was anatomically specific, marking the medial border of auditory 814 cortex (within the circular sulcus) and revealing a potential anatomical index of ‘processing style’ 815 (from more to less tonotopically mapped) along anterolateral superior temporal gyrus. We found 816 this pattern to hold true in the data from the current study as well as in an independent cohort 817 scanned with quite different tonotopic stimuli and with multiparameter maps acquired on a 818 different scanner model, with different sequence settings (Fig. 8c). Although there was a 819 relatively reliable pattern of R1-tonotopy correspondence at a group level, there was some 820 notable individual variation in local shared R1/tonotopy gradients relative to gyral anatomy. 821 Thus, these patterns may be more useful than curvature for establishing areal borders on an 822 individual subject basis, particularly when there is no obvious sharp change in a single measure 823 (for discussion see also Glasser et al., 2016). Such work holds promise for generating novel 824 hypotheses for more intensively characterized species like mouse, ferret, or marmoset, 825 particularly in tandem with imaging techniques that that can cover multiple cortical areas 826 simultaneously. 827 828 Future Directions. In the current study, we limited our investigation to broadly defined auditory 829 cortex, where there was good evidence for systematic tonotopic representation from a number 830 of previous studies (Talavage and Edmister, 2004; Hackett, 2007; Moerel et al., 2013; 2014; 831 Saenz and Langers, 2014; Leaver and Rauschecker, 2016). In future research it will be 832 32 informative to examine interactions with several frontal regions whose potential analogues are 833 known to have direct feedforward and feedback connections in macaque monkeys (Romanski 834 and Goldman-Rakic, 2002), and where in ferret there are clear modulatory influences on 835 primary and non-primary auditory cortex during learning (Atiani et al., 2014; Shamma and Fritz, 836 2014). Similar to recent work in vision (Klein et al., 2014; Puckett and DeYoe, 2015), it will also 837 be useful to establish the shape of the attentional population receptive field, and how this varies 838 across auditory areas and relates to stimulus-driven auditory population receptive field size 839 (Thomas et al., 2015). Finally, following on from our own pilot work, it will be exciting to explore 840 whether higher-level auditory regionalization may follow along some of the 'fault lines' revealed 841 by shared local tonotopic and myelin gradients, and whether or not more sophisticated and fine- 842 grained spectral attentional manipulations may reveal a relationship between the degree of 843 attentional malleability and underlying cortical architecture and circuitry. 844 informative to examine interactions with several frontal regions whose potential analogues are 833 known to have direct feedforward and feedback connections in macaque monkeys (Romanski 834 and Goldman-Rakic, 2002), and where in ferret there are clear modulatory influences on 835 primary and non-primary auditory cortex during learning (Atiani et al., 2014; Shamma and Fritz, 836 2014). Similar to recent work in vision (Klein et al., 2014; Puckett and DeYoe, 2015), it will also 837 be useful to establish the shape of the attentional population receptive field, and how this varies 838 across auditory areas and relates to stimulus-driven auditory population receptive field size 839 (Thomas et al., 2015). Finally, following on from our own pilot work, it will be exciting to explore 840 whether higher-level auditory regionalization may follow along some of the 'fault lines' revealed 841 by shared local tonotopic and myelin gradients, and whether or not more sophisticated and fine- 842 grained spectral attentional manipulations may reveal a relationship between the degree of 843 attentional malleability and underlying cortical architecture and circuitry. 844 845 33 33 Ahveninen J, Chang W-T, Huang S, Keil B, Kopco N, Rossi S, Bonmassar G, Witzel T, Polimeni 849 JR (2016) Intracortical depth analyses of frequency-sensitive regions of human auditory 850 cortex using 7TfMRI. NeuroImage 143:116–127. 851 Atiani S, David SV, Elgueda D, Locastro M, Radtke-Schuller S, Shamma SA, Fritz JB (2014) 852 Emergent Selectivity for Task-Relevant Stimuli in Higher-Order Auditory Cortex. Neuron 853 82:486–499. 854 Atiani S, David SV, Elgueda D, Locastro M, Radtke-Schuller S, Shamma SA, Fritz JB (2014) 852 Emergent Selectivity for Task-Relevant Stimuli in Higher-Order Auditory Cortex. Neuron 853 82:486–499. 854 Barascud N, Pearce MT, Griffiths TD, Friston KJ, Chait M (2016) Brain responses in humans 855 reveal ideal observer-like sensitivity to complex acoustic patterns. Proc Natl Acad Sci 856 USA 113:E616–E625. 857 Barton B, Venezia JH, Saberi K, Hickok G, Brewer AA (2012) Orthogonal acoustic dimensions 858 define auditory field maps in human cortex. Proc Natl Acad Sci USA. 859 Brewer AA, Barton B (2016) Maps of the Auditory Cortex. Annu Rev Neurosci 39:385–407. 860 Calford MB, Semple MN (1995) Monaural inhibition in cat auditory cortex. J Neurophysiol 861 73:1876–1891. 862 Callaghan MF, Helms G, Lutti A, Mohammadi S, Weiskopf N (2014) A general linear 863 relaxometry model of R1 using imaging data. Magn Reson Med 73:1309–1314. 864 Carey D, Caprini F, Allen M, Lutti A, Weiskopf N, Rees G, Callaghan MF, Dick F (2017) 865 Quantitative MRI provides markers of intra-, inter-regional, and age-related differences in 866 young adult cortical microstructure. bioRxiv, 139568. 867 34 Cox RW (2012) AFNI: what a long strange trip it's been. NeuroImage 62:743–747. 868 Cox RW (2012) AFNI: what a long strange trip it's been. NeuroImage 62:743–747. 868 Da Costa S, van der Zwaag W, Marques JP, Frackowiak RSJ, Clarke S, Saenz M (2011) 869 Human Primary Auditory Cortex Follows the Shape of Heschl's Gyrus. Journal of 870 Neuroscience 31:14067–14075. 871 Da Costa S, Van Der Zwaag W, Miller LM, Clarke S, Saenz M (2013) Tuning in to sound: 872 frequency-selective attentional filter in human primary auditory cortex. Journal of 873 Neuroscience 33:1858–1863. 874 Dale AM, Sereno MI (1993) Improved localization of cortical activity by combining EEG and 875 MEG with MRI cortical surface reconstruction: a linear approach. J Cogn Neurosci 876 5:162–176. 877 David SV, Fritz JB, Shamma SA (2012) Task reward structure shapes rapid receptive field 878 plasticity in auditory cortex. Proc Natl Acad Sci USA 109:2144–2149. 879 De Martino F, Moerel M, Ugurbil K, Goebel R, Yacoub E, Formisano E (2015a) Frequency 880 preference and attention effects across cortical depths in the human primary auditory 881 cortex. Proc Natl Acad Sci USA 112:16036–16041. 882 De Martino F, Moerel M, Xu J, van de Moortele PF, Ugurbil K, Goebel R, Yacoub E, Formisano 883 E (2015b) High-Resolution Mapping of Myeloarchitecture In Vivo: Localization of 884 Auditory Areas in the Human Brain. Cerebral Cortex 25:3394–3405. 885 Dick F, Tierney AT, Lutti A, Josephs O, Sereno MI, Weiskopf N (2012) In Vivo Functional and 886 Myeloarchitectonic Mapping of Human Primary Auditory Areas. Journal of Neuroscience 887 32:16095–16105. 888 35 Dinse J, Härtwich N, Waehnert MD, Tardif CL, Schäfer A, Geyer S, Preim B, Turner R, Bazin PL 889 (2015) A cytoarchitecture-driven myelin model reveals area-specific signatures in human 890 primary and secondary areas using ultra-high resolution in-vivo brain MRI. NeuroImage 891 114:71–87. 892 Engel SA (2012) The development and use of phase-encoded functional MRI designs. 893 NeuroImage 62:1195–1200. 894 Engel SA, Rumelhart DE, Wandell BA, Lee AT, Glover GH, Chichilnisky EJ, Shadlen MN (1994) 895 fMRI of human visual cortex. Nature 369:525. 896 Feinberg DA, Moeller S, Smith SM, Auerbach E, Ramanna S, Glasser MF, Miller KL, Ugurbil K, 897 Yacoub E (2010) Multiplexed Echo Planar Imaging for Sub-Second Whole Brain FMRI 898 and Fast Diffusion Imaging Valdes-Sosa PA, ed. PLoS ONE 5:e15710. 899 Feinberg DA, Setsompop K (2013) Ultra-fast MRI of the human brain with simultaneous multi- 900 slice imaging. Journal of Magnetic Resonance 229:90–100. 901 Fischl B, Sereno MI, Tootell RB, Dale AM (1999) High-resolution intersubject averaging and a 902 coordinate system for the cortical surface. Hum Brain Mapp 8:272–284. 903 Fritz JB, David SV, Radtke-Schuller S, Yin P, Shamma SA (2010) Adaptive, behaviorally gated, 904 persistent encoding of task-relevant auditory information in ferret frontal cortex. Nature 905 Neuroscience 13:1011–1019. 906 Fritz JB, Elhilali M, David SV, Shamma SA (2007a) Does attention play a role in dynamic 907 receptive field adaptation to changing acoustic salience in A1? Hear Res 229:186–203. 908 Fritz JB, Elhilali M, Shamma SA (2005) Differential dynamic plasticity of A1 receptive fields 909 during multiple spectral tasks. J Neurosci 25:7623–7635. 910 36 Fritz JB, Elhilali M, Shamma SA (2007b) Adaptive changes in cortical receptive fields induced 911 by attention to complex sounds. J Neurophysiol 98:2337–2346. 912 Glasser MF, Coalson TS, Robinson EC, Hacker CD, Harwell J, Yacoub E, Ugurbil K, Andersson 913 J, Beckmann CF, Jenkinson M, Smith SM, Van Essen DC (2016) A multi-modal 914 parcellation of human cerebral cortex. Nature:1–11. 915 Green TJ, McKeown JD (2001) Capture of attention in selective frequency listening. J Exp 916 Psychol Hum Percept Perform 27:1197–1210. 917 Greenberg GZ (1968) Frequency-Response Characteristic of Auditory Observers Detecting 918 Signals of a Single Frequency in Noise: The Probe-Signal Method. J Acoust Soc Am 919 44:1513. 920 Greve DN, Fischl B (2009) Accurate and robust brain image alignment using boundary-based 921 registration. NeuroImage 48:63–72. 922 Hackett TA, Preuss TM, Kaas JH (2001) Architectonic identification of the core region in 923 auditory cortex of macaques, chimpanzees, and humans. J Comp Neurol 441:197–222. 924 auditory cortex of macaques, chimpanzees, and humans. J Comp Neurol 441:197–222. 924 Hackett TA (2007) Organization and Correspondence of the Auditory Cortex of Humans and 925 Nonhuman Primates. Evolution of the Nervous System, JH Kaas, Ed:109–119.Hagler D, 926 Saygin A, Sereno M (2006) Smoothing and cluster thresholding for cortical surface- 927 based group analysis of fMRI data. NeuroImage 33:1093–1103. 928 Hackett TA (2007) Organization and Correspondence of the Auditory Cortex of Humans and 925 Nonhuman Primates. Evolution of the Nervous System, JH Kaas, Ed:109–119.Hagler D, 926 Saygin A, Sereno M (2006) Smoothing and cluster thresholding for cortical surface- 927 based group analysis of fMRI data. NeuroImage 33:1093–1103. 928 Hagler DJ Jr, Sereno MI (2006) Spatial maps in frontal and prefrontal cortex. NeuroImage 929 29:567–577. 930 Hagler DJ, Riecke L, Sereno MI (2007) Parietal and superior frontal visuospatial maps activated 931 by pointing and saccades. NeuroImage 35:1562–1577. 932 Hagler DJ, Riecke L, Sereno MI (2007) Parietal and superior frontal visuospatial maps activated 931 by pointing and saccades. NeuroImage 35:1562–1577. 932 37 Helms G, Dathe H, Dechent P (2008) Quantitative FLASH MRI at 3T using a rational 933 approximation of the Ernst equation. Magn Reson Med 59:667–672. 934 Helms G, Draganski B, Frackowiak R, Ashburner J, Weiskopf N (2009) Improved segmentation 935 of deep brain grey matter structures using magnetization transfer (MT) parameter maps. 936 NeuroImage 47:194–198. 937 Herdener M, Esposito F, Scheffler K, Schneider P, Logothetis NK, Uludağ K, Kayser C (2013) 938 Spatial representations of temporal and spectral sound cues in human auditory cortex. 939 Cortex 49:2822–2833. 940 Herrmann B, Henry MJ (2013) Frequency-specific adaptation in human auditory cortex depends 941 on the spectral variance in the acoustic stimulation. Journal of …. 942 Humphries C, Liebenthal E, Binder JR (2010) Tonotopic organization of human auditory cortex. 943 NeuroImage 50:1202–1211. 944 Idemaru K, Holt LL (2011) Word recognition reflects dimension-based statistical learning. J Exp 945 Psychol Hum Percept Perform 37:1939–1956. 946 Kastner S, Ungerleider LG (2000) Mechanisms of visual attention in the human cortex. Annu 947 Rev Neurosci 23:315–341. 948 Klein BP, Harvey BM, Dumoulin SO (2014) Attraction of Position Preference by Spatial Attention 949 throughout Human Visual Cortex. Neuron 84:227–237. 950 Kuchibhotla KV, Gill JV, Lindsay GW, Papadoyannis ES, Field RE, Sten TAH, Miller KD, 951 Froemke RC (2017) Parallel processing by cortical inhibition enables context-dependent 952 behavior. Nature Neuroscience 20:62–71. 953 Helms G, Dathe H, Dechent P (2008) Quantitative FLASH MRI at 3T using a rational 933 approximation of the Ernst equation. Magn Reson Med 59:667–672. 934 Helms G, Draganski B, Frackowiak R, Ashburner J, Weiskopf N (2009) Improved segmentation 935 of deep brain grey matter structures using magnetization transfer (MT) parameter maps. 936 NeuroImage 47:194–198. 937 Herdener M, Esposito F, Scheffler K, Schneider P, Logothetis NK, Uludağ K, Kayser C (2013) 938 Spatial representations of temporal and spectral sound cues in human auditory cortex. 939 Cortex 49:2822–2833. 940 Herrmann B, Henry MJ (2013) Frequency-specific adaptation in human auditory cortex depends 941 on the spectral variance in the acoustic stimulation. Journal of …. 942 Herrmann B, Henry MJ (2013) Frequency-specific adaptation in human auditory cortex depends 941 on the spectral variance in the acoustic stimulation. Journal of …. 942 Humphries C, Liebenthal E, Binder JR (2010) Tonotopic organization of human auditory cortex. 943 NeuroImage 50:1202–1211. 944 Idemaru K, Holt LL (2011) Word recognition reflects dimension-based statistical learning. J Exp 945 Psychol Hum Percept Perform 37:1939–1956. 946 Kastner S, Ungerleider LG (2000) Mechanisms of visual attention in the human cortex. Annu 947 Rev Neurosci 23:315–341. 948 Klein BP, Harvey BM, Dumoulin SO (2014) Attraction of Position Preference by Spatial Attention 949 throughout Human Visual Cortex. Neuron 84:227–237. 950 Kuchibhotla KV, Gill JV, Lindsay GW, Papadoyannis ES, Field RE, Sten TAH, Miller KD, 951 Froemke RC (2017) Parallel processing by cortical inhibition enables context-dependent 952 behavior. Nature Neuroscience 20:62–71. 953 38 Kuehn E, Dinse J, Jakobsen E, Long X, Schäfer A, Bazin P-L, Villringer A, Sereno MI, Margulies 954 DS (2017) Body Topography Parcellates Human Sensory and Motor Cortex. Cerebral 955 Cortex:1–16. 956 Langers DRM, Krumbholz K, Bowtell RW, Hall DA (2014) Neuroimaging paradigms for tonotopic 957 mapping (I): the influence of sound stimulus type. NeuroImage 100:650–662. 958 Langers DRM, van Dijk P (2012) Mapping the tonotopic organization in human auditory cortex 959 with minimally salient acoustic stimulation. Cerebral Cortex 22:2024–2038. 960 Leaver AM, Rauschecker JP (2016) Functional Topography of Human Auditory Cortex. Journal 961 of Neuroscience 36:1416–1428. 962 Lutti A, Dick F, Sereno MI, Weiskopf N (2014) Using high-resolution quantitative mapping of R1 963 as an index of cortical myelination. NeuroImage 93 Pt 2:176–188. 964 Lutti A, Hutton C, Finsterbusch J, Helms G, Weiskopf N (2010) Optimization and validation of 965 methods for mapping of the radiofrequency transmit field at 3T. Magn Reson Med 966 64:229–238. 967 Lutti A, Stadler J, Josephs O, Windischberger C, Speck O, Bernarding J, Hutton C, Weiskopf N 968 (2012) Robust and fast whole brain mapping of the RF transmit field B1 at 7T. PLoS 969 ONE 7:e32379. 970 Merzenich MM, Brugge JF (1973) Representation of the cochlear partition of the superior 971 temporal plane of the macaque monkey. Brain Res 50:275–296. 972 Moerel M, De Martino F, Formisano E (2012) Processing of Natural Sounds in Human Auditory 973 Cortex: Tonotopy, Spectral Tuning, and Relation to Voice Sensitivity. Journal of 974 Neuroscience 32:14205–14216. 975 Kuehn E, Dinse J, Jakobsen E, Long X, Schäfer A, Bazin P-L, Villringer A, Sereno MI, Margulies 954 DS (2017) Body Topography Parcellates Human Sensory and Motor Cortex. Cerebral 955 Cortex:1–16. 956 Langers DRM, Krumbholz K, Bowtell RW, Hall DA (2014) Neuroimaging paradigms for tonotopic 957 mapping (I): the influence of sound stimulus type. NeuroImage 100:650–662. 958 Langers DRM, van Dijk P (2012) Mapping the tonotopic organization in human auditory cortex 959 with minimally salient acoustic stimulation. Cerebral Cortex 22:2024–2038. 960 Leaver AM, Rauschecker JP (2016) Functional Topography of Human Auditory Cortex. Journal 961 of Neuroscience 36:1416–1428. 962 Lutti A, Dick F, Sereno MI, Weiskopf N (2014) Using high-resolution quantitative mapping of R1 963 as an index of cortical myelination. NeuroImage 93 Pt 2:176–188. 964 Lutti A, Hutton C, Finsterbusch J, Helms G, Weiskopf N (2010) Optimization and validation of 965 methods for mapping of the radiofrequency transmit field at 3T. Magn Reson Med 966 64:229–238. 967 Lutti A, Stadler J, Josephs O, Windischberger C, Speck O, Bernarding J, Hutton C, Weiskopf N 968 (2012) Robust and fast whole brain mapping of the RF transmit field B1 at 7T. PLoS 969 ONE 7:e32379. 970 Merzenich MM, Brugge JF (1973) Representation of the cochlear partition of the superior 971 temporal plane of the macaque monkey. Brain Res 50:275–296. 972 Merzenich MM, Brugge JF (1973) Representation of the cochlear partition of the superior 971 temporal plane of the macaque monkey. Brain Res 50:275–296. 972 Moerel M, De Martino F, Formisano E (2012) Processing of Natural Sounds in Human Auditory 973 Cortex: Tonotopy, Spectral Tuning, and Relation to Voice Sensitivity. Journal of 974 Neuroscience 32:14205–14216. 975 39 Moerel M, De Martino F, Formisano E (2014) An anatomical and functional topography of 976 human auditory cortical areas. Frontiers in Neuroscience 8. 977 Moerel M, De Martino F, Santoro R, Ugurbil K, Goebel R, Yacoub E, Formisano E (2013) 978 Processing of natural sounds: characterization of multipeak spectral tuning in human 979 auditory cortex. Journal of Neuroscience 33:11888–11898. 980 Moore BCJ, Hafter ER, Glasberg BR (1996) The probe-signal method and auditory-filter shape: 981 Results from normal- and hearing-impaired subjects. J Acoust Soc Am 99:542–552. 982 Morel A, Garraghty PE, Kaas JH (1993) Tonotopic organization, architectonic fields, and 983 connections of auditory cortex in macaque monkeys. J Comp Neurol 335:437–459. 984 Morosan P, Rademacher J, Schleicher A, Amunts K, Schormann T, Zilles K (2001) Human 985 primary auditory cortex: cytoarchitectonic subdivisions and mapping into a spatial 986 reference system. NeuroImage 13:684–701. 987 O'Connell MN, Barczak A, Schroeder CE, Lakatos P (2014) Layer Specific Sharpening of 988 Frequency Tuning by Selective Attention in Primary Auditory Cortex. Journal of 989 Neuroscience 34:16496–16508. 990 Oh J, Kwon JH, Yang PS, Jeong J (2013) Auditory imagery modulates frequency-specific areas 991 in the human auditory cortex. J Cogn Neurosci 25:175–187. 992 Paltoglou AE, Sumner CJ, Hall DA (2009) Examining the role of frequency specificity in the 993 enhancement and suppression of human cortical activity by auditory selective attention. 994 Hear Res 257:106–118. 995 Petkov CI, Kang X, Alho K, Bertrand O, Yund EW, Woods DL (2004) Attentional modulation of 996 human auditory cortex. Nature Neuroscience 7:658–663. 997 40 Pierpaoli C (2010) Quantitative Brain MRI. Topics in Magnetic Resonance Imaging, 21:63 998 Preibisch C, Deichmann R (2009) Influence of RF spoiling on the stability and accuracy of T1 999 mapping based on spoiled FLASH with varying flip angles. Magn Reson Med 61:125– 1000 135. 1001 Puckett AM, DeYoe EA (2015) The Attentional Field Revealed by Single-Voxel Modeling of fMRI 1002 Time Courses. Journal of Neuroscience 35:5030–5042. 1003 Puckett AM, DeYoe EA (2015) The Attentional Field Revealed by Single-Voxel Modeling of fMRI 1002 Time Courses. Journal of Neuroscience 35:5030–5042. 1003 Rao AA, Talavage TM (2005) Reliability of phase-encode mapping in the presence of spatial 1004 non-stationarity of response latency. NeuroImage 28:563–578. 1005 Rao AA, Talavage TM (2005) Reliability of phase-encode mapping in the presence of spatial 1004 non-stationarity of response latency. NeuroImage 28:563–578. 1005 Riecke L, Peters JC, Valente G, Kemper VG, Formisano E, Sorger B (2016) Frequency- 1006 Selective Attention in Auditory Scenes Recruits Frequency Representations Throughout 1007 Human Superior Temporal Cortex. Cerebral Cortex. 1008 Romanski LM, Goldman-Rakic PS (2002) An auditory domain in primate prefrontal cortex. 1009 Nature Neuroscience 5:15–16. 1010 Saenz M, Langers DRM (2014) Tonotopic mapping of human auditory cortex. Hear Res 1011 307:42–52. 1012 Saygin AP, Sereno MI (2008) Retinotopy and attention in human occipital, temporal, parietal, 1013 and frontal cortex. Cereb Cortex 18:2158–2168. 1014 Scharf B (1989) Spectral specificity in auditory detection: The effect of listening on hearing. 1015 Journal of the Acoustical Society of Japan (E). 1016 Scharf B, Quigley S, Aoki C, Peachey N, Reeves A (1987) Focused auditory attention and 1017 frequency selectivity. Percept Psychophys 42:215–223. 1018 41 Schönwiesner M, Zatorre RJ (2009) Spectro-temporal modulation transfer function of single 1019 voxels in the human auditory cortex measured with high-resolution fMRI. Proc Natl Acad 1020 Sci USA 106:14611–14616. 1021 Schwarzkopf DS, Song C, Rees G (2011) The surface area of human V1 predicts the subjective 1022 experience of object size. Nature Neuroscience 14:28–30. 1023 Sereno MI, Dale AM, Reppas JB, Kwong KK, Belliveau JW, Brady TJ, Rosen BR, Tootell RB 1024 (1995) Borders of multiple visual areas in humans revealed by functional magnetic 1025 resonance imaging. Science (New York, NY) 268:889–893. 1026 Sereno MI, Huang R-S (2006) A human parietal face area contains aligned head-centered 1027 visual and tactile maps. Nature Neuroscience 9:1337–1343. 1028 Sereno MI, Lutti A, Weiskopf N, Dick F (2013) Mapping the human cortical surface by combining 1029 quantitative T(1) with retinotopy. Cerebral Cortex 23:2261–2268. 1030 Shamma S, Fritz J (2014) Adaptive auditory computations. Curr Opin Neurobiol 25:164–168. 1031 Smith SM, Jenkinson M, Woolrich MW, Beckmann CF, Behrens TEJ, Johansen-Berg H, 1032 Bannister PR, De Luca M, Drobnjak I, Flitney DE, Niazy RK, Saunders J, Vickers J, 1033 Zhang Y, De Stefano N, Brady JM, Matthews PM (2004) Advances in functional and 1034 structural MR image analysis and implementation as FSL. NeuroImage 23:S208–S219. 1035 Stüber C, Morawski M, Schäfer A, Labadie C, Wähnert M, Leuze C, Streicher M, Barapatre N, 1036 Reimann K, Geyer S, Spemann D, Turner R (2014) Myelin and iron concentration in the 1037 human brain: A quantitative study of MRI contrast. NeuroImage. 1038 42 Sutter ML, Schreiner CE, McLean M, O'Connor KN, Loftus WC (1999) Organization of Inhibitory 1039 Frequency Receptive Fields in Cat Primary Auditory Cortex. J Neurophysiol 82:2358– 1040 2371. 1041 Sweet RA, Dorph-Petersen K-A, Lewis DA (2005) Mapping auditory core, lateral belt, and 1042 parabelt cortices in the human superior temporal gyrus. J Comp Neurol 491:270–289. 1043 Talavage TM, Edmister WB (2004) Nonlinearity of FMRI responses in human auditory cortex. 1044 Hum Brain Mapp 22:216–228. 1045 Talavage TM, Sereno MI, Melcher JR, Ledden PJ, Rosen BR, Dale AM (2004) Tonotopic 1046 organization in human auditory cortex revealed by progressions of frequency sensitivity. 1047 J Neurophysiol 91:1282–1296. 1048 Tardif CL, Gauthier CJ, Steele CJ, Bazin P-L, Schäfer A, Schaefer A, Turner R, Villringer A 1049 (2016) Advanced MRI techniques to improve our understanding of experience-induced 1050 neuroplasticity. NeuroImage 131:55–72. 1051 Tardif CL, Schäfer A, Waehnert M, Dinse J, Turner R, Bazin P-L (2015) Multi-contrast multi- 1052 scale surface registration for improved alignment of cortical areas. NeuroImage 1053 111:107–122. 1054 Thomas JM, Huber E, Stecker GC, Boynton GM, Saenz M, Fine I (2015) Population receptive 1055 field estimates of human auditory cortex. NeuroImage 105:428–439. 1056 Turner R (2015) Myelin imaging. In: Brain mapping: An encyclopedic reference, pp 137–142. 1057 Academic Press: Elsevier. 1058 43 43 Wallace M, Cronin M, Bowtell RW, Scott I, Palmer AR, Gowland P (2016) Histological Basis of 1059 Laminar MRI Patterns in High Resolution Images of Fixed Human Auditory Cortex. 1060 Frontiers in Neuroscience. 1061 Wehr M, Zador AM (2003) Balanced inhibition underlies tuning and sharpens spike timing in 1062 auditory cortex. Nature 426:442–446. 1063 Woods DL, Stecker GC, Rinne T, Herron TJ, Cate AD, Yund EW, Liao I, Kang X (2009) 1064 Functional maps of human auditory cortex: effects of acoustic features and attention. 1065 PLoS ONE 4:e5183. 1066 1067 Wallace M, Cronin M, Bowtell RW, Scott I, Palmer AR, Gowland P (2016) Histological Basis of 1059 Laminar MRI Patterns in High Resolution Images of Fixed Human Auditory Cortex. 1060 Frontiers in Neuroscience. 1061 Wehr M, Zador AM (2003) Balanced inhibition underlies tuning and sharpens spike timing in 1062 auditory cortex. Nature 426:442–446. 1063 Woods DL, Stecker GC, Rinne T, Herron TJ, Cate AD, Yund EW, Liao I, Kang X (2009) 1064 Functional maps of human auditory cortex: effects of acoustic features and attention. 1065 PLoS ONE 4:e5183. 1066 44 44 Figure 1. Stimuli and design overview. (A) In a representative 12.8s Tonotopy block a 1070 neutral verbal prompt “hear” precedes 14 four-tone mini-sequences sampled around one of 5 1071 center frequencies. The task is to detect the 1-3 mini-sequence repeats embedded within the 1072 block. The grey box highlights a single mini-sequence repeat. (B) A single Attention-Tonotopy 1073 (Attn-tono) block includes two simultaneous streams of mini-sequences with distinct center 1074 frequencies. Mini-sequence repeats occur in each stream. A verbal prompt (“high” or “low”) 1075 directs listeners to attend to one stream and report mini-sequence repeats in that stream, while 1076 ignoring repeats in the unattended stream. Two randomly-ordered orientation tones at the 1077 center frequency of each stream alert listeners to the frequency neighborhood of the upcoming 1078 streams. (C) A single 64s cycle of Stepped Attn-tono blocks includes five 12.8s blocks that 1079 step up (shown), or down, in center frequency. In this single cycle, the frequency band to which 1080 attention is directed by the verbal prompt (indicated with “high”/”low” above each block) is 1081 acoustically matched with the Tonotopy cycle shown in (B), but there are always competing 1082 unattended mini-sequences in a distinct frequency band. (D) A single 64s cycle of 1083 Randomized Attn-tono blocks is acoustically identical to the Stepped Attn-tono cycle in (C), 1084 except that half of the verbal prompts have been swapped, and therefore no longer cue 1085 attention to frequency with consistent phase lags. (E) The distinction between Stepped and 1086 Randomized Attn-tono blocks is highlighted by examining the first three (of eight) cycles of a 1087 Stepped (top) versus Randomized (bottom) Attn-tono run. The focus of attention is color 1088 coded in the frequency-band-specific manner shown in (A). For the Stepped condition (top), 1089 there is a consistent relationship between the stimulus phase lag and the attended frequency 1090 across cycles within a run. Thus, for voxels that show a consistently higher response at one 1091 attended frequency band compared to all others, there will be a periodic response at 8 1092 cycles/run at a given phase lag corresponding to the particular frequency band attended. For 1093 45 the Randomized condition (bottom), there is no consistent relationship, providing a control 1094 condition for Fourier analyses because frequency-band-directed attention is aperiodic across a 1095 run. (F) The stimulus phase lag with the highest periodic BOLD signal amplitude is determined 1096 for each voxel, mapped to a color scale, and then painted onto the cortical surface patch. BOLD 1097 signal amplitude is mapped to the color’s saturation. (Note: In these panels A-D, stimulus 1098 intensity is adjusted across the spectrum to aid visual presentation of energy across frequency 1099 bands. See Methods for details on actual intensity across frequency bands). 1100 1101 Figure 2. Group activation for Tonotopy and Attention-Tonotopy conditions, with R1 1102 contours showing putative auditory core. The left-most panel shows cortical-surface-based 1103 group-averaged R1, projected on the lateral inflated surface of one subject. (The left hemisphere 1104 is mirror-reversed to align cortical maps for visual comparison). For tonotopic map display, a 1105 patch of cortex including the entire temporal plane (shown in purple on the inflated surface) was 1106 cut and flattened. Panels A-C show this region enlarged, with isocontour lines showing 1107 quantitative R1 values for the group-averaged putative auditory core, and color maps showing 1108 group-averaged best frequency as a function of (A) Tonotopy, (B) Attn-tono (Stepped), and (C) 1109 Attn-tono (Randomized Control) conditions. The stars are fiduciary points to assist in visual 1110 comparisons of maps across conditions; the outline of Heschl's gyrus is in yellow dashed lines 1111 (in (A), from the individual subject whose cortical patch was used). Consistent with previous 1112 work, the tonotopic map is characterized by two pairs of three interlacing best-frequency 1113 ‘fingers,’ with the high-frequency fingers (red/orange colormap) showing greatest frequency 1114 preference medially and extending laterally, where they meet interdigitated lower-frequency 1115 fingers (green/yellow colormap) extending lateral to medial, with the longest ‘middle’ lower- 1116 frequency finger extending about halfway into auditory core. This pattern is evident in Fourier- 1117 analysis-derived maps of the ‘stepped’ Attn-tono condition but not in the ‘randomized control’ 1118 Attn-tono condition, for which the attentional response was phase-cancelled. All maps are 1119 46 statistically masked by overall activation to tonotopy stimuli in each hemisphere (cluster- 1120 corrected p < 10-8, and gently shaded to show relative amplitude). 1121 Figure 3. Individual subjects’ Tonotopy and Attention-Tonotopy maps. Each subject’s 1123 Tonotopic and Attn-Tono (Stepped) Fourier-analysis-derived maps are displayed on the same 1124 subject’s flattened superior temporal cortical patch. R1 isocontours around presumptive auditory 1125 core are shown in white, with thick solid lines depicting the lowest-valued (outermost) R1 1126 isocontour, thin solid lines depicting the highest (innermost) R1 isocontour, and dashed lines 1127 showing intermediate values. (R1 values differ somewhat across individuals). Dashed black lines 1128 indicate the outline of the cortical-surface-morphed 'TE1.0' label, where the area inside the line 1129 contains vertices estimated to have a p > 0.3 probability of falling within primary auditory cortex 1130 based on the Morosan et al., (2001) postmortem probability atlas (see Methods). Activation 1131 maps are gently shaded to show changes in response amplitude, but are unthresholded for 1132 comparison with individual maps from previous studies, e.g., Da Costa et al., (2011). 1133 1134 Figure 4. Comparison of responses in regression-based ‘Winner-Takes-All’ maps, 1135 Tonotopy and Attn-Tono. The color maps projected onto the right (top panels) and left (bottom 1136 panels) hemisphere cortical patches (same as patches shown in purple in Figure 2) show the 1137 cross-subject average best frequency band (‘Winner-Takes-All,’ or WTA) for 'stepped' tonotopic 1138 (left) 'stepped' attn-tono (middle) and 'randomized’ attn-tono conditions. Note that with the 1139 regression-based approach, the 'randomized' condition is also expected to evoke strong 1140 attentionally-driven tonotopic maps. The dotted yellow line depicts the outermost R1 contour 1141 (0.66 sec-1) around presumptive auditory core as shown in Figure 2. 1142 1143 Figure 5. Comparison of Tonotopy and Attn-tono maps. (A) ‘Concordance maps’ are 1144 rendered in heatscale on the inflated hemispheres to illustrate the similarity in best frequency 1145 Figure 5. Comparison of Tonotopy and Attn-tono maps. (A) ‘Concordance maps’ are 1144 rendered in heatscale on the inflated hemispheres to illustrate the similarity in best frequency 1145 47 between tonotopic and attn-tono maps (the latter averaged over stepped and randomized 1146 blocks). These maps were calculated in two stages. First, in each subject’s native EPI space, a 1147 voxel was coded as ‘1’ if tonotopy and attn-tono stimuli evoked the same best frequency, and 1148 otherwise coded as ‘0’. Second, for each subject, the concordance maps were resampled to the 1149 individual’s cortical surface, and projected onto the unit icosahedron for cross-subject surface- 1150 based averaging, thereby creating a composite measure of agreement between tonotopy and 1151 attn-tono maps, weighted by the consistency of this agreement across subjects. The 1152 concordance maps are statistically masked with a cross-subject t-map, calculated versus 1153 chance agreement (p=0.20) with a surface cluster correction of p < 0.001 (vertex-wise p<0.01, 1154 cluster threshold surface area> 74 mm2, (Hagler et al., 2006). To demonstrate the extent of 1155 tonotopically-mapped cortex that is similarly mapped through spectrally-directed attention, the 1156 phase-encoded tonotopy cortical patches from Figure 2a are overlaid with the outline of the 1157 thresholded concordance map shown by the yellow dotted line. The white solid outline shows 1158 the Bonferroni-corrected ROI-wise correspondence outline from the 'ROI quilt' in panel B (B) 1159 The shading in each small ROI patch shows the z-score for the partial fit between tonotopy and 1160 attn-tono responses to each frequency band (with subjects as a random factor). ROIs with 1161 significant z-scores (Bonferroni-corrected p-value threshold of p < 0.05) are indicated by the thin 1162 white outline. 1163 Figure 6. Comparison of responses in regression-based ‘Loser-Takes-All’ maps, 1165 Tonotopy and Attn-tono. (A) The colormaps projected onto the same cortical patches as 1166 Figures 2 and 4 show cross-subject group-average maps that depict the frequency band that 1167 drives the least activation compared to all other frequency bands (‘Loser-Takes-All’, or LTA) in 1168 tonotopy and attn-tono (stepped plus randomized blocks) conditions and in right and left 1169 hemispheres. As in Figure 4, the presumptive auditory core shown by the dashed yellow line 1170 depicting the outermost R1 contour (0.66 sec-1). (B) The tonotopy versus attn-tono LTA 1171 48 concordance map was created as in Figure 5a; note that the midpoint of the heatscale has been 1172 lowered slightly compared to Figure 5a, reflecting the overall somewhat lower concordance in 1173 the LTA maps compared to WTA. The dotted yellow R1 isocontour is the same as Figure 4. 1174 1175 Figure 7. Comparison of maps when best frequency is attended versus ignored. The 1176 heatscale (t-values, thresholded as in Figure 5a) depicts the cross-subject cortical-surface- 1177 based average difference in activation when the subject-specific best frequency band of each 1178 voxel was attended versus ignored. The dotted green R1 isocontour estimating auditory core is 1179 as in Figure 4. 1180 1181 Figure 8. Local normalized covariance between R1 values and tonotopic and attn- tono 1182 response amplitude. The heatscale value at each vertex represents the normalized spatial 1183 covariance within a 4mm (2D) radius between R1 and the amplitude of the tonotopic or attn- 1184 tono signal (e.g., the amplitude of the Fourier component at the stimulus frequency of 8 1185 cycles/run). (A) The cross-subject (N=8) cortical-surface-based average normalized covariance 1186 between R1 and tonotopic amplitude. (B) The R1 versus tonotopy normalized covariance in an 1187 independent cohort (N=6), using data previously acquired with a different tonotopy protocol 1188 (bandpass-filter-swept non-linguistic vocalizations) and on a different scanner (Siemens 3T 1189 Trio); full protocol as described in Dick et al., (2012). (C) The average normalized covariance 1190 between R1 and attn-tono amplitude in the current cohort. 1191 1192 concordance map was created as in Figure 5a; note that the midpoint of the heatscale has been 1172 lowered slightly compared to Figure 5a, reflecting the overall somewhat lower concordance in 1173 the LTA maps compared to WTA. The dotted yellow R1 isocontour is the same as Figure 4. 1174 1175 concordance map was created as in Figure 5a; note that the midpoint of the heatscale has been 1172 lowered slightly compared to Figure 5a, reflecting the overall somewhat lower concordance in 1173 the LTA maps compared to WTA. The dotted yellow R1 isocontour is the same as Figure 4. 1174 1175 concordance map was created as in Figure 5a; note that the midpoint of the heatscale has been 1172 lowered slightly compared to Figure 5a, reflecting the overall somewhat lower concordance in 1173 the LTA maps compared to WTA. The dotted yellow R1 isocontour is the same as Figure 4. 1174 1175 Figure 7. Comparison of maps when best frequency is attended versus ignored. The 1176 heatscale (t-values, thresholded as in Figure 5a) depicts the cross-subject cortical-surface- 1177 based average difference in activation when the subject-specific best frequency band of each 1178 voxel was attended versus ignored. The dotted green R1 isocontour estimating auditory core is 1179 as in Figure 4. 1180 Figure 8. Local normalized covariance between R1 values and tonotopic and attn- tono 1182 response amplitude. The heatscale value at each vertex represents the normalized spatial 1183 covariance within a 4mm (2D) radius between R1 and the amplitude of the tonotopic or attn- 1184 tono signal (e.g., the amplitude of the Fourier component at the stimulus frequency of 8 1185 cycles/run). (A) The cross-subject (N=8) cortical-surface-based average normalized covariance 1186 between R1 and tonotopic amplitude. (B) The R1 versus tonotopy normalized covariance in an 1187 independent cohort (N=6), using data previously acquired with a different tonotopy protocol 1188 (bandpass-filter-swept non-linguistic vocalizations) and on a different scanner (Siemens 3T 1189 Trio); full protocol as described in Dick et al., (2012). (C) The average normalized covariance 1190 between R1 and attn-tono amplitude in the current cohort. 1191 49 49
https://openalex.org/W2944189233	https://europepmc.org/articles/pmc6554677?pdf=render	English	null	A Model for Apoptotic-Cell-Mediated Adaptive Immune Evasion via CD80–CTLA-4 Signaling	Frontiers in pharmacology	2,019	cc-by	11,766	A Model for Apoptotic-Cell-Mediated Adaptive Immune Evasion via CD80–CTLA-4 Signaling Abraam M. Yakoub1* and Stefan Schülke 2 1Department of Molecular and Cellular Physiology, School of Medicine, Stanford University, Stanford, CA, United States, 2 Vice President’s Research Group: Molecular Allergology, Paul-Ehrlich-Institut, Langen, Germany Abraam M. Yakoub1* and Stefan Schülke 2 1Department of Molecular and Cellular Physiology, School of Medicine, Stanford University, Stanford, CA, United States, 2 Vice President’s Research Group: Molecular Allergology, Paul-Ehrlich-Institut, Langen, Germany Apoptotic cells carry a plethora of self-antigens but they suppress eliciting of innate and adaptive immune responses to them. How apoptotic cells evade and subvert adaptive immune responses has been elusive. Here, we propose a novel model to understand how apoptotic cells regulate T cell activation in different contexts, leading mostly to tolerogenic responses, mainly via taking control of the CD80–CTLA-4 coinhibitory signal delivered to T cells. This model may facilitate understanding of the molecular mechanisms of autoimmune diseases associated with dysregulation of apoptosis or apoptotic cell clearance, and it highlights potential therapeutic targets or strategies for treatment of multiple immunological disorders. Edited by: Maria Gerosa, University of Milan, Italy Keywords: apoptosis, immunotolerance, autoimmunity, costimulatory pathway, CD80, CTLA-4, coinhibitory pathway, immune evasion Reviewed by: Satish Ramalingam, SRM Institute of Science and Technology, India Roberta Gualtierotti, University of Milan, Italy Reviewed by: Satish Ramalingam, SRM Institute of Science and Technology, India Roberta Gualtierotti, University of Milan, Italy APOPTOSIS Apoptosis, or programmed cell death (PCD), is the physiological form of cell death that plays an important role in tissue homeostasis and regeneration, as well as maintenance of robust organ functions. While cell death by necrosis may have immunostimulatory and inflammatory effects (Sauter et al., 2000), apoptosis shows no immunostimulatory capacities, and may serve beneficial functions to the host (Sauter et al., 2000; Mahajan et al., 2016). *Correspondence: Abraam M. Yakoub abraamyakoub@gmail.com Our understanding of many aspects of apoptosis has constantly increased over the past ~30 years, due to tremendous effort and a myriad of studies using various model systems. For example, mouse models, due to the power of mouse genetics and with the availability of gene-targeting approaches, have been a mainstay tool to understand both the various functions of apoptosis-related genes in development and the association between gene functions or apoptosis states and disease phenotypes in mammals (see, for instance, Reyes et al., 2010; Gómez-Sintes et al., 2011; Yamaguchi et al., 2011; Wu et al., 2015). Given that the apoptosis machinery is evolutionarily conserved, from worms to mammals, other eukaryotic models have also been used such as drosophila, especially considering the ease of genetic screens and availability of a large number of fly lines (see, for instance, Richardson and Kumar, 2002; Gullaud et al., 2003; Denton et al., 2008; Xu et al., 2009). Even yeast, which, while it lacks main regulators of mammalian apoptosis such as caspases and the B cell lymphoma 2 (Bcl-2) family members, was used to study apoptosis via heterologous expression of many such genes, given the advantages of the yeast model (e.g., easy manipulation via molecular biology or genetics, low cost, and the availability of powerful tools such as the yeast two-hybrid system) (Fleury et al., 2002; Kazemzadeh et al., 2012). In addition to the in vivo models, Specialty section: This article was submitted to Inflammation Pharmacology, a section of the journal Frontiers in Pharmacology Received: 29 August 2018 Accepted: 06 May 2019 Published: 31 May 2019 PERSPECTIVE published: 31 May 2019 doi: 10.3389/fphar.2019.00562 PERSPECTIVE Citation: In the intrinsic pathway, activation of apoptosis is triggered by either developmental signals or genotoxic substances resulting in the release of many proteins including cytochrome C from the mitochondria by pro-apoptotic members of the Bcl-2 family (Nagata and Tanaka, 2017). The released cytochrome C subsequently mediates the formation of apoptosomes in the respective cell’s cytosol, which are multiprotein complexes consisting of cytochrome C, pro-caspase 9, and apoptotic protease-activating factor 1 (APAF1) that process pro-caspase 9 to its mature form (Liu et al., 1996; Zou et al., 1997, 1999). Mature caspase 9 finally mediates the maturation of inactive pro-caspase 3 to its active form caspase 3 (Nagata and Tanaka, 2017). In the extrinsic pathway of apoptosis, binding of FasL (Fas Ligand, expressed on the surface of the apoptosis-inducing cell) to Fas (CD95, tumor necrosis factor receptor superfamily member 6) on the cell destined to undergo apoptosis results in a conformational change in the Fas trimer allowing for the formation of the death-inducing signaling complex (DISC) (Nagata and Tanaka, 2017). DISC is a multiprotein complex containing the Fas-associated death domain protein (FADD) and pro-caspase 8 (Chinnaiyan et al., 1995; Kischkel et al., 1995; Muzio et al., 1996). DISC activation results in the production of mature caspase 3 by DISC-matured caspase 8 (Nagata and Tanaka, 2017). Finally, caspase 3 activated by both apoptosis pathways triggers the apoptosis program via the cleavage of >500 cellular substrates (Nagata and Tanaka, 2017). While FasL expression is restricted to cytotoxic T lymphocytes, T helper type-2 (Th2) cells, and Natural Killer (NK) cells (Kägi et al., 1994; Lowin et al., 1994), Fas is expressed by most cell types (Nagata and Tanaka, 2017). Therefore, FasL-Fas interaction-induced apoptosis is very important for tissue homeostasis. Besides FasL, other ligands such as tumor necrosis factor-alpha (TNF-α), lymphotoxin-alpha (LT-α), TNF-like protein-1A (TL1A), and Apo2L/TNF-related apoptosis- inducing ligand (TRAIL) can also trigger Fas-dependent apoptosis via the extrinsic pathway (Yamada et al., 2017). Besides IL-10, ACs were shown to induce the production of many anti-inflammatory cytokines such as transforming growth factor beta (TGF-β), platelet activating factor (PAF), and prostaglandin E2 (PGE2) (Voll et al., 1997b; Cvetanovic and Ucker, 2004). In addition, macrophage exposure to ACs caused a reduction in the macrophages’ expression of the pro-inflammatory and immunostimulatory cytokines tumor necrosis factor (TNF)-α, IL-12, IL-1β, IL-18, and granulocyte-macrophage colony-stimulating factor (GM-CSF) (Fadok et al., 1998; Kim et al., 2005). Citation: Yakoub AM and Schülke S (2019) A Model for Apoptotic-Cell-Mediated Adaptive Immune Evasion via CD80–CTLA-4 Signaling. Front. Pharmacol. 10:562. doi: 10.3389/fphar.2019.00562 May 2019 \| Volume 10 \| Article 562 1 Frontiers in Pharmacology \| www.frontiersin.org A Model for Apoptotic Immune Evasion Yakoub and Schülke Early, in 1997, a pioneering study (Voll et al., 1997a) showed that peripheral blood-derived macrophages exposed to ACs exhibited enhanced production of the immunosuppressive cytokine interleukin (IL)-10, which is an important immune regulatory molecule that prevents inflammatory immune responses, tissue damage, and the development of autoimmunity. Recently, ACs were shown to induce upregulation of the transcription factor aryl hydrocarbon receptor (AhR) in a Toll-like receptor (TLR) 9-dependent manner, which enhanced production of IL-10 to mediate AC-dependent immunosuppression (Shinde et al., 2018). Consequently, AhR knockout induced autoimmune responses and systemic lupus erythematosus (SLE) disease in a mouse model (Shinde et al., 2018). However, it is important to note that, while IL-10 is mainly considered to have anti-inflammatory effects on a wide range of target cells, recent findings suggest a more complex modulatory function of this important cytokine. Because of its role as an important B cell growth and differentiation factor (that promotes B cell proliferation and IgG production), IL-10 was suggested to contribute to the pathology of SLE via activation of autoreactive B cells (reviewed in Geginat et al., 2016). IL-10 levels were shown to increase in SLE patients and polymorphisms in the IL-10 promoter were strongly associated with SLE development (Peng et al., 2013). In line with these findings, neutralization of IL-10 blocked autoantibody production in SLE patients (Llorente et al., 1995). However, both the source of the pathogenic IL-10 production in SLE patients and its possible contribution to other autoimmune diseases remain to be further characterized (Geginat et al., 2016). in vitro models have also been frequently used to understand the signaling pathways or molecular interactions that regulate apoptosis at the cellular level, in physiological or disease conditions (Calissano et al., 2009; Spencer and Sorger, 2011). Importantly, with the advent of stem cell technologies and in vitro differentiation methods, many human (stem cell-derived) cell types, including neurons, were used to understand apoptosis-related molecular disease mechanisms in the human genetic background (Csöbönyeiová et al., 2016; Fang et al., 2018). Induction of cell death by apoptosis in mammals is initiated by two major signaling cascades: the “extrinsic” and “intrinsic” pathways of apoptosis (Nagata and Tanaka, 2017). Citation: Therefore, ACs are able to modulate the activation state of, and cytokine secretion from, antigen-presenting cells (APCs) which influences both innate and subsequent adaptive immune responses to the ACs. This immune modulation also has consequences for T cell activation upon encountering ACs. For example, suppression of macrophage-derived IL-12 production may prevent the differentiation of self-reactive T helper type-1 (Th1) CD4+ cells and autoimmunity (Trembleau et al., 1995), while AC-induced IL-10 represses the expression of MHC-II and costimulatory molecules required for antigen presentation and subsequent T cell activation (Couper et al., 2008). APOPTOTIC CELLS AND ADAPTIVE IMMUNITY It was initially thought that apoptotic cells (ACs) might be immunologically null, however a plethora of evidence has since then indicated that ACs are immunologically active, exerting, in most cases, anti-inflammatory and immunosuppressive effects. The Route to T Cell Activation Frontiers in Pharmacology \| www.frontiersin.org An Integrative Model of Apoptotic-Cell- Mediated Immune Evasion and Tolerance It was suggested that CTLA-4 and PD-1 may play redundant or complementary functions that differentially target different stages of tolerance (priming, activation, or reactivation of T cells, respectively) (Linsley and Ledbetter, 1993; June et al., 1994; Dahl et al., 2000).h g pi Although CD80 can bind to and activate CD28, significant evidence suggests that it also contributes a strong coinhibitory function. In fact, the binding of CD80 to the coinhibitory receptor CTLA-4 occurs with higher affinity than its binding to the costimulatory receptor CD28 (KD = 0.2 and 4 μM, respectively) (van der Merwe and Davis, 2003; Collins et al., 2005; Butte et al., 2008). Furthermore, the crystal structure of the CD80-CTLA-4 complex showed that CD80 homodimers bind bivalent CTLA-4 homodimers in an unusually stable, high-avidity complex (Ikemizu et al., 2000; Stamper et al., 2001). CTLA-4-mediated coinhibitory signaling is critical for negative regulation of T cell activation and proliferation, as evidenced by the severe lymphoproliferation and multi-organ inflammatory lymphocytic infiltrates observed in mice lacking CTLA-4 signaling (Tivol et al., 1995) or in cancer patients receiving anti-CTLA-4 antibodies (Pardoll, 2012). The role of the costimulatory/coinhibitory molecules in immune responses to ACs in vivo (in mice) has also been suggested by the finding that antigen-coupled ACs (derived from splenocytes) induced T cell tolerance via enhanced IL-10 and PD-L1 expression on AC-ingesting macrophages. These ACs also enhanced Treg activation that maintained immunotolerance in that model (Kushwah et al., 2010). PD-L1 upregulation was dependent on IL-10, as IL-10 neutralization with antibodies reduced the PD-L1 response to ACs. Even when the TCR and CD28 are ligand-activated, CTLA-4 activation can inhibit cell cycle progression and cause proliferative arrest of T cells by suppressing IL-2 production (Krummel and Allison, 1996; Walunas et al., 1996). It thus seems that CTLA-4 has a superdominant, overarching role in the regulation of T cell activation. CTLA-4 selectively reverses CD28-mediated costimulation (Walunas et al., 1996; Tai et al., 2007). Moreover, since CTLA-4 is a higher affinity CD80/86 binding partner, CTLA-4 can compete out CD28 for CD80/CD86 binding, also leading to suppression of T cell activation. The Route to T Cell Activation Upon T cell receptor (TCR) activation by ligand binding, such as by specific-antigen-bound major histocompatibility complex May 2019 \| Volume 10 \| Article 562 Frontiers in Pharmacology \| www.frontiersin.org 2 A Model for Apoptotic Immune Evasion Yakoub and Schülke (MHC) on the surface of APCs, the TCR-associated CD3 chains become tyrosine-phosphorylated leading to recruitment of kinases and scaffold proteins and formation of a supramolecular complex that triggers signaling pathways and transcriptional cascades responsible for T cell differentiation and clonal expansion, as well as effector cell generation (Rathmell et al., 2003; Smith-Garvin et al., 2009; Marko et al., 2010; Schultze et al., 2012). Among those signaling and transcriptional events are the upregulation of the glucose receptor Glut 1 (Frauwirth et al., 2002) and the glutamine receptors Snat1 and Snat2 (Carr et al., 2010), and the activation of the MAPK and PI3K/ Akt pathways (Kannan et al., 2012). Collectively, these events are required to fulfill the metabolic needs of the proliferating T cells and to support cell cycle progression and cytokine production (Appleman et al., 2002).fi ligands were later discovered and reported to modulate APC-T cell interaction (for review, see Kow and Mak, 2013; Attanasio and Wherry, 2016; Schildberg et al., 2016). ligands were later discovered and reported to modulate APC-T cell interaction (for review, see Kow and Mak, 2013; Attanasio and Wherry, 2016; Schildberg et al., 2016). An Integrative Model of Apoptotic-Cell- Mediated Immune Evasion and Tolerance Strong evidence indicates an essential role for the coinhibitory pathway in suppressing adaptive immune responses. First, the role of the coinhibitory signaling in regulating self-tolerance or autoimmunity is supported by the finding that various coinhibitory ligands are expressed, besides APCs, on non-hematopoietic cells which was suggested to play a role in maintaining tissue tolerance by suppressing self-reactive T cells in the periphery (Anderson et al., 2016; Schildberg et al., 2016; Ward-Kavanagh et al., 2016; Janakiram et al., 2017). Notably, some tumors and infectious pathogens evade immune recognition by exploiting such natural tolerance mechanisms (Odorizzi and Wherry, 2012; Pardoll, 2012; Wang et al., 2013; Attanasio and Wherry, 2016; Baumeister et al., 2016). However, TCR ligation alone is insufficient to trigger or maintain robust T cell activation, as this process is tightly regulated by a complex array of costimulatory and coinhibitory ligands and receptors (Esensten et al., 2016). The prototypical costimulatory receptor is CD28, while the prototypical coinhibitory receptors are cytotoxic T lymphocyte antigen-4 (CTLA-4) and PD-1 (Buchbinder and Desai, 2016). Shared ligands between both types of receptors include CD80 (B7-1) and CD86 (B7-2). Stimulation of CD28 potentiates and sustains IL-2 production from T cells and prevents peripheral immunotolerance development (Bour-Jordan et al., 2011; Kow and Mak, 2013). These T cells activated by CD28-B7 signaling then mature and differentiate, subsequently inducing B cell proliferation and differentiation into plasma cells producing antigen-specific antibodies (Kow and Mak, 2013). y ) Moreover, CTLA-4-mediated coinhibition was shown to be essential for terminating T cell activation as CTLA-4−/− mice develop massive lymphoproliferation and early death (Tivol et al., 1995; Waterhouse et al., 1995). CTLA-4 was also suggested as a master switch for peripheral tolerance (Bluestone, 1997). Importantly, the essential role of CTLA-4 in autoimmune regulation was further highlighted by the fact that blockade of CTLA-4 signaling in multiple animal models resulted in aggravation of autoimmune diseases (Karandikar et al., 1996; Luhder et al., 1998; Chitnis et al., 2001). Moreover, CTLA-4 gene single-nucleotide polymorphisms (SNPs) in humans were associated with autoimmune disorders. For example, a SNP in the 6.1-kb (kilobase) 3′ region of the CTLA-4 gene was associated with higher risk of Grave’s disease, autoimmune hypothyroidism, and type 1 diabetes mellitus (Ueda et al., 2003). Blockade of the CTLA-4- or PD-1-mediated coinhibitory signaling accelerated cardiac allograft rejection in C57BL/6 mice receiving BALB/c hearts (Ito et al., 2005). Frontiers in Pharmacology \| www.frontiersin.org An Integrative Model of Apoptotic-Cell- Mediated Immune Evasion and Tolerance In another study, dendritic cells exposed to ACs exhibited reduced T cell proliferation and activation and reduced lipopolysaccharide (LPS)-triggered upregulation of the costimulatory molecule CD86 (Stuart et al., 2002). In that study, ACs did not significantly affect IL-10 levels secreted by dendritic cells, and even IL-10 neutralization by soluble IL-10R did not affect expression of the costimulatory molecules on the dendritic cells (Stuart et al., 2002). Even further, ACs could inhibit LPS-induced activation of bone marrow-derived dendritic cells derived from IL-10-deficient mice; and neutralizing another anti-inflammatory cytokine, TGF-β1, could not suppress the inhibition of dendritic cell maturation by ACs (Stuart et al., 2002), contrary to what was suggested (Chen et al., 2001). In total, these reports suggest that the effect of ACs on adaptive immune responses cannot, or cannot completely, be attributed to secondary effects of the cytokine secretion modulated by ACs, and that ACs may have direct effects on the machinery regulating adaptive immune responses. CD80 binding to CTLA-4 conveys an inhibitory signal to T cell activation that overrides costimulatory signals, counteracting the initiation of T cell activation and proliferation and indeed inducing their apoptosis (Walunas et al., 1994). Moreover, other mechanisms could also contribute to AC-mediated suppression of adaptive immune responses. For example, ACs (derived from dendritic cells) engulfed by dendritic cells induced TGF-β1 secretion and differentiation of naïve T cells into Foxp3+ Tregs (Kushwah et al., 2010). Naïve T cells can differentiate upon antigen recognition into effector T cell subsets such as Th1, Th2, and Th17, or into immunosuppressive Tregs. AC-ingesting dendritic cells suppressed the development of the effector Th17 cells, but enhanced the development of Tregs where dendritic cell-T cell interaction and the costimulatory/coinhibitory signaling were suggested to play a role in Treg induction (Yamazaki et al., 2003; Torchinsky et al., 2009). Similar to the results in macrophages showing upregulation of CD80 by ACs (Yakoub et al., 2018), ACs were reported to induce CD80 and CD86 expression on in vitro differentiated human dendritic cells, which involved both soluble factors secreted and cell-cell contact to achieve the full effects of the ACs (Johansson et al., 2007; Pathak et al., 2012). Notably, however, these effects of ACs on CD80/86 expression on dendritic cells required “pre-activation” of the ACs with anti-CD3 and anti-CD28 antibodies, whereas non-pre-activated ACs showed no effect on CD80/86 expression on dendritic cells (Johansson et al., 2007; Pathak et al., 2012). An Integrative Model of Apoptotic-Cell- Mediated Immune Evasion and Tolerance Besides CD28 and CTLA-4, other costimulatory and coinhibitory receptors and Since ACs were reported to induce IL-10, and given that IL-10 is mostly an anti-inflammatory cytokine as discussed above, which is important for the induction of tolerance and suppression of dendritic cell maturation (Faulkner et al., 2000; Corinti et al., 2001), it seemed plausible to suggest that adaptive immune responses to ACs could be mediated by IL-10 effects. However, that hypothesis has been difficult to May 2019 \| Volume 10 \| Article 562 3 A Model for Apoptotic Immune Evasion Yakoub and Schülke there were also instances where ACs were reported to have immunostimulatory effects (Hoffmann et al., 2000; Ignatius et al., 2000; Feng et al., 2002; Buttiglieri et al., 2003; Goldszmid et al., 2003; Ishii et al., 2003; Casares et al., 2005). It is thus possible to propose a model (Figure 1) whereby exposure to ACs in a non-inflammatory/non-immunostimulatory context (that does not involve immunogenic stimuli that trigger T cell activation) mounts a tolerogenic response via T cell inhibition through the coinhibitory pathway, whereas exposure to activated ACs in an immunostimulatory context mounts an immunogenic response via T cell activation through the costimulatory pathway. In support of this model is our finding that under non-immunostimulatory conditions, the AC-induced CD80 upregulation on macrophages was coupled with CD28 downregulation on T cells (Yakoub et al., 2018), which may possibly enhance the coinhibitory functions of CD80 as it binds the coinhibitory receptor CTLA-4 with much higher affinity and stability than it does the costimulatory receptor CD28 as described above. fully establish due to different conclusions from various studies. For example, ACs were shown to induce IL-10 production by monocytes (Voll et al., 1997a), but not macrophages (Fadok et al., 1998). But antigen-coupled ACs enhanced IL-10 and PD-L1 expression in AC-ingesting macrophages and induced T cell tolerance in mice (Getts et al., 2011). Importantly, however, while statistically significant, the IL-10-induced PD-L1 upregulation in that study was subtle when compared to the IL-10-neutralized control (only ~20% increase in PD-L1 mean fluorescence intensity (MFI) over the control). Conversely, using macrophage cell lines and mouse primary macrophages, we found that the upregulation of another costimulatory/ coinhibitory molecule, CD80, was more pronounced (on average > 4-fold upregulation, relative to the unstimulated control) (Yakoub et al., 2018). Frontiers in Pharmacology \| www.frontiersin.org An Integrative Model of Apoptotic-Cell- Mediated Immune Evasion and Tolerance These results suggest a differential AC response between dendritic cells and macrophages, as such AC activation was not required to modulate CD80 levels on macrophages (Yakoub et al., 2018). In a similar concept, “heat-stressed,” but not unstressed, ACs induced the costimulatory molecules CD40, CD80, and CD86 on dendritic cells and secretion of the immunostimulatory IL-12, resulting in enhanced T cell responses (Feng et al., 2002).f y ) CD80 expressed on T cells was also shown to bind PD-L1 on APCs with an affinity greater than that of CD80-CD28 binding (Butte et al., 2007; Rollins and Gibbons Johnson, 2017), which may negatively regulate T cell activation. Similarly, CD80 on APCs, which ACs upregulate (Yakoub et al., 2018), was suggested to bind PD-L1 on T cells (Schildberg et al., 2016), which may downregulate T cell activation, if not directly, indirectly by competing out CD28 and thus reducing the costimulatory signal that is essential for T cell activation and sustenance of the adaptive immune response. Notably, PD-L1 expression on parenchymal tissues including pancreatic islets mediated tolerance and inhibited self-reactive CD4 T cells (Keir et al., 2006); and interference with CD80-PD-L1 binding enhanced activation of CD4 and CD8 T cells in vivo and accelerated development of autoimmune diabetes in NOD mice (Paterson et al., 2011). Interestingly, it was also proposed that CD80/86 binding to CTLA-4 and PD-L1 on T cells enhances T cell motility, reducing T cell-APC contacts and the strength of the immune synapse, while enhancing contacts with, and activation of, Tregs (Dilek et al., 2013). Moreover, binding of PD-L1 on ACs to PD-1 on T cells is also possible (Kushwah et al., 2010), which may also strengthen the coinhibitory signal While the immunosuppressive or tolerogenic effects of ACs are established by a plethora of evidence (Kabelitz and Janssen, 1997; Steinman et al., 2000; Fadok et al., 2001; Ferguson et al., 2002; Liu et al., 2002; Stuart et al., 2002; Morelli et al., 2003; Rovere-Querini et al., 2004; Gray et al., 2007), May 2019 \| Volume 10 \| Article 562 Frontiers in Pharmacology \| www.frontiersin.org 4 A Model for Apoptotic Immune Evasion Yakoub and Schülke ülke A Model for Apoptotic Immune Evasio A B C A model for the regulation of T cell-mediated adaptive immune responses to apoptotic cells. (A) At the resting state, no activation of the TCR ulatory pathway takes place. An Integrative Model of Apoptotic-Cell- Mediated Immune Evasion and Tolerance (B) In immunostimulatory contexts (such as AC ingestion by APCs in the presence of LPS, TLR ligands, or al antigens), TCR binding to AC antigens presented in the context of MHC-I/II and the costimulatory signaling mediated by binding of CD28 by (or CD86 and CD80) take place. Thus, activation of the costimulatory pathway ensures effector T cell (Teff) activation and proliferation, leading of an immune response to AC antigens. (C) In non-immunostimulatory contexts, such as AC ingestion by APCs in the absence of additional ntiating stimuli (e.g., TLR ligands or mycobacterial antigens), the costimulatory signaling mediated by CD28 is downregulated. Rather, CD80 is which binds mainly to CTLA-4 (CD80 binds CTLA-4 with much higher affinity than CD28), initiating the coinhibitory signaling that leads to Teff and apoptosis. Even if some costimulatory signal is conveyed (by binding of some CD80 molecules to CD28), the CTLA-4-mediated ignal predominates and usually overrides costimulation. Other signals are possible (e.g., binding of CD80 to PD-L1 on T cells or binding of Cs to PD-1 on T cells) and might have some contribution to the overall coinhibitory signaling delivered to T cells, although their significance A B C A B C A A A B C FIGURE 1 \| A model for the regulation of T cell-mediated adaptive immune responses to apoptotic cells. (A) At the resting state, no activation of the TCR or the costimulatory pathway takes place. (B) In immunostimulatory contexts (such as AC ingestion by APCs in the presence of LPS, TLR ligands, or mycobacterial antigens), TCR binding to AC antigens presented in the context of MHC-I/II and the costimulatory signaling mediated by binding of CD28 by mainly CD86 (or CD86 and CD80) take place. Thus, activation of the costimulatory pathway ensures effector T cell (Teff) activation and proliferation, leading to mounting of an immune response to AC antigens. (C) In non-immunostimulatory contexts, such as AC ingestion by APCs in the absence of additional immunopotentiating stimuli (e.g., TLR ligands or mycobacterial antigens), the costimulatory signaling mediated by CD28 is downregulated. Rather, CD80 is upregulated, which binds mainly to CTLA-4 (CD80 binds CTLA-4 with much higher affinity than CD28), initiating the coinhibitory signaling that leads to Teff suppression and apoptosis. Even if some costimulatory signal is conveyed (by binding of some CD80 molecules to CD28), the CTLA-4-mediated coinhibitory signal predominates and usually overrides costimulation. Frontiers in Pharmacology \| www.frontiersin.org Sjögren’s Syndrome Sjögren’s Syndrome (SS) is an autoimmune disease targeting the salivary and lacrimal glands resulting in chronic dryness of mouth and eyes (Ainola et al., 2018). In SS, apoptotic particles blebbing from apoptotic epithelial cells, possibly caused by defects in the production of sex hormones, that contain typical SS autoantigens such as hY1RNA were shown to contribute to disease pathology (Ainola et al., 2018). In line with this increase in apoptosis rates, enhanced levels of both Fas and FasL were reported in salivary gland tissues, but not in lacrimal gland tissues or peripheral blood An Integrative Model of Apoptotic-Cell- Mediated Immune Evasion and Tolerance Other signals are possible (e.g., binding of CD80 to PD-L1 on T cells or binding of PD-L1 on APCs to PD-1 on T cells) and might have some contribution to the overall coinhibitory signaling delivered to T cells, although their significance and relative contribution have yet to be established. Overall, the coinhibitory signaling to T cells triggers the inhibition of Teff functions; and in this context, differentiation of Tregs may also be enhanced. B B B B C C FIGURE 1 \| A model for the regulation of T cell-mediated adaptive immune responses to apoptotic cells. (A) At the resting state, no activation of the TCR or the costimulatory pathway takes place. (B) In immunostimulatory contexts (such as AC ingestion by APCs in the presence of LPS, TLR ligands, or mycobacterial antigens), TCR binding to AC antigens presented in the context of MHC-I/II and the costimulatory signaling mediated by binding of CD28 by mainly CD86 (or CD86 and CD80) take place. Thus, activation of the costimulatory pathway ensures effector T cell (Teff) activation and proliferation, leading to mounting of an immune response to AC antigens. (C) In non-immunostimulatory contexts, such as AC ingestion by APCs in the absence of additional immunopotentiating stimuli (e.g., TLR ligands or mycobacterial antigens), the costimulatory signaling mediated by CD28 is downregulated. Rather, CD80 is upregulated, which binds mainly to CTLA-4 (CD80 binds CTLA-4 with much higher affinity than CD28), initiating the coinhibitory signaling that leads to Teff suppression and apoptosis. Even if some costimulatory signal is conveyed (by binding of some CD80 molecules to CD28), the CTLA-4-mediated coinhibitory signal predominates and usually overrides costimulation. Other signals are possible (e.g., binding of CD80 to PD-L1 on T cells or binding of PD-L1 on APCs to PD-1 on T cells) and might have some contribution to the overall coinhibitory signaling delivered to T cells, although their significance and relative contribution have yet to be established. Overall, the coinhibitory signaling to T cells triggers the inhibition of Teff functions; and in this context, differentiation of Tregs may also be enhanced. Frontiers in Pharmacology \| www.frontiersin.org May 2019 \| Volume 10 \| Article 562 5 A Model for Apoptotic Immune Evasion Yakoub and Schülke to T cells, although its significance needs to be established as previously discussed. the immunosuppressive properties of ACs may be exploited for therapeutic purposes. Rheumatoid Arthritis Rheumatoid Arthritis (RA) is a chronic systemic autoimmune disorder characterized by progressive inflammatory bone and joint destruction. Immunologically, the autoimmune responses are mediated by either circulating autoantibodies directed against citrullinated peptides and rheumatoid factor or complement protein C3 (Abdolmaleki et al., 2018). The destructive autoimmune responses in RA are maintained by IL-6 and TNF-α secreted by tissue macrophages triggering the MMP and RANK-ligand- supported activation of osteoclasts (An et al., 2016). In RA patients, the sustained joint inflammation is maintained by an abnormal state of aberrant cell survival caused by perpetual T cell activation resulting in stimulation and proliferation of fibroblasts which was termed “apoptosis resistance” (Malemud, 2018). In line with these observations, anti-apoptotic proteins were shown to be upregulated in RA synovial fluids (reviewed in Williams et al., 2018). Therefore, induction of apoptosis has the potential to reduce joint damage and to further modulate autoimmune responses in RA by modulating the coinhibitory signaling to T cells as previously described. Experimental apoptosis induction has therefore been considered as potential therapeutic avenue in RA, e.g., by targeting intracellular apoptotic inhibitory molecules, but thus far has not reached clinical trials (Williams et al., 2018). p While the distinction between APC types (dendritic cells or macrophages) in terms of the type of immune response (immunogenic vs. tolerogenic) to ACs cannot be completely settled given the varying reports thus far in this regard, there is still some preliminary evidence to propose that macrophages might mainly mediate tolerogenic responses to ACs while dendritic cells might mainly mediate the immunogenic responses. Dendritic cells and macrophages exhibit distinct locales and may thus mediate differential, locale-specific, AC responses (T cell activation or inhibition). For example, in the intestinal lamina propria, both APC types present commensal microbe and dietary antigens to T cells, with dendritic cells inducing effector Th17 T cells, and macrophages inducing Tregs (Denning et al., 2007). Tregs induced by AC-presenting macrophages showed induced anti-inflammatory cytokine production and reduced immunostimulatory cytokines, and displayed an anergic phenotype after restimulation with the antigen (Denning et al., 2007). Because macrophages are more abundant than dendritic cells in the lamina propria, T cell tolerance thus becomes the predominant response in that context. An Integrative Model of Apoptotic-Cell- Mediated Immune Evasion and Tolerance Whether ACs would induce a tolerogenic or immunogenic response may depend on presence of immunostimulatory conditions or cues, including: (1) the type of cell death induced in the ACs, e.g., caspase-dependent or independent (Larmonier et al., 2002); (2) type of the apoptosis-inducing agent or drug used in the experimental model (Casares et al., 2005; Obeid et al., 2007); (3) the stage of apoptosis (early vs. late) in the ACs (Weyd et al., 2013); (4) type of the apoptotic corpse (ACs or apoptotic blebs) ingested by the APC (Fransen et al., 2009); (5) cell-type of the ACs ingested by the APCs (Kushwah et al., 2010); (6) secretion of T cell immuostimulatory cytokines, such as TNF-α, or TLR ligands (Clayton et al., 2003); (7) type and ratio of the APCs (dendritic cells or macrophages) present in a particular tissue site (Denning et al., 2007); or (8) presence of potentially immunogenic or immunogenizing infectious pathogen (e.g., mycobacterial) antigens in the tissue microenvironment (Espinosa-Cueto et al., 2017). Systemic Lupus Erythematosus y p y SLE is a chronic systemic autoimmune disorder characterized by the presence of circulating nuclear antigens, including DNA and nucleosomes, and of autoantibodies against these nuclear antigens (Poon et al., 2014). SLE affects the skin, lungs, kidneys, and central nervous system. Impaired engulfment of ACs by phagocytes leads to accumulation of ACs in the lymph nodes and blood and in the skin after UV exposure. Impaired AC clearance eventually leads to secondary necrosis, which allows intracellular (self-) antigens, normally hidden within the AC to be exposed extracellularly, and thus recognizing these self-antigens by the immune system, causing the production of autoantibodies and autoimmunity. While mostly suggested to be caused by defective AC engulfment, super-stimulatory APCs that lead to hyperactive T cells were also suggested to recapitulate SLE in a mouse model (Zhu et al., 2005). Thus, targeting APC activity which is normally modulated by ACs in normal physiological conditions might be a successful strategy for therapy development for SLE. Rheumatoid Arthritis Importantly, macrophages were shown to be essential for clearing tumor ACs introduced into mice and eliciting tolerogenic responses to the ACs (Asano et al., 2011); as ablation of the spleen marginal zone macrophages in these mice diminished the immunosuppressive potential of the ACs and enabled triggering of an immune response to ACs (McGaha et al., 2011). It thus seems possible that the immunosuppressive or tolerogenic response to ACs is mainly mediated by macrophages as the APCs. Cancer Anti-tumor chemotherapeutic agents lead to production of massive cytotoxicity and generation of ACs. Given the adaptive immunosuppressive and tolerogenic effects of ACs, it is plausible to hypothesize that the apoptosis induced, and the ACs produced, by cancer treatments contribute to tumor survival indirectly by dampening immune responses to cancer cell antigens carried on these ACs. Thus, functionally blocking ACs in cancer may help reduce these undesirable effects of antineoplastic agents on anti-tumor immunity. In fact, some therapeutic strategies to target the coinhibitory molecules that mediate the adaptive immune responses to ACs have been investigated for their possible beneficial effects on anti-tumor immunity (Chen, 2004; Hirano et al., 2005; Curran et al., 2010). Diabetes Mellitus Type 1 diabetes mellitus can be caused by autoimmune responses to pancreatic beta-cell antigens resulting in insulin deficiency and hyperglycemia. While it was suggested that inefficient clearing of apoptotic pancreatic cells resulting in the release of damage-associated molecular patterns (DAMPs) in combination with autoantigens may contribute to the pathology of type 1 diabetes (Heimberg et al., 2001; O’Brien et al., 2006), ACs were also suggested as a tool to induce tolerance to beta- cell self-antigens. Indeed, ACs (apoptotic beta-cell infusion) could suppress beta-cell antigen-specific CD4+ T cell proliferation and delay the onset of diabetes in the diabetes-susceptible, autoimmune (NOD) mice (Xia et al., 2007; Marin-Gallen et al., 2010). Therefore, ACs show a promising potential for the treatment of type 1 diabetes. Although the costimulatory/coinhibitory pathway has been a tempting target for disease therapy, some challenges remain and are yet to be overcome in the future to enable full harnessing of the therapeutic potential of this pathway. For example, some therapies targeting the costimulatory/coinhibitory pathway have even been discontinued in medical practice due to limited effectiveness (Smith et al., 2013), or have failed clinical trials at early stages (Khoury and Sayegh, 2004). Some proposed therapies can also pose significant risks (Frebel and Oxenius, 2013). This all reflects the fact that we have not yet reached complete and thorough understanding of the costimulation/ coinhibition pathways and their intricate interactions with diseases. In conclusion, this perspective proposes a model to understand how ACs regulate the costimulatory/coinhibitory signaling pathways of regulating T cell activation in order to suppress adaptive immune responses, which may facilitate harnessing these molecular mechanisms in therapy development for various immunopathological conditions. Autoimmune Lymphoproliferative Syndrome y Autoimmune Lymphoproliferative Syndrome (ALPS) is an inherited autoimmune disorder characterized by spleno- and hepatomegaly, lymphadenopathy, autoimmune lesions in multiple organs, as well as autoimmune hemolytic anemia, thrombocytopenia, or leukocytopenia, caused by cell-type specific autoantibody production (Turbyville and Rao, 2010; Price et al., 2014; Yamada et al., 2017). In ALPS patients, these symptoms are caused by spontaneous mutations in the Fas, FasL, or caspase 10 genes, resulting in a defective apoptosis of antigen-activated T- and B cells in the periphery, and the impaired limitation of immune responses (Yamada et al., 2017). Since the pathology of ALPS is caused by a disruption of lymphocyte apoptosis, it is plausible to speculate that a decrease in production of ACs (which exert immunoinhibitory/anti-inflammatory effects as discussed) may contribute to the autoimmune pathology in ALPS patients. The role of ACs in ALPS pathogenesis and its possible exploitation for therapy is thus an interesting area for future research. CONCLUDING REMARKS While ACs have been investigated as tolerogenic or immunosuppressive “vaccines,” understanding the molecular mechanisms of the ACs’ immunomodulating effects, especially their interaction with the costimulatory/coinhibitory pathway, may encourage attempts at targeting specific molecules that ACs exploit in mediating their effects. In general, the costimulatory/coinhibitory pathway has been explored as target for therapeutic purposes. For example, targeting of either CD28, PD-1, PD-L1, or CTLA-4 has been investigated and some therapies targeting this machinery are already in clinical use (e.g., the anti-CTLA-4 antibody Ipilimumab for the treatment of advanced metastatic melanoma, and the anti-PD-1 antibodies Pembrolizumab and Nivolumab for the treatment of advanced melanoma, advanced non-small cell lung cancer, and metastatic renal cell carcinoma) (Dilek et al., 2013). ROLE OF APOPTOTIC-CELL-MEDIATED IMMUNOSUPPRESSION IN DISEASES An abundance of ACs (~70 billion) is produced daily in humans and defective clearance of these ACs has been associated with diseases, primarily autoimmune conditions. Therefore, we will briefly discuss some disease conditions, where we propose that May 2019 \| Volume 10 \| Article 562 Frontiers in Pharmacology \| www.frontiersin.org 6 A Model for Apoptotic Immune Evasion Yakoub and Schülke Donor’s ACs given to transplant patients may tolerize or repress the recipient’s immune responses to the transplant, prevent graft- versus-host reactions, and enhance graft survival (Kleinclauss et al., 2006; Wang et al., 2006, 2009; Bittencourt et al., 2011). lymphocytes of patients with SS, implicating Fas-mediated apoptosis in the destruction of salivary gland tissue (Ishimaru et al., 2001; Bolstad et al., 2003). The presence of both autoantigens and adjuvanting nucleic acids in these apoptotic particles was shown to stimulate plasmacytoid dendritic cells in salivary glands via TLR7 and TLR9, resulting in the activation of autoreactive T and B cells (Ainola et al., 2018). However, Ishimaru et al. reported that mice treated with an anti-murine FasL antibody (to suppress Fas-mediated apoptosis) unexpectedly showed exacerbations of the autoimmune lesions in both salivary and lacrimal glands (Ishimaru et al., 2001). Therefore, both the role of apoptosis in the pathology of SS and the therapeutic potential of ACs in the treatment of SS are poorly understood, warranting further investigations. REFERENCES Abdolmaleki, F., Farahani, N., Gheibi Hayat, S. M., Pirro, M., Bianconi, V., Barreto, G. E., et al. (2018). The role of efferocytosis in autoimmune diseases. Front. Immunol. 9:1645. doi: 10.3389/fimmu.2018.01645 Casares, N., Pequignot, M. O., Tesniere, A., Ghiringhelli, F., Roux, S., Chaput, N., et al. (2005). Caspase-dependent immunogenicity of doxorubicin-induced tumor cell death. J. Exp. Med. 202, 1691–1701. doi: 10.1084/jem.20050915 i Ainola, M., Porola, P., Takakubo, Y., Przybyla, B., Kouri, V. P., Tolvanen, T. A., et al. (2018). Activation of plasmacytoid dendritic cells by apoptotic particles— mechanism for the loss of immunological tolerance in Sjögren’s syndrome. Clin. Exp. Immunol. 191, 301–310. doi: 10.1111/cei.13077 Chen, L. (2004). Co-inhibitory molecules of the B7-CD28 family in the control of T cell immunity Nat Rev Immunol 4 336 347 doi 10 1038/nri1349 Chen, L. (2004). Co-inhibitory molecules of the B7-CD28 family in the control of T-cell immunity. Nat. Rev. Immunol. 4, 336–347. doi: 10.1038/nri1349 Chen, W., Frank, M. E., Jin, W., and Wahl, S. M. (2001). TGF- released by apoptotic T cells contributes to an immunosuppressive milieu. Immunity 14, 715–725. doi: 10.1016/S1074-7613(01)00147-9 An, J., Hao, D., Zhang, Q., Chen, B., Zhang, R., Wang, Y., et al. (2016). Natural products for treatment of bone erosive diseases: the effects and mechanisms on inhibiting osteoclastogenesis and bone resorption. Int. Immunopharmacol. 36, 118–131. doi: 10.1016/j.intimp.2016.04.024 apoptotic T cells contributes to an immunosuppressive milieu. Immunity 14, 715–725. doi: 10.1016/S1074-7613(01)00147-9 Chinnaiyan, A. M., O’Rourke, K., Tewari, M., and Dixit, V. M. (1995). FADD, a novel death domain-containing protein, interacts with the death domain of Fas and initiates apoptosis. Cell 81, 505–512. doi: 10.1016/0092-8674(95)90071-3 Anderson, A. C., Joller, N., and Kuchroo, V. K. (2016). Lag-3, Tim-3, and TIGIT: co-inhibitory receptors with specialized functions in immune regulation. Immunity 44, 989–1004. doi: 10.1016/j.immuni.2016.05.001 Chitnis, T., Najafian, N., Abdallah, K. A., Dong, V., Yagita, H., Sayegh, M. H., et al. (2001). CD28-independent induction of experimental autoimmune encephalomyelitis. J. Clin. Invest. 107, 575–583. doi: 10.1172/JCI11220 Appleman, L. J., van Puijenbroek, A. A., Shu, K. M., Nadler, L. M., and Boussiotis, V. A. (2002). CD28 costimulation mediates down-regulation of p27kip1 and cell cycle progression by activation of the PI3K/PKB signaling pathway in primary human T cells. J. Immunol. 168, 2729–2736. doi: 10.4049/ jimmunol.168.6.2729 Clayton, A. R., Prue, R. L., Harper, L., Drayson, M. T., and Savage, C. O. (2003). REFERENCES Increased salivary gland tissue expression of Fas, Fas ligand, cytotoxic T lymphocyte-associated antigen 4, and programmed cell death 1 in primary Sjogren’s syndrome. Arthritis Rheum. 48, 174–185. doi: 10.1002/art.10734 Cvetanovic, M., and Ucker, D. S. (2004). Innate immune discrimination of apoptotic cells: repression of proinflammatory macrophage transcription is coupled directly to specific recognition. J. Immunol. 172, 880–889. doi: 10.4049/jimmunol.172.2.880 y p y g p g p y Sjogren’s syndrome. Arthritis Rheum. 48, 174–185. doi: 10.1002/art.10734 Bour-Jordan, H., Esensten, J. H., Martinez-Llordella, M., Penaranda, C., Stumpf, M., Bour-Jordan, H., Esensten, J. H., Martinez-Llordella, M., Penaranda, C., Stumpf, M., and Bluestone, J. A. (2011). Intrinsic and extrinsic control of peripheral T-cell tolerance by costimulatory molecules of the CD28/B7 family. Immunol. Rev. 241, 180–205. doi: 10.1111/j.1600-065X.2011.01011.x Dahl, A. M., Klein, C., Andres, P. G., London, C. A., Lodge, M. P., Mulligan, R. C., et al. (2000). Expression of bcl-xL restores cell survival, but not proliferation off effector differentiation, in CD28-deficient T lymphocytes. J. Exp. Med. 191, 2031–2038. doi: 10.1084/jem.191.12.2031 Buchbinder, E. I., and Desai, A. (2016). CTLA-4 and PD-1 pathway similarities, differences, and implications of their inhibition. Am. J. Clin. Oncol. 39, 98–106. doi: 10.1097/COC.0000000000000239 Denning, T. L., Wang, Y. C., Patel, S. R., Williams, I. R., and Pulendran, B. (2007). Lamina propria macrophages and dendritic cells differentially induce regulatory and interleukin 17-producing T cell responses. Nat. Immunol. 8, 1086–1094. doi: 10.1038/ni1511 Butte, M. J., Keir, M. E., Phamduy, T. B., Freeman, G. J., and Sharpe, A. H. (2007). PD-L1 interacts specifically with B7-1 to inhibit T cell proliferation. Immunity 27, 111–122. doi: 10.1016/j.immuni.2007.05.016 Denton, D., Mills, K., and Kumar, S. (2008). Methods and protocols for studying cell death in Drosophila. Methods Enzymol. 446, 17–37. doi: 10.1016/ S0076-6879(08)01602-9 Butte, M. J., Peña-Cruz, V., Kim, M. J., Freeman, G. J., and Sharpe, A. H. (2008). Interaction of human PD-L1 and B7-1. Mol. Immunol. 45, 3567–3572. doi: 10.1016/j.molimm.2008.05.014 Dilek, N., Poirier, N., Hulin, P., Coulon, F., Mary, C., Ville, S., et al. (2013). Targeting CD28, CTLA-4 and PD-L1 costimulation differentially controls immune synapses and function of human regulatory and conventional T-cells. PLoS One 8:e83139. doi: 10.1371/journal.pone.0083139 Buttiglieri, S., Galetto, A., Forno, S., De Andrea, M., and Matera, L. (2003). Influence of drug-induced apoptotic death on processing and presentation of tumor antigens by dendritic cells. Int. J. Cancer 106, 516–520. doi: 10.1002/ ijc.11243 Esensten, J. H., Helou, Y. A., Chopra, G., Weiss, A., and Bluestone, J. Transplantationt After transplantation, immunosuppressive drugs are given to the patients, to induce tolerance and prevent graft rejection. However, these drugs show many undesirable and potentially dangerous side effects. Thus, ACs were suggested to be used as possibly side effect-free tolerogenic tools (Morelli and Larregina, 2010). May 2019 \| Volume 10 \| Article 562 Frontiers in Pharmacology \| www.frontiersin.org 7 A Model for Apoptotic Immune Evasion Yakoub and Schülke ACKNOWLEDGMENTS We would like to thank Dr. R. Gibbons Johnson (University of Minnesota) for critically reading the initial manuscript and providing useful feedback. We would like to thank Dr. R. Gibbons Johnson (University of Minnesota) for critically reading the initial manuscript and providing useful feedback. AY conceived and wrote the first draft of this manuscript. SS reviewed, critiqued, and revised the first draft. AY and SS rewrote and finally revised the final manuscript. AY conceived and wrote the first draft of this manuscript. SS reviewed, critiqued, and revised the first draft. AY and SS rewrote and finally revised the final manuscript. Carr, E. L., Kelman, A., Wu, G. S., Gopaul, R., Senkevitch, E., Aghvanyan, A., et al. (2010). Glutamine uptake and metabolism are coordinately regulated by ERK/MAPK during T lymphocyte activation. J. Immunol. 185, 1037–1044. doi: 10.4049/jimmunol.0903586 REFERENCES Dendritic cell uptake of human apoptotic and necrotic neutrophils inhibits CD40, CD80, and CD86 expression and reduces allogeneic T cell responses: relevance to systemic vasculitis. Arthritis Rheum. 48, 2362–2374. doi: 10.1002/art.11130 Asano, K., Nabeyama, A., Miyake, Y., Qiu, C. H., Kurita, A., Tomura, M., et al. (2011). CD169-positive macrophages dominate antitumor immunity by crosspresenting dead cell-associated antigens. Immunity 34, 85–89. doi: 10.1016/j.immuni.2010.12.011 Collins, M., Ling, V., and Carreno, B. M. (2005). The B7 family of immune- regulatory ligands. Genome Biol. 6:223. doi: 10.1186/gb-2005-6-6-223 Corinti, S., Albanesi, C., la Sala, A., Pastore, S., and Girolomoni, G. (2001). Regulatory activity of autocrine IL-10 on dendritic cell functions. J. Immunol. 166, 4312–4318. doi: 10.4049/jimmunol.166.7.4312 Attanasio, J., and Wherry, E. J. (2016). Costimulatory and coinhibitory receptor pathways in infectious disease. Immunity 44, 1052–1068. doi: 10.1016/j. immuni.2016.04.022 Couper, K. N., Blount, D. G., and Riley, E. M. (2008). IL-10: the master regulator of immunity to infection. J. Immunol. 180, 5771–5777. doi: 10.4049/ jimmunol.180.9.5771 Baumeister, S. H., Freeman, G. J., Dranoff, G., and Sharpe, A. H. (2016). Coinhibitory pathways in immunotherapy for cancer. Annu. Rev. Immunol. 34, 539–573. doi: 10.1146/annurev-immunol-032414-112049 Csöbönyeiová, M., Danišovič, L., and Polák, S. (2016). Induced pluripotent stem cells for modeling and cell therapy of Parkinson’s disease. Neural Regen. Res. 11, 727–728. doi: 10.4103/1673-5374.182692 Bittencourt, M. C., Perruche, S., Contassot, E., Fresnay, S., Baron, M. H., Angonin, R., et al. (2011). Intravenous injection of apoptotic leukocytes enhances bone marrow engraftment across major histocompatibility barriers. Blood 98, 224–230. doi: 10.1182/blood.V98.1.224 Curran, M. A., Montalvo, W., Yagita, H., and Allison, J. P. (2010). PD-1 and CTLA-4 combination blockade expands infiltrating T cells and reduces regulatory T and myeloid cells within B16 melanoma tumors. Proc. Natl. Acad. Sci. USA 107, 4275–4280. doi: 10.1073/pnas.0915174107 Bluestone, J. A. (1997). Is CTLA-4 a master switch for peripheral T cell tolerance? J. Immunol. 158, 1989–1993. Bluestone, J. A. (1997). Is CTLA-4 a master switch for peripheral T cell tolerance? J. Immunol. 158, 1989–1993. Bolstad, A. I., Eiken, H. G., Rosenlund, B., Alarcon-Riquelme, M. E., and Jonsson, R. (2003). Increased salivary gland tissue expression of Fas, Fas ligand, cytotoxic T lymphocyte-associated antigen 4, and programmed cell death 1 in primary Sjogren’s syndrome. Arthritis Rheum. 48, 174–185. doi: 10.1002/art.10734 Bolstad, A. I., Eiken, H. G., Rosenlund, B., Alarcon-Riquelme, M. E., and Jonsson, R. (2003). REFERENCES doi: 10.1016/j.biocel.2012.08.023 Geginat, J., Larghi, P., Paroni, M., Nizzoli, G., Penatti, A., Pagani, M., et al. (2016). The light and the dark sides of Interleukin-10 in immune-mediated diseases and cancer. Cytokine Growth Factor Rev. 30, 87–93. doi: 10.1016/j. cytogfr.2016.02.003 Karandikar, N. J., Vanderlugt, C. L., Walunas, T. L., Miller, S. D., and Bluestone, J. A. (1996). CTLA-4 (a negative regulator of autoimmune disease). J. Exp. Med. 184, 783–788. doi: 10.1084/jem.184.2.783 Getts, D. R., Turley, D. M., Smith, C. E., Harp, C. T., McCarthy, D., Feeney, E. M., et al. (2011). Tolerance induced by apoptotic antigen-coupled leukocytes is induced by PD-L1+ and IL-10-producing splenic macrophages and maintained by T regulatory cells. J. Immunol. 187, 2405–2417. doi: 10.4049/ jimmunol.1004175 Kazemzadeh, L., Cvijovic, M., and Petranovic, D. (2012). Boolean model of yeast apoptosis as a tool to study yeast and human apoptotic regulations. Front. Physiol. 3:446. doi: 10.3389/fphys.2012.00446 Keir, M. E., Liang, S. C., Guleria, I., Latchman, Y. E., Qipo, A., Albacker, L. A., et al. (2006). Tissue expression of PD-L1 mediates peripheral T cell tolerance. J. Exp. Med. 203, 883–895. doi: 10.1084/jem.20051776h Goldszmid, R. S., Idoyaga, J., Bravo, A. I., Steinman, R., Mordoh, J., and Wainstok, R. (2003). Dendritic cells charged with apoptotic tumor cells induce long-lived protective CD4+ and CD8+ T cell immunity against B16 melanoma. J. Immunol. 171, 5940–5947. doi: 10.4049/jimmunol.171.11.5940 Khoury, S. J., and Sayegh, M. H. (2004). The roles of the new negative T cell costimulatory pathways in regulating autoimmunity. Immunity 20, 529–538. doi: 10.1016/S1074-7613(04)00116-5 Kim, S., Chung, E. Y., and Ma, X. (2005). Immunological consequences of macrophage-mediated clearance of apoptotic cells. Cell Cycle 4, 231–234. doi: 10.4161/cc.4.2.1421 Gómez-Sintes, R., Hernández, F., Lucas, J. J., and Avila, J. (2011). GSK-3 mouse models to study neuronal apoptosis and neurodegeneration. Front. Mol. Neurosci. 4:45. doi: 10.3389/fnmol.2011.00045 Kischkel, F. C., Hellbardt, S., Behrmann, I., Germer, M., Pawlita, M., Krammer, P. H., et al. (1995). Cytotoxicity-dependent APO 1 (Fas/CD95)-associated proteins from a death-inducing signaling complex (DISC) with the receptor. EMBO J. 14, 5579–5588. doi: 10.1002/j.1460-2075.1995.tb00245.x Gray, M., Miles, K., Salter, D., Gray, D., and Savill, J. (2007). Apoptotic cells protect mice from autoimmune inflammation by the induction of regulatory B cells. Proc. Natl. Acad. Sci. USA 104, 14080–14085. doi: 10.1073/ pnas.0700326104 Kleinclauss, F., Perruche, S., Masson, E., de Carvalho Bittencourt, M., Biichle, S., Remy-Martin, J. P., et al. (2006). Intravenous apoptotic spleen cell infusion induces a TGF-β-dependent regulatory T-cell expansion. REFERENCES A. (2016). CD28 costimulation: from mechanism to therapy. Immunity 44, 973–988. doi: 10.1016/j.immuni.2016.04.020 Calissano, P., Matrone, C., and Amadoro, G. (2009). Apoptosis and in vitro Alzheimer disease neuronal models. Commun. Integr. Biol. 2, 163–169. May 2019 \| Volume 10 \| Article 562 Frontiers in Pharmacology \| www.frontiersin.org 8 A Model for Apoptotic Immune Evasion Yakoub and Schülke Hoffmann, T. K., Meidenbauer, N., Dworacki, G., Kanaya, H., and Whiteside, T. L. (2000). Generation of tumor-specific T-lymphocytes by cross-priming with human dendritic cells ingesting apoptotic tumor cells. Cancer Res. 60, 3542–3549. Espinosa-Cueto, P., Magallanes-Puebla, A., Castellanos, C., and Mancilla, R. (2017). Dendritic cells that phagocytose apoptotic macrophages loaded with mycobacterial antigens activate CD8 T cells via cross-presentation. PLoS One 12:e0182126. doi: 10.1371/journal.pone.0182126 Fadok, V. A., Bratton, D. L., and Henson, P. M. (2001). Phagocyte receptors for apoptotic cells. Recognition, uptake, and consequences. J. Clin. Invest. 108, 957–962. doi: 10.1172/JCI200114122 Ignatius, R., Marovich, M., Mehlhop, E., Villamide, L., Mahnke, K., Cox, W. I., et al. (2000). Canarypox virus-induced maturation of dendritic cells is mediated by apoptotic cell death and tumor necrosis factor α secretion. J. Virol. 74, 11329–11338. doi: 10.1128/JVI.74.23.11329-11338.2000 Fadok, V. A., Bratton, D. L., Konowal, A., Freed, P. W., Westcott, J. Y., and Henson, P. M. (1998). Macrophages that have ingested apoptotic cells in vitro inhibit proinflammatory cytokine production through autocrine/paracrine mechanisms involving TGF-β, PGE2, and PAF. J. Clin. Invest. 101, 890–898. doi: 10.1172/JCI1112 Ikemizu, S., Gilbert, R. J., Fennelly, J. A., Collins, A. V., Harlos, K., Jones, E. Y., et al. (2000). Structure and dimerization of a soluble form of B7-1. Immunity 12, 51–60. doi: 10.1016/S1074-7613(00)80158-2 Ishii, S., Hiroishi, K., Eguchi, J., and Mitamura, K. (2003). Dendritic cell maturation induced by delivery of ultraviolet-mediated apoptotic colorectal cancer cell lines. Anticancer Res. 23, 2457–2463. Fang, Y., Gao, T., Zhang, B., and Pu, J. (2018). Recent advances: decoding Alzheimer’s disease with stem cells. Front. Aging Neurosci. 10:77. doi: 10.3389/ fnagi.2018.00077 Ishimaru, N., Yanagi, K., Ogawa, K., Suda, T., Saito, I., and Hayashi, Y. (2001). Possible role of organ-specific autoantigen for Fas ligand-mediated activation- induced cell death in murine Sjogren’s syndrome. J. Immunol. 167, 6031–6037. doi: 10.4049/jimmunol.167.10.6031 Faulkner, L., Buchan, G., and Baird, M. (2000). Interleukin-10 does not affect phagocytosis of particulate antigen by bone marrow-derived dendritic cells but does impair antigen presentation. Immunology 99, 523–531. doi: 10.1046/j. 1365-2567.2000.00018.x Ito, T., Ueno, T., Clarkson, M. R., Yuan, X., Jurewicz, M. REFERENCES M., Yagita, H., et al. (2005). Analysis of the role of negative T cell costimulatory pathways in CD4 and CD8 T cell-mediated alloimmune responses in vivo. J. Immunol. 174, 6648–6656. doi: 10.4049/jimmunol.174.11.6648 Feng, H., Zeng, Y., Graner, M. W., and Katsanis, E. (2002). Stressed apoptotic tumor cells stimulate dendritic cells and induce specific cytotoxic T cells. Blood 100, 4108–4115. doi: 10.1182/blood-2002-05-1389fi Janakiram, M., Shah, U. A., Liu, W., Zhao, A., Schoenberg, M. P., and Zang, X. (2017). The third group of the B7-CD28 immune checkpoint family: HHLA2, TMIGD2, B7x, and B7-H3. Immunol. Rev. 276, 26–39. doi: 10.1111/imr.12521 Ferguson, T. A., Herndon, J., Elzey, B., Griffith, T. S., Schoenberger, S., and Green, D. R. (2002). Uptake of apoptotic antigen-coupled cells by lymphoid dendritic cells and cross-priming of CD8+ T cells produce active immune unresponsiveness. J. Immunol. 168, 5589–5595. doi: 10.4049/jimmunol. 168.11.5589 Johansson, U., Walther-Jallow, L., Smed-Sörensen, A., and Spetz, A. L. (2007). Triggering of dendritic cell responses after exposure to activated, but not resting, apoptotic PBMCs. J. Immunol. 179, 1711–1720. doi: 10.4049/ jimmunol.179.3.1711 Fleury, C., Pampin, M., Tarze, A., and Mignotte, B. (2002). Yeast as a model to study apoptosis? Biosci. Rep. 22, 59–79. doi: 10.1023/A:1016013123094f j June, C. H., Bluestone, J. A., Nadler, L. M., and Thompson, C. B. (1994). The B7 and CD28 receptor families. Immunol. Today 15, 321–331. doi: 10.1016/0167-5699(94)90080-9 Fransen, J. H., Hilbrands, L. B., Ruben, J., Stoffels, M., Adema, G. J., van der Vlag, J., et al. (2009). Mouse dendritic cells matured by ingestion of apoptotic blebs induce T cells to produce interleukin-17. Arthritis Rheum. 60, 2304–2313. doi: 10.1002/art.24719 Kabelitz, D., and Janssen, O. (1997). Antigen-induced death of T-lymphocytes. Front. Biosci. 2, d61–d77. doi: 10.2741/A175 Frauwirth, K. A., Riley, J. L., Harris, M. H., Parry, R. V., Rathmell, J. C., Plas, D. R., et al. (2002). The CD28 signaling pathway regulates glucose metabolism. Immunity 16, 769–777. doi: 10.1016/S1074-7613(02)00323-0h Kägi, D., Vignaux, F., Ledermann, B., Bürki, K., Depraetere, V., Nagata, S., et al. (1994). Fas and perforin pathways as major mechanisms of T cell-mediated cytotoxicity. Science 265, 528–530. doi: 10.1126/science.7518614 Frebel, H., and Oxenius, A. (2013). The risks of targeting co-inhibitory pathways to modulate pathogen-directed T cell responses. Trends Immunol. 34, 193–199. doi: 10.1016/j.it.2012.12.002 Kannan, A., Huang, W., Huang, F., and August, A. (2012). Signal transduction via the T cell antigen receptor in naïve and effector/memory T cells. Int. J. Biochem. Cell Biol. 44, 2129–2134. REFERENCES R., and Gibbons Johnson, R. M. (2017). CD80 expressed by CD8+ T cells contributes to PD-L1-induced apoptosis of activated CD8+ T cells. J. Immunol. Res., 7659462. doi: 10.1155/2017/7659462fi i Malemud, C. J. (2018). Defective T-cell apoptosis and T-regulatory cell dysfunction in rheumatoid arthritis. Cells 7:E223. doi: 10.3390/cells7120223 Rovere-Querini, P., Capobianco, A., Scaffidi, P., Valentinis, B., Catalanotti, F., Giazzon, M., et al. (2004). HMGB1 is an endogenous immune adjuvant released by necrotic cells. EMBO Rep. 5, 825–830. doi: 10.1038/sj.embor.7400205 Marin-Gallen, S., Clemente-Casares, X., Planas, R., Pujol-Autonell, I., Carrascal, J., Carrillo, J., et al. (2010). Dendritic cells pulsed with antigen-specific apoptotic bodies prevent experimental type 1 diabetes. Clin. Exp. Immunol. 160, 207–214. doi: 10.1111/j.1365-2249.2009.04082.x Sauter, B., Albert, M. L., Francisco, L., Larsson, M., Somersan, S., and Bhardwaj, N. (2000). Consequences of cell death: exposure to necrotic tumor cells, but not primary tissue cells or apoptotic cells, induces the maturation of immunostimulatory dendritic cells. J. Exp. Med. 191, 423–434. doi: 10.1084/ jem.191.3.423 Marko, A. J., Miller, R. A., Kelman, A., and Frauwirth, K. A. (2010). Induction of glucose metabolism in stimulated T lymphocytes is regulated by mitogen- activated protein kinase signaling. PLoS One 5:e15425. doi: 10.1371/journal. pone.0015425 j Schildberg, F. A., Klein, S. R., Freeman, G. J., and Sharpe, A. H. (2016). Coinhibitory pathways in the B7-CD28 ligand-receptor family. Immunity 44, 955–972. doi: 10.1016/j.immuni.2016.05.002 McGaha, T. L., Chen, Y., Ravishankar, B., van Rooijen, N., and Karlsson, M. C. (2011). Marginal zone macrophages suppress innate and adaptive immunity to apoptotic cells in the spleen. Blood 117, 5403–5412. doi: 10.1182/blood-2010-11-320028 Schultze, S. M., Hemmings, B. A., Niessen, M., and Tschopp, O. (2012). PI3K/ AKT, MAPK and AMPK signalling: protein kinases in glucose homeostasis. Expert Rev. Mol. Med. 14:e1. doi: 10.1017/S1462399411002109 Morelli, A. E., and Larregina, A. T. (2010). Apoptotic cell-based therapies against transplant rejection: role of recipient’s dendritic cells. Apoptosis 15, 1083–1097. doi: 10.1007/s10495-010-0469-9 Shinde, R., Hezaveh, K., Halaby, M. J., Kloetgen, A., Chakravarthy, A., da Silva Medina, T., et al. (2018). Apoptotic cell-induced AhR activity is required for immunological tolerance and suppression of systemic lupus erythematosus in mice and humans. Nat. Immunol. 19, 571–582. doi: 10.1038/s41590-018-0107-1 Morelli, A. E., Larregina, A. T., Shufesky, W. J., Zahorchak, A. F., Logar, A. J., Papworth, G. D., et al. (2003). Internalization of circulating apoptotic cells by splenic marginal zone dendritic cells. Dependence on complement receptors and effect on cytokine production. Blood 101, 611–620. REFERENCES doi: 10.1146/ annurev.iy.11.040193.001203 Peng, H., Wang, W., Zhou, M., Li, R., Pan, H. F., and Ye, D. Q. (2013). Role of interleukin-10 and interleukin-10 receptor in systemic lupus erythematosus. Clin. Rheumatol. 32, 1255–1266. doi: 10.1007/s10067-013-2294-3 Liu, K., Iyoda, T., Saternus, M., Kimura, Y., Inaba, K., and Steinman, R. M. (2002). Immune tolerance after delivery of dying cells to dendritic cells in situ. J. Exp. Med. 196, 1091–1097. doi: 10.1084/jem.20021215 Poon, I. K., Lucas, C. D., Rossi, A. G., and Ravichandran, K. S. (2014). Apoptotic cell clearance: basic biology and therapeutic potential. Nat. Rev. Immunol. 14, 166–180. doi: 10.1038/nri3607 Liu, X., Kim, C. N., Yang, J., Jemmerson, R., and Wang, X. (1996). Induction of apoptotic program in cell-free extracts: requirement for dATP and cytochrome c. Cell 86, 147–157. doi: 10.1016/S0092-8674(00)80085-9 Price, S., Shaw, P. A., Seitz, A., Joshi, G., Davis, J., Niemela, J. E., et al. (2014). Natural history of autoimmune lymphoproliferative syndrome associated with Fas gene mutations. Blood 123, 1989–1999. doi: 10.1182/blood-2013-10-535393 Llorente, L., Zou, W., Levy, Y., Richaud-Patin, Y., Wijdenes, J., Alcocer-Varela, J., et al. (1995). Role of interleukin 10 in the B lymphocyte hyperactivity and autoantibody production of human systemic lupus erythematosus. J. Exp. Med. 181, 839–844. doi: 10.1084/jem.181.3.839 Rathmell, J. C., Fox, C. J., Plas, D. R., Hammerman, P. S., Cinalli, R. M., and Thompson, C. B. (2003). Akt-directed glucose metabolism can prevent Bax conformation change and promote growth factor-independent survival. Mol. Cell Biol. 23, 7315–7328. doi: 10.1128/MCB.23.20.7315-7328.2003 Lowin, B., Hahne, M., Mattmann, C., and Tschopp, J. (1994). Cytolytic T cell cytotoxicity is mediated through perforin and Fas lytic pathways. Nature 370, 650–652. doi: 10.1038/370650a0 Reyes, N. A., Fisher, J. K., Austgen, K., Vanden Berg, S., Huang, E. J., and Oakes, S. A. (2010). Blocking the mitochondrial apoptotic pathway preserves motor neuron viability and function in a mouse model of amyotrophic lateral sclerosis. J. Clin. Invest. 120, 3673–3679. doi: 10.1172/JCI42986l Luhder, F., Hoglund, P., Allison, J. P., Benoist, C., and Mathis, D. (1998). Cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) regulates the unfolding of autoimmune diabetes. J. Exp. Med. 187, 427–432. doi: 10.1084/jem.187.3.427i Richardson, H., and Kumar, S. (2002). Death to flies: drosophila as a model system to study programmed cell death. J. Immunol. Methods 265, 21–38. doi: 10.1016/S0022-1759(02)00068-6 Mahajan, A., Herrmann, M., and Muñoz, L. E. (2016). Clearance deficiency and cell death pathways: a model for the pathogenesis of SLE. Front. Immunol. 7:35. doi: 10.3389/fimmu.2016.00035 Rollins, M. REFERENCES doi: 10.1182/ blood-2002-06-1769 Smith, S. M., Schöder, H., Johnson, J. L., Jung, S. H., Bartlett, N. L., Cheson, B. D., et al. (2013). The anti-CD80 primatized monoclonal antibody, galiximab, is well-tolerated but has limited activity in relapsed Hodgkin lymphoma: cancer and Leukemia Group B 50602 (Alliance). Leuk. Lymphoma 54, 1405–1410. doi: 10.3109/10428194.2012.744453 Muzio, M., Chinnaiyan, A. M., Kischkel, F. C., O’Rourke, K., Shevchenko, A., Ni, J., et al. (1996). FLICE, a novel FADD-homologous ICE/CED 3 like protease, is recruited to the CD95 (Fas/APO 1) death-inducing signaling complex. Cell 85, 817–827. doi: 10.1016/S0092-8674(00)81266-0 Smith-Garvin, J. E., Koretzky, G. A., and Jordan, M. S. (2009). T cell activation. Annu. Rev. Immunol. 27, 591–619. doi: 10.1146/annurev.immunol.021908.132706 Spencer, S. L., and Sorger, P. K. (2011). Measuring and modeling apoptosis in single cells. Cell 144, 926–939. doi: 10.1016/j.cell.2011.03.002 Smith-Garvin, J. E., Koretzky, G. A., and Jordan, M. S. (2009). T cell activation. Annu. Rev. Immunol. 27, 591–619. doi: 10.1146/annurev.immunol.021908.132706 Nagata, S., and Tanaka, M. (2017). Programmed cell death and the immune system. Nat. Rev. Immunol. 17, 333–340. doi: 10.1038/nri.2016.153 Spencer, S. L., and Sorger, P. K. (2011). Measuring and modeling apoptosis in single cells. Cell 144, 926–939. doi: 10.1016/j.cell.2011.03.002 O’Brien, B. A., Geng, X., Orteu, C. H., Huang, Y., Ghoreishi, M., Zhang, Y., et al. (2006). A deficiency in the in vivo clearance of apoptotic cells is a feature of the NOD mouse. J. Autoimmun. 26, 104–115. doi: 10.1016/j.jaut.2005.11.006 Obeid, M., Tesniere, A., Ghiringhelli, F., Fimia, G. M., Apetoh, L., Perfettini, J. L., et al. (2007). Calreticulin exposure dictates the immunogenicity of cancer cell death. Nat. Med. 13, 54–61. doi: 10.1038/nm1523 O’Brien, B. A., Geng, X., Orteu, C. H., Huang, Y., Ghoreishi, M., Zhang, Y., et al. (2006). A deficiency in the in vivo clearance of apoptotic cells is a feature of the NOD mouse. J. Autoimmun. 26, 104–115. doi: 10.1016/j.jaut.2005.11.006 Stamper, C. C., Zhang, Y., Tobin, J. F., Erbe, D. V., Ikemizu, S., Davis, S. J., et al. (2001). Crystal structure of the B7-1/CTLA-4 complex that inhibits human immune responses. Nature 410, 608–611. doi: 10.1038/35069118h Obeid, M., Tesniere, A., Ghiringhelli, F., Fimia, G. M., Apetoh, L., Perfettini, J. L., et al. (2007). Calreticulin exposure dictates the immunogenicity of cancer cell death. Nat. Med. 13, 54–61. doi: 10.1038/nm1523 Steinman, R. M., Turley, S., Mellman, I., and Inaba, K. (2000). The induction of tolerance by dendritic cells that have captured apoptotic cells. J. Exp. Med. REFERENCES Cell Death Differ. 13, 41–52. doi: 10.1038/sj.cdd.4401699 Gullaud, M., Delanoue, R., and Silber, J. (2003). A Drosophila model to study the functions of TWIST orthologs in apoptosis and proliferation. Cell Death Differ. 10, 641–651. doi: 10.1038/sj.cdd.4401222 f Heimberg, H., Heremans, Y., Jobin, C., Leemans, R., Cardozo, A. K., Darville, M., et al. (2001). Inhibition of cytokine-induced NF-κB activation by adenovirus- mediated expression of a NF-κB super-repressor prevents β-cell apoptosis. Diabetes 50, 2219–2224. doi: 10.2337/diabetes.50.10.2219 Kow, N. Y., and Mak, A. (2013). Costimulatory pathways: physiology and potential therapeutic manipulation in systemic lupus erythematosus. Clin. Dev. Immunol. 2013:245928. doi: 10.1155/2013/245928 Krummel, M. F., and Allison, J. P. (1996). CTLA-4 engagement inhibits IL-2 accumulation and cell cycle progression upon activation of resting T cells. J. Exp. Med. 183, 2533–2540. doi: 10.1084/jem.183.6.2533 Hirano, F., Kaneko, K., Tamura, H., Dong, H., Wang, S., Ichikawa, M., et al. (2005). Blockade of B7-H1 and PD-1 by monoclonal antibodies potentiates cancer therapeutic immunity. Cancer Res. 65, 1089–1096. May 2019 \| Volume 10 \| Article 562 Frontiers in Pharmacology \| www.frontiersin.org 9 A Model for Apoptotic Immune Evasion Yakoub and Schülke Kushwah, R., Wu, J., Oliver, J. R., Jiang, G., Zhang, J., Siminovitch, K. A., et al. (2010). Uptake of apoptotic DC converts immature DC into tolerogenic DC that induce differentiation of Foxp3+ Treg. Eur. J. Immunol. 40, 1022–1035. doi: 10.1002/eji.200939782 Paterson, A. M., Brown, K. E., Keir, M. E., Vanguri, V. K., Riella, L. V., Chandraker, A., et al. (2011). The programmed death-1 ligand 1:B7-1 pathway restrains diabetogenic effector T cells in vivo. J. Immunol. 187, 1097–1105. doi: 10.4049/jimmunol.1003496 Larmonier, N., Billerey, C., Rébé, C., Parcellier, A., Moutet, M., Fromentin, A., et al. (2002). An atypical caspase-independent death pathway for an immunogenic cancer cell line. Oncogene 21, 6091–6100. doi: 10.1038/ sj.onc.1205738 Pathak, S. K., Sköld, A. E., Mohanram, V., Persson, C., Johansson, U., and Spetz, A. L. (2012). Activated apoptotic cells induce dendritic cell maturation via engagement of Toll-like receptor 4 (TLR4), dendritic cell-specific intercellular adhesion molecule 3 (ICAM-3)-grabbing nonintegrin (DC-SIGN), and β2 integrins. J. Biol. Chem. 287, 13731–13742. doi: 10.1074/jbc.M111.336545 p p p via engagement of Toll-like receptor 4 (TLR4), dendritic cell-specific intercellular adhesion molecule 3 (ICAM-3)-grabbing nonintegrin (DC-SIGN), and β2 integrins. J. Biol. Chem. 287, 13731–13742. doi: 10.1074/jbc.M111.336545 Linsley, P. S., and Ledbetter, J. A. (1993). The role of the CD28 receptor during T cell responses to antigen. Annu. Rev. Immunol. 11, 191–212. REFERENCES 191, 411–416. doi: 10.1084/jem.191.3.411 Odorizzi, P. M., and Wherry, E. J. (2012). Inhibitory receptors on lymphocytes: insights from infections. J. Immunol. 188, 2957–2965. doi: 10.4049/ jimmunol.1100038 Stuart, L. M., Lucas, M., Simpson, C., Lam, J., Savill, J., and Lacy-Hulbert, A. (2002). Inhibitory effects of apoptotic cell ingestion upon endotoxin-driven myeloid dendritic cell maturation. J. Immunol. 168, 1627–1635. doi: 10.4049/ jimmunol.168.4.1627 Pardoll, D. M. (2012). The blockade of immune checkpoints in cancer immunotherapy. Nat. Rev. Cancer 12, 252–264. doi: 10.1038/nrc3239 May 2019 \| Volume 10 \| Article 562 Frontiers in Pharmacology \| www.frontiersin.org 10 A Model for Apoptotic Immune Evasion Yakoub and Schülke Tai, X., Van Laethem, F., Sharpe, A. H., and Singer, A. (2007). Induction of autoimmune disease in CTLA-4−/− mice depends on a specific CD28 motif that is required for in vivo costimulation. Proc. Natl. Acad. Sci. USA 104, 13756–13761. doi: 10.1073/pnas.0706509104 Waterhouse, P., Penninger, J. M., Timms, E., Wakeham, A., Shahinian, A., Lee, K. P., et al. (1995). Lymphoproliferative disrorders with early lethality in mice deficient in Ctla-4. Science 270, 985–988. doi: 10.1126/science.270.5238.985 i Weyd, H., Abeler-Dörner, L., Linke, B., Mahr, A., Jahndel, V., Pfrang, S., et al. (2013). Annexin A1 on the surface of early apoptotic cells suppresses CD8+ i Weyd, H., Abeler-Dörner, L., Linke, B., Mahr, A., Jahndel, V., Pfrang, S., et al. (2013). Annexin A1 on the surface of early apoptotic cells suppresses CD8+ T cell immunity. PLoS One 8:e62449. doi: 10.1371/journal.pone.0062449 Tivol, E. A., Borriello, F., Schweitzer, A. N., Lynch, W. P., Bluestone, J. A., and Sharpe, A. H. (1995). Loss of CTLA-4 leads to massive lymphoproliferation and fatal multiorgan tissue destruction, revealing a critical negative regulatory role of CTLA-4. Immunity 3, 541–547. doi: 10.1016/1074-7613(95)90125-6 Williams, B., Dharmapatni, A., and Crotti, T. (2018). Intracellular apoptotic pathways: a potential target for reducing joint damage in rheumatoid arthritis. Inflamm. Res. 67, 219–231. doi: 10.1007/s00011-017-1116-5 Torchinsky, M. B., Garaude, J., Martin, A. P., and Blander, J. M. (2009). Innate immune recognition of infected apoptotic cells directs TH17 cell differentiation. Nature 458, 78–82. doi: 10.1038/nature07781 l Wu, Y., Wang, F., Fu, M., Wang, C., Quon, M. J., and Yang, P. (2015). Cellular stress, excessive apoptosis, and the effect of metformin in a mouse model of type 2 diabetic embryopathy. Diabetes 64, 2526–2536. doi: 10.2337/db14-1683 Trembleau, S., Penna, G., Bosi, E., Mortara, A., Gately, M. K., and Adorini, L. (1995). REFERENCES Interleukin 12 administration induces T helper type 1 cells and accelerates autoimmune diabetes in NOD mice. J. Exp. Med. 181, 817–821. doi: 10.1084/jem.181.2.817 Xia, C. Q., Peng, R., Qiu, Y., Annamalai, M., Gordon, D., and Clare-Salzler, M. J. (2007). Transfusion of apoptotic beta-cells induces immune tolerance to beta-cell antigens and prevents type 1 diabetes in NOD mice. Diabetes 56, 2116–2123. doi: 10.2337/db06-0825i Turbyville, J. C., and Rao, V. K. (2010). The autoimmune lymphoproliferative syndrome: a rare disorder providing clues about normal tolerance. Autoimmun. Rev. 9, 488–493. doi: 10.1016/j.autrev.2010.02.007 Xu, D., Woodfield, S. E., Lee, T. V., Fan, Y., Antonio, C., and Bergmann, A. (2009). Genetic control of programmed cell death (apoptosis) in drosophila. Fly 3, 78–90.f j Ueda, H., Howson, J. M., Esposito, L., Heward, J., Snook, H., Chamberlain, G., et al. (2003). Association of the T-cell regulatory gene CTLA4 with susceptibility to autoimmune disease. Nature 423, 506–511. doi: 10.1038/nature01621 van der Merwe, P. A., and Davis, S. J. (2003). Molecular interactions mediating T cell antigen recognition. Annu. Rev. Immunol. 21, 659–684. doi: 10.1146/ annurev.immunol.21.120601.141036 Ueda, H., Howson, J. M., Esposito, L., Heward, J., Snook, H., Chamberlain, G., et al. (2003). Association of the T-cell regulatory gene CTLA4 with susceptibility Ueda, H., Howson, J. M., Esposito, L., Heward, J., Snook, H., Chamberlain, G., et al. (2003). Association of the T-cell regulatory gene CTLA4 with susceptibility to autoimmune disease. Nature 423, 506–511. doi: 10.1038/nature01621 Yakoub, A. M., Schulz, R., Seiffert, M., and Sadek, M. (2018). autoantigen- harboring apoptotic cells hijack the coinhibitory pathway of T cell activation. Sci. Rep. 8:10533. doi: 10.1038/s41598-018-28901-0 van der Merwe, P. A., and Davis, S. J. (2003). Molecular interactions mediating T cell antigen recognition. Annu. Rev. Immunol. 21, 659–684. doi: 10.1146/ annurev.immunol.21.120601.141036 T cell antigen recognition. Annu. Rev. Immunol. 21, 659–684. doi: 10.1146/ annurev.immunol.21.120601.141036 Yamada, A., Arakaki, R., Saito, M., Kudo, Y., and Ishimaru, N. (2017). Dual role of Fas/FasL-mediated signal in peripheral immune tolerance. Front. Immunol. 8:403. doi: 10.3389/fimmu.2017.00403 Voll, R. E., Herrmann, M., Roth, E. A., Stach, C., Kalden, J. R., and Girkontaite, I. (1997a). Immunosuppressive effects of apoptotic cells. Nature 390, 350–351. i Yamaguchi, Y., Shinotsuka, N., Nonomura, K., Takemoto, K., Kuida, K., Yosida, H., et al. (2011). Live imaging of apoptosis in a novel transgenic mouse highlights its role in neural tube closure. J. Cell Biol. 195, 1047–1060. doi: 10.1083/jcb.201104057 f Voll, R. E., Roth, E. REFERENCES A., Girkontaite, I., Fehr, H., Herrmann, M., Lorenz, H. M., et al. (1997b). Histone-specific Th0 and Th1 clones derived from systemic lupus erythematosus patients induce double-stranded DNA antibody production. Arthritis Rheum. 40, 2162–2171. Yamazaki, S., Iyoda, T., Tarbell, K., Olson, K., Velinzon, K., Inaba, K., et al. (2003). Direct expansion of functional CD25+ CD4+ regulatory T cells by antigen- processing dendritic cells. J. Exp. Med. 198, 235–247. doi: 10.1084/jem.20030422 Walunas, T. L., Bakker, C. Y., and Bluestone, J. A. (1996). CTLA-4 ligation blocks CD28- dependent T cell activation. J. Exp. Med. 183, 2541–2550. doi: 10.1084/jem.183.6.2541 Zhu, J., Liu, X., Xie, C., Yan, M., Yu, Y., Sobel, E. S., et al. (2005). T cell hyperactivity in lupus as a consequence of hyperstimulatory antigen-presenting cells. J. Clin. Invest. 115, 1869–1878. doi: 10.1172/JCI23049 Walunas, T. L., Lenschow, D. J., Bakker, C. Y., Linsley, P. S., Freeman, G. J., Green, J. M., et al. (1994). CTLA-4 can function as a negative regulator of T cell activation. Immunity 1, 405–413. doi: 10.1016/1074-7613(94)90071-X Zou, H., Henzel, W. J., Liu, X., Lutschg, A., and Wang, X. (1997). Apaf 1, a human protein homologous to C. elegans CED 4, participates in cytochrome c dependent activation of caspase 3. Cell 90, 405–413. doi: 10.1016/S0092-8674(00)80501-2 Wang, Z., Larregina, A. T., Shufesky, W. J., Perone, M. J., Montecalvo, A., Zahorchak, A. F., et al. (2006). Use of the inhibitory effect of apoptotic cells on dendritic cells for graft survival via T-cell deletion and regulatory T cells. Am. J. Transplant. 6, 1297–1311. doi: 10.1111/j.1600-6143.2006.01308.x Zou, H., Li, Y., Liu, X., and Wang, X. (1999). An APAF 1. Cytochrome c multimeric complex is a functional apoptosome that activates procaspase 9. J. Biol. Chem. 274, 11549–11556. doi: 10.1074/jbc.274.17.11549 Wang, X. F., Lei, Y., Chen, M., Chen, C. B., Ren, H., and Shi, T. D. (2013). PD-1/PDL1 and CD28/CD80 pathways modulate natural killer T cell function to inhibit hepatitis B virus replication. J. Viral Hepat. 20, 27–39. doi: 10.1111/ jvh.12061 Conflict of Interest Statement: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Conflict of Interest Statement: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Wang, Z., Shufesky, W. J., Montecalvo, A., Divito, S. J., Larregina, A. Frontiers in Pharmacology \| www.frontiersin.org REFERENCES T., and Morelli, A. E. (2009). In situ-targeting of dendritic cells with donor-derived apoptotic cells restrains indirect allorecognition and ameliorates allograft vasculopathy. PLoS One 4:e4940. doi: 10.1371/journal.pone.0008442h Copyright © 2019 Yakoub and Schülke. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Ward-Kavanagh, L. K., Lin, W. W., Šedý, J. R., and Ware, C. F. (2016). The TNF receptor superfamily in co-stimulating and co-inhibitory responses. Immunity 44, 1005–1019. doi: 10.1016/j.immuni.2016.04.019 May 2019 \| Volume 10 \| Article 562 Frontiers in Pharmacology \| www.frontiersin.org Frontiers in Pharmacology \| www.frontiersin.org 11
https://openalex.org/W2953840854	https://www.nature.com/articles/s41598-019-46210-y.pdf	English	null	BRCA1 regulates the cancer stem cell fate of breast cancer cells in the context of hypoxia and histone deacetylase inhibitors	Scientific reports	2,019	cc-by	8,060	BRCA1 regulates the cancer stem cell fate of breast cancer cells in the context of hypoxia and histone deacetylase inhibitors Hoon Kim1,2 Qun Lin1 & ZhongYun 1 Received: 24 August 2018 Accepted: 21 June 2019 Published: xx xx xxxx Hoon Kim1,2, Qun Lin1 & Zhong Yun 1 Cancer cell stemness is essential for enabling malignant progression and clonal evolution. Cancer cell fate is likely determined by complex mechanisms involving both cell-intrinsic pathways and stress signals from tumor microenvironment. In this study, we examined the role of the tumor suppressor BRCA1 and hypoxia in the regulation of cancer cell stemness using genetically matched breast cancer cell lines. We have found that BRCA1, a multifunctional protein involved in DNA repair and epigenetic regulation, plays a critical role in the regulation of cancer stem cell (CSC)-like characteristics. Reconstitution of BRCA1 resulted in significant decrease of the CSC-like populations in breast cancer cells whereas down-regulation of BRCA1 resulted in significant increase of the CSC-like populations. Furthermore, the BRCA1-reconstituted tumor cells are more sensitive to the histone deacetylase (HDAC) inhibitor-induced loss of stemness than the BRCA1-deficient cells are. Surprisingly, hypoxia preferentially blocks HDAC inhibitor-induced differentiation of the BRCA1-reconstituted breast cancer cells. In light of the increasing numbers of clinical trials involving HDAC inhibitors in human cancers, our observations strongly suggest that the BRCA1 status and tumor hypoxia should be considered as potentially important clinical parameters that may affect the therapeutic efficacy of HDAC inhibitors. Tumor cells, even cell lines in vitro, consist of mixed populations some of which are capable of tumor initiation and possess stem cell-like characteristics. These tumor-initiating cells (TICs), also interchangeably termed as can- cer stem cells (CSC), are thought to be the major cause of therapy resistance and tumor recurrence1. The cell fate of tumor cells, just like that of normal stem or progenitor cells, is subject to tight regulations by intrinsic genetic and epigenetic factors, as well as by their niche microenvironment. p g y The tumor suppressor gene BRCA1 is frequently mutated in human cancers including breast cancer, ovarian cancer and prostate cancer2,3. BRCA1 protein plays a critical role in error-free DNA repair and its mutation is associated with global chromosome instability and tumor formation4–6. BRCA1 has also been found to play an important role in chromatin remodeling and gene transcription, indicating that BRCA1 may have pleiotropic functions during tumor development7–9. Interestingly, BRCA1 has been shown to be required for differentiation of mammary stem/progenitor cells to luminal epithelial cells10,11, suggesting that BRCA1 constitutes an important intrinsic pathway involved in cell fate determination. www.nature.com/scientificreports www.nature.com/scientificreports www.nature.com/scientificreports 1Department of Therapeutic Radiology, Yale School of Medicine, New Haven, CT, 06510, USA. 2Present address: Oncology & Immunology Research Team II, LG Chem. LG Science Park, E8 30 Magokjungang 10-ro Gangseo-gu, Seoul, 07796, Korea. Correspondence and requests for materials should be addressed to Z.Y. (email: zhong.yun@yale.edu) BRCA1 regulates the cancer stem cell fate of breast cancer cells in the context of hypoxia and histone deacetylase inhibitors Hoon Kim1,2 Qun Lin1 & ZhongYun 1 As an emerging concept, tumor microenvironment can potentially provide a unique niche for CSCs to survive and continuously propagate12–14. Increasing evidence shows that hypoxia, a condition of oxygen deficiency and a hallmark of tumor microenvironment (TME), up-regulates CSC-related genes, promotes self-renewal and sup- presses cell differentiation15,16. A number of in vitro studies have shown that hypoxia or hypoxia-sensing pathways play a significant role in the maintenance of the CSC phenotype in breast cancer cells17–23. Hypoxia is also impli- cated in increased CSC-like populations in breast cancer xenografts treated by antiangiogenic agents24. We have recently found direct evidence that CSC-like population of breast cancer cells are significantly enriched in the hypoxic regions in vivo25. Interestingly, it has been shown that BRCA1 transcription is strongly repressed under hypoxic conditions26,27, suggesting that inadequate BRCA1 expression and functions can be found in the hypoxic 1Department of Therapeutic Radiology, Yale School of Medicine, New Haven, CT, 06510, USA. 2Present address: Oncology & Immunology Research Team II, LG Chem. LG Science Park, E8 30 Magokjungang 10-ro Gangseo-gu, Seoul, 07796, Korea. Correspondence and requests for materials should be addressed to Z.Y. (email: zhong.yun@yale.edu) Scientific Reports \| (2019) 9:9702 \| https://doi.org/10.1038/s41598-019-46210-y 1 www.nature.com/scientificreports/ Figure 1. BRCA1 suppresses ALDH1A expression and activity in breast cancer cells. (A) Wild-type BRCA1 was reconstituted in HCC1937 human breast cancer cell line by retroviral transduction (Lane 2) with the empty vector as control (Lane 1). In addition, the BRCA1-deficient MDA-MB-436 (Lane 3) and the BRCA1- competent SKBR-3 (Lane 4) were used as control for Western blots. NSB: non-specific protein band. (B) The clonogenic potential of HCC1937 ± BRCA1 cells (n = 6). (C,D) The aldehyde dehydrogenase activity measured by the AldeRed assay with representative flow cytometry data shown in (C) and quantitation in (D, n = 3). The DEAB treated cells were used as negative control. (E) Expression levels of three ALDH1A isoforms in HCC1937 ± BRCA1 cells measured by qRT-PCR (n = 3). Figure 1. BRCA1 suppresses ALDH1A expression and activity in breast cancer cells. (A) Wild-type BRCA1 was reconstituted in HCC1937 human breast cancer cell line by retroviral transduction (Lane 2) with the Figure 1. BRCA1 suppresses ALDH1A expression and activity in breast cancer cells. (A) Wild-type BRCA1 was reconstituted in HCC1937 human breast cancer cell line by retroviral transduction (Lane 2) with the empty vector as control (Lane 1). BRCA1 regulates the cancer stem cell fate of breast cancer cells in the context of hypoxia and histone deacetylase inhibitors Hoon Kim1,2 Qun Lin1 & ZhongYun 1 In addition, the BRCA1-deficient MDA-MB-436 (Lane 3) and the BRCA1- competent SKBR-3 (Lane 4) were used as control for Western blots. NSB: non-specific protein band. (B) The clonogenic potential of HCC1937 ± BRCA1 cells (n = 6). (C,D) The aldehyde dehydrogenase activity measured by the AldeRed assay with representative flow cytometry data shown in (C) and quantitation in (D, n = 3). The DEAB treated cells were used as negative control. (E) Expression levels of three ALDH1A isoforms in HCC1937 ± BRCA1 cells measured by qRT-PCR (n = 3). tumor microenvironment in solid tumors. These findings suggest that hypoxia and downregulation of BRCA1 could synergize to enhance and/or maintain stem cell characteristics of cancer cells. In this study, we examined the role of BRCA1 in the regulation of breast cancer cell stemness. Reconstitution of BRCA1 expression in the BRCA1-mutated HCC1937 cells resulted in a decrease of the CSC-like populations. On the other hand, down-regulation of BRCA1 in SKBR3 breast cancer cells significantly increased the CSC-like populations. Hypoxia facilitated the enrichment of the CSC-like populations in both BRCA1-competent and BRCA1-deficient breast cancer cells. Furthermore, we found that the BRCA1-reconstituted tumor cells were more sensitive than the BRCA1-mutated cells to histone deacetylase (HDAC) inhibitor-induced differentia- tion. Interestingly, hypoxia significantly blocked HDAC inhibitor-induced differentiation, especially, of the BRCA1-competent breast cancer cells. Our data strongly suggest that BRCA1 does not only regulate cancer cell fate but also affect how cancer cells respond to tumor microenvironmental stresses and therapeutic drugs. Results C 2C, reconstitution of BRCA1 in HCC1937 resulted in significant suppression of the total CD44 expression (p < 0.0001) including the CD44s (p = 0.0001) and CD44v5/6 (p = 0.0002) variants. Using a CD44 promoter-luciferase reporter containing the 2 kb CD44 promoter/enhancer region, we found that the tran- scription activity of the CD44 promoter is significantly (p = 0.0022) down-regulated in the BRCA1-reconstituted HCC1937 cells (Fig. 2D). It has been shown that BRCA1 can interact with the histone deacetylase complex34, thus affecting gene transcription. However, using chromatin immunoprecipitation, we found that ectopic expression of BRCA1 in HCC1937 cells did not reduce the level of histone H3 acetylation of CD44 promoter, nor that of ALDH1A promoter. These observations suggest that BRCA1 is likely to be indirectly involved in down-regulation of the transcription of CD44 and ALDH1A. Nonetheless, the results presented above indicate that restoration of BRCA1 in BRCA1-deficient breast cancer cells is sufficient to decrease their cancer stemness. Knockdown of BRCA1 enhances cancer stem cell-like characteristics. In addition to genetic mutations, BRCA1 expression can be down-regulated under stress conditions such as hypoxia27,35. To deter- mine whether down-regulation of BRCA1 could affect breast cancer stem cell fate, we knocked-down BRCA1 in the BRCA1-competent human breast cancer cell line SKBR3 using RNA interference (Fig. 3A). Interestingly, siBRCA1 treatment resulted in significant increase in the expression of cancer stem cell-associated mark- ers including CD44, ALDH1A3, and OCT4, while leading to down-regulation of CD24 (p < 0.0001 for all pair-wise comparisons, Fig. 3B). The transcription activity of the 2-kb CD44 promoter was moderately, but significantly (p = 0.0254), increased in the siBRCA1-treated SKBR3 cells (Fig. 3C). Consistent with the gene expression data, we found that the CD44+ population was significantly (p = 0.0472) increased upon siBRCA1 treatment (Fig. 3D,E). Knocking-down BRCA1 also significantly (p < 0.0001) increased ALDH+ population in the siBRCA1-treated tumor cells (Fig. 3F,G). ( g , ) To determine whether BRCA1 plays a broad role in the regulation of cancer cell fate in general, we knocked-down BRCA1 expression in human neuroblastoma SK-N-BE(2)C cells (Fig. 4A). We found that siBRCA1-treatment resulted in a significant increase (p < 0.0001) of clonogenic growth of SK-N-BE(2)C cells (Fig. 4B), suggesting increased stemness. Consistently, the stemness-associated CD44+ population was signif- icantly increased (p = 0.0165) in the siBRCA1-treated neuroblastoma cells (Fig. 4C). Results C In contrast, both ALDH1A2 and ALDH1A3 were expressed at much lower levels than ALDH1A1 and their expression was not significantly affected by BRCA1. The transmembrane glycoprotein CD44 is another commonly used cell surface marker of cancer stem cells32,33. Ectopic expression of BRCA1 in HCC1937 cells led to approximately a ten-fold decrease (p < 0.0001) in the CD44+ population (Fig. 2A,B). To examine the effects of BRCA1 in CD44 expression, we used a panel of qRT-PCR primers specific for all CD44 variants (CD44), the standard isoform (CD44s), and the CD44v5/6 var- iant. As shown in Fig. 2C, reconstitution of BRCA1 in HCC1937 resulted in significant suppression of the total CD44 expression (p < 0.0001) including the CD44s (p = 0.0001) and CD44v5/6 (p = 0.0002) variants. Using a CD44 promoter-luciferase reporter containing the 2 kb CD44 promoter/enhancer region, we found that the tran- scription activity of the CD44 promoter is significantly (p = 0.0022) down-regulated in the BRCA1-reconstituted HCC1937 cells (Fig. 2D). It has been shown that BRCA1 can interact with the histone deacetylase complex34, thus affecting gene transcription. However, using chromatin immunoprecipitation, we found that ectopic expression of BRCA1 in HCC1937 cells did not reduce the level of histone H3 acetylation of CD44 promoter, nor that of ALDH1A promoter. These observations suggest that BRCA1 is likely to be indirectly involved in down-regulation of the transcription of CD44 and ALDH1A. Nonetheless, the results presented above indicate that restoration of BRCA1 in BRCA1-deficient breast cancer cells is sufficient to decrease their cancer stemness. ALDH activities in breast cancer cells29–31. Among its three isoforms, we found that ALDH1A1 was significantly (p = 0.0012) down-regulated in BRCA1-reconstituted cells. In contrast, both ALDH1A2 and ALDH1A3 were expressed at much lower levels than ALDH1A1 and their expression was not significantly affected by BRCA1. p p gi yf y The transmembrane glycoprotein CD44 is another commonly used cell surface marker of cancer stem cells32,33. Ectopic expression of BRCA1 in HCC1937 cells led to approximately a ten-fold decrease (p < 0.0001) in the CD44+ population (Fig. 2A,B). To examine the effects of BRCA1 in CD44 expression, we used a panel of qRT-PCR primers specific for all CD44 variants (CD44), the standard isoform (CD44s), and the CD44v5/6 var- iant. As shown in Fig. Results C BRCA1 suppresses cancer stem cell-like characteristics of human breast cancer cells. To exam- ine the role of BRCA1 in the regulation of breast cancer cell stemness, we created a genetically matched pair of human breast cancer cell lines using the HCC1937 cell line derived from a Grade 3 primary ductal carcinoma with a loss-of-function mutation in the BRCA1 gene (insertion C at nucleotide 5382). The HCC1937BRCA1 cell line was generated by infection of retrovirus containing the full-length wild-type BRCA1 and the control line was made using the empty vector-containing viruses (Fig. 1A). Reconstitution with the wild-type BRCA1 significantly (p < 0.0001) suppressed the clonogenic potential of HCC1937 cells (Fig. 1B), an important charac- teristics of cancer stemness. We further determined the impact of BRCA1 on breast cancer stemness using the ALDH activity assay as a readout for the endogenous ALDH activities, a widely used functional assay of breast cancer stemness28,29. As shown in Fig. 1C,D, ectopic expression of BRCA1 in HCC1937 cells resulted in approx- imately 50% decrease of ALDH activities (p = 0.0032). ALDH1 has been shown to be the major contributor of Scientific Reports \| (2019) 9:9702 \| https://doi.org/10.1038/s41598-019-46210-y 2 www.nature.com/scientificreports/ Figure 2. BRCA1 suppresses CD44 expression in breast cancer cells. The cell surface CD44 levels in HCC1937 ± BRCA1 cells were analyzed by flow cytometry with representative flow cytometry data shown in (A) and quantitation in (B, n = 3). (C) Expression levels of the total CD44 (including all isoforms/variants), CD44s, and the CD44v5/6 variant in HCC1937 ± BRCA1 cells measured by qRT-PCR (n = 3). (D) CD44 promoter activity in HCC1937 ± BRCA1 cells was measured using luciferase assay (n = 4). Figure 2. BRCA1 suppresses CD44 expression in breast cancer cells. The cell surface CD44 levels in HCC1937 ± BRCA1 cells were analyzed by flow cytometry with representative flow cytometry data shown in (A) and quantitation in (B, n = 3). (C) Expression levels of the total CD44 (including all isoforms/variants), CD44s, and the CD44v5/6 variant in HCC1937 ± BRCA1 cells measured by qRT-PCR (n = 3). (D) CD44 promoter activity in HCC1937 ± BRCA1 cells was measured using luciferase assay (n = 4). ALDH activities in breast cancer cells29–31. Among its three isoforms, we found that ALDH1A1 was significantly (p = 0.0012) down-regulated in BRCA1-reconstituted cells. Results C On the other hand, the cell population with high levels of the differentiation-associated CD24 (CD24++) were significantly decreased (p < 0.001), whereas the low expressor population (CD24+) were significantly increased (p < 0.001), as a con- sequence of BRCA1 knockdown (Fig. 4D). We further found that the stem cell-associated genes CD44, SOX2, MSI1, and ASCL1 were all upregulated, whereas CD24 was suppressed, upon siBRCA1 treatment (Fig. 4E). Collectively, these data have clearly demonstrated that down-regulation of BRCA1 can enhance cancer cell stem- ness in multiple tumor types. Scientific Reports \| (2019) 9:9702 \| https://doi.org/10.1038/s41598-019-46210-y 3 3 www.nature.com/scientificreports/ Figure 3. Down-regulation of BRCA1 promotes breast cancer stem cell characteristics. (A) BRCA1 was down-regulated by RNA inference in the BRCA1-competent SKBR-3 breast cancer cells, which was confirmed by Western blot and qRT-PCR (n = 3). (B) Expression of breast cancer stem cell-associated genes in SKBR3 ± siBRCA1 cells were measured by qRT-PCR (n = 3). (C) CD44 promoter activity in SKBR3 ± siBRCA1 cells was measured using luciferase assay (n = 3). The cell surface CD44 levels in SKBR3 ± siBRCA1 cells were analyzed by flow cytometry with representative flow cytometry data shown in (D) and quantitation in (E, n = 3). The ALDH activity in SKBR3 ± siBRCA1 cells were analyzed by flow cytometry with representative flow cytometry data shown in (F) and quantitation in (G, n = 3). The DEAB treated cells were used as negative control. Figure 3. Down-regulation of BRCA1 promotes breast cancer stem cell characteristics. (A) BRCA1 g g p ( ) was down-regulated by RNA inference in the BRCA1-competent SKBR-3 breast cancer cells, which was confirmed by Western blot and qRT-PCR (n = 3). (B) Expression of breast cancer stem cell-associated genes in SKBR3 ± siBRCA1 cells were measured by qRT-PCR (n = 3). (C) CD44 promoter activity in SKBR3 ± siBRCA1 cells was measured using luciferase assay (n = 3). The cell surface CD44 levels in SKBR3 ± siBRCA1 cells were analyzed by flow cytometry with representative flow cytometry data shown in (D) and quantitation in (E, n = 3). The ALDH activity in SKBR3 ± siBRCA1 cells were analyzed by flow cytometry with representative flow cytometry data shown in (F) and quantitation in (G, n = 3). The DEAB treated cells were used as negative control. Hypoxia increases ALDH activities independent of BRCA1. Results C Hypoxia is a hallmark of tumor microen- vironment (TME) in solid tumors. Tumor hypoxia is an independent prognostic factor for advanced disease progression and poor patient survival36–38. Studies from us and others have shown that hypoxia can facilitate the maintenance or enrichment of the cell population with stem cell characteristics16,25,39–42. We asked whether the BRCA1 status would affect breast cancer cell stemness under hypoxic conditions. Using the genetically matched pair of HCC1937 ± BRCA1 cells, we found that hypoxia significantly increased the ALDH+ populations of both parental (Vector) cells and the BRCA1-reconstituted HCC1937 cells in comparison to their respective nor- moxia (20% O2) controls (Fig. 5A,B). However, because BRCA1 strongly suppressed the ALDH activities, the hypoxia-increased ALDH+ population of HCC1937 + BRCA1 cells still did not reach the level of HCC1937/ Vector control cells under normoxia (Fig. 5A,B). These results suggest that BRCA1, i.e. the genetic background, is a profoundly powerful determinant of breast cancer cell fate as compared to hypoxia, i.e. the microenvironment. Consistently, hypoxia strongly increased the ALDH+ populations of BRCA1-profiient SKBR3 (siCtrl) and the BRCA1-knocked-down SKBR3 (siBRCA1) cells (Fig. 5C). Nonetheless, much higher ALDH+ populations were observed in the siBRCA1-treated SKBR3, which further supports BRCA1 as an important regulator of breast cancer cell stemness. BRCA1-deficient breast cancer cells are less sensitive to HDAC inhibition. In recent years, histone deacetylase (HDAC) inhibitors have emerged as an important class of anti-cancer drugs by inducing transcriptional and other epigenetic stresses in cancer cells43,44. In light of the role of BRCA1 in the regulation of both coding and non-coding RNAs34,45, we hypothesized that the status of BRCA1 would affect tumor cell Scientific Reports \| (2019) 9:9702 \| https://doi.org/10.1038/s41598-019-46210-y 4 www.nature.com/scientificreports/ www.nature.com/scientificreports entificreports/ Figure 4. Down-regulation of BRCA1 promotes cancer cell stemness in neuroblastoma cells. BRCA1 was down-regulated by RNA inference in the SK-N-BE(2)C human neuroblastoma cells (Western blot shown in A). Effects of BRCA1 on neuroblastoma cell clonogenicity was measured by the clonogenic assay (B, n = 6). The cell surface levels of CD44 (C) and CD24 (D) in BE(2)C ± siBRCA1 cells were analyzed by flow cytometry (n = 3). Expression of neuroendocrine cancer stem cell-associated genes in BE(2)C ± siBRCA1 cells were measured by qRT-PCR (E, n = 3). Figure 4. Down-regulation of BRCA1 promotes cancer cell stemness in neuroblastoma cells. BRCA1 was down-regulated by RNA inference in the SK-N-BE(2)C human neuroblastoma cells (Western blot shown in A). Results C Effects of BRCA1 on neuroblastoma cell clonogenicity was measured by the clonogenic assay (B, n = 6). The cell surface levels of CD44 (C) and CD24 (D) in BE(2)C ± siBRCA1 cells were analyzed by flow cytometry (n = 3). Expression of neuroendocrine cancer stem cell-associated genes in BE(2)C ± siBRCA1 cells were measured by qRT-PCR (E, n = 3). response to histone deacetylases. We treated the genetically matched pair of parental (Vector) cells and the BRCA1-reconstituted HCC1937 breast cancer cells with SAHA44, a broad spectrum HDAC inhibitors and an FDA approved anti-cancer drug (Vorinostat), under normoxia and hypoxia, respectively. Using the ALDH activ- ity assay, we found that SAHA strongly decreased ALDH activities in the BRCA1-reconstituted HCC1937 cells with the ALDH+ populations decreasing from >20% (untreated control, open circle on axis, Fig. 6A) to ≤5% under normoxia (open squares, Fig. 6A, p values shown in chart) without significantly affecting the BRCA1 protein levels (Supplementary Fig. 1). In contrast, SAHA induced moderate decrease of the ALDH+ HCC1937 parental (Vector) populations from >40% (untreated control, open circle on axis, Fig. 6B) to approximately 30% (open squares, Fig. 6B, p values shown in chart). response to histone deacetylases. We treated the genetically matched pair of parental (Vector) cells and the BRCA1-reconstituted HCC1937 breast cancer cells with SAHA44, a broad spectrum HDAC inhibitors and an FDA approved anti-cancer drug (Vorinostat), under normoxia and hypoxia, respectively. Using the ALDH activ- ity assay, we found that SAHA strongly decreased ALDH activities in the BRCA1-reconstituted HCC1937 cells with the ALDH+ populations decreasing from >20% (untreated control, open circle on axis, Fig. 6A) to ≤5% under normoxia (open squares, Fig. 6A, p values shown in chart) without significantly affecting the BRCA1 protein levels (Supplementary Fig. 1). In contrast, SAHA induced moderate decrease of the ALDH+ HCC1937 parental (Vector) populations from >40% (untreated control, open circle on axis, Fig. 6B) to approximately 30% (open squares, Fig. 6B, p values shown in chart). ( p q g p ) When treated with SAHA under the hypoxic condition, the ALDH+ HCC1937-BRCA1 population was both strongly and significantly increased as compared to their corresponding normoxia controls over 3 days of con- tinuous drug treatment (closed triangle versus open squares, Fig. 6A). Results C On the other hand, hypoxia was able to significantly increase and/or maintain the ALDH+ HCC1937-Vector population in the presence of SAHA for 2 days only (closed triangle versus open squares, Fig. 6B). The hypoxia effects diminished by the third day (Fig. 6B). Scientific Reports \| (2019) 9:9702 \| https://doi.org/10.1038/s41598-019-46210-y 5 www.nature.com/scientificreports/ These results suggest that BRCA1-competent breast cancer cells are more sensitive, whereas BR Figure 5. BRCA1 affects hypoxia-induced breast cancer cell stemness. HCC1937 ± BRCA1 cells w maintained under ambient tissue culture conditions (20% O2) or under hypoxia (1% O2). Their AL activities were measured at the indicated time. The representative flow data are shown in (A) and q in (B, n = 3). SKBR3 ± siBRCA1 cells were maintained under normoxia or hypoxia and their ALD (ALDEFLUOR) were measured at the indicated time with the flow data shown in (C) and quantita n = 2). The DEAB treated cells were used as negative control. Figure 5. BRCA1 affects hypoxia-induced breast cancer cell stemness. HCC1937 ± BRCA1 cells were either maintained under ambient tissue culture conditions (20% O2) or under hypoxia (1% O2). Their ALDH activities were measured at the indicated time. The representative flow data are shown in (A) and quantitation in (B, n = 3). SKBR3 ± siBRCA1 cells were maintained under normoxia or hypoxia and their ALDH activities (ALDEFLUOR) were measured at the indicated time with the flow data shown in (C) and quantitation in (D, n = 2). The DEAB treated cells were used as negative control. ( % 2) yp ( % 2)h ctivities were measured at the indicated time. The representative flow data are shown in (A) and quantitation n (B, n = 3). SKBR3 ± siBRCA1 cells were maintained under normoxia or hypoxia and their ALDH activities ALDEFLUOR) were measured at the indicated time with the flow data shown in (C) and quantitation in (D, = 2). The DEAB treated cells were used as negative control. These results suggest that BRCA1-competent breast cancer cells are more sensitive, whereas BRCA1-deficient breast cancer cells are less sensitive, to SAHA-induced loss of stemness under normoxic conditions. Consistent with these findings, SAHA has been shown to induce morphological changes in human breast cancer cells to Scientific Reports \| (2019) 9:9702 \| https://doi.org/10.1038/s41598-019-46210-y 6 www.nature.com/scientificreports/ Figure 6. BRCA1 status and hypoxia determine breast cancer cell response to the histone deacetylase inhibitor SAHA. Results C HCC1937 ± BRCA1 cells were treated with 1 μM SAHA under either normoxia or hypoxia. ALDH activities (ALDEFLUOR) were measured at the indicated time following incubation (n = 3). The DEAB treated cells were used as negative control. Figure 6. BRCA1 status and hypoxia determine breast cancer cell response to the histone deacetylase inhibitor SAHA. HCC1937 ± BRCA1 cells were treated with 1 μM SAHA under either normoxia or hypoxia. ALDH activities (ALDEFLUOR) were measured at the indicated time following incubation (n = 3). The DEAB treated cells were used as negative control. resemble morphology of differentiated cells with flat appearance and decreased nucleus-to-cytoplasm ratio46. Surprisingly, hypoxia has much stronger protective effects in the stemness of BRCA1-competent breast cancer cells. In light of the increased clinical efforts to use SAHA and other HDAC inhibitors for breast cancer therapy, our data suggest that both the BRCA1 status and the hypoxic tumor microenvironment are potential important parameters that may affect the clinical efficacy of HDAC inhibitors. resemble morphology of differentiated cells with flat appearance and decreased nucleus-to-cytoplasm ratio46. Surprisingly, hypoxia has much stronger protective effects in the stemness of BRCA1-competent breast cancer cells. In light of the increased clinical efforts to use SAHA and other HDAC inhibitors for breast cancer therapy, our data suggest that both the BRCA1 status and the hypoxic tumor microenvironment are potential important parameters that may affect the clinical efficacy of HDAC inhibitors. Scientific Reports \| (2019) 9:9702 \| https://doi.org/10.1038/s41598-019-46210-y Methods C ll li Cell lines, cell culture, and hypoxia. HCC-1937 and SKBR3 human breast cancer cells (ATCC) were maintained in RPMI1640. SK-N-BE(2)C human neuroblastoma cells were maintained in MEM/F12 (1:1) sup- plemented with 10% fetal bovine serum. The experimental conditions for hypoxia were the same as described previously52,53. Briefly, the hypoxia culture media were supplemented with 25 mM HEPES at pH7.4 to strengthen the pH buffering capacity. Cells were incubated at 1% O2 in a hypoxia workstation (Ruskinn Invivo 400). f HCC-1937Vector and HCC-1937BRCA1 cell lines were generated by infection with retrovirus containing either pLZRS vector or pLZRS-BRCA1. For RNAi-mediated knockdown of BRCA1, SKBR3 cells were transiently transfected with Dharmacon human BRCA1 siRNA (BRCA1 ON-TARGET plus SMART pool) and negative control siRNA (ON-TARGET plus non-targeting pool) using the DharmaFECT™ transfection reagent according to manufacturer’s instruction. Chemical reagent. SAHA (suberoylanilide hydroxamic acid or Vorinostat, CAS #149647-78-9) was pur- chased from Cayman Chemicals. A 50 mM stock solution was prepared in DMSO. The working concentration was used at 1 μM. Analysis of cell surface markers by flow cytometry. Cell preparation, incubation with antibod- ies, and flow cytometry were performed using our previously published protocol25. Antibodies used for flow cytometry were purchased from eBiosciences (ThermoScientific): anti-CD24-PE-Cyanine 7 (antibody dilu- tion of 1:25, catalogue #25-0247-42), anti-CD44-FITC (antibody dilution of 1:50, catalogue #11-0441-82), anti-CD44-PerCP-Cy5.5 (antibody dilution of 1:50, catalogue #45-0441-82). Isotype controls included mouse IgGκ-PE-Cyanine 7 (antibody dilution of 1:25, catalogue #25-4714-42), rat IgG2b-FITC (antibody dilution of 1:50, catalogue #11-4031-82), and rat IgG2b-PerCP-Cy5.5 (antibody dilution of 1:50, catalogue #45-4031-80). Cells were incubated with antibodies for 30 minutes on ice, filtered, and then analyzed on a BD LSR II flow cytometer. Singlets were gated for analysis, based on forward scatter (FSC) and side scatter (SSC) profiles. The instruments were calibrated daily. FACS data were analyzed using the FlowJo™ software. Aldehyde dehydrogenase (ALDH) activity assay. Cells were incubated using either the AldeRed ALDH Detection Assay kit (EMD Millipore, #SCR150) or the ALDEFLUOR kit (STEMCELL Technologies, #C01700) according to the manufacturer’s instructions. Aldehyde dehydrogenase (ALDH) activity assay. Cells were incubated using either the AldeRed ALDH Detection Assay kit (EMD Millipore, #SCR150) or the ALDEFLUOR kit (STEMCELL Technologies C01700) according to the manufacturer’s instructions. Clonogenic assay. The clonogenic assay is based on our previously published protocols52,54. www.nature.com/scientificreports/ www.nature.com/scientificreports/ The interaction between BRCA1 and HDACs suggests that the BRCA1 status may affect tumor cell response to HDAC inhibitors. We have found that SAHA, an FDA-approved HDAC inhibitor, decreases the CSC-like population in both BRCA1-deficient and -competent breast cancer cells. Nonetheless, the BRCA1-reconstituted HCC1937 cells are much more sensitive to SAHA-induced loss of stemness than BRCA1-deficient HCC1937 parental cells. Given its ability to interact with HDACs34, BRCA1 might synergize with SAHA to create significant epigenetic stresses, which results in the loss of breast cancer cell stemness. Although the mechanisms remain to be examined, these data suggest that the BRCA1 status may determine tumor response to HDAC inhibitors.ii gg y p Consistent with our previous findings, hypoxia, the hallmark of tumor microenvironment, significantly increases the CSC-like population in both BRCA1-deficient and -competent HCC1937 cells. However, the abso- lute increase in the CSC-like population in the BRCA1-reconstituted tumor cells under hypoxia remains small as compared to that in the BRCA1-mutated parental cells. These results suggest that BRCA1-deficiency may confer tumor cells with more efficient adaptability to changes in tumor microenvironment. However, both surprisingly and interestingly, hypoxia can significantly block SAHA-induced loss of stemness in the BRCA1-reconstituted HCC1937 cells, suggesting that hypoxia has the potential to disrupt the synergy between BRCA1 and SAHA in inducing breast cancer cell differentiation. gf Currently, SAHA and other HDAC inhibitors have been actively tested in a large number of clinical trials involving breast cancers and many other cancers (ClinicalTrials.gov). In light of our findings, it is prudent to con- sider the BRCA1 status and tumor hypoxia as potentially important clinical parameters affecting the therapeutic efficacy of HDAC inhibitors. Discussion Cancer cell stemness is essential for enabling malignant progression and clonal evolution. The cell fate of cancer cells is likely to be determined by both cell-intrinsic pathways and by stress signals from tumor microenviron- ment. In this study, we have found that the tumor suppressor BRCA1, widely known as a key player in DNA repair, plays a critical role in the regulation of CSC-like characteristics. Ectopic expression of BRCA1 resulted in significant decrease of the CSC-like populations in human breast cancer cells whereas down-regulation of BRCA1 resulted in significant increase of the CSC-like population. Similar results are obtained from human neuroblas- toma cells, suggesting BRCA1 has the potential to affect cell fate determinant in many tumor types. These results are consistent with the finding that BRCA1 is required for luminal differentiation of human mammary stem/ progenitor cells10. p g The exact mechanisms by which BRCA1 regulates cell fate remain to be delineated. BRCA1 is a multi-functional protein47. In addition to its classical role in homology-dependent DNA repair, BRCA1 has the potential to regulate gene transcription47,48 and to affect the epigenetic landscape via interacting with HDAC com- plexes34 and modulating mRNA expression45 among others. We have found that expression of the CSC-associated markers CD44 and ALDH1A is significantly increased in tumor cells with mutated or down-regulated BRCA1. However, the BRCA1 status did not significantly affect the level of histone H3 acetylation of CD44 promoter and ALDH1A promoter, nor the activity of the CD24 and ALDH1 promoters (data not shown). We also found that reconstitution of BRCA1 resulted in significant increase of the breast cancer cell population expressing high levels of the differentiation-associated CD24 cell surface protein (Supplementary Fig. 2) without significantly affect CD24 mRNA levels. These data suggest that transcription of CD44, ALDH1A and CD24 may be differen- tially regulated by BRCA1-related epigenetic pathways, given the consensus finding that BRCA1 does not bind to DNA in a sequence-specific manner47,48. Nonetheless, our observations are consistent with previous findings that deletion or loss-of-function of BRCA1 facilitates the expansion of mammary progenitor cells49 and promotes the development of basal-like mammary tumors50,51. Scientific Reports \| (2019) 9:9702 \| https://doi.org/10.1038/s41598-019-46210-y 7 www.nature.com/scientificreports/ www.nature.com/scientificreports/ Gene Forward Primer Reverse Primer CD44 (all variants) 5′-TGCCGCTTTGCAGGTGTAT-3′ 5′-GGCCTCCGTCCGAGAGA-3′ CD44s (NM_001001391 and NM_001202557) 5′-TACTGATGATGACGTGAGCA-3′ 5′-GAATGTGTCTTGGTCTCTGGT-3′ CD44v5/6 (NM_001001389) 5′-GTAGACAGAAATGGCACCAC-3′ 5′-CAGCTGTCCCTGTTGTCGAA-3′ BRCA1 5′-TCCTTCCTTGCAGGAAACCAGTCTC-3′ 5′-TGAGGTTGTATCCGCTGCTTTGTCC-3′ MSI1 5′-CTCCAAAACAATTGACCCTAAGGT-3′ 5′-GACAGCCCCCCCACAAAG-3′ ASCL1 5′-CACTGACTTTTGCTGCTGCTTCT-3′ 5′-TGGCGCTCGCGTGTG-3′ ALDH1A1 5′-GCACGCCAGACTTACCTGTC-3′ 5′-CCTCCTCAGTTGCAGGATTAAAG-3′ ALDH1A2 5′-AGGCCCTCCTCGCTCAC-3′ 5′-TCTGCCCCAGAATGAGCTC-3′ ALDH1A3 5′-GCATGAGCCCATTGGTGTCT-3′ 5′-CGCAGGCTTCAGGACCAT-3′ OCT4 5′-GAGAACCGAGTGAGAGGCAACC-3′ 5′-CATAGTCGCTGCTTGATCGCTTG-3′ CD24 5′-AAACAACAACTGGAACTTCAAGTAACTC-3′ 5′-GGTGGTGGCATTAGTTGGATTT-3′ SOX2 5′-TGCGAGCGCTGCACAT-3′ 5′-GCAGCGTGTACTTATCCTTCTTCA-3′ HPRT 5′-TATGGCGACCCGCAGCCCT-3′ 5′-GGTGGTGGCATTAGTTGGATTT-3′ Table 1. List of qRT-PCR primers. Table 1. List of qRT-PCR primers. CD44 promoter activity luciferase assay. The reporter CD44P pGL3, a gift from Dr. Robert Weinberg (Addgene plasmid # 19122), contains 2021 bp of the human CD44 promoter/enhancer upstream of the trans- lation initiation site. CD44P pGL3 reporter plasmid was co-transfected with 0.2 μg Renilla Luciferase plasmid DNA (Promega, Mannheim, Germany) as a control for transfection efficiency. After 48 hr incubation, transfected cells were lysed using the Dual-Luciferase Reporter Assay System (Promega). Promoter-driven luciferase activity was measured on the Llumat LB 9507 machine (EG&G Berthold Co.) and normalized to the Renilla Luciferase activity. Each experiment was carried out in triplicates and repeated three times. Statistics. Two group comparisons were analyzed by 2-tailed Student’s t-test (GraphPad Prism 7). Significant difference was declared if p < 0.05. References 1. Visvader, J. E. & Lindeman, G. J. Cancer stem cells in solid tumours: accumulating evidence and unresolved questions. Nat Rev Cancer 8, 755–768, https://doi.org/10.1038/nrc2499 (2008). p g 2. Miki, Y. et al. A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1. Science 266, 66–71 (1994). p g 2. Miki, Y. et al. A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1. Science 266, 66–71 (1994). 3. Petrucelli, N., Daly, M. B. & Feldman, G. L. Hereditary breast and ovarian cancer due to mutations in BRCA1 and BRCA2. Gene p g . Miki, Y. et al. A strong candidate for the breast and ovarian cance p g 2. Miki, Y. et al. A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1. Science 266, 66–71 (1994). g p y g 3. Petrucelli, N., Daly, M. B. & Feldman, G. L. Hereditary breast and ovarian cancer due to mutations in BRCA1 and BRCA2. Gene Med 12, 245–259, https://doi.org/10.1097/GIM.0b013e3181d38f2f (2010).h 4. Wu, J., Lu, L. Y. & Yu, X. The role of BRCA1 in DNA damage response. Protein Cell 1, 117–123, https://doi.org/10.1007/s13238-010- 0010-5 (2010). h Ch ll l h l d d A l C ll ( ) 5. Moynahan, M. E., Chiu, J. W., Koller, B. H. & Jasin, M. Brca1 controls homology-directed DNA repair. Mol Cell 4, 511–518 (1999). y , , , J , , J , gy p , ( ) 6. Deng, C. X. & Wang, R. H. Roles of BRCA1 in DNA damage repair: a link between development and cancer. Hum Mol Genet 12 Spec No 1, R113–123 (2003). 7. Bochar, D. A. et al. BRCA1 is associated with a human SWI/SNF-related complex: linking chromatin remodeling to breast cancer. Cell 102, 257–265 (2000). 8. Miyake, T., Hu, Y. F., Yu, D. S. & Li, R. A functional comparison of BRCA1 C-terminal domains in transcription activation and chromatin remodeling. J Biol Chem 275, 40169–40173, https://doi.org/10.1074/jbc.M007138200 (2000).h 9. Mullan, P. B., Quinn, J. E. & Harkin, D. P. The role of BRCA 9. Mullan, P. B., Quinn, J. E. & Harkin, D. P. The role of BRCA1 in transcriptional regulation and cell cycle control. Oncogene 25, 5854–5863, https://doi.org/10.1038/sj.onc.1209872 (2006). h 5854–5863, https://doi.org/10.1038/sj.onc.1209872 (2006). p g j 10. Liu, S. et al. BRCA1 regulates human mammary stem/progenitor cell fate. Proc. Natl. Acad. Sci. USA 105, 1680–1685, https://doi. y Data presented in this manuscript will be made available upon request. y Data presented in this manuscript will be made available upon request. Methods C ll li Briefly, tumor cells were plated at 600 cells/well for MDA-MB-231 cells and at 1000 cells/well for SK-N-BE(2)C cells in 6-well plates and incubated for 10 to 14 days. Colonies were stained with Crystal Violet. Plating Efficiency (PE) = lnum- bers of colonies (≥50 cells/colony) divided by numbers of cells plated × 100%. Western blot analysis. Whole cell lysates were prepared in RIPA buffer. Proteins were separated by SDS-PAGE under reducing conditions. BRCA1 protein was detected using mouse monoclonal anti-BRCA1 (SC- 6954, 1:100, Santa Cruz Biotechnology). Protein bands were visualized using ECL substrates (BioRad, #170-5061) and imaged on Kodak X-OMAT 2000A. Real-time quantitative reverse transcription PCR (RT-qPCR). Total cellular RNA was isolated with the TRizol reagent (Invitrogen). First-strand cDNA was synthesized using High Capacity cDNA Reverse Transcription Kit (Applied Biosystems). Levels of gene expression were performed by qRT-PCR on StepOne Plus (Applied Biosystems) using iTaq Universal SYBR Green Supermix (Bio-Rad) under the following conditions: initiation at 95 °C × 30 seconds, 40 cycles at 95 °C × 15 seconds, and 60 °C × 60 seconds. The house keeping gene HPRT was used as a control for normalization. Specificity of the primers (Table 1) was confirmed by a single peak on the dissociation curve. 8 Scientific Reports \| (2019) 9:9702 \| https://doi.org/10.1038/s41598-019-46210-y www.nature.com/scientificreports/ www.nature.com/scientificreports/ 21. Zhang, C. et al. Hypoxia induces the breast cancer stem cell phenotype by HIF-dependent and ALKBH5-mediated m(6) A-demethylation of NANOG mRNA. Proc. Natl. Acad. Sci. USA 113, E2047–2056, https://doi.org/10.1073/pnas.1602883113 (2016). 22 Zh C t l H i i d ibl f l l i f i b ZNF217 d ALKBH5 di d d l i y p g p 22. Zhang, C. et al. Hypoxia-inducible factors regulate pluripotency factor expression by ZNF217- and ALKBH5-mediated modul of RNA methylation in breast cancer cells. Oncotarget 7, 64527–64542, https://doi.org/10.18632/oncotarget.11743 (2016). y g p g g 23. Zhang, H. et al. HIF-1 regulates CD47 expression in breast cancer cells to promote evasion of phagocytosis and maintenance of cancer stem cells. Proc. Natl. Acad. Sci. USA 112, E6215–6223, https://doi.org/10.1073/pnas.1520032112 (2015). y g p g g 23. Zhang, H. et al. HIF-1 regulates CD47 expression in breast cancer cells to promote evasion of phagocytosis and maintenan 23. Zhang, H. et al. HIF 1 regulates CD47 expression in breast cancer cells to promote evasion of phagocytosis and maintenan cancer stem cells. Proc. Natl. Acad. Sci. USA 112, E6215–6223, https://doi.org/10.1073/pnas.1520032112 (2015). l l b ll h f h l d p g p 24. Conley, S. J. et al. Antiangiogenic agents increase breast cancer stem cells via the generation of tum USA 109, 2784–2789, https://doi.org/10.1073/pnas.1018866109 (2012).h 4. Conley, S. J. et al. Antiangiogenic agents increase breast cancer stem cells via the generation of tumor hypoxia. Proc. Natl. Acad. Sci USA 109, 2784–2789, https://doi.org/10.1073/pnas.1018866109 (2012).h p g p ( ) 5. Kim, H., Lin, Q., Glazer, P. M. & Yun, Z. The hypoxic tumor microenvironment in vivo selects the cancer stem cell fate of breas cancer cells. Breast Cancer Res 20, 16, https://doi.org/10.1186/s13058-018-0944-8 (2018). p g 6. Bindra, R. S. et al. Hypoxia-induced down-regulation of BRCA1 expression by E2Fs. Cancer Res. 65, 11597–11604, https://doi org/10.1158/0008-5472.CAN-05-2119 (2005). org/10.1158/0008-5472.CAN-05-2119 (2005). 7. Lu, Y., Chu, A., Turker, M. S. & Glazer, P. M. Hypoxia-induced epigenetic regulation and silencing of the BRCA1 promoter. Mol. Cell g 7. Lu, Y., Chu, A., Turker, M. S. & Glazer, P. M. Hypoxia-induced epigenetic regulation and silencing of the BRCA1 promoter. Mol. Cell Biol. 31, 3339–3350, https://doi.org/10.1128/MCB.01121-10 (2011). 28. Badve, S. & Nakshatri, H. Breast-cancer stem cells-beyond semantics. Lancet Oncol 13, e43–48, https://doi.org/10.1016/S1470- 2045(11)70191-7 (2012). 9. Ginestier, C. et al. www.nature.com/scientificreports/ Rev. Mol. Cell Biol 9, 285–296, https://doi.org/10.1038/nrm2354 (2008). p g 0. Lin, Q., Lee, Y. J. & Yun, Z. Differentiation arrest by hypoxia. J. Biol. Chem. 281, 30678–30683, https://doi.org/10.1074/jbc C600120200 (2006). 1. Yun, Z. & Lin, Q. Hypoxia and regulation of cancer cell stemness. Adv. Exp. Med. Biol. 772, 41–53, https://doi.org/10.1007/978-1 4614-5915-6_2 (2014). 2. Mohyeldin, A., Garzón-Muvdi, T. & Quiñones-Hinojosa, A. Oxygen in Stem Cell Biology: A Critical Component of the Stem Cel Niche. Cell Stem Cell 7, 150–161, https://doi.org/10.1016/j.stem.2010.07.007 (2010). 3. Falkenberg, K. J. & Johnstone, R. W. Histone deacetylases and their inhibitors in cancer, neurological diseases and immune disorders. Nat Rev Drug Discov 13, 673–691, https://doi.org/10.1038/nrd4360 (2014). g p g 4. West, A. C. & Johnstone, R. W. New and emerging HDAC inhibitors for cancer treatment. J. Clin. Invest. 124, 30–39, https://doi org/10.1172/JCI69738 (2014).h g 5. Chang, S. & Sharan, S. K. The role of epigenetic transcriptional regulation in BRCA1-mediated tumor suppression. Transcription 4 24–28, https://doi.org/10.4161/trns.22600 (2013).hf p g 46. Munster, P. N. et al. The histone deacetylase inhibitor suberoylanilide hydroxamic acid induces differentiation of human breast cancer cells. Cancer Res. 61, 8492–8497 (2001).h ( ) 47. Starita, L. M. & Parvin, J. D. The multiple nuclear functions of BRCA1: transcription, ubiquitination and DNA repair. Curr. Opin. Cell Biol. 15, 345–350 (2003). 48. Rosen, E. M., Fan, S. & Ma, Y. BRCA1 regulation of transcription. Cancer Lett. 236, 175–185, https://doi.org/10.10 canlet.2005.04.037 (2006).f 49. Furuta, S. et al. Depletion of BRCA1 impairs differentiation but enhances proliferation of mammary epithelial cells. Proc. Natl. Acad. Sci. USA 102, 9176–9181, https://doi.org/10.1073/pnas.0503793102 (2005). p g p 50. Bouwman, P. & Jonkers, J. Mouse models for BRCA1 associated tumorigenesis: from fundamental insights to preclinical utility. Cell Cycle 7, 2647–2653, https://doi.org/10.4161/cc.7.17.6266 (2008).h Cycle 7, 2647–2653, https://doi.org/10.4161/cc.7.17.6266 (2008).h y p g 1. Shakya, R. et al. The basal-like mammary carcinomas induced by Brca1 or Bard1 inactivation implicate the BRCA1/BARD1 heterodimer in tumor suppression. Proc. Natl. Acad. Sci. USA 105, 7040–7045, https://doi.org/10.1073/pnas.0711032105 (2008). 52. Kim, Y., Lin, Q., Zelterman, D. & Yun, Z. Hypoxia-regulated delta-like 1 homologue enhances cancer cell stemness and tumorigenicity. Cancer Res. 69, 9271–9280, https://doi.org/10.1158/0008-5472.CAN-09-1605 (2009).f g y p g 53. Lin, Q., Cong, X. & Yun, Z. Differential hypoxic regulation of hypoxia-inducible factors 1alpha and 2alpha. Mol Cancer Res 9, 757–765, https://doi.org/10.1158/1541-7786.MCR-11-0053 (2011). p g 4. Li, Q., Lin, Q. & Yun, Z. Hypoxia-activated cytotoxicity of benznidazole against clonogenic tumor cells. www.nature.com/scientificreports/ Cancer biology & therapy 17 1266–1273, https://doi.org/10.1080/15384047.2016.1250988 (2016). www.nature.com/scientificreports/ ALDH1 is a marker of normal and malignant human mammary stem cells and a predictor of poor clinica outcome. Cell Stem Cell 1, 555–567, https://doi.org/10.1016/j.stem.2007.08.014 (2007).hi 30. Kapucuoglu, N. et al. The clinicopathological and prognostic significance of CD24, CD44, CD133, ALDH1 expressions in invasive ductal carcinoma of the breast: CD44/CD24 expression in breast cancer. Pathol. Res. Pract. 211, 740–747, https://doi.org/10.1016/j. prp.2015.05.011 (2015). p p ( ) 31. Ricardo, S. et al. Breast cancer stem cell markers CD44, CD24 and ALDH1: expression distribution within intrinsic mole subtype. J. Clin. Pathol. 64, 937–946, https://doi.org/10.1136/jcp.2011.090456 (2011). yp p g j p 32. Senbanjo, L. T. & Chellaiah, M. A. CD44: A Multifunctional Cell Surface Adhesion Receptor Is a Regulator of Progression and Metastasis of Cancer Cells. Front Cell Dev Biol 5, 18, https://doi.org/10.3389/fcell.2017.00018 (2017).h p g 33. Yan, Y., Zuo, X. & Wei, D. Concise Review: Emerging Role of CD44 in Cancer Stem Cells: A Promising Biomarker and Therapeutic Target. Stem Cells Transl Med 4, 1033–1043, https://doi.org/10.5966/sctm.2015-0048 (2015). 4. Yarden, R. I. & Brody, L. C. BRCA1 interacts with components of the histone deacetylase complex. Proc. Natl. Acad. Sci. USA 96 4983–4988 (1999). 35. Neumeister, V. M. et al. Hypoxia-induced protein CAIX is associated with somatic loss of BRCA1 protein and pathway activity in triple negative breast cancer. Breast Cancer Res. Treat. 136, 67–75, https://doi.org/10.1007/s10549-012-2232-0 (2012).hi 36. Dhani, N., Fyles, A., Hedley, D. & Milosevic, M. The clinical significance of hypoxia in human cancers. Semin. Nucl. Med. 45, 110–121, https://doi.org/10.1053/j.semnuclmed.2014.11.002 (2015).i h 110–121, https://doi.org/10.1053/j.semnuclmed.2014.11.002 (201 p g j 7. Vaupel, P. & Mayer, A. Hypoxia in cancer: significance and impact on clinical outcome. Cancer Metastasis Rev. 26, 225–239, https:/ doi.org/10.1007/s10555-007-9055-1 (2007). 37. Vaupel, P. & Mayer, A. Hypoxia in cancer: significance and impact on clinical outcome. Cancer Metastasis Rev. 26, 225–239, ht doi.org/10.1007/s10555-007-9055-1 (2007). 38 Hockel M et al Association between tumor hypoxia and malignant progression in advanced cancer of the uterine cervix Ca p , y , yp gi p , , p doi.org/10.1007/s10555-007-9055-1 (2007). 8. Hockel, M. et al. Association between tumor hypoxia and malignant progression in advanced cancer of the uterine cervix. Cance g 8. Hockel, M. et al. Association between tumor hypoxia and malignant progression in advanced cancer of the uterine cervix. Cance Res. 56, 4509–4515 (1996).h ( ) 9. Simon, M. C. & Keith, B. The role of oxygen availability in embryonic development and stem cell function. Nat. Acknowledgementsh g The authors thank Dr. Peter Glazer for reagents and helpful discussions. This work is in part supported by an NIH grant (R01 CA178254 to ZY). References org/10.1073/pnas.0711613105 (2008).f g p 1. Kubista, M., Rosner, M., Kubista, E., Bernaschek, G. & Hengstschlager, M. Brca1 regulates in vitro differentiation of mammary epithelial cells. Oncogene 21, 4747–4756, https://doi.org/10.1038/sj.onc.1205580 (2002). p g p g j 2. Borovski, T., De Sousa, E. M. F., Vermeulen, L. & Medema, J. P. Cancer stem cell niche: the place to be. Cancer Res 71, 634–639 https://doi.org/10.1158/0008-5472.CAN-10-3220 (2011). p g 13. Filatova, A., Acker, T. & Garvalov, B. K. The cancer stem cell niche(s): the crosstalk between glioma stem cells and their microenvironment. Biochim Biophys Acta 1830, 2496–2508, https://doi.org/10.1016/j.bbagen.2012.10.008 (2013). p y p g 14. Nie, D. Cancer stem cell and niche. Front Biosci (Schol Ed.) 2, (184–193 (2010). p y p g j g 4. Nie, D. Cancer stem cell and niche. Front Biosci (Schol Ed.) 2, (184–193 (2010). 5. Ungefroren, H., Sebens, S., Seidl, D., Lehnert, H. & Hass, R. Interaction of tumor cells with the microenvironment. Cell Commun Signal 9, 18, https://doi.org/10.1186/1478-811x-9-18 (2011). g p g 6. Lin, Q. & Yun, Z. Impact of the hypoxic tumor microenvironment on the regulation of cancer stem cell characteristics. Cance biology & therapy 9, 949–956, https://doi.org/10.4161/cbt.9.12.12347 (2010). gy py p g 17. Carcereri de Prati, A. et al. Metastatic Breast Cancer Cells Enter Into Dormant State and Express Cancer Stem Cells Phenotype Under Chronic Hypoxia. J. Cell. Biochem. 118, 3237–3248, https://doi.org/10.1002/jcb.25972 (2017). yp p g j ( ) 8. Iriondo, O. et al. Distinct breast cancer stem/progenitor cell populations require either HIF1alpha or loss of PHD3 to expand under hypoxic conditions. Oncotarget 6, 31721–31739, https://doi.org/10.18632/oncotarget.5564 (2015). yp g p g g 19. Shiraishi, A. et al. Hypoxia promotes the phenotypic change of aldehyde dehydrogenase activity of breast cancer stem cells. Cancer Sci 108, 362–372, https://doi.org/10.1111/cas.13147 (2017). , , p g ( ) 20. Xiang, L. et al. Hypoxia-inducible factor 1 mediates TAZ expression and nuclear localization to induce the breast cancer stem cell phenotype. Oncotarget 5, 12509–12527, https://doi.org/10.18632/oncotarget.2997 (2014). Scientific Reports \| (2019) 9:9702 \| https://doi.org/10.1038/s41598-019-46210-y 9 9 Author Contributions H.K., Q.L. and Z.Y. conceived and designed the study, and acquired data. All authors contributed to data analysis and manuscript preparation. Scientific Reports \| (2019) 9:9702 \| https://doi.org/10.1038/s41598-019-46210-y 10 www.nature.com/scientificreports/ Additional Information Supplementary information accompanies this paper at https://doi.org/10.1038/s41598-019-46210-y. Competing Interests: The authors declare no competing interests. Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2019 Additional Information Additional Information Supplementary information accompanies this paper at https://doi.org/10.1038/s41598-019-46210-y. dd t o a o at o Supplementary information accompanies this paper at https://doi.org/10.1038/s41598-019-46210-y. Competing Interests: The authors declare no competing interests. Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and i i i l ffili i Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps an institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2019 Scientific Reports \| (2019) 9:9702 \| https://doi.org/10.1038/s41598-019-46210-y 11
https://openalex.org/W4310910018	https://pure.eur.nl/files/78005195/Deep_reinforcement_learning_for_cerebral_anterior_vessel_tree_extraction_from_3D_CTA_images.pdf	English	null	Deep reinforcement learning for cerebral anterior vessel tree extraction from 3D CTA images	Medical image analysis	2,023	cc-by	16,138	Jiahang Su a,∗, Shuai Li a, Lennard Wolff a, Wim van Zwam b, Wiro J. Niessen a,c, Aad van der Lugt a, Theo van Walsum a Jiahang Su a,∗, Shuai Li a, Lennard Wolff a, Wim van Zwam b, Wiro J. Niessen a,c, Aad van der Lugt a, Theo van Walsum a Jiahang Su a,∗, Shuai Li a, Lennard Wolff a, Wim van Zwam b, Wiro J. Niessen a,c Aad van der Lugt a, Theo van Walsum a a Department of Radiology & Nuclear Medicine, Erasmus MC, University Medical Center Rotterdam, The Netherlands b Department of Radiology & Nuclear Medicine, Maastricht UMC, Cardiovascular Research Institute Maastricht, The Netherlands c Faculty of Applied Sciences, Delft University of Technology, The Netherlands ∗Corresponding author. E-mail address: j.su@erasmusmc.nl (J. Su). A R T I C L E I N F O A R T I C L E I N F O MSC: 41A05 41A10 65D05 65D17 Keywords: Deep reinforcement learning Tracking Brain vessel Bifurcation detection 3D CTA CNN Extracting the cerebral anterior vessel tree of patients with an intracranial large vessel occlusion (LVO) is relevant to investigate potential biomarkers that can contribute to treatment decision making. The purpose of our work is to develop a method that can achieve this from routinely acquired computed tomography angiography (CTA) and computed tomography perfusion (CTP) images. To this end, we regard the anterior vessel tree as a set of bifurcations and connected centerlines. The method consists of a proximal policy optimization (PPO) based deep reinforcement learning (DRL) approach for tracking centerlines, a convolutional neural network based bifurcation detector, and a breadth-first vessel tree construction approach taking the tracking and bifurcation detection results as input. We experimentally determine the added values of various components of the tracker. Both DRL vessel tracking and CNN bifurcation detection were assessed in a cross validation experiment using 115 subjects. The anterior vessel tree formation was evaluated on an independent test set of 25 subjects, and compared to interobserver variation on a small subset of images. The DRL tracking result achieves a median overlapping rate until the first error (1.8 mm off the reference standard) of 100, [46, 100] % on 8032 vessels over 115 subjects. The bifurcation detector reaches an average recall and precision of 76% and 87% respectively during the vessel tree formation process. The final vessel tree formation achieves a median recall of 68% and precision of 70%, which is in line with the interobserver agreement. 1. Introduction commonly used to visualize cerebral vasculature. In this manuscript, we are focusing on CT-based imaging, which is a common modality for the workup of stroke patients. CT imaging techniques for cerebral vessels in clinical practice are CTA (both single phase and multiphase) and computed tomography perfusion (CTP) (Potter et al., 2019). CTP is an imaging protocol where a series of 3D CTA images is acquired. 1.1. Clinical background Medical Image Analysis 84 (2023) 102724 Medical Image Analysis 84 (2023) 102724 1.1. Clinical background The cerebral vessel network is a complex network that feeds the brain tissue. Diseases affecting the cerebral vessel network, such as an intracranial ischemic stroke caused by a large vessel occlusion, may have severe consequences. In patients with an intracranial large vessel occlusion (LVO), the occlusion often occurs in the anterior circulation, which consists of the middle cerebral artery (MCA) vessel tree and anterior cerebral artery (ACA) vessel tree. Knowing the precise anterior vessel tree with its topology could provide additional information for the treatment planning of LVO and potentially contribute to the treat- ment decision making process. The complete cerebral vessel centerline annotation with anterior vessel trees is shown in Fig. 1. The purpose of our work is to extract the anterior vessel trees as distal as possible given the initial direction vectors of each tree. Available online 9 December 2022 1361-8415/© 2022 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). https://doi.org/10.1016/j.media.2022.102724 Received 17 June 2022; Received in revised form 24 November 2022; Accepted 2 December 2022 Available online 9 December 2022 1361-8415/© 2022 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license ( https://doi.org/10.1016/j.media.2022.102724 Received 17 June 2022; Received in revised form 24 November 2022; Accepted 2 December 2022 Available online 9 December 2022 1361-8415/© 2022 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Received 17 June 2022; Received in revised form 24 November 2022; Accepted 2 December 2022 1.2. Related work Traditionally, vessel tree extraction starts with obtaining a vessel segmentation or geometric model (Damseh et al., 2020). Extensive reviews on vessel segmentation and vessel shape extraction methods have been presented by Kirbas and Quek (2004), Lesage et al. (2009), In clinical practice, 3D modalities such as magnetic resonance an- giography (MRA) and computed tomography angiography (CTA) are Medical Image Analysis 84 (2023) 102724 J. Su et al. Nomenclature 2D Two dimensional 3D Three dimensional A1 Proximal segment of Anterior Cerebral Artery tree A2C Advantage Actor Critic ACA Anterior Cerebral Artery Bf Bifurcation CCS Curve to Curve Similarity CNN Convolutional Neural Network CT Computed Tomography CTA Computed Tomography Angiography CTP Computed Tomography Perfusion CTT Correct Tree Topology DQN Deep Q-network DRL Deep Reinforcement Learning GAE Generalized Advantage Estimation HU Hounsfield Units IQR Interquartile Range JS Jiahang Su LSTM Long Short-Term Memory LVO Large Vessel Occlusion M Medical student M1 Proximal segment of Middle Cerebral Artery tree MCA Middle Cerebral Artery MDP Markov Decision Process MRA Magnetic Resonance Angiography OR Overlap Rate PPO Proximal Policy Optimization PReLU Parametric Rectified Linear Unit RCNN Region-based Convolutional Neural Net- works ReLU Rectified Linear Unit RNN Recurrent Neural Network SLO Scanning Laser Ophthalmoscopy TRPO Trust Region Policy Optimization orientation and radius from cardiac CT angiography, Guo et al. (2019) use a multi-task CNN method to obtain an estimate of a centerline distance map and an endpoint confidence map from coronary CTA images and subsequently a minimal path method is used to compute the centerline. Reinforcement learning is a branch of machine learning algorithms that have been studied in domains such as gaming and control. Deep reinforcement learning (DRL), which is the fusion of reinforcement learning and deep learning, has been shown to be able to solve complex image processing challenges by exploiting image features (Arulku- maran et al., 2017). Such DRL based methods are gradually entering the medical imaging field (Zhou et al., 2021). Tracking is a preferred application direction for DRL approaches, as DRL learns the optimal policy by maximizing the accumulated reward over time, which maps nicely to the problem of finding a path through an nD image. The optimal policy determines the best actions for an agent in each state over time. There are two main streams in DRL methods, value-based and policy gradient based. In value-based methods e.g. 1.2.2. Bifurcation detection Bifurcation detection plays an important role in determining the vessel tree. Some feature based machine learning methods have been presented in the past. For example, Baboiu and Hamarneh (2012) utilized the scale-space features of vessel structures to construct a bifurcation detector on 2D and 3D synthetic images. Kalaie and Gooya (2017) used Gaussian profiles in cross sections of bifurcation points to discriminate vessels from bifurcation points in 2D retinal images. Jeon (2021) performs bifurcation detection using intensity based spatial clustering methods for 3D coronary arteries extraction from CTA im- ages. Recently, deep learning methods have been presented by Zheng et al. (2015), who utilized a two-stage CNN approach to detect carotid artery bifurcations from 3D CTA images, and Hervella et al. (2020), who used a multi-instance heat map with a U-net architecture Ron- neberger et al. (2015) to estimate vessel crossings and bifurcation points from 2D retina images. The latter application was also addressed by Zhao et al. (2020) utilizing a two-stage refinement approach, in which an RCNN (Girshick et al., 2014) based network first obtained a rough estimation of bifurcations and in a 2nd stage, the same network architecture was applied to address the errors from the first network. 1.2. Related work deep Q network (Mnih et al., 2013) the policy 𝜋is directly obtained by approximating the state–action value 𝑄𝜃(𝑠, 𝑎) as a function of state 𝑠and action 𝑎or state value 𝑉𝜃(𝑠) as a function of a state 𝑠. Here, 𝜃represents the network parameters. For example, Zhang et al. (2018) apply a deep Q network (Mnih et al., 2013) (DQN) approach with a discrete action space given by a six-connected neighborhood as action space and a point-to-curve reward function for thoracic aorta centerline tracking in 3D contrast-enhanced and none contrast-enhanced CT images. Zhang et al. (2020) similarly utilizes a double-DQN (Van Hasselt et al., 2016) with a discrete 26-connected neighborhood as the action space to track the coronary centerline. In this work, the bifurcations were detected using a standalone detector. Li et al. (2021) utilize the DQN approach with predefined orientation as discrete action space to track coronary centerlines. In addition, the value-based methods are known to often lead to an overestimation of the action value (Van Hasselt et al., 2016). In policy gradient methods, a parameterized policy is learned di- rectly using a stochastic optimization algorithm (Sutton and Barto, 2018). Trust region policy optimization (TRPO) (Schulman et al., 2015a) and proximal policy optimization (PPO) (Schulman et al., 2017) are two well-known methods for such optimization. Dai et al. (2019) utilize PPO for learning the policy for neuronal tracking in 2D neuronal microscopy images. The above-mentioned DRL-based tracking approach has demonstrated superior performance in comparison with a benchmark using CNN based methods. and Moccia et al. (2018). After obtaining a vessel segmentation from the input image, various methods can be used to obtain the vessel tree. Lee et al. (1994) used a thinning algorithm, Moriconi et al. (2018) used geodesics in combination with a minimum spanning tree, Robben et al. (2016) used a graph-based tracking approach to obtain the Circle of Willis and the proximal segment of each arterial trees from 3D MRA images and Zhang et al. (2021) utilize the general confluence constraint with minimal arborescence on a directed graph. Others build the tree in a more explicit way from a set of bifurcations and connected centerlines, an approach followed by Zhang et al. (2020) and Jeon (2021). In these approaches, vessel tracking and bifurcation detection are two essential ingredients. 2.1.1. State and action The agent at time 𝑡has an associated state, denoted as 𝑠𝑡. Next to the current position in the image, the agent state contains image information from previous steps and the corresponding displacement vector array. Let 𝑝be any discrete 3D position in a CTA volume. The image information in the state consists of three 21 × 21 × 21 sub-volumes centered at respectively the agent position at 𝑝𝑡and two previous agent positions 𝑝𝑡−1, 𝑝𝑡−2. This size is sufficient to capture the local neighborhood, even for large vessels and bifurcations. In addition, the state contains a sub-volume of the binary agent path centered at 𝑝𝑡, which represents the voxels the agent went through. As a vessel can be viewed as a tubular structure with varying radius, any independent point in the track through a vessel can be tracked in two directions. Using the image information from the three previous steps and the binary agent path can help the agent track the vessel in a directed way. The dynamic displacement vector array at time 𝑡consists of the directional vectors (steps of the agent) from the current position till the initial position [𝑎𝑡, 𝑎𝑡−1, … , 𝑎0]. The dynamic displacement vector provides additional directional information on previous steps. Our aim is to extract the anterior vessel tree from 3D CT images, independent of the acquisition protocol. For such a task, to the best of our knowledge, most vessel tree extraction methods and the subsequent tracking and bifurcation steps have focused on coronary artery tree extraction from 3D cardiac CTA images and vessel tree extraction from 2D retina scanning laser ophthalmoscope (SLO) images. None of the existing works presents a method to extract the cerebral anterior vessel tree from 3D CTA images or methods that perform tree extraction beyond the proximal anterior vessel tree. Our work has several contributions. First, we assess a DRL approach that, to the best of our knowledge, has not been applied yet for cerebral vessel tracking approaches. Second, the DRL method was adapted to this application. More specifically, we introduce a novel curve-to-curve distance based reward function and a network architecture (CNN with recurrent neural network (RNN)) for cerebral vessel tracking, and we assess the impact of various features of the methods in an ablation study. In addition, we performed a comparison with an existing base- line method. 1.2.1. Tracking In conventional methods, vessel segmentation approaches start with centerline tracking. Many tracking methods have been described in the previously mentioned reviews. Nowadays, deep learning based methods have replaced conventional approaches for many image pro- cessing problems. Also, combinations of convolutional neural net- works (CNNs) and conventional approaches have become popular. For example, Wolterink et al. (2019) extract the coronary artery centerline with an iterative tracking approach utilizing a 3D CNN to estimate the 2 J. Su et al. Medical Image Analysis 84 (2023) 102724 Fig. 1. Example of brain vessel annotation with colored anterior vessel trees for a subject with a large vessel occlusion in the left M2 segment. White represents veins and posterior cerebral artery trees, red is the right MCA tree, yellow is the right ACA tree, green is the left ACA tree, blue is the left MCA tree, a: the right sagittal view; b: the coronal view; c: the left sagittal view. Fig. 1. Example of brain vessel annotation with colored anterior vessel trees for a subject with a large vessel occlusion in the left M2 segment. White represents veins and posterior cerebral artery trees, red is the right MCA tree, yellow is the right ACA tree, green is the left ACA tree, blue is the left MCA tree, a: the right sagittal view; b: the coronal view; c: the left sagittal view. 2.1.1. State and action Finally, all experiments have been performed with a large set of clinical data. The agent discrete action space is defined as the 26 connected neighboring voxels of the current position 𝑝𝑡in an agent path. The remainder of this manuscript is organized as follows. In Sec- tion 2, we first present a DRL PPO based directed vessel centerline tracking model, followed by a CNN-based bifurcation detection, and breadth-first tree formation methods. The data set, data annotations, and prepossessing are described in Section 3. The implementation, hyperparameter optimizations and experimental results can be found in Section 4, followed by discussion in Section 5 and conclusions in Section 6. We represent the set of actions by scaled displacement vectors as follows: 𝛥𝑝𝑡= 𝛼× {𝑥, 𝑦, 𝑧\| 𝑥, 𝑦, 𝑧∈{−1, 0, 1}}, (1) (1) As a consequence, an agent action 𝑎𝑡can make a step of at least two voxels per episode step if 𝛼= 2. 2. Method The reward function is an essential component for the network to achieve the optimal policy. The ultimate goal of the vessel tracking training task is to learn the agent tracking the vessel centerline by maximizing 𝑅, the accumulated discounted instant reward 𝑟𝑡: The proposed cerebral anterior vessel tree extraction method starts with a DRL centerline tracking approach starting from the root of the tree (with an initial direction vector). Along the tracked path (agent path), a CNN based bifurcation detector is used to identify the bifurcation points. New tracks are generated from the bifurcation points detected along the tracks. In the tree formation, stopping criteria are applied to remove the spurious parts of the tracked paths, and a breadth-first approach is used to construct the tree from the tracked paths and the bifurcation points. 𝑅= 𝑇−1 ∑ 𝑡=0 𝛾𝑡𝑟𝑡+1 (2) 𝑅= 𝑇−1 ∑ 𝑡=0 𝛾𝑡𝑟𝑡+1 (2) The instant reward 𝑟𝑡is used to measure the policy performance of the agent at time 𝑡, and thus is relevant for policy optimization. We would like the agent to follow the reference path 𝐺𝑡, which in our case is a voxelized representation of the annotated vessel track. 2.1.3. Policy In our vessel tracking application, the optimal policy is obtained us- ing the PPO method with the advantage actor critic (A2C) (Mnih et al., 2016) framework. The A2C architecture consists of two networks: an actor network and a critic network. The critic network 𝑉𝑣(𝑠) estimates the state value and actor network 𝑄𝑤(𝑠, 𝑎) learns the state–action value suggested by the critic. The advantage function 𝐴(𝑠𝑡, 𝑎𝑡) (Eq. (4)) is introduced to measure the state action pair at time 𝑡: The architecture of the actor and critic networks are similar, they both consist of a CNN and an RNN. The CNN model is similar to the architecture proposed by Wolterink et al. (2019) but with a parametric rectified linear activation function (PReLU) (He et al., 2015) instead of a rectified linear activation function (ReLU) (Maas et al., 2013). The proposed CNN architecture is based on a dilated network (Yu and Koltun, 2015) architecture. Each CNN block consists of a series of dilated convolutions, with dilation kernels of size 1. For each convo- lutional layer, instance normalization was applied. The detailed CNN architecture of the actor and critic can be found in Table 1. The critic CNN takes a ternary ground truth sub-volume centered at 𝑝𝑡as additional input. The dimension of this sub-volume is 21 × 21 × 21 voxels, and each voxel has a value of −1, 0, or 1: the paired ground truth centerline for the current episode has value 1, other centerlines are labeled as −1, and the remaining voxels have value 0. This setup ensures a decrease in instant reward if the agent moves to the wrong vessel. (4) 𝐴(𝑠𝑡, 𝑎𝑡) = 𝑄𝑤(𝑠𝑡, 𝑎𝑡) −𝑉𝑣(𝑠𝑡) . (4) It determines how much better a selected action is compared to the expected value of all possible actions. In this application, generalized advantage estimation (GAE) (Schulman et al., 2015b) is applied to further reduce the variance of the temporal difference error of the advantage function. In policy gradient DRL methods, stochastic optimization algorithms such as gradient ascent are sensitive to the gradient update step size. To accommodate for the step size issue, trust region policy optimization (TRPO) (Schulman et al., 2015a) was introduced using KL-divergence (Kullback and Leibler, 1951) constraints to stabilize the gradient update step between old and new policies. However, the use of KL-divergence yields costly computations on multiple Hessian-vector products (Engstrom et al., 2020). 2.1. Directed vessel tracking We, therefore, use the difference of the curve-to-curve distance between the agent path and reference standard between time 𝑡and 𝑡−1 to measure the reward of the corresponding action. The curve- to-curve distance was introduced in Walsum et al. (2008) to quantify the distance between coronary centerlines. The curve-to-curve distance is computed by approximating the surface area spanned between two A reinforcement learning problem can be modeled as a Markov De- cision Process (MDP) (Bertsekas et al., 2011). We, therefore, introduce the key ingredients of our DRL tracker as the components of an MDP. The environment in our case is the normalized 3D CTA image. The other components, such as state and action, are introduced below. 3 J. Su et al. Medical Image Analysis 84 (2023) 102724 Fig. 2. Example of curve-to-curve distance and CCS metric, the red dotted line is the reference standard and the blue dotted line is the actual track, the summation of lengths of the red lines between the reference standard and track is the computed surface distance; a: the anterior vessel tree with highlighted color-coded vessel centerline (corresponding to the colored frame). b: example of 𝐿0. c: example of success tracked case, CCS = 0.87, OR = 1. d: example track at time 𝑡, CCS = 0.58, OR = 0.5. e: example of miss tracked case, CCS = 0.58, OR = 0.09. Fig. 2. Example of curve-to-curve distance and CCS metric, the red dotted line is the reference standard and the blue dotted line is the actual track, the summation of lengths of the red lines between the reference standard and track is the computed surface distance; a: the anterior vessel tree with highlighted color-coded vessel centerline (corresponding to the colored frame). b: example of 𝐿0. c: example of success tracked case, CCS = 0.87, OR = 1. d: example track at time 𝑡, CCS = 0.58, OR = 0.5. e: example of miss tracked case, CCS = 0.58, OR = 0.09. constraints by a simple but efficient regularization mechanism in objective function as follows: curves via the summation of the Euclidean length of all corresponding point-to-point connections. The optimal (minimal area) correspondence between two curves was determined via a Dijkstra minimum cost algorithm (Dijkstra et al., 1959). Fig. 2 shows examples of surface area computation (the summation of red lines) in various scenarios. 2.1. Directed vessel tracking (5) 𝐽(𝜃) = E[min(𝜌(𝑡)𝐴(𝑠𝑡, 𝑎𝑡), clip(𝜌(𝑡), 1 −𝜀, 1 + 𝜀)𝐴(𝑠𝑡, 𝑎𝑡))] where 𝜌(𝑡) is defined as the likelihood ratio between policies at two time points: In addition, we would like to enforce the agent to follow the refer- ence standard 𝐺𝑡. Therefore, a binary overlap metric, which measures the overlap between the reference path 𝐺𝑡and the agent path, is part of the instant reward. 𝜌(𝑡) = 𝜋𝑡(𝑠𝑡, 𝑎𝑡) 𝜋𝑡−1(𝑠𝑡−1, 𝑎𝑡−1) . (6) 𝜌(𝑡) = 𝜋𝑡(𝑠𝑡, 𝑎𝑡) 𝜋𝑡−1(𝑠𝑡−1, 𝑎𝑡−1) . (6) We propose an instant reward function that combines binary over- lap and the curve-to-curve distance. The binary overlap metric provides extra positive (+1) feedback when the agent path and ground truth are overlapping at position 𝑝𝑡. Denoting the surface distance in position 𝑝𝑡 as 𝐿𝑡, and the binary overlap at position 𝑝𝑡as 𝐵𝑡, the proposed instant reward is defined as follows: 2.1.4. Network architecture The proposed network architecture is shown in Fig. 3. The archi- tecture consists of an actor network and a critic network. The critic network helps the agent (actor) with learning the optimal state–action value during the training process. Assuming that the policy is optimal after training, only actor network is required for obtaining the tracks in the inference stage. 𝑟𝑡= { 𝐵𝑡−log(𝜖+ \|𝐿𝑡−𝐿𝑡−1\|), if 𝐿𝑡−𝐿𝑡−1 < 0; 𝐵𝑡+ log(𝜖+ \|𝐿𝑡−𝐿𝑡−1\|), otherwise. (3) (3) The agent state consists of sub-volumes sampled from three consec- utive points along the agent path and one sub-volume containing the binary agent path. In the CNN part, the three sub-volumes generally cannot cover the whole path of the agent. A recurrent neural net- work (RNN) (Hochreiter and Schmidhuber, 1997) was therefore used to provide extra temporal information along the agent path. 2.3. Tree formation point 𝑡consists of the paired instant rewards (𝑟𝑡), the log of instant reward (𝑙𝑜𝑔(𝑟𝑡)), the intensity value of agent path (𝐼𝑡) and the mean intensity value of a two voxel wide region around the agent path (𝐼𝑡): The anterior vessel tree can be viewed as an arborescence with a given root. We apply a breadth-first tree formation for the anterior vessel tree extraction. The tracking agent and bifurcation detector described in the previous paragraph can be applied to obtain a series of tracks and bifurcation points. Starting from the root, we obtain the first track. Subsequently, we prune the tracked path based on stopping criteria, and recursively and in a breadth-first manner, find bifurcations along this pruned tracked path, and start new tracks from these bifurcations. In this latter step, new paths are tracked from the bifurcation point in several directions, and then failed or overlapping tracks are removed. 𝑥= ⎛ ⎜ ⎜ ⎜ ⎜⎝ 𝑟𝑡 𝑙𝑜𝑔(𝑟𝑡) 𝐼𝑡 𝐼𝑡 𝑟𝑡−1 𝑙𝑜𝑔(𝑟𝑡−1) 𝐼𝑡−1 𝐼𝑡−1 ⋮ ⋮ ⋮ ⋮ 𝑟0 𝑙𝑜𝑔(𝑟0) 𝐼0 𝐼0 ⎞ ⎟ ⎟ ⎟ ⎟⎠ (7) (7) The outputs of the CNN and RNN are concatenated and fed into a fully connected layer. The final output of the actor network is a (log) softmax layer, which determines the direction of the agent in the next state. The final output of the critic network is a scalar value, which approximates the value of the input state. In an attempt to get a more robust approach, increasing the true positive tracks, while reducing the missed ones and false positives, we also introduce an ensemble method with five different vessel tracking models. The final fully connected layer in the actor network is absent in the critic network. As the latter network needs to learn the state value, the final fully connected layer is not necessary. 2.1.3. Policy The actor’s input is 21 × 21 × 21 × 4 and the critic’s input is 21 × 21 × 21 × 5. Each convolutional layer is followed by an instance normalization and PReLU. Layer 1 2 3 4 5 6 7 Kernel width 3 3 3 3 1 1 7 Dilation rate 1 1 2 3 4 1 1 Channels 32 32 32 32 64 64 26 2.1.3. Policy The PPO method (Schulman et al., 2017) which was later developed to approximate the KL-divergence The RNN in both the actor and critic network consists of two LSTM cells. The input of the RNN in the actor network is the action history vector; the input vector 𝑥of the RNN in the critic network at time 4 J. Su et al. Medical Image Analysis 84 (2023) 102724 J. Su et al. Fig. 3. The architecture of the advantage actor critic (A2C) network. Both actor and critic networks consist of CNN and an RNN. The RNN part consists of two LSTM cells. The input of the actor CNN (A) consists of three CTA cubes and one binary agent path cube (21 × 21 × 21 voxels). The three CTA cubes are centered at the current position 𝑝𝑡and two previous positions 𝑝𝑡−1, 𝑝𝑡−2. The binary agent path cube is centered in the current position 𝑝𝑡. The CNN input of the critic network (𝐴⊕𝐵) consists of A and a ternary ground truth cube (B) that is centered at the current position 𝑝𝑡. The RNN input of the critic network is a vector array (x) that consists of paired local features along the track. The RNN input of the actor network is a vector [𝑎𝑡, 𝑎𝑡−1, … , 𝑎1, 𝑎0] that contains the action history. Medical Image Analysis 84 (2023) 102724 Fig. 3. The architecture of the advantage actor critic (A2C) network. Both actor and critic networks consist of CNN and an RNN. The RNN part consists of two LSTM cells. The input of the actor CNN (A) consists of three CTA cubes and one binary agent path cube (21 × 21 × 21 voxels). The three CTA cubes are centered at the current position 𝑝𝑡and two previous positions 𝑝𝑡−1, 𝑝𝑡−2. The binary agent path cube is centered in the current position 𝑝𝑡. The CNN input of the critic network (𝐴⊕𝐵) consists of A and a ternary ground truth cube (B) that is centered at the current position 𝑝𝑡. The RNN input of the critic network is a vector array (x) that consists of paired local features along the track. The RNN input of the actor network is a vector [𝑎𝑡, 𝑎𝑡−1, … , 𝑎1, 𝑎0] that contains the action history. The network architecture of actor/critic CNN. 3. Data Application of the bifurcation detection model (Section 2.2) for all the voxels of a tracked path yields a probability of a bifurcation being present at each voxel of the track. The detector may already give some probability to voxels that are close to a bifurcation, The probabilities in voxels from near a bifurcation to the bifurcation increase, and have a peak at the bifurcation. We, therefore, extract the bifurcation points by taking the max probability voxel from a consecutive series of voxels that consists of at least two voxels with a probability larger than 0.5. The threshold allows for smaller bifurcations to be detected. 2.2. Bifurcation detection Each convolutional layer is followed by a switchable normalization, a dropout rate of 0.2, and a PReLU activation function. Layer 1 2 3 4 5 6 7 Kernel width 3 3 3 3 1 1 7 Dilation rate 1 1 2 3 4 1 1 Channels 32 32 32 32 64 64 26 Table 2 The network architecture of bifurcation CNN. The input size is 21×21×21. Each convolutional layer is followed by a switchable normalization, a dropout rate of 0.2, and a PReLU activation function. Fig. 4. Processing of the anterior vessel trees data preparation; a. shows the example of the anterior vessel tree structure, the MCA tree is presented in blue and ACA trees are presented in green; b. shows the anterior vessel trees and thresholded (0.5) MCA and ACA mask, the green mask is the MCA mask, the red mask is the ACA mask; c. shows the segment labels. The proximal segments are in red, the middle segments are in blue, the distal segments are in green and the skull segments are in yellow. 2.3.4. Ensemble method In addition to tracking with one model, the combined policy and models ensemble method aims to further improve the performance by combining results from multiple tracking models in two different ways. The first track of each tree is essential for tree formation. To ensure a good initial track, we use majority voting for the first track: at each tracking step, we initiate five agents to generate the candidate actions and select the action (i.e. next voxel) that is closest to the average of the actions of each of the models. After the first track, the output of all models is used to reduce missed vessels in bifurcations. The trackers in the ensemble method use the same stopping criteria as the other trackers (see Section 2.3.1). More detailed information on the data selected is presented in Table 3. 3.1. Data overview In our study, we used CTA and CTP image data from two sources: MR CLEAN registry (Jansen et al., 2018) and Erasmus MC. The MR CLEAN registry is an ongoing multi-center registry for stroke patients that underwent endovascular treatment for LVO in The Netherlands since March 2014. Seventeen centers are involved in this study. Data selection criteria applied when selecting the images were: (𝑖) slice thickness ≤1.5 mm; (𝑖𝑖) slice spacing ≤1.5 mm; (𝑖𝑖𝑖) for CTA, the contrast acquisition phase has to be peak arterial, equilibrium or early venous phase according to the definition of Rodriguez-Luna et al. (2014); (𝑖𝑣) brain coverage has to be at least half of the brain, the insula region needs to present in both hemispheres; (𝑣) no large motion artifact along the time axis for CTP data; (𝑣𝑖) CTP data has to contain full cardiac cycle from early arterial phase to late venous phase. 2.3.3. Bifurcation path tracking After bifurcation detection, new paths must be tracked from the bifurcation. This may be done using an estimate of the candidate directions based on the vessel features. We take a two-step approach. In the first step, we start a tracker in all candidate directions regardless of the presence of vessels. Then, we remove unwanted tracks that do not satisfy the eligibility criteria. The candidate directions are determined from 26 connected neighborhoods. In order to prevent the tracked paths from following the original track backward, tracks are only started in the forward direction with regard to the current position (i.e. if the inner product of the current path and direction vector of the new path is positive). After tracking these candidate directions, spurious tracks will be removed based on the amount of overlap with existing tracks and the length of the track. If the track has an overlap with a previously tracked path that is larger than or equal to 95 %, the tracked path will be ignored. Short tracked paths (less than 4.5 mm) will also be ignored. 1594 subjects were included in the MR CLEAN registry from March 2014 to June 2016. In a previous study (Su et al., 2020), we selected 270 images from this set that matched the inclusion criteria. Of these, 49 were manually selected such that there is variation in vendor, image quality, and acquisition phase. In our study, we used these 49 images (with annotations), and we randomly selected another 26 from the 270 images. From the 63 registry subjects with CTP images matching our se- lection criteria, we randomly selected 35 subjects. In addition, from 58 stroke patients with an LVO that were admitted from Jan. 2018 to March 2020 at the Erasmus MC, we randomly selected 15 subjects from the 58 subjects (out of a total of 335) that matched our selection criteria. 2.2. Bifurcation detection Stopping criteria are used to determine the termination point of a track; i.e. the track will be pruned if one of the following conditions is met: To be able to build a vascular tree, a bifurcation detector was developed that can be applied to each path that was generated by the tracker; from bifurcations, new paths can be tracked. • the track runs into the brain tissue • the track runs into the dilated skull, We approach bifurcation detection as a binary classification prob- lem. The structure of the model used for bifurcation detection is similar to the dilated network architecture in the CNN blocks of the A2C networks, but with only one image cube 𝑝𝑡as input. Each convolutional layer is followed by switchable normalization (Luo et al., 2018) and a dropout rate of 0.2; we use binary cross-entropy as the loss function. The detailed architecture of the bifurcation detector can be found in Table 2. • the length of the track exceeds a maximal length. • the track reaches the border of images in z-axis (for half brain coverage cases). • the track runs into a different arterial territory. e.g. forming ACA tree but running to MCA territory. Fig. 4b shows the example of the relationship between the anterior vessel trees and the arterial territory map. 5 J. Su et al. Medical Image Analysis 84 (2023) 102724 Table 2 The network architecture of bifurcation CNN. The input size is 21×21×21. Each convolutional layer is followed by a switchable normalization, a dropout rate of 0.2, and a PReLU activation function. Layer 1 2 3 4 5 6 7 Kernel width 3 3 3 3 1 1 7 Dilation rate 1 1 2 3 4 1 1 Channels 32 32 32 32 64 64 26 Fig. 4. Processing of the anterior vessel trees data preparation; a. shows the example of the anterior vessel tree structure, the MCA tree is presented in blue and ACA trees are presented in green; b. shows the anterior vessel trees and thresholded (0.5) MCA and ACA mask, the green mask is the MCA mask, the red mask is the ACA mask; c. shows the segment labels. The proximal segments are in red, the middle segments are in blue, the distal segments are in green and the skull segments are in yellow. Table 2 The network architecture of bifurcation CNN. The input size is 21×21×21. 3.2. Data annotation The tracks of the anterior vessel tree for training the DRL were manually annotated by the first author of this manuscript (JS) for all 125 subjects. This annotation was done under the supervision of an experienced radiologist (20 years of experience) and a physician (5 years of experience). The annotation task was defined as labeling the track of every anterior vessel from the ICA-top to the most distal part 6 Medical Image Analysis 84 (2023) 102724 J. Su et al. Table 3 The data distribution of 125 subjects. 3.2. Data annotation Properties Data division (n = 125) Set A (n = 20) Set B (n = 20) Set C (n = 20) Set D (n = 20) Set E (n = 20) Validation (n = 10) Test (n = 15) Image modalities Numbers of CTA (n = 75) 13 10 14 15 12 2 9 Numbers of CTP (n = 50) 7 10 6 5 8 8 6 Brain coverage More than half (n = 25) 2 6 3 5 5 2 2 Complete (n = 100) 18 14 17 15 15 8 13 Slice thickness [0.50-0.75)mm (n = 59) 10 9 9 11 8 6 6 [0.75-1.00)mm (n = 32) 5 5 7 3 6 1 5 [1.00-1.50]mm (n = 34) 5 6 4 6 6 3 4 Acquisition phase Early arterial phase (n = 19) 4 3 3 1 3 0 5 Peak arterial phase (n =37) 6 6 4 11 6 1 3 Equilibrium phase (n = 69) 10 11 13 8 11 9 7 Occlusion location ICA (n = 20) 4 4 5 2 2 2 1 M1 (n = 60) 9 12 11 7 10 4 7 M2 and above (n = 38) 6 3 3 9 8 3 6 Others (n = 7) 1 1 1 2 0 1 1 Complexity Low: < 50 (n = 40) 7 5 8 7 7 1 5 Medium:[50-100) (n = 65) 12 12 10 9 9 6 7 High: ≥100 (n = 20) 1 3 2 4 4 3 3 Proximal segments Numbers (n = 417) 67 65 65 70 64 34 52 Intensity (HU) 311 ±94 306 ±84 280 ±87 312 ±92 321 ±120 256 ±105 274 ±77 Length (mm) 26 ±17 24 ±15 26 ±14 25 ±15 24 ±13 21 ±12 24 ±16 Middle segments Number (n = 3814) 607 693 577 650 564 288 435 Intensity (HU) 161 ±76 161 ±70 152 ±68 153 ±68 167 ±80 153 ±73 143 ±61 Length (mm) 29 ±23 29 ±26 29 ±26 31 ±25 31 ±24 29 ±25 31 ±24 Distal segments Number (n = 2436) 343 390 336 432 410 200 325 Intensity (HU) 129 ±66 134 ±61 130 ±64 121 ±54 148 ±84 116 ±51 127 ±65 Length (mm) 27 ±22 29 ±26 25 ±22 28 ±24 28 ±23 27 ±25 28 ±27 Near skull segments Number (n = 1404) 170 237 216 220 255 112 194 Intensity (HU) 153 ±60 149 ±61 144 ±47 139 ±45 149 ±56 128 ±46 127 ±38 Length (mm) 63 ±40 64 ±40 61 ±39 62 ±38 61 ±37 61 ±40 57 ±39 Near skull segments Numbers (n = 595) 121 171 147 94 62 0 0 (Added extra) Intensity(HU) 204 ±81 140 ±56 173 ±83 148 ±63 181 ±95 NA NA Length(mm) 65 ±30 60 ±30 67 ±39 63 ±30 57 ±26 NA NA in spatial resolution, all CTA and CTP images were resampled to a 0.45 mm isotropic grid using cubic B-spline interpolation and were normalized to a range of 0 to 1 using min–max normalization. 3.5. Segment label generation To permit an analysis of the results (i.e. distal versus proximal), we define four categories (labels) for the vessels. proximal, near skull seg- ments, middle segments (majority of the segment inside the MCA/ACA mask), and distal segments (remaining segments). The proximal seg- ments consisted of M1 and A1, which are defined as the segments between the given root till the first bifurcation. Near skull segments were defined as the vessel segments falling into the dilated skull (thresholded at 1000 HU, dilated with a 5 × 5 × 5 kernel) for con- secutive five points. The middle segments were defined using the MCA and ACA mask, i.e. those segments where the majority of the points in the segment were inside these masks. The above-mentioned segments labels and the thresholded MCA and ACA maps are shown in Figs. 4b, 3.3. Interobserver variation on annotation For such a difficult annotation task, it is relevant to assess the in- terobserver variation in the annotation. To this end, a second observer, a medical student (M) was asked to perform the same annotation. For this, three subjects from the CTP category were selected, two subjects with middle complexity and one with high complexity. The tree overlap was calculated using a dilated binary spherical shape with a radius of four voxels with observer M as the reference standard, in the same way as the overlap is computed for the method. For the three subjects, the tree overlap was 0.50, 0.70, and 0.88 respectively, with an average overlap of 0.69. The vessels where the annotation differed were mainly low intensity vessels, and vessels close to the skull. 3.2. Data annotation of the MCA and ACA vessel tree in the CTA or CTP maximal intensity projection (MIP) images. Each tree always starts from the ICA-top of the corresponding side, if there is no ICA-top occlusion. The annotation of a track was discontinued when the artery was not clearly visible anymore. In our study, the vessel diameter varies from 3.6 ± 0.45 mm at the ICA top (Rai et al., 2013) to 0.45 mm in the most distal vessels. For this annotation task, we used an in-house developed annotation tool used in Su et al. (2020). In order to increase the number of segments near the skull (which are difficult vessels to track), veins near the skull were annotated in 10 out of 75 subjects (randomly selected). These veins have an appearance that is similar to arteries. The annotated vessel tracks were transformed into an arborescence structure from the root based on the connectivity as shown in Fig. 4a. In this way, all the bifurcations and the connected segments (vessel segment) are known. During this transformation process, two adjacent bifurcations with a distance of less than 1.8 mm were merged into one trifurcation. The vessel segment representation was subsequently transformed into a voxelized 26-connected representation for use in the training process of the directed vessel tracking model. The bifurcations were used for the bifurcation models. For the tree formation, the initial direction vectors were generated from two manually annotated starting points from the ICA top to the proximal points of each tree. The user has to click two points for the initial direction vector. 3.4. Data preparation In this study, we use both the CTA and the frame with maximal vessel volume from the CTP series. In order to minimize the variation 7 J. Su et al. J. Su et al. Medical Image Analysis 84 (2023) 102724 4c. The segment labels are only used in the analysis of the proposed method. 4c. The segment labels are only used in the analysis of the proposed method. with 𝐿0 as the max surface distance at the initial point of the reference standard curve, see also Fig. 2b. This metric is 1 in case of complete overlap and is negative when the track runs in the opposite position. 4.2. Directed vessel tracking assessment The assessment of the directed vessel tracking consists of an ablation study, comparisons with a baseline DQN model and the baseline DQN with our own reward function. In addition, we assess the impact of using all data. For the final model, we also present an analysis per vessel category. We use overlap rate and average distance as metrics, where the overlap rate (OR) is defined as the part of the track that matches the reference standard until the first error (distance larger than 1.8 mm). The average distance refers to the average curve-to- curve distance in world coordinates between the overlapping part of the tracking path and the reference standard. Statistical significance was assessed using a paired t-test on all metrics, and a value lower than 𝑝= 0.05 was considered statistically significant. The vessel tracking results are reported using a median with an interquartile range (IQR). The hyper-parameters of the PPO were: discount factor 𝛾= 0.9; PPO clip value 𝜖= 0.2; GAE parameter 𝜆= 0.95. The model weights were updated 10 times per mini batch. The initial learning rate for Adam was 1e−5. The learning rate is halved when the validation score does not improve for five epochs in a row. The lower bound of the learning rate was 1e−6. During training, the only augmentation applied is the random flipping of both 3D CTA image and centerline segments along the x- and 𝑦-axis with a probability of 0.5. We use a two-stage training: first, the data from the anterior vessel tree for learning the general anterior vessel tree track is used until the model stops converging; in this stage, the target of a track was a bifurcation. In the second stage, we added the extra segments close to the skull to better learn to track vessels running in the vicinity of the skull. Also, the target is set to five points beyond the bifurcation with both branches with equal probability. In this way, the agent could learn to track beyond bifurcations. 4.1.2. Bifurcation detection The data in this study is used for training and testing DRL directed vessel tracking and bifurcation detection model, and for assessing tree formation. In the directed vessel tracking and bifurcation detection experiments, we use the same data distribution: in both cases, we use 100 images in a five-fold cross validation setup, where the data was stratified on complexity (defined as the number of segments per subject). The ten subjects with extra annotations of segments near the skull were evenly distributed over each fold. The validation set in the cross-validation consisted of another ten subjects (from the remaining 25), and this set was kept the same for all cross-validation experiments. The remaining fifteen subjects were used as an independent test set. The tree formation assessments were performed on the validation and test sets. 1.8 mm means 4 voxels, and with a scaling (stepsize) of 2, we thus allow an error of maximal two steps. The model implementation and training for bifurcation detection were similar to the PPO, both using Pytorch and the same GPU. The training samples are sub-volumes that are at randomly shifted positions along the reference centerlines. This random offset is to simulate a track resulting from the vessel tracking. The random offsets were ranging from 1 to 3 voxels. The training label was obtained by dilating the ground-truth bifurcation points with a binary spherical kernel with a radius of four voxels, in order to increase the size of true positives and make the classification problem less imbalanced. During training, the number of true positive samples was equal to the number of true negative samples. In the training phase, the initial learning rate was 1e−3 for the Adam optimizer. The learning rate reduction scheme was identical to the PPO training. The learning rate lower bound was 1e−6 as well. The image characteristics and segment labels of each fold, valida- tion, and test set are listed in Table 3. In the following, we will name the models based on the fold. For instance, Set A is the test set of tracking model Tr-A and Bifurcation model (Bf-A) in 5 fold cross validation setting. The tracking model Tr-S and bifurcation model (Bf-S) use set A to E for training, and a test set (n = 15) for testing. The validation set (n = 10) is the same for all models. 4.1.3. Tree formation h f i The tree formation method was implemented in python. For the stopping criteria, tracking into brain tissue was defined as an intensity value along the track less than 50 HU for three consecutive points, and tracking into the skull is defined as tracking into a dilated skull (thresh- old at 1000 HU, dilated with 5 × 5 × 5 kernel) for three consecutive points. The maximum length for a track was 330 mm, which is 1.2 times the maximal length of root-to-leaf distance in the training set of 100 subjects. 4.1.1. Directed vessel centerline tracking The proposed PPO based directed vessel tracking approach was implemented in PyTorch (Paszke et al., 2019). The model training and validation were done on NVIDIA A40 GPUs. The PPO training uses episodic learning with a random start position from the first position to five voxels before a bifurcation point of a segment. The length of each episode was therefore arbitrary with a minimal length of five voxels. Each mini batch consists of 8 episodes, in which one episode corresponds to one segment. The training was stopped when one of the following criteria was met: (1) the total length of the track exceeds 1.5 × the length of the reference standard, (2) the agent is off the reference standard segment for a distance of 1.8 mm, (3) the target was reached (4) at the beginning of the training, if the agent has gone into the direction that is opposite to the direction of the initial vector. 4.1. Implementation 4.1. Implementation Probability density maps of the MCA and ACA regions, as well as a hemisphere map, were obtained using earlier described atlases (Peter et al., 2017) that were registered to the images. After transformation to the CTA or CTP images, the MCA and ACA arterial territory maps were thresholded at 0.5 to obtain a binary mask. The hemisphere map consists of three values, indicating the left and right brain hemispheres and the background. 4.1.1. Directed vessel centerline tracking 4.2.1. Ablation study h bl Curve to curve similarity (CCS) metrics of the ablation study models on network architecture and training scheme (part one) and model Tr-A (All models are trained using set B to E, validated on set validation and tested on set A). i i il i ( ) i f h bl i d d l k • the RNN module, by comparing the complete model with a model that uses only a CNN part in the A2C network (CNN only model), • the RNN module with less image information, by training a model that uses image information from only one time point with RNN module (One time point model), • scaled action space, by removing the scaling factor, (No scaling model) • two-stage training, by adding all data and bifurcation extension throughout the training process (Single stage model). For reference, we also train a model that contains all compo- nents model Tr-A. The ablation study part two investigates the impact of various reward functions with the same model configuration as model Tr-A, by comparing our model with the following rewards: • curve-to-curve distance only, by using the log curve-to-curve distance function (CC distance model), • the point to curve distance reward of Li et al. (2021) (Li reward) • the point to curve distance reward of Li et al. (2021) (Li reward) Fig. 5 shows the learning curves with CCS metric and Table 4 con- tains the test results for the ablation study of network architecture and the training scheme. The median OR values in the ablation study are 100 %, but the variation in Q1 values demonstrates that the distributions are different. The Q3 CCS of most models is above 90 %, except the No scaling model (83%). The difference between Tr-A and the other configurations is statistically significant for both CCS and OR metrics. In addition, statistically significant differences were found between the No scaling configuration and the other configurations in terms of average distance. Fig. 5. Curve to curve similarity (CCS) metrics of the ablation study models on network architecture and training scheme (part one) and model Tr-A (All models are trained using set B to E, validated on set validation and tested on set A). ablation study. In addition, we also used a baseline architecture with our own reward function(DQN + ours) to further assess the added value of this reward function in a different DRL method. 4.2.2. Comparison with DQN baseline We also compared our method with a popular DQN method. Our baseline model was introduced by Li et al. (2021). This method uses the same network architecture as Wolterink et al. (2019), which is similar to the CNN architecture of our actor network but with one time point and ReLU activation function. To do a fair comparison, we limit the action space of the baseline DQN methods to the same action space (26 connected neighborhoods). The hyperparameters and optimizer settings are the same as for the PPO training. The data used is the same as in the 4.2.4. Directed vessel tracking performance 4.2.1. Ablation study h bl The performance of baseline DQN and (DQN + ours) are shown in Table 5 and the learning curves are shown in Fig. 6. Fig. 6 shows the learning curves and Table 5 contains the results of the reward functions ablation study. Statistically significant differences were found between the Tr-A and the other configurations in all of the cases except for the average distance of Li reward. 4.2.3. Amount of training data We also investigated whether the current amount data training data is sufficient for our application. For this purpose, we trained a model (model S) with all 100 subjects from set A to E with the same training scheme as was used in the five fold cross validation setup. Table 6 shows the test results of model S and five models on the same independent test set. None of the differences is statistically significant. 4.2.1. Ablation study h bl In the ablation study, we investigate the importance of our network architecture design choices, training scheme, and reward function. In addition, we compare our reward function with an existing reward from Li et al. (2021). We, therefore, divided our ablation study into two parts. In part one, the ablation study focused on investigating the added value of network architecture and training schemes. In part two, the ablation study investigates the impact of the reward functions. The network architecture may have a substantial impact on the performance of DRL methods, and similarly, the activation function has been shown to have an impact on the performance (Henderson et al., 2018). The ablation study part one focuses therefore on investigating the added value of: To monitor the episodic training process, we use a curve-to-curve similarity (CCS) metric, defined as follows: 𝐶𝐶𝑆= 1 −𝐿𝑡 𝐿0 (8) 8 J. Su et al. J. Su et al. Medical Image Analysis 84 (2023) 102724 Table 4 Evaluation result of ablation in network architecture and training scheme (part one) and proposed method on set A. Model CCS (%) Average distance (mm) Overlap rate (%) CNN only 75 [29, 91] 0.64 [0.57, 0.74] 100 [36, 100] One time point 78 [40, 93] 0.63 [0.56, 0.72] 100 [46, 100] No scaling 38 [ 3, 80] 0.56 [0.49, 0.66] 72 [18, 100] ReLU 79 [40, 93] 0.62 [0.55, 0.72] 100 [44, 100] Single stage 80 [36, 93] 0.61 [0.54, 0.72] 100 [41, 100] Model Tr-A 82 [44, 93] 0.65 [0.58, 0.75] 100 [50, 100] Table 5 Evaluation result of reward function ablation study(CC distance and Li reward), baseline DQN, DQN with our reward function(DQN + ours) and proposed method on set A. 4.2.1. Ablation study h bl Model CCS (%) Average distance (mm) Overlap rate (%) CC distance 71 [34, 88] 0.75 [0.66, 0.85] 100 [40, 100] Li reward 53 [21, 85] 0.64 [0.56, 0.73] 61 [28, 100] Baseline DQN 26 [13, 53] 0.69 [0.59, 0.78] 29 [13, 58] DQN + ours 44 [15, 79] 0.73 [0.63, 0.85] 44 [16, 100] Model Tr-A 82 [44, 93] 0.65 [0.58, 0.75] 100 [50, 100] module, by comparing the complete model with a model only a CNN part in the A2C network (CNN only model), module with less image information, by training a model image information from only one time point with RNN One time point model), tion space, by removing the scaling factor, (No scaling stead of ReLU (ReLU model), e training, by adding all data and bifurcation extension ut the training process (Single stage model). ce, we also train a model that contains all compo- -A. on study part two investigates the impact of various ns with the same model configuration as model Tr-A, our model with the following rewards: curve distance only, by using the log curve-to-curve function (CC distance model), to curve distance reward of Li et al. (2021) (Li reward) ws the learning curves with CCS metric and Table 4 con- results for the ablation study of network architecture ng scheme. The median OR values in the ablation study ut the variation in Q1 values demonstrates that the re different. The Q3 CCS of most models is above 90 %, scaling model (83%). The difference between Tr-A and igurations is statistically significant for both CCS and addition, statistically significant differences were found o scaling configuration and the other configurations in ge distance. Fig. 5. Curve to curve similarity (CCS) metrics of the ablation study models on network architecture and training scheme (part one) and model Tr-A (All models are trained using set B to E, validated on set validation and tested on set A). ablation study. In addition, we also used a baseline architecture with our own reward function(DQN + ours) to further assess the added value of this reward function in a different DRL method. The performance of b li DQN d (DQN + ) h i T bl 5 d th l i Table 5 Evaluation result of reward function ablation study(CC distance and Li reward), baseline DQN, DQN with our reward function(DQN + ours) and proposed method on set A. Fig. 5. Table 7 Table 7 Table 7 The evaluation result of the directed tracking from bifurcation to bifurcation using five fold cross validation over 100 subjects. Fig. 6. Curve to curve similarity (CCS) metrics of the rewards ablation study, baseline DQN, DQN with our reward function(DQN + ours) and model Tr-A. All models are trained using set B to E, validated on set validation and tested on set A. Fig. 7. Curve to curve similarity metrics for 5 fold cross validation. Fig. 7. Curve to curve similarity metrics for 5 fold cross validation. Fig. 7. Curve to curve similarity metrics for 5 fold cross validation. Fig. 6. Curve to curve similarity (CCS) metrics of the rewards ablation study, baseline DQN, DQN with our reward function(DQN + ours) and model Tr-A. All models are trained using set B to E, validated on set validation and tested on set A. standard, with random samples from the randomly shifted voxels along the annotated tracks (see. Section 4.1.2). This was done in a five fold cross validation setup (models Bf-A – Bf-E), and the results are shown in Fig. 8. The test accuracy of all five models with different sets converges to about 0.82. The accuracy on the validation set was 0.79, slightly lower than the test performance. this, we used 100 images from the five fold cross validation. The learning curves of the five fold cross validation are shown in Fig. 7. The test result of the models with their corresponding set is listed in Table 7. Variations in the results may be caused by differences in the validation datasets. Therefore, we compute the result for all models with the same independent test set. These results are listed in Table 6. It shows only minor differences between the results, and there are no statistically significant differences between all listed models (including model S). The second assessment focused on bifurcation detection on paths that were obtained from the DRL-based tracker. For this, 100 subjects from the five fold cross validation and 15 subjects from the independent test set were used. The directed vessel tracking model was combined with the corre- sponding bifurcation detection model. e.g. for the testing set A, tracking model A and bifurcation model A were used. In addition to the five fold cross validation models, we trained a bifurcation model Bf-S with 100 subjects. Table 7 The average precision and recall for the different sets with corresponding models are listed in Table 9. The recall and precision of bifurcation detection during tree formation are around 76% and 87% respectively, and there is no statistically significant difference between the models. For the same test set and models, we also analyze the tracking performance with regard to the segment labels. The result can be found in Table 8. They show that the performance of directed vessel tracking depends on how distal the vessels are. 4.2.4. Directed vessel tracking performance The last tracking experiment focuses on the generalizability of the tracking model and further analysis of the tracking performance. For 9 Medical Image Analysis 84 (2023) 102724 J. Su et al. Table 6 The evaluation result of the five models and model S from bifurcation to bifurcation on 15 independent test subjects. Model Tr CCS (%) Average distance (mm) Overlap rate (%) Model Tr-A 83 [44, 93] 0.60 [0.54, 0.70] 100 [46, 100] Model Tr-B 82 [40, 93] 0.60 [0.54, 0.72] 100 [46, 100] Model Tr-C 84 [49, 93] 0.60 [0.54, 0.70] 100 [47, 100] Model Tr-D 82 [34, 92] 0.60 [0.54, 0.70] 100 [41, 100] Model Tr-E 84 [41, 93] 0.61 [0.55, 0.71] 100 [46, 100] Model Tr-S 85 [50, 93] 0.59 [0.54, 0.70] 100 [46, 100] Table 7 The evaluation result of the directed tracking from bifurcation to bifurcation using five fold cross validation over 100 subjects. Model Tr CCS (%) Average distance (mm) Overlap rate (%) Model Tr-A 82 [44, 93] 0.65 [0.58, 0.75] 100 [50, 100] Model Tr-B 76 [32, 92] 0.65 [0.56, 0.76] 100 [38, 100] Model Tr-C 81 [37, 93] 0.65 [0.56, 0.76] 100 [44, 100] Model Tr-D 78 [35, 92] 0.66 [0.57, 0.75] 100 [44, 100] Model Tr-E 85 [46, 94] 0.64 [0.56, 0.74] 100 [55, 100] Fig. 6. Curve to curve similarity (CCS) metrics of the rewards ablation study, baseline DQN, DQN with our reward function(DQN + ours) and model Tr-A. All models are trained using set B to E, validated on set validation and tested on set A. Fig. 7. Curve to curve similarity metrics for 5 fold cross validation. standard, with random samples from the randomly shifted voxels along the annotated tracks (see. Section 4.1.2). This was done in a five fold J. Su et al. J. Su et al. Medical Image Analysis 84 (2023) 102724 Table 6 The evaluation result of the five models and model S from bifurcation to bifurcation on 15 independent test subjects. Table 6 Table 6 a e 6 The evaluation result of the five models and model S from bifurcation to bifurcation on 15 independent test subjects p test subjects. Model Tr CCS (%) Average distance (mm) Overlap rate (%) Model Tr-A 83 [44, 93] 0.60 [0.54, 0.70] 100 [46, 100] Model Tr-B 82 [40, 93] 0.60 [0.54, 0.72] 100 [46, 100] Model Tr-C 84 [49, 93] 0.60 [0.54, 0.70] 100 [47, 100] Model Tr-D 82 [34, 92] 0.60 [0.54, 0.70] 100 [41, 100] Model Tr-E 84 [41, 93] 0.61 [0.55, 0.71] 100 [46, 100] Model Tr-S 85 [50, 93] 0.59 [0.54, 0.70] 100 [46, 100] 4.2.4. Directed vessel tracking performance Model Tr CCS (%) Average distance (mm) Overlap rate (%) Model Tr-A 83 [44, 93] 0.60 [0.54, 0.70] 100 [46, 100] Model Tr-B 82 [40, 93] 0.60 [0.54, 0.72] 100 [46, 100] Model Tr-C 84 [49, 93] 0.60 [0.54, 0.70] 100 [47, 100] Model Tr-D 82 [34, 92] 0.60 [0.54, 0.70] 100 [41, 100] Model Tr-E 84 [41, 93] 0.61 [0.55, 0.71] 100 [46, 100] Model Tr-S 85 [50, 93] 0.59 [0.54, 0.70] 100 [46, 100] Table 7 The evaluation result of the directed tracking from bifurcation to bifurcation using five fold cross validation over 100 subjects. Model Tr CCS (%) Average distance (mm) Overlap rate (%) Model Tr-A 82 [44, 93] 0.65 [0.58, 0.75] 100 [50, 100] Model Tr-B 76 [32, 92] 0.65 [0.56, 0.76] 100 [38, 100] Model Tr-C 81 [37, 93] 0.65 [0.56, 0.76] 100 [44, 100] Model Tr-D 78 [35, 92] 0.66 [0.57, 0.75] 100 [44, 100] Model Tr-E 85 [46, 94] 0.64 [0.56, 0.74] 100 [55, 100] e similarity (CCS) metrics of the rewards ablation study, baseline reward function(DQN + ours) and model Tr-A. All models are E, validated on set validation and tested on set A. Fig. 7. Curve to curve similarity metrics for 5 fold cross valida standard, with random samples from the randomly shifted vo the annotated tracks (see. Section 4.1.2). This was done in cross validation setup (models Bf-A – Bf-E), and the results ar 4.4.4. Impact of imaging characteristics on tree formation 4.4.4. Impact of imaging characteristics on tree formation To get more insight into the performance of the methods, and how these depend on the properties of the images (CTA vs CTP, acquisition phase, number of vessels, etc.), we present the tree formation results for each of these categories in Table 11. It shows that the method performs better in the CTA images compared to CTP and that proximal vessels are better detected than distal vessels. For the other categories (acquisition phase, tree type, or complexity), there are no apparent differences in the results. The topology was assessed using the correct tree topology (CTT) ratio, which quantifies the fraction of points of the tree for which the path to the root is similar (i.e. within 1.8 mm distance everywhere) to the path to the root of the corresponding point in the reference standard. 4.4.2. Consistency test of ensemble method We also perform a consistency test with our final tree formation en- semble method, to assess the impact of changes in the starting positions. For this experiment, we used images with moderate complexity (n = 13) from the test and validation set. The input of the methods was varied by randomly shifting the starting position within a range of five voxels along the given starting centerline. We ran the methods five times with a randomly shifted initial vector. The first result is used as the baseline of the tree formation consistency test. We then compute the precision and recall for the other four results. Both average precision and recall are above 99%. Fig. 8. Learning curve of bifurcation model 5 fold cross validation with 100 subjects. 5. Discussion In this work, we developed and assessed a method to construct tree models of the anterior vessel arteries from 3D CTA and CTP images of the brain. The method consists of a deep reinforcement learning tracking approach, a CNN-based classifier to detect bifurcations along the tracks, and a breadth-first tree construction algorithm that uses the 4.4.1. Single model vs ensemble method 4.3. Bifurcation detection assessment Segments (type) CCS (%) Average distance (mm) Overlap rate (%) Proximal 91 [84, 94] 0.61 [0.55, 0.69] 100 [100, 100] Middle 88 [62, 94] 0.63 [0.55, 0.73] 100 [84, 100] Distal 80 [44, 91] 0.65 [0.56, 0.74] 100 [59, 100] Near skull 44 [12, 82] 0.68 [0.59, 0.79] 46 [18, 81] Table 9 The evaluation result of bifurcation detection and bifurcation inference for tree formation over 115 subjects. Model Bf Bf-A Bf-B Bf-C Bf-D Bf-E Bf-S Precision (%) 86 87 88 87 86 87 Recall (%) 78 77 76 76 75 77 based on the different segment labels on 115 independent subjects. Fig. 8. Learning curve of bifurcation model 5 fold cross validation with 100 subjects. models) both the 10 validation and 15 test images were used. The tree formation results are shown in the top row of Table 11. The precision and recall are reported with median and IQR. However, the CTT was reported using the mean and standard deviation since the median and IQR in almost all items are 100, [100, 100], except near skull segments. In general, the performance of the ensemble model is better than the performance of the single model, in terms of precision and recall. When compared with a single model, the true positive rate of the ensemble model increases by 14, [6, 30] %. Fig. 8. Learning curve of bifurcation model 5 fold cross validation with 100 subjects. In addition, we determined the intensity distribution of false pos- itive and false negative vessels with regard to our reference standard for the ensemble method over 25 subjects. The example results of the ensemble method can be found in Fig. 9. The false positives have a mean and standard deviation of 134 ±91 HU, and the false negatives have a mean and standard deviation of 114 ±62 HU. 4.4. Tree formation In the tree formation evaluation, we assess the performance of the combined tracker and bifurcation detector. In addition, we investigate whether the ensemble method would improve over the single model method, and the consistency of the tree formation method and we as- sess the impact of image characteristics. We performed a visual analysis of the false positive and false negative branches. Also, we performed interobserver analysis of tree formation between two annotators as shown in Table 10. 4.4.3. Interobserver analysis on tree formation 4.3. Bifurcation detection assessment The bifurcation detection was tested in two ways. First, the per- formance was assessed in the training setup with a dilated reference 10 J. Su et al. J. Su et al. Medical Image Analysis 84 (2023) 102724 Table 8 Analysis based on the different segment labels on 115 independent subjects. Segments (type) CCS (%) Average distance (mm) Overlap rate (%) Proximal 91 [84, 94] 0.61 [0.55, 0.69] 100 [100, 100] Middle 88 [62, 94] 0.63 [0.55, 0.73] 100 [84, 100] Distal 80 [44, 91] 0.65 [0.56, 0.74] 100 [59, 100] Near skull 44 [12, 82] 0.68 [0.59, 0.79] 46 [18, 81] Table 9 The evaluation result of bifurcation detection and bifurcation inference for tree formation over 115 subjects. Model Bf Bf-A Bf-B Bf-C Bf-D Bf-E Bf-S Precision (%) 86 87 88 87 86 87 Recall (%) 78 77 76 76 75 77 Fig. 8. Learning curve of bifurcation model 5 fold cross validation with 100 subjects. models) both the 10 validation and 15 test images were formation results are shown in the top row of Table 11. and recall are reported with median and IQR. However reported using the mean and standard deviation since th IQR in almost all items are 100, [100, 100], except near s In general, the performance of the ensemble model is b performance of the single model, in terms of precision an compared with a single model, the true positive rate of model increases by 14, [6, 30] %. In addition, we determined the intensity distribution itive and false negative vessels with regard to our refer for the ensemble method over 25 subjects. The example ensemble method can be found in Fig. 9. The false po mean and standard deviation of 134 ±91 HU, and the f have a mean and standard deviation of 114 ±62 HU. 4.4.2. Consistency test of ensemble method We also perform a consistency test with our final tree semble method, to assess the impact of changes in the star For this experiment, we used images with moderate comp from the test and validation set The input of the metho Table 8 Analysis based on the different segment labels on 115 independent subjects. 4.4.3. Interobserver analysis on tree formation For the interobserver analysis part, we would like to compare the tree formation result for the three subjects that were annotated by both observers with the annotated versions. The precision and recall regarding to the different observers were shown in Table 10. Two features are relevant for the tree formation method: (1) the completeness of the tree and (2) the topology of the tree (whether connectivity is correct). The completeness of the tree is measured with precision and recall. To this end, the reference standard and extracted tree were dilated using a binary spherical shape with a radius of four voxels (allowing a maximum distance of 1.8 mm). True positives are points along the extracted tree that are in the dilated reference standard, false positives are points of the extracted tree that are not being covered in dilated reference standard, and false negatives are points along the reference standard that are not in the dilated extracted tree. Table 11 Properties Single Model Ensemble Precision (%) Recall (%) CTT (%) Precision (%) Recall (%) CTT (%) Overall 63 [52, 74] 59 [46, 70] 90 ±31 70 [57, 81] 68 [54, 81] 87 ±34 Image modalities CTA 66 [52, 79] 64 [51, 73] 95 ±22 72 [58, 83] 77 [63, 86] 92 ±27 CTP 62 [52, 71] 51 [41, 63] 85 ±35 69 [57, 72] 61 [47, 78] 82 ±38 Acquisition phase Early arterial phase 77 [67, 87] 67 [64, 73] 94 ±23 82 [77, 86] 77 [70, 81] 94 ±23 Peak arterial phase 66 [57, 78] 58 [49, 68] 95 ±22 70 [58, 84] 80 [58, 85] 90 ±30 Equilibrium phase 60 [51, 69] 52 [42, 67] 87 ±34 66 [55, 74] 62 [48, 81] 83 ±38 Complexity Low 57 [48, 71] 73 [62, 83] 95 ±22 62 [56, 73] 77 [63, 87] 88 ±33 Middle 64 [56, 78] 62 [52, 81] 86 ±35 72 [56, 82] 66 [48, 81] 85 ±36 High 68 [58, 72] 58 [41, 64] 94 ±25 71 [67, 78] 66 [57, 76] 90 ±31 Tree type MCA 62 [51, 72] 57 [43, 67] 88 ±33 73 [64, 83] 60 [52, 78] 84 ±37 ACA 66 [56, 79] 59 [48, 72] 92 ±28 65 [55, 80] 70 [60, 86] 89 ±31 Segment label Proximal 99 [68, 100] 99 [68, 100] 100 ±0 92 [71, 100] 90 [69, 100] 100 ±0 Middle 60 [42, 76] 59 [45, 70] 97 ±16 69 [54, 79] 67 [57, 84] 98 ±16 Distal 61 [45, 82] 47 [38, 64] 80 ±41 74 [59, 88] 67 [49, 79] 78 ±42 Near skull 39 [27, 59] 43 [18, 75] 63 ±49 66 [48, 87] 64 [35, 75] 50 ±51 tracker and bifurcation detector to construct the tree from the tracking and bifurcation detection results. When evaluated on a test set of 25 subjects, the median precision and recall w.r.t. the manual annotation is 68% and 69%. and multiple time points serve as a means to include information from previous steps in the model. From the results it is clear that using RNN with one CNN time point has a better result than CNN only with more time points; apparently, these elements of the model permit to focus on different temporal aspects relevant for the tracking. Adding an RNN module would enhance the tracking ability. Table 11 e 11 evaluation result of tree formation. The precision and recall are reported with median and IQR. However, the CTT were reported using th median and IQR in almost all items are 100, [100, 100], except near skull segments. on. The precision and recall are reported with median and IQR. However, the CTT were reported using the mean and standard deviation ms are 100, [100, 100], except near skull segments. The evaluation result of tree formation. The precision and recall are reported with median and IQR. However, the CTT were reported using the mean and standard deviation since the median and IQR in almost all items are 100, [100, 100], except near skull segments. 4.4.1. Single model vs ensemble method For the comparison of the single model (Model Tr–S, trained on 100 images) and the ensemble method (using the five cross-validation 11 J. Su et al. Medical Image Analysis 84 (2023) 102724 Fig. 9. Examples of ensemble method tree formation result in the test set. Those subjects were selected based on the median value of recall. Green denotes true positive segments, blue is false positive, and yellow is false negative. a&b: frontal and sagittal view of a subject with middle complexity, a precision of 0.85, and recall of 0.77. c&d: frontal and sagittal view of a subject with high complexity, a precision of 0.83, and recall of 0.76. Fig. 9. Examples of ensemble method tree formation result in the test set. Those subjects were selected based on the median value of recall. Green denotes true positive segments, blue is false positive, and yellow is false negative. a&b: frontal and sagittal view of a subject with middle complexity, a precision of 0.85, and recall of 0.77. c&d: frontal and sagittal view of a subject with high complexity, a precision of 0.83, and recall of 0.76. Table 10 The tree formation result using different reference standard. Observer JS Observer M Agreement Precision Recall Precision Recall (%) (%) (%) (%) Subject 1 0.78 0.57 0.45 0.49 0.50 Subject 2 0.59 0.72 0.53 0.68 0.70 Subject 3 0.71 0.76 0.71 0.54 0.88 Table 10 The tree formation result using different reference standard. Table 11 After annotating the vessels of the CTP in the MIP image, we decided to use the maximum volume CTP image for training and tracking instead of the MIP. Whereas the MIP is convenient for annotation, its appearance is different from CTA images, because of the higher background intensities and more noisy appearance caused by the MIP. This is a limitation of our study, we however preferred including the (suboptimal) annotations over leaving out the CTP, or redoing the annotation. Most existing DRL tracking methods use a DQN approach with a customized reward function; for instance, Li et al. (2021) utilize a DQN tracking approach with the same network architecture as the state-of- art method of Wolterink et al. (2019) and a point-to-curve distance based instance reward for coronary tracking in a 3D CTA image. The tracking performance of the method by Li et al. (2021) outperforms the CNN tracking result of Wolterink et al. (2019). We, therefore, used this method as our baseline method. The baseline methods perform less well in our application. Zhang et al. (2018) utilize the CNN with one time point (different architecture) to track the thoracic aorta. The reward function is similar to Li et al. (2021). Zhang et al. (2020) utilize the same CNN architecture with double-DQN methods and the dot product between the reference standard and agent path of 𝑡and 𝑡−1 as the reward function for coronary centerline tracking. The corresponding point between the reference standard and the corresponding agent location was determined based on the shortest distance. The above- mentioned architectures are less complex and are sufficient for tracking the coronary arteries and aorta. The brain vasculature, as shown in Fig. 1, is different from coronary arteries and aorta, e.g. there is more bending and there is a large variety in curvatures in cerebral vessels compared to coronary vessels, and they are much smaller than the aorta. As a consequence, during tracking, a CNN with one time point only might not be able to provide sufficient information for the network to make accurate direction estimation. In addition, the agent might have difficulty finding the correspondence between the agent’s position and the reference standard. The final result of the method is a tree representation of the brain vasculature, and this tree representation will be the basis for subse- quent works. We intend to use this tree to e.g. Table 11 (2021) aim to optimize the distance between the agent position and corresponding curves. Using the curve-to-curve distance reward function, the instant reward at 𝑡is approximated by the difference in surface area between the agent path at 𝑡and 𝑡−1 with regard to the complete reference standard. Therefore, the agent moves along the centerline because it minimizes the surface distance between the agent path and reference standard. This is particularly handy in the cases of sharp turning vessel structures, e.g helix shape, where the correspondence between agent position and corresponding reference standard point, such as used in other approaches, cannot be found in an accurate way using the closest Euclidean distance (e.g using point to curve distance). Therefore, the curve-to-curve distance fits better with our application. average interobserver overlap of 0.69. The intensity distribution in the branches where the observers disagree is in the range of the low intensity vessels that consist of distal and near skull segments. Though the annotations are not perfect, we assume that for the training, because of the large number of vessels, the errors in the annotations would not greatly impact the models, which is also suggested by the results. Errors in the annotation also impact the quantitative results. We therefore also compare the tree formation results with the annotations from two different observers in three datasets. The differences again are mainly in the vessels that have a low intensity, which is harder to annotate, and often are distal vessels. Also, there was a variation in vessels close to the skull. These errors are similar to the errors of the automated method, suggesting that these low-intensity vessels are difficult for both humans and the automated method. From the analysis of the tree formation result with respect to several imaging characteristics, it followed that only the imaging modality has a significant impact on the quantitative results: the tree formation method on CTP performs less than on CTA. The difference between the validation and test set (the performance on the validation set was always worse than on the test set), may therefore be caused by the higher percentage of CTP images in the validation set. The worse performance may be caused by the use of the maximum volume image for the tracking, whereas the annotation was done in the MIP image, and the blurring was caused by the interpolation after the alignment to the first frame. Table 11 For the tracking, we investigate the policy gradient based proximal policy optimization DRL approach to perform the directed tracking. The average tracking performance on directed tracking from bifurcation to bifurcation over 100 subjects with 7026 vessels was 100 [46, 100]. Bifurcation detection is another essential ingredient, it was tested both in the training scenario and during the single model tree formation with the same 100 subjects. The accuracy of bifurcation detection over 100 test subjects was 82%. During the single model tree formation setting, the precision and recall were on average 87% and 76% respectively. At the final ensemble tree formation method test on 25 independent subjects, the overall precision and recall were 68 [54, 81] and 70 [57, 81]. Excluding the scaling of the step size (i.e. taking smaller steps) yields improved accumulated reward but lower CCS and OR. An ex- planation could be that, first, the frequency of calculating the instant reward is at least two times more than the rest of the configuration. Furthermore, in the no scaling variant, the distance between 𝑝𝑡and 𝑝𝑡−2 is 0.9 mm. This leads to very similar image information at the three time points, which hampers the full exploitation of temporal information. The performance of this no scaling model is similar to the DQN with our reward function(DQN+ours), which is the configuration using CNN with one time point. An ablation study was performed to assess the added value of various design choices made. From the results, it follows that most of the choices (including time information via RNN and multiple time points, activation function, and training strategy) have a small but statistically significant impact on the final tracking results. Both RNN In the reward ablation study, CC distance performed second best, by comparing the result of CC distance and our proposed method Tr-A model, adding binary overlap in the reward function would improve the average distance. The point-to-curve distance reward has a higher 12 J. Su et al. Medical Image Analysis 84 (2023) 102724 average distance but lower CCS and OR in comparison with our pro- posed reward. This result can be observed from both DQN methods and PPO methods. The curve-to-curve distance reward aims to optimize the surface area between the reference standard and agent path at 𝑡and 𝑡−1, whereas the point-to-curve distance used by the work of Zhang et al. (2018) and Li et al. Table 11 find lesions (occlusions in more distal arteries, such as M2, which is still a challenging task Luijten et al., 2021), better quantification of collateral status, and possibly also linking/registering the 3D vascular information from CTA and CTP to interventional DSA images for improving image guidance during endovascular treatment. The tree formation method performed reason- ably well in the proximal and middle segments. In the distal and near skull segments, the performances were also in the interobserver ranges. For the DSA to CTA mapping and distal occlusion detection, it is likely that such tree formation performance is sufficient. While we were finalizing this manuscript, the work of Chen et al. (2021) demonstrated the possibility of walking outside of the classical reinforcement learning problem formulation, which is either value- based or policy-based. In contrast, they regard the reinforcement learn- ing problem as a sequence prediction problem. Similarly, we could use our trained agent and our reward with the sequential framework in an offline reinforcement learning fashion, which may improve our tracking performance. To what extent and whether the transformer would be the best sequential model for our reward function could be further investigated. 6. Conclusion In this study, we developed and assessed a method to construct a brain vessel tree from a CTA or CTP image, using a starting point and direction vector. The method consists of a DRL-based tracker, a CNN-based bifurcation detector, and a breadth-first tree building. The tracker performance on segments from bifurcation to bifurcation has a median overlap of 100 %, and the bifurcation detector has a mean accuracy of 82 %. The combination of both components in an ensemble tree building algorithm yields trees that have a 69 % overlap with manual annotations, which is in the interobserver variation range. When training DL models, data is essential: both the amount of data and the quality. To investigate whether adding may further improve the results, we compared the results of the cross-validation models with a model trained on the full dataset. The results show that the additional 25% of data does not significantly improve the results, suggesting that the amount of data is sufficient for the task. References Moriconi, S., Zuluaga, M.A., Jäger, H.R., Nachev, P., Ourselin, S., Cardoso, M.J., 2018. Inference of cerebrovascular topology with geodesic minimum spanning trees. IEEE Trans. Med. Imaging 38 (1), 225–239. Arulkumaran, K., Deisenroth, M.P., Brundage, M., Bharath, A.A., 2017. A brief survey of deep reinforcement learning. arXiv preprint arXiv:1708.05866. Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., et al., 2019. Pytorch: An imperative style, high-performance deep learning library. Adv. Neural Inf. Process. Syst. 32. Baboiu, D.-M., Hamarneh, G., 2012. Vascular bifurcation detection in scale-space. In: 2012 IEEE Workshop on Mathematical Methods in Biomedical Image Analysis. IEEE, pp. 41–46. Bertsekas, D.P., et al., 2011. Dynamic Programming and Optimal Control, third ed. vol. ii, Athena Scientific, Belmont, MA. Peter, R., Emmer, B.J., van Es, A.C., van Walsum, T., 2017. Cortical and vascular probability maps for analysis of human brain in computed tomography images. In: 2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017). IEEE, pp. 1141–1145. Chen, L., Lu, K., Rajeswaran, A., Lee, K., Grover, A., Laskin, M., Abbeel, P., Srinivas, A., Mordatch, I., 2021. Decision transformer: Reinforcement learning via sequence modeling. Adv. Neural Inf. Process. Syst. 34, 15084–15097. Potter, C.A., Vagal, A.S., Goyal, M., Nunez, D.B., Leslie-Mazwi, T.M., Lev, M.H., 2019. CT for treatment selection in acute ischemic stroke: a code stroke primer. Radiographics 39 (6), 1717–1738. Dai, T., Dubois, M., Arulkumaran, K., Campbell, J., Bass, C., Billot, B., Uslu, F., De Paola, V., Clopath, C., Bharath, A.A., 2019. Deep reinforcement learning for subpixel neural tracking. In: International Conference on Medical Imaging with Deep Learning. PMLR, pp. 130–150. Rai, A.T., Hogg, J.P., Cline, B., Hobbs, G., 2013. Cerebrovascular geometry in the anterior circulation: an analysis of diameter, length and the vessel taper. J. Neurointerventional Surg. 5 (4), 371–375. Damseh, R., Delafontaine-Martel, P., Pouliot, P., Cheriet, F., Lesage, F., 2020. Laplacian flow dynamics on geometric graphs for anatomical modeling of cerebrovascular networks. IEEE Trans. Med. Imaging 40 (1), 381–394. Robben, D., Türetken, E., Sunaert, S., Thijs, V., Wilms, G., Fua, P., Maes, F., Suetens, P., 2016. Simultaneous segmentation and anatomical labeling of the cerebral vasculature. Med. Image Anal. 32, 201–215. Dijkstra, E.W., et al., 1959. A note on two problems in connexion with graphs. Numer. Math. 1 (1), 269–271. References Rodriguez-Luna, D., Dowlatshahi, D., Aviv, R.I., Molina, C.A., Silva, Y., Dzialowski, I., Lum, C., Czlonkowska, A., Boulanger, J.-M., Kase, C.S., et al., 2014. Venous phase of computed tomography angiography increases spot sign detection, but intracerebral hemorrhage expansion is greater in spot signs detected in arterial phase. Stroke 45 (3), 734–739. Engstrom, L., Ilyas, A., Santurkar, S., Tsipras, D., Janoos, F., Rudolph, L., Madry, A., 2020. Implementation matters in deep policy gradients: A case study on PPO and TRPO. arXiv preprint arXiv:2005.12729. Girshick, R., Donahue, J., Darrell, T., Malik, J., 2014. Rich feature hierarchies for accurate object detection and semantic segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp. 580–587. Ronneberger, O., Fischer, P., Brox, T., 2015. U-net: Convolutional networks for biomedical image segmentation. In: International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, pp. 234–241. S h l J L i S Abb l P J d M M i P 2015 T i Guo, Z., Bai, J., Lu, Y., Wang, X., Cao, K., Song, Q., Sonka, M., Yin, Y., 2019. Deep- centerline: A multi-task fully convolutional network for centerline extraction. In: International Conference on Information Processing in Medical Imaging. Springer, pp. 441–453. Schulman, J., Levine, S., Abbeel, P., Jordan, M., Moritz, P., 2015a. Trust region policy optimization. In: International Conference on Machine Learning. PMLR, pp. 1889–1897. Schulman, J., Moritz, P., Levine, S., Jordan, M., Abbeel, P., 2015b. High-dimensional continuous control using generalized advantage estimation. arXiv preprint arXiv: 1506.02438. He, K., Zhang, X., Ren, S., Sun, J., 2015. Delving deep into rectifiers: Surpassing human-level performance on imagenet classification. In: Proceedings of the IEEE International Conference on Computer Vision. pp. 1026–1034. Schulman, J., Wolski, F., Dhariwal, P., Radford, A., Klimov, O., 2017. Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347. Henderson, P., Islam, R., Bachman, P., Pineau, J., Precup, D., Meger, D., 2018. Deep reinforcement learning that matters. In: Proceedings of the AAAI Conference on Artificial Intelligence. 32, (1). Á Su, J., Wolff, L., van Es, A.C.M., Van Zwam, W., Majoie, C., Dippel, D.W., Van Der Lugt, A., Niessen, W.J., Van Walsum, T., 2020. Automatic collateral scoring from 3D CTA images. IEEE Trans. Med. Imaging 39 (6), 2190–2200. Hervella, Á.S., Rouco, J., Novo, J., Penedo, M.G., Ortega, M., 2020. Deep multi-instance heatmap regression for the detection of retinal vessel crossings and bifurcations in Hervella, Á.S., Rouco, J., Novo, J., Penedo, M.G., Ortega, M., 2020. Data availability Lee, T.-C., Kashyap, R.L., Chu, C.-N., 1994. Building skeleton models via 3-D medial surface axis thinning algorithms. CVGIP, Graph. Models Image Process. 56 (6), 462–478. The authors do not have permission to share data. Lesage, D., Angelini, E.D., Bloch, I., Funka-Lea, G., 2009. A review of 3D vessel lumen segmentation techniques: Models, features and extraction schemes. Med. Image Anal. 13 (6), 819–845. Declaration of competing interest The authors declare the following financial interests/personal rela- tionships which may be considered as potential competing interests: Wim van Zwam reports financial support was provided by Cerenovus. Wim van Zwam reports financial support was provided by Stryker Neurovascular. Aad van der Lugt reports financial support was provided by Siemens Healthineers. Aad van der Lugt reports financial support was provided by GE Healthcare. Aad van der Lugt reports financial support was provided by Philips Healthcare. Annotation of anterior vessel trees in brain CT images is a tedious and difficult task, among others caused by the coexistence of arterial and venous structures in the whole brain. In Fig. 1, the anterior vessel tree is only a small part of the complete brain vessel annotation. Un- der such circumstances, kissing vessels, i.e. locations (mostly distally) where vessels (almost) touch are common. This explains the moderate 13 J. Su et al. Medical Image Analysis 84 (2023) 102724 Acknowledgments Li, Z., Xia, Q., Hu, Z., Wang, W., Xu, L., Zhang, S., 2021. A deep reinforced tree-traversal agent for coronary artery centerline extraction. In: International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, pp. 418–428. We thank all MR CLEAN (Multicenter Randomized Clinical Trial of Endovascular Treatment for Acute Ischemic Stroke in the Nether- lands) Registry investigators. MR CLEAN Registry was funded through the CONTRAST consortium, which acknowledges the support from the Netherlands Cardiovascular Research Initiative, an initiative of the Dutch Heart Foundation (CVON2015-01: CONTRAST), and from the Brain Foundation Netherlands (HA2015.01.06). The collaboration project is additionally financed by the Ministry of Economic Affairs by means of the PPP Allowance made available by the Top Sec- tor Life Sciences & Health to stimulate public-private partnerships (LSHM17016). This work was funded in part through unrestricted fund- ing by Stryker, Medtronic and Cerenovus. The funding sources were not involved in study design, monitoring, data collection, statistical analyses, interpretation of results, or manuscript writing. Luijten, S.P., Wolff, L., Duvekot, M.H., van Doormaal, P.-J., Moudrous, W., Kerkhoff, H., a Nijeholt, G.J.L., Bokkers, R.P., Lonneke, S., Hofmeijer, J., et al., 2021. Diagnostic performance of an algorithm for automated large vessel occlusion detection on CT angiography. J. Neurointerventional Surg.. Luo, P., Ren, J., Peng, Z., Zhang, R., Li, J., 2018. Differentiable learning-to-normalize via switchable normalization. arXiv preprint arXiv:1806.10779. Maas, A.L., Hannun, A.Y., Ng, A.Y., et al., 2013. Rectifier nonlinearities improve neural network acoustic models. In: Proc. Icml. 30, (1), Atlanta, Georgia, USA, p. 3. Mnih, V., Badia, A.P., Mirza, M., Graves, A., Lillicrap, T., Harley, T., Silver, D., Kavukcuoglu, K., 2016. Asynchronous methods for deep reinforcement learning. In: International Conference on Machine Learning. PMLR, pp. 1928–1937. Mnih, V., Kavukcuoglu, K., Silver, D., Graves, A., Antonoglou, I., Wierstra, D., Riedmiller, M., 2013. Playing atari with deep reinforcement learning. arXiv preprint arXiv:1312.5602. Moccia, S., De Momi, E., El Hadji, S., Mattos, L.S., 2018. Blood vessel segmentation algorithms—review of methods, datasets and evaluation metrics. Comput. Methods Programs Biomed. 158, 71–91. Zheng, Y., Liu, D., Georgescu, B., Nguyen, H., Comaniciu, D., 2015. 3D deep learning for efficient and robust landmark detection in volumetric data. In: International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, pp. 565–572. Zhou, S.K., Le, H.N., Luu, K., Nguyen, H.V., Ayache, N., 2021. Deep reinforcement learning in medical imaging: A literature review. Med. Image Anal. 73, 102193. Zhang, Z., Marin, D., Drangova, M., Boykov, Y., 2021. Confluent vessel trees with accurate bifurcations. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 9573–9582. Zhao, H., Sun, Y., Li, H., 2020. Retinal vascular junction detection and classification via deep neural networks. Comput. Methods Programs Biomed. 183, 105096. Zhang, P., Wang, F., Zheng, Y., 2018. Deep reinforcement learning for vessel centerline tracing in multi-modality 3D volumes. In: International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, pp. 755–763. Zhang, Z., Marin, D., Drangova, M., Boykov, Y., 2021. Confluent vessel trees with accurate bifurcations. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 9573–9582. Zhang, P., Wang, F., Zheng, Y., 2018. Deep reinforcement learning for vessel centerline tracing in multi-modality 3D volumes. In: International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, pp. 755–763. Zhao, H., Sun, Y., Li, H., 2020. Retinal vascular junction detection and classification via deep neural networks. Comput. Methods Programs Biomed. 183, 105096. J. Su et al. References Deep multi-instance heatmap regression for the detection of retinal vessel crossings and bifurcations in eye fundus images. Comput. Methods Programs Biomed. 186, 105201. h i S S h idh b 199 h l C 9 (8) Sutton, R.S., Barto, A.G., 2018. Reinforcement Learning: An Introduction. MIT Press. l il i f l i i h d bl Van Hasselt, H., Guez, A., Silver, D., 2016. Deep reinforcement learning with double q-learning. In: Proceedings of the AAAI Conference on Artificial Intelligence. vol. 30, (1). Hochreiter, S., Schmidhuber, J., 1997. Long short-term memory. Neural Comput. 9 (8), 1735–1780. Jansen, I.G., Mulder, M.J., Goldhoorn, R.-J.B., 2018. Endovascular treatment for acute ischaemic stroke in routine clinical practice: prospective, observational cohort study (MR CLEAN Registry). Bmj 360. Walsum, T.v., Schaap, M., Metz, C.T., Giessen, A.G., Niessen, W.J., 2008. Averaging centerlines: mean shift on paths. In: International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, pp. 900–907. Jeon, B., 2021. Deep recursive Bayesian tracking for fully automatic centerline extraction of coronary arteries in CT images. Sensors 21 (18), 6087. Wolterink, J.M., van Hamersvelt, R.W., Viergever, M.A., Leiner, T., Išgum, I., 2019. Coronary artery centerline extraction in cardiac CT angiography using a CNN-based orientation classifier. Med. Image Anal. 51, 46–60. Kalaie, S., Gooya, A., 2017. Vascular tree tracking and bifurcation points detection in retinal images using a hierarchical probabilistic model. Comput. Methods Programs Biomed. 151, 139–149. Yu, F., Koltun, V., 2015. Multi-scale context aggregation by dilated convolutions. arXiv preprint arXiv:1511.07122. Kirbas, C., Quek, F., 2004. A review of vessel extraction techniques and algorithms. ACM Comput. Surv. 36 (2), 81–121. Zhang, Y., Luo, G., Wang, W., Wang, K., 2020. Branch-aware double DQN for centerline extraction in coronary ct angiography. In: International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, pp. 35–44. Kullback, S., Leibler, R.A., 1951. On information and sufficiency. Ann. Math. Stat. 22 (1), 79–86. 14 J. Su et al. Medical Image Analysis 84 (2023) 102724 Medical Image Analysis 84 (2023) 102724 Zhou, S.K., Le, H.N., Luu, K., Nguyen, H.V., Ayache, N., 2021. Deep reinforcement learning in medical imaging: A literature review. Med. Image Anal. 73, 102193. 15 15
https://openalex.org/W3119690087	https://link.springer.com/content/pdf/10.1007/978-3-030-55878-9_11.pdf	English	null	Sustainability in a Digital Age as a Trigger for Organizational Development in Education	Springer eBooks	2,021	cc-by	7,161	N. Grünberger (*) · P. Szucsich University College of Teacher Education Vienna, Vienna, Austria e-mail: nina.gruenberger@phwien.ac.at; petra.szucsich@phwien.ac.at Chapter 11 Sustainability in a Digital Age as a Trigger for Organizational Development in Education Nina Grünberger and Petra Szucsich Nina Grünberger and Petra Szucsich 1 In the German-speaking discourse, the terms Mediatisierung und Medialisierung are common (see, e.g., Bettinger and Aßmann 2017). The team digitization is rather used as a political and economic term than as a scientific concept. © The Author(s) 2021 D. Ifenthaler et al. (eds.), Digital Transformation of Learning Organizations, https://doi.org/10.1007/978-3-030-55878-9_11 11.1 Introduction 2013 2017 2019 1 In the digital age, organizations are, amongst other factors, to a great extent also determined by digitality. New structures, new infrastructures, new jobs and job descriptions as well as new forms of collaboration within educational organizations and between educational organizations with other institutions are needed. In short, the digital age makes a fundamental process of organization development in the educational context necessary (see, e.g., Eickelmann 2010; Grünberger and Münte- Goussar 2017; Schiefner-Rohs 2017; Tulodziecki et al. 2018). 1 In the German-speaking discourse, the terms Mediatisierung und Medialisierung are common (see, e.g., Bettinger and Aßmann 2017). The team digitization is rather used as a political and economic term than as a scientific concept. 189 190 N. Grünberger and P. Szucsich N. Grünberger and P. Szucsich Following a holistic perspective, we can see that digitization affects our living environment much more than previously anticipated. What we have to take into account are the connections “of media technologies, their materiality, hardware and energy, with the geophysical nature” (Parikka 2015, p. 8). Developing, producing, using and disposing of digital media has an ecological, economic and social impact on Planet Earth and our society. Digitization shapes our landscape, our environment and social relationships; sometimes, much more than we could ever have imagined. p g As a development parallel to the digital age, individual and political efforts of climate and environmental protection are growing. According to that, educational concepts such as Education for Sustainability (EfS) or Education for Sustainable Development (ESD) are getting more attention (Chang et al. 2020, p. 1). This might be the result of, on the one hand, political efforts, especially by the United Nations (2015) and their formulation of “Sustainable Development Goals” (short: SDGs) and, on the other hand, the enormous efforts of social movements. Often these social movements grow from a common concern shared by several people into a global movement, like, for instance the Fridays for Future initiative. This means that we can already see some slight changes in the political agenda and efforts towards a sustainable approach, such as “The Green European Deal” by the European Commission (2019). 2 This chapter was basically written in March 2020 during the world-wide crisis triggered by the pandemic of COVID-19. It is clear that this pandemic will have an enormous transformational impact on all organizations, including educational organizations. In many countries, schools are closed, and teaching in schools up to higher education suddenly had to take place in digital form. The restrictions on air traffic, mobility and manufacturing at a global scale may have a positive effect on the fight against climate change. According to these assumptions, it is obvious that the pandemic of COVID-19 will have enormous implications on the topic of this chapter, as it addresses social and organizational transformations by digitization as well as a sustainable approach. However, at the current moment, the consequences of the whole crisis cannot be assessed and in the near future, there will certainly be a lot of research. However, this chapter is not going to include aspects concerning the pandemic as these would just be speculations. 11.1 Introduction Similar to digitization, sustainability is a rather vague term: on the one hand, it describes the maintenance of good conditions for “environmental, economic and social well-being for current and future generations”; on the other hand, it focuses on ensuring the existence on our planet, which means not to “damage” or not to “be harmful to the environment” (Chang et al. 2020, p. 1). 2019 2019 2019 2 11 Sustainability in a Digital Age as a Trigger for Organizational Development… 191 This chapter focuses on the interrelatedness between digitization and efforts con cerning a sustainable behaviour for a better future, especially in regard to climate change issues. In the first chapter, some theoretical concepts and empirical research results are explained. Furthermore, we discuss the peculiarity of digitization as well sustainability as so-called wicked problems and crisis-like situations. From this, we derive implications for transformational processes of educational organizations by becoming more digital and more sustainable. We finally conclude with some indi cations for further research in the fields of transformational processes of educational organizations, education for sustainability as well as education in consideration of digitization. 11.2 Sustainability in a Digital Age Sustainability is a frequently used term, especially in recent years, triggered, for example, by the activities of the Friday for Future initiative and ongoing debates on climate change. From a political perspective, the effort of the United Nations (2015) concerning the SDGs sets an operational framework for further political decisions. Taking into account that the ongoing climate crisis asks for a fundamental change and therefore rapid actions, the European Commission (2019) formulated “The Green European Deal”. In Austria, for example, environmental and climate protec tion plays an important role in the current policy of the Austrian Federal Government (Österreichische Bundesregierung 2020). In its origin, the term sustainability comes from to sustain which means to bear or to suffer as well as to keep up (Harper 2001a). This explanation of the term’s origin already points out the two poles of, on the one hand, striving not to destroy our planet and, on the other hand, trying to preserve the beauty and the balance of our natural habitat for future generations (Harper 2001b). Furthermore, sustainabil ity always includes a social, economic and ecological perspective and their inter connections with each other. At the moment, we can detect a lot of effort concerning environmental and especially climate protection. As Jonathan Franzen (2020) puts it a bit cynically: Maybe we should not concentrate that much on climate protection as we have already lost this fight. He advocates focusing on the giant negative effects of climate change such as droughts and heat, famine, even more exploitation of human beings and our Planet Earth as well as a giant refugee crisis. According to him, all efforts of a so-called green new deal (e.g. Naomi Klein 2019) or a green deal, for instance, the one proposed by the European Commission (2019), are 192 N. Grünberger and P. Szucsich N. Grünberger and P. Szucsich useless and just wasted time and effort. From his perspective, we can see that a unidirectional focus on climate protection and climate change is problematic. As mentioned above, we always have to consider the three aspects of sustainability: ecological, economic and social aspects by regarding past, present and future developments. p Apart from the argument that we are fighting in an already lost “war,” there is another challenge: The trend of “going green” has become popular as people take steps in order to help preserve the Earth for future generations. 11.2 Sustainability in a Digital Age The development of neo-ecology also concerns ecological, economic as well as societal commitment. It changes markets slowly but noticeably, not only for companies but for consumers as well, as an increasing number of people want to consume in a fair and conscious way. For companies and institutions “going green” is necessary to stay attractive to customers. However, the label “green” does not necessarily mean that the company is really concerned about the environment and acts in an eco-friendly way, as real sustainability may not make business sense. Another example of a development that strives to combine the two aspects of digitization and sustainability is “Green IT,” the practice of using computers and ICT resources in a more efficient and environ mentally responsible way. Green IT is “a collection of strategic and tactical initia tives which either: (1) directly reduce the ‘carbon footprint’ of the organisation’s computing operation; (2) use the services of IT to help reduce the organisation’s overall carbon footprint; (3) incentivise and support greener behaviour by the organ isation’s employees, customers and suppliers; (4) ensure the sustainability of the resources used by IT” (Hird 2010, p. 16). Green IT could, for example, consider the reduction of the server performance at night or at weekends to reduce power con sumption. Research has shown that algorithms, which adjust the performance of a wireless network according to utilization, could achieve an energy reduction of 15% on average (Umweltbundesamt Deutschland 2019, p. 19). According to Schratz and Steiner-Löffler (1999), we need educational institu tions in the digital age that are less concerned about learning issues (reproduction of knowledge) than about life issues (transformation of knowledge). Unfortunately, these questions are often regarded as an irritation rather than a challenge or as some thing that is not mentioned in the curriculum and therefore not covered at all. “Going green” in the context of an educational institution could include using recycled paper or reducing the amount of paper used in the first place. At scientific conferences, for example, it has become popular not to print the programme and other print media for sustainability reasons. However, current discussions revolve around the world’s digital carbon footprint, emphasizing the constantly increasing greenhouse gas emissions caused by the transmission of data via the internet and the consumption of electricity by using digital media. 11.2 Sustainability in a Digital Age This is because the process of transmitting or streaming data requires millions of physical servers in data centres around the world, all spending a lot of energy. Concepts for a more sustainable use of information and communications technology (ICT) often recommend a reduction of something: we should reduce the amount of transmitted data, the power con sumption, the consumption of streaming services such as Netflix or Spotify, the 1 Sustainability in a Digital Age as a Trigger for Organizational Development… 11 193 numerous replacements of digital devices, etc. It is clear that these concepts depend on the social acceptance of an over-all reduction. However, another example shows the ambivalence and enormous complexity of the topic: Digital technology can also be used for climate protection. Some extrapo lations point out that ICT tools may even reduce CO2 emissions by up to 20% by the year 2030 (Clausen et al. 2019, p. 1). As one example, the block chain technol ogy is being discussed as a possible means of reducing the emissions of CO2. Again, a lot of parameters need to be considered if you want to decide whether an ICT tool is sustainable or not. For instance video-conferencing tools are often referred to as a more sustainable alternative to business trips (especially when traveling by plane). However, research results have revealed that video-conferencing tools reduce busi ness trips just at first sight. On second glance, it has to be taken into account that fewer business trips result in more free time for new projects and new invitations for meetings and, eventually, lead to new occasions for business trips and other forms of required energy (e.g. electricity, amounts of data saved on servers, paper) (Clausen et al. 2019). Digital technology might also be a means of enhancing environmental und cli mate protection. On the one hand, digital technology is used for collecting climate relevant data from around the world. Consequently, digital technology is a main tool for understanding climate change and for monitoring its development (Umweltbundesamt Deutschland 2019, p. 49f). Or, as Chun (2015) puts it: Our idea of climate change is calculated and illustrated by algorithms. Computers collect data, put them in correlation and point out trends of climate change. But they are and always will be calculations and hypotheses and not a blueprint of reality (Chun 2015, p. 678f). 11.2 Sustainability in a Digital Age However, nowadays, it is far from that. Large corporations like Google, Amazon or Alibaba have the power over the internet. Thinking of other countries from the global south, we have to take into account that internet access is a question of infrastructure like computer hardware, software and wireless network as well as a question of language as most of the information shared on the internet is in English, followed by Russian and German. Smaller languages and minorities have problems to be represented on the World Wide Web. Therefore, standards to represent these minorities are urgently required (Norbert Klein 2018). Still, we can see a huge social gap offline as well as online. As mentioned above, Jonathan Franzen (2020) emphasizes that we should not focus on climate protection that much, as we have lost this war already. Therefore, Jesse Ribot suggests focusing much more on “climate-related crises”: “I am definitely not writing about the causes of climate change. I am not writing about smokestacks or drivers in New Jersey or Beijing or anything like that. Rather, I mean the causes of the crises themselves. The causes of hunger, famine, dislocation, economic loss; that is, the outcomes that happen when climate trends or events hit the ground.” (Ribot 2019, p. 34) As Ribot points out, these climate-related challenges affect “vulnerability,” which is closely related to crises, because “without vulnerability there is no crisis [and …] vulnerability here is the predisposition, in some way or another, to damage”. When vulnerability comes together with hazard and with specific moments like a climate situation, it can easily turn into a social crisis. But: “[…] climate-related crises therefore do not merely fall from the sky when there is a climate event. They are socially produced via conditions on the ground” (Ribot 2019, p. 34). As Ribot points out, these climate-related challenges affect “vulnerability,” which is closely related to crises, because “without vulnerability there is no crisis [and …] vulnerability here is the predisposition, in some way or another, to damage”. When vulnerability comes together with hazard and with specific moments like a climate situation, it can easily turn into a social crisis. But: “[…] climate-related crises therefore do not merely fall from the sky when there is a climate event. They are socially produced via conditions on the ground” (Ribot 2019, p. 34). As sustainability is a far-reaching concept, so is digitization. 11.2 Sustainability in a Digital Age On the other hand, digital media and digital visualizations help to make something abstract like climate, climate change and global developments vis ible and tangible for researchers and citizens. Therefore, we could say that digitiza tion helps us to gather better and more information and awareness about climate change on a local and global scale. Consequently, we can rethink our behaviour and strive to live a more sustainable life (Umweltbundesamt Deutschland 2019, p. 50). However, in contrast to all aspects mentioned above, we must not forget that digital technology itself represents an ecological, economic and social challenge at all stages in the life cycle of digital media: from technology development, the pro duction process, from transport up to the use of digital media and their disposal and/ or up- or recycling. In short, digital technology poses a problem for climate protec tion. In 2019, around 80% of people living in developed countries had a smartphone and used it almost every day. However, a large number of cell phones “can only communicate via networks based on 2G technology, which does not allow using the Internet” (The Shift Project 2019, p. 42). These cell phones are going to be replaced soon. And that is a lot of devices with a lot of energy used for their production and raw materials needed for developing, producing, using and disposing of them. Furthermore, “the number of smartphones will rise from 1.7 billion in 2013 to 5.8 billion in 2020, with a growth of 11% a year” (The Shift Project 2019, p. 34). These arguments concern both, the ecological and economic perspective, but we also have to consider the social perspective: Digital media, which are, for example, used in 194 N. Grünberger and P. Szucsich N. Grünberger and P. Szucsich western countries are produced in China, the raw materials are mostly mined in African countries, where people have to cope with exploitation, child labour and human trafficking. These are practical thoughts concerning the “real” world: this is the “blueprint of reality” mentioned above. We have to think about post- and neo- colonial exploitations (e.g. Castro Varela and Dhawan 2005; Thiébaud et al. 2018) by developed countries and eventually accept our responsibility. According to this, we can see similarities when we consider the development of the internet: The internet was originally built as a power-free space with equal access for all users (Jörissen and Verständig 2017). 11.3 About Crises and Wicked Problems One peculiarity of digitization as well sustainability is that they are difficult to define and are often regarded in the context of crisis-like situations. As a modern society, we currently seem to face a lot of crises. But crises are not necessarily nega tive. They are conditio sine qua non and stimuli sui generis of learning processes and of pedagogy as a research discipline (Schneider-Taylor 2009, p. 104). The ety mological origin of the word crisis reaches back to the sixteenth century. The Greek word krísís literally means separation or decision. A crisis certainly is a kind of turning point (Kluge 2011, Abschn. Krise; Schneider-Taylor 2009, p. 109f). As Koselleck (1973, p. 141) points out, the word crisis was first commonly used in medicine as a decision point between life and death. Therefore, a crisis is often related to the fear of death. In addition to that, the word crisis is closely related to the word criticism, which asks for proof and valuation of an issue and can thus be the starting point of a crisis (O’Mahony 2014, S. 250; Pfeifer 1989, p. 934f). As mentioned above, in order to overcome a crisis, traditional structures are being questioned. Additionally, as part of the crisis, socially accepted systems of norms and values have to change as well. After a while, step by step, a new system of structures and a new system of norms and values will be constructed. Another important issue is that with the developments of digitization and climate change, we do not only face crises, but so-called wicked problems, with challenges which reflect the structural interrelatedness of media, digitization, economy, mankind, ecology and habitat. As stated above, wicked problems cannot “be clearly defined with pro posed and testable possible solutions”. They have no “definitive formulation,” “there is no way of determining when a solution has been found; solutions are not true or false but rather good or bad” and “there is no immediate or ultimate test of a solution because any possible solution modifies and changes the problem”. Therefore, wicked problems cannot be solved. The aim must be to understand a wicked problem more and more, to raise awareness and “to learn how to live with it” (Peters 2018, p. 429). When talking about sustainability and climate change, the term “wicked prob lem” is commonly used (e.g. Peters 2018). 11.2 Sustainability in a Digital Age However, the social transformation triggered by digitization is not only a “more of something.” For example digital technology in everyday life does not only mean reading and writing with the help of digital media. It is a transformation which is changing social struc tures. Furthermore, digital technology is not only black and white. According to Kerres (2018), it should not be regarded additively, as an additional aspect to our lives but as an integral part. Taking matters a step further, as digital technology makes up an integral part of our lives, thinking and behaving in a “sustainable” and environmentally responsible way should be integral, too. Both, building holistic structures to cope with digital technology and behaving sustainably in organizations such as schools and other educational institutions, ask for a fundamental transfor mation process. Society expects citizens to be able to deal with change 11 Sustainability in a Digital Age as a Trigger for Organizational Development… 195 11 constructively, both privately and in the dynamic context of global, multicultural change. According to Fullan, educational organizations are the only social institu tions that have the potential to make a significant contribution to this goal (Fullan 1999). However, a transformational process like that leads to questioning well- known, traditional structures and values. Apart from that, speaking of fundamental changes, we often use the term “crises”. 11.3 About Crises and Wicked Problems Considering the discourse of digitiza tion, this is, however, not the case. This may seem strange as the speed of the digital development makes it simply impossible to keep up with concepts about possible implications digital technology may have on humans, animals and our Planet Earth, in general. When considering both aspects, we could even speak of “wicked and 196 N. Grünberger and P. Szucsich interwoven challenges.” These challenges have in common that the conditions and many things around them are constantly changing, while people are struggling to solve them. This has various consequences, for instance consequences for social structures: First of all, we have to accept the fact that we do not know and are not able to anticipate how things might develop. Therefore, some projects are open- ended, which means that we may find answers at all levels which will then just lead to even more questions than before. However, crisis-like situations can also be regarded as opportunities for structural changes, for example changes in power structures, and for a more collaborative process of research. It is obvious that this approach cannot follow just one perspective. It clearly requires a holistic, interdis ciplinary approach. In addition to that, it has to consider future changes in the short and long run. And, at the same time, it is obvious that this might somehow seem uncomfortable and awkward as uncertainty usually implicates fear and frustration. 11.4 Transformations of Educational Organizations? Furthermore, they are sup posed to know to what extent the use of digital technologies damages the environment or contributes to environmental protection (Bundesgesetzblatt für die Republik Österreich 2018). This means, the ecological aspect is mentioned, if only in one point. p To sum up, we can at least see some effort of awareness raising towards ecologi cal aspects in the educational agenda. In addition to that, the efforts of institutions like the German “Wissenschaftlicher Beirat für Globale Umweltveränderungen” (WBGU 2018, 2019) and more or less private institutions like the “Rat für digitale Ökologie” (https://ratfuerdigitaleoekologie.org/) have to be mentioned. The global perspective is taken into consideration, for example, by the United Nation’s (2015) formulation of the SDGs: The SDG 9 targets to “build resilient infrastructure, pro mote inclusive and sustainable industrialization and foster innovation.” The sub- goal (9.c) aims to “significantly increase access to information and communications technology and strive to provide universal and affordable access to the Internet in least developed countries by 2020.” As a first result, the UN postulated for 2019 “almost all people around the world now live within range of a mobile-cellular net work signal, with 90 per cent living within range of a 3G-quality or higher network […]” (United Nations 2015). However, this can be interpreted differently as well: The formulation of the SDG 9 may be (ab)used by ICT companies from western countries to create new jobs to install their infrastructure to the global South without taking heterogeneous cultures, natural environments and social structures into con sideration at all. It is obvious that bringing ICT to “least developed countries” is not a good thing per se but has to follow participative processes by regarding the spe cific conditions of these countries. After having considered educational policies and the political effort, the question arises what all that means for educational organizations? From our point of view, the social structures of schools as well as of other educational institutions have to change according to “wicked problems,” which cannot be solved and cause discom fort as well as uncertainty in regard to future developments. One important aspect is that the traditional relationship between teachers and students changes as hierar chies are generally levelled down. A lot of local initiatives have emerged, which influence general structures from bottom up. Another important point is that, espe cially in educational organizations, learners (e.g. 11.4 Transformations of Educational Organizations? How will these “wicked and interwoven challenges” mentioned before transform whole organizations, especially schools and other educational institutions? As stated above, neo-ecology is a trend that will shape the 2020s, as environmental awareness is something companies as well as consumers simply cannot ignore any more. Buzzwords and phrases like energy efficiency, clean energy, greenwashing, “going green,” “Green your product!” are widely used and environmental awareness has become a social movement. As we know, this could be a political or marketing strategy and/or real engagement for climate and environmental protection. In addi tion to that, we also know that “going green” mainly focuses on ecological aspects and does not primarily consider equal rights and social fairness on a local and global scale. The economic perspective is still the predominant one in our neo-liberal world as it is getting more and more detached from social and ecological aspects. However, as previously mentioned, the consistent reference of ecology, economy and social justice and the relationship among them is urgently necessary, not only in the discourse of digitization. In the previous chapters, we have already discussed some political initiatives concerning digitization and sustainability. Another example is the DigComp – Concept of the European Commission, which provides a framework for major skills that are important in a digitalized world. This framework is often used as a basis for developing educational programmes. It contains one paragraph saying “4.4 Protecting the environment. To be aware of the environmental impact on digital technologies and their use.” (Carretero et al. 2017, p. 17) Furthermore, the German strategy of the Standing Conference of the Ministers of Education and Cultural Affairs (KMK) called “Education in the Digital World” claims to protect nature and the environment (“4.4. Natur und Umwelt schützen”) (Deutsche Kultusministerkonferenz 2016). In Austria, a new curriculum was developed and put into practice in 2018 for a new school subject called “Basic Digital Education” 11 Sustainability in a Digital Age as a Trigger for Organizational Development… 197 11 (Digitale Grundbildung) in secondary schools, which can either be taught as an individual subject or be integrated in various already existing subjects. In accor dance with this new curriculum, children learn about the dynamics and meaning of values, norms and different interests with regard to the use of digital media in vari ous contexts (economic, religious, political, cultural). 11.4 Transformations of Educational Organizations? pupils and students) as well as teachers can put governing bodies under pressure and thus convince them to take action. We can also see a lot of citizen projects, which first started out locally and then grew into bigger or even global movements such as the Fridays for Future ini tiative. Fridays for Future has managed to put pressure on educational policy mak ers to allow the participation on the Fridays for Future demonstrations and to pay more attention to environmental and climate protection in general. All these initia tives have something in common: They have very flat hierarchies. One possible 198 N. Grünberger and P. Szucsich N. Grünberger and P. Szucsich reason for that could be, that all citizens have to act and think like researchers to solve or to better understand a certain “wicked problem.” As there is not the one right answer or the one approach to find a solution, we all are unknowing but at least trying and learning and are hopefully doing research.l In the context of flat hierarchies, teachers become more like coaches who support learners in individualized learning scenarios. In this special learning environment, subjects, time and space disperse and new paths for cross-curricular learning open up. Learning is no longer confined to the classroom but takes place in companies, on excursions, in museums or exhibitions. The learning time is no longer rigorously squeezed into a 45- or 50-min. cycle, because cross-curricular projects allow stu dents to work on their topic for a longer time and take their own breaks sensibly. Apart from that, a “wicked situation” like the ones mentioned above requires new forms of research for new solutions in a collaborative and interdisciplinary way. We need to use all our knowledge, imagination, creativity and technology to strive to solve the big challenges of our society. Thus, schools and other educational institutions should allow employees and learners to come together. They should grant them time and space to think about current issues critically and creatively. As a consequence, so-called communities of practice can emerge that share a concern or passion for something they do and, as they interact regularly, they learn how to do it even better and more effectively (Wenger 1998). Professional practice is based on the capacity to reflect on things as one step in the circle of continuous learning. 11.4 Transformations of Educational Organizations? Teachers are experts in many fields like the development of innovative learning designs or the integration of ICT in class. Much of this knowledge, however, is implicit. To share, discuss and reflect on this tacit knowledge with others, teachers often need support to make their competencies visible to themselves and others. In communities of practice, often supported by the supervision and counseling of a university or college, teachers become aware of their competencies and can thus learn and profit from each other’s experience-based knowledge. In this context, learning means learning how to understand the complex develop ments of our present and future society and trying to find solutions. This approach is not new, as it was already used by the educationalist Wolfgang Klafki (2007). Klafki wrote a didactic concept that describes ways to cope with so-called arche typal, revolutionary key issues (‘epochaltypischen Schlüsselproblemen’). According to him, learning does not aim at developing a verifiable growth of competencies. Consequently, this understanding of learning processes requires open curricula and alternative forms of assessments. In addition to that, this approach is very practice- oriented: After having found a possible solution to a complex problem, this knowl edge must be put into action. Without action, the best solution is redundant. But, again, these processes should not happen within a traditional hierarchical structure. All participants of an educational organization – schools and other educational institutions can be regarded as learning organizations in this context – should have the possibility to co-create the organization with the board committees as well as to co-create visions of possible problem-solving strategies and a sustainable future. Another aspect, which is being discussed in several approaches of “global citi zenship education”, “service learning” or “civic education” (Schlicht and 11 Sustainability in a Digital Age as a Trigger for Organizational Development… 199 11 Slepcevic-Zach 2016; Sporer and Bremer 2016, p. 356), is a multiple perspectives approach. It is important for educational institutions to start projects cooperating with non-educational partners in order to get a wider perspective and see the big picture. On the one hand, this can be accomplished on a local scale: Local institu tions and communities often have needs in specific areas, which may be solved or at least worked on by teachers and learners in cooperation with the local groups. 11.4 Transformations of Educational Organizations? On the other hand, schools profit a lot from external experts who produce new ideas and can enhance the collective understanding of various issues. On a global scale, coop eration or networking with other people, maybe even from different continents, can initiate discussions, raise attention for social inequality and can thus promote toler ance and problem-solving skills. To sum up, being aware of the transformational power of digital technology, on the one hand, and, on the other hand, of the necessity to cope with environmental and climate protection by acting in a sustainable and environmentally responsible way at all levels – ecologically, economically and socially – can be a trigger for fundamental transformational changes of educational organizations. These changes concern traditional educational structures as teacher-student relationships, time, space as well as the curricula and teaching and learning processes. Becoming aware of these “wicked problems” may consequently also have an impact on individual people and other (learning) organizations on a local and global scale and – in the long run – on national and international political agendas as well. 11.5 Conclusion In this chapter, we discussed the significance of digitization for climate change as well as a possible means against the further destruction of the environment. Digitization and the idea of sustainability have in common that they can both be seen as “wicked problems” and crisis situations. Thus, both can also be regarded as opportunities and triggers for a necessary transformation of learning organizations to be prepared for future times and for main challenges we, as a society, have to face. Without any doubt, digitization represents a comprehensive challenge for today’s society. And without a doubt, environmental and climate protection is a very com plex issue on a national and international level this century and beyond. But, instead of discussing and meeting these challenges separately, we should deal with them in an interdisciplinary approach, which allows people and institutions to undergo a transformational process that enables them to live and act according to the premises of sustainability. In this context, sustainability has to be understood within the triad of ecological, economic and social aspects. On behalf of this mindset, educational organizations are required to rethink their traditional social structures, economic outcomes and ecological behaviour. As this chapter has shown, sustainability and environmental responsibility include extensive transformational changes in educa tional organizations and transformational learning processes of all the people involved. Institutions and their people have to think out of the box: Digitization and 200 N. Grünberger and P. Szucsich N. Grünberger and P. Szucsich sustainability are targeting a global perspective, and educational organizations therefore need to open up, reconsider traditional structures and values and include local and global perspectives as well. After all, thinking out of the box may be the one way to face the discussed grand challenges of our society. In the future, more research has to be carried out in the context of digitization and protecting our environment and climate, in the context of digitization and Education for Sustainability (EfS) as well as digitization and transformations of learning orga nizations with a specific focus on sustainability. It is obvious that the challenges in the research fields mentioned above can only be met by inter- and transdisciplinary research methodologies. 11.5 Conclusion Acknowledgements The research results discussed in this chapter are based on the project OEHA (oeha.phwien.ac.at), funded by the “Innovationsstiftung für Bildung” and the “Austrian agency for international mobility and cooperation in education, science, research and culture” (OeAD) from September 2019 to August 2020. Acknowledgements The research results discussed in this chapter are based on the project OEHA (oeha.phwien.ac.at), funded by the “Innovationsstiftung für Bildung” and the “Austrian agency for international mobility and cooperation in education, science, research and culture” (OeAD) from September 2019 to August 2020. References Bettinger, P., & Aßmann, S. (2017). Das diskursive Feld um Mediatisierung und Mediensozialisation. Eine Analyse deutschsprachiger Fachzeitschriften. In D. Hoffmann, F. Krotz, & W. Reißmann (Hrsg.), Mediatisierung und Mediensozialisation. Prozesse—Räume—Praktiken (S. 79–S. 100). Wiesbaden: Springer VS. Bundesgesetzblatt für die Republik Österreich. (2018, April 19). Änderung der Verordnung über die Lehrpläne der Neuen Mittelschulen sowie der Verordnung über die Lehrpläne der allge meinbildenden höheren Schulen. Retrieved June 6, 2020, from https://www.ris.bka.gv.at/ Dokumente/BgblAuth/BGBLA_2018_II_71/BGBLA_2018_II_71.rtf Carretero, S., Vuorikari, R., & Punie, Y. (2017). DigComp 2.1. The digital competence frame work for citizens with eight proficiency levels and examples of use. Luxembourg: Publications Office of the European Union. Retrieved June 6, 2020, from http://publications.jrc.ec.europa. eu/repository/bitstream/JRC106281/web-digcomp2.1pdf_(online).pdf Castro Varela, M. do M. do M., & Dhawan, N. (Hrsg.). (2005). Postkoloniale Theorie—Eine kri tische Einführung. Bielefeld: transcript. Chang, C.-H., Kidman, G., & Wi, A. (2020). Issues in teaching and learning of education for sus tainability. Theory into practice. London; New York: Routledge. Chun, W. H. K. (2015). On hypo-real models or global climate change: A challenge for the human ities. Critical Inquiry, 41(3), 675–703. Retrieved June 6, 2020, from https://www.jstor.org/ stable/10.1086/680090 Clausen, J., Schramm, S., & Hintermann, R. (2019). CliDiTrans Werkstattbericht 3-2: Virtuelle Konferenzen und Online-Zusammenarbeit in Unternehmen: Effektiver Klimaschutz oder Mythos? Retrieved June 6, 2020, from https://www.borderstep.de/wp-content/uploads/2019/08/ AP3-2Telekonferenzen_20190821.pdf Deutsche Kultusministerkonferenz. (2016). Bildung in der digitalen Welt. Strategie de Kultusministerkonferenz. Retrieved June 6, 2020, from www.kmk.org/fileadmin/Dateien/pdf Deutsche Kultusministerkonferenz. (2016). Bildung in der digitalen Welt. Strategie der Kultusministerkonferenz. Retrieved June 6, 2020, from www.kmk.org/fileadmin/Dateien/pdf/ PresseUndAktuelles/2016/Bildung_digitale_Welt_Webversion.pdf PresseUndAktuelles/2016/Bildung_digitale_Welt_Webversion.pdf Eickelmann, B. (2010). Digitale Medien in Schule und Unterricht erfolgreich implementiere Eine empirische Analyse aus Sicht der Schulentwicklungsforschung. Münster: Waxmann. European Commission. (2019). The European green deal. Retrieved June 6, 2020, from https:// ec.europa.eu/info/sites/info/files/european-green-deal-communication_en.pdf 11 Sustainability in a Digital Age as a Trigger for Organizational Development… 11 Sustainability in a Digital Age as a Trigger for Organizational Development… 201 11 Franzen, J. (2020). Wann hören wir auf, uns etwas vorzumachen? Hamburg: Rowohlt. Franzen, J. (2020). Wann hören wir auf, uns etwas vorzumachen? Hamburg: Rowohlt Fullan, M. (1999). Die Schule als lernendes Unternehmen. Stuttgart: Klett-Cotta. Fullan, M. (1999). Die Schule als lernendes Unternehmen. Stuttgart: Klett-Cotta. Grünberger, N., & Münte-Goussar, S. (2017). ‹Medienbildung in der Schule› oder ‹Schule im Medium›. MedienPädagogik: Zeitschrift für Theorie und Praxis der Medienbildung; Heft 27: Tagungsband: Spannungsfelder und blinde Flecken. Medienpädagogik zwischen Emanzipationsanspruch und Diskursvermeidung. https://doi.org/10.21240/ mpaed/27/2017.04.05.X Han, B.-C. (2013). Im Schwarm. Ansichten des Digitalen (4. unveränderte Auflage). References Berlin: Matthes uns Seitz. Harper, D. (2001a). Sustain. In Online etymology dictionary. Retrieved June 6, 2020, from https:// www.etymonline.com/search?q=sustain Harper, D. (2001b). Sustainability. In Online etymology dictionary. Retrieved June 6, 2020, from https://www.etymonline.com/word/sustainability Hird, G. (2010). Green IT in practice: How one company is approaching the greening of it IT. Retrieved June 6, 2020, from http://search.ebscohost.com/login.aspx?direct=true&db=nleb k&AN=391092&site=ehost-live Jörissen, B., & Verständig, D. (2017). Code, Software und Subjekt. Zur Relevanz der Critical Software Studies für eine nicht-reduktionistisches Verständnis „digitaler Bildung“. In R. Biermann, & D. Verständig (Hrsg.), Das umkämpfte Netz. Macht- und medienbildungstheo retische Analysen zum Digitalen (S. 37–S. 50). Wiesbaden: Springer VS. Kerres, M. (2018). Mediendidaktik: Konzeption und Entwicklung digitaler Lerna erweiterte Auflage., 5th expanded edition). Boston, Massachusetts: De Gruyter. Klafki, W. (2007). Neue Studien zur Bildungstheorie und Didaktik. Zeitgemässe Allgemeinbildun und kritisch-konstruktive Didaktik (6. neu ausgestattete Aufl.). Weinheim und Basel: Beltz. Klein, N. (2018). Computergestützter Kulturkolonialismus: Die Ökonomie zur Nutzung vieler Sprachen in Datensammlungen und im Internet. südostasien – Zeitschrift für Politik • Kultur • Dialog. Retrieved June 6, 2020, from https://doi.org/10.11588/SOA.2005.4.5680 Klein, N. (2019). Warum nur ein Green New Deal unseren Planeten retten kann. Hamburg: Hoffmann und Campe. Kluge, F. (2011). Etymologisches Wörterbuch der deutschen Sprache (25. durchges. und bearbe itete Aufl. (bearbeitet von Seebold, Elma)). Berlin: DeGuryter. Koselleck, R. (1973). Kritik und Krise: Eine Studie zur Pathogenese der bürgerlichen Welt (11 Aufl.). Berlin: Suhrkamp Verlag. O’Mahony, P. (2014). Europe, crisis, and critique: Social theory and transnational society. European Journal of Social Theory, 17(3), 238–257. Retrieved 6 June 2020, from https://doi. org/10.1177/1368431014530931 p f org/10.1177/1368431014530931 Österreichische Bundesregierung. (2020). Aus Verantwortung für Österreich. Regierungsprogramm 2020–2024. Retrieved June 6, 2020, from https://gruene.at/themen/demokratie-verfassung/ regierungsuebereinkommen-tuerkis-gruen/regierungsuebereinkommen.pdf Parikka, J. (2015). A geology of media. Minneapolis/London: Electronic Mediations Peters, K. E. (2018). Living with the wicked problem of climate change. Zygon, 53(2), 427–442. Retrieved 6 June 2020, from https://doi.org/10.1111/zygo.12400 Peters, K. E. (2018). Living with the wicked problem of climate chan Retrieved 6 June 2020, from https://doi.org/10.1111/zygo.12400 Petry, T. (2019). Digital Leadership: Erfolgreiches Führen in Zeiten der Digital Economy (2 Auflage 2019). Haufe. Pfeifer, W. (1989). Etymologisches Wörterbuch des Deutschen (Bd. 2). Berlin: Akademie-Verlag. Ribot, J. (2019). Social causality of our common climate crisis. Towards a sociodicy for the anthro pocene. In T. Haller, T. Breu, T. De Moor, C. Rohr, & H. Znoj (Hrsg.), The commons in a Glocal world: Global connections and local responses (S. 34–S. 53). London, New York: Routledge, Taylor & Francis Group. Schiefner-Rohs, M. (2017). Medienbildung in der Schule. Blinde Flecken und Spannungsfelder in einer Kultur der Digitalität. MedienPädagogik. Zeitschrift für Theorie und Praxis der Medienbildung, 27(Tagungsband: Spannungsfelder und blinde Flecken. Medienpädagogik N. Grünberger and P. Szucsich 202 zwischen Emanzipationsanspruch und Diskursvermeidung), 153–172. Retrieved 6 June 2020, from http://www.medienpaed.com/article/view/594/557 zwischen Emanzipationsanspruch und Diskursvermeidung), 153–172. Retrieved 6 June 2020, from http://www.medienpaed.com/article/view/594/557 Schlicht, J., & Slepcevic-Zach, P. (2016). Research-Based Learning und Service Learning als Varianten problembasierten Lernens. ZFHE, JG. 11(Nr. 3), 85–105. Retrieved June 6, 2020, from https://www.zfhe.at/index.php/zfhe/article/view/948 Schneider-Taylor, B. (2009). Krise als Kairos im Bildungsprozess. Einige Anmerkungen zu den nachhaltig-krisenhaften Momenten in der Gangstruktur des Lernens. In R. Kubac, C. Rabl, & E. Sattler (Hrsg.), Weitermachen? Einsätze theoretischer Erziehungswissenschaft (S. 103–S. 111). Würzburg: Königshausen & Neumann.l Schratz, M., & Steiner-Löffler, U. (1999). Die Lernende Schule. Arbeitsbuch pädagogische Schulentwicklung. Weinheim und Basel: Beltz. Sporer, T., & Bremer, C. (2016). Offene Bildungsressourcen für das Lernen durch Verantwortung in Schule, Hochschule und Zivilgesellschaft. In J. Wachtler, M. Ebner, O. Gröblinger, M. Kopp, & E. Bratengeyer (Hrsg.), Digitale Medien: Zusammenarbeit in der Bildung (S. 356–S. 357). Retrieved June 6, 2020, from https://www.waxmann.com/fileadmin/media/ zusatztexte/3490Volltext.pdf zusatztexte/3490Volltext.pdf The Shift Project. (2019). Lean ICT. Towards digital sobriety. Report of the working group directed by Hugues Ferreboeuf. Retrieved June 6, 2020, from https://theshiftproject.org/wp- content/uploads/2019/03/Lean-ICT-Report_The-Shift-Project_2019.pdf Thiébaud, E., Hilty, L. M., Schluep, M., Böni, H. W., & Faulstich, M. (2018). Where do our resources go? Indium, neodymium, and gold flows connected to the use of electronic equip ment in Switzerland. Sustainability, 10(8), 2658. Retrieved 6 June 2020, from https://doi. org/10.3390/su10082658 Tulodziecki, G., Grafe, S., & Herzig, B. (2018). Medienbildung in Schule und Unterricht Grundlagen und Beispiele (1.). Stuttgart: UTB GmbH. Umweltbundesamt Deutschland. (2019). Digitalisierung nachhaltig gestalten. Ein Impulspapier des Umweltbundesamtes. Retrieved June 6, 2020, from https://www.umweltbundesamt.de/ sites/default/files/medien/376/publikationen/uba_fachbroschuere_digitalisierung_nachhaltig_ gestalten_0.pdf United Nations. (2015). Sustainable development goals. Retrieved 24 January 2020, from https:// www.un.org/sustainabledevelopment/sustainable-development-goals/ WBGU (Wissenschaftlicher Beirat für Globale Umweltveränderungen der Deutschen Bundesregierung). (2018). Digitalisierung Worüber wir jetzt reden müssen. Retrieved 6 June 2020, from https://www.wbgu.de/fileadmin/user_upload/wbgu/publikationen/factsheets/digi- talisierung.pdf WBGU (Wissenschaftlicher Beitrat für Globale Umweltveränderungen der Deutschen Bundesregierung). (2019). Unsere gemeinsame digitale Zukunft—Empfehlungen. Retrieved 6 June 2020, from https://www.wbgu.de/fileadmin/user_upload/wbgu/publikationen/haupt- gutachten/hg2019/pdf/WBGU_HGD2019_Empfehlungen.pdf É Wenger, É. (1998). Communities of practice: Learning, meaning, and identity. New York: Cambridge University Press. Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made. The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
https://openalex.org/W3163357366	https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0267704&type=printable	English	null	Detection of infiltrating fibroblasts by single-cell transcriptomics in human kidney allografts	bioRxiv (Cold Spring Harbor Laboratory)	2,020	cc-by	15,933	OPEN ACCESS Citation: Suryawanshi H, Yang H, Lubetzky M, Morozov P, Lagman M, Thareja G, et al. (2022) Detection of infiltrating fibroblasts by single-cell transcriptomics in human kidney allografts. PLoS ONE 17(6): e0267704. https://doi.org/10.1371/ journal.pone.0267704 RESEARCH ARTICLE Detection of infiltrating fibroblasts by single- cell transcriptomics in human kidney allografts Hemant Suryawanshi1, Hua Yang2, Michelle Lubetzky2,3, Pavel Morozov1, Mila Lagman2, Gaurav TharejaID4, Alicia Alonso5, Carol Li2, Catherine Snopkowski2, Aziz Belkadi4, Franco B. Mueller2, John R. Lee2,3, Darshana M. Dadhania2,3, Steven P. Salvatore6, Surya V. Seshan6, Vijay K. Sharma2, Karsten Suhre4, Manikkam Suthanthiran2,3, Thomas Tuschl1, Thangamani MuthukumarID2,3* a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 1 Laboratory of RNA Molecular Biology, The Rockefeller University, New York, NY, United States of America, 2 Division of Nephrology and Hypertension, Department of Medicine, Weill Cornell Medical College, New York, NY, United States of America, 3 Department of Transplantation Medicine, New York Presbyterian Hospital-Weill Cornell Medical College, New York, NY, United States of America, 4 Department of Physiology and Biophysics, Weill Cornell Medical College in Qatar, Doha, Qatar, 5 Epigenomics Core Facility, Weill Cornell Medical College, New York, NY, United States of America, 6 Division of Renal Pathology, Department of Pathology and Laboratory Medicine, Weill Cornell Medical College, New York, NY, United States of America * hsuryawans@rockefeller.edu (HS); ttuschl@rockefeller.edu (TT); mut9002@med.cornell.edu (TM) PLOS ONE RESEARCH ARTICLE PLOS ONE PLOS ONE Abstract We tested the hypothesis that single-cell RNA-sequencing (scRNA-seq) analysis of human kidney allograft biopsies will reveal distinct cell types and states and yield insights to deci- pher the complex heterogeneity of alloimmune injury. We selected 3 biopsies of kidney cor- tex from 3 individuals for scRNA-seq and processed them fresh using an identical protocol on the 10x Chromium platform; (i) HK: native kidney biopsy from a living donor, (ii) AK1: allo- graft kidney with transplant glomerulopathy, tubulointerstitial fibrosis, and worsening graft function, and (iii) AK2: allograft kidney after successful treatment of active antibody-medi- ated rejection. We did not study T-cell-mediated rejections. We generated 7217 high-quality single cell transcriptomes. Taking advantage of the recipient-donor sex mismatches revealed by X and Y chromosome autosomal gene expression, we determined that in AK1 with fibrosis, 42 months after transplantation, more than half of the kidney allograft fibro- blasts were recipient-derived and therefore likely migratory and graft infiltrative, whereas in AK2 without fibrosis, 84 months after transplantation, most fibroblasts were donor-organ- derived. Furthermore, AK1 was enriched for tubular progenitor cells overexpressing profi- brotic extracellular matrix genes. AK2, eight months after successful treatment of rejection, contained plasmablast cells with high expression of immunoglobulins, endothelial cell elabo- ration of T cell chemoattractant cytokines, and persistent presence of cytotoxic T cells. In addition to these key findings, our analysis revealed unique cell types and states in the kid- ney. Altogether, single-cell transcriptomics yielded novel mechanistic insights, which could pave the way for individualizing the care of transplant recipients. Introduction Molecular approaches complement conventional histopathology and have propelled precision transplantation medicine to the bedside [1–3]. Single-cell RNA-sequencing (scRNA-seq) pro- vides hitherto unavailable opportunities to study cell types and cell states at an unprecedented level of precision [4–6]. Our goal was to investigate the utility of scRNA-seq at an individual patient level to address important conundrums in clinical transplantation. Given the complex heterogeneity of alloimmune rejection, we tested the hypothesis that single-cell transcrip- tomics—by enabling molecular phenotyping of the host infiltrating cells and donor parenchy- mal cells—will yield novel mechanistic insights, especially in the context of antibody-mediated injury, for individualizing the care of transplant recipients. Immune rejection of the allograft remains a significant challenge despite the use of potent immunosuppressive drugs [7–9]. Rejection episodes restrict the benefits of transplantation and negatively impact long-term kidney allograft survival [10]. Treatment of rejection is con- strained by the limited therapeutic armamentarium focused predominantly on the adaptive arm of the immune system and despite improvement in clinical and laboratory parameters, seldom achieves histological remission [10, 11]. Also, despite anti-rejection therapy, it is possi- ble that allograft injury persists at a molecular level and perpetuates allograft dysfunction. It is tempting to speculate that effective treatment of the lingering immune injury may improve the long-term outcome of kidney transplant recipients. This, however, requires better understand- ing of the complex immune interactions between the recipient genome and the genome of the organ donor. We studied two clinico-pathological scenarios: (i) chronic persistent tissue injury and wors- ening allograft function and (ii) resolved acute tissue injury following successful treatment of an episode of active antibody-mediated rejection. These results were compared to the single- cell transcriptomes of cells isolated from a native kidney used for living-donor kidney trans- plantation. We did not study T-cell-mediated rejection. We resolved 12 clusters of major cell types at the first level of single-cell gene expression analysis, with a subset of cell clusters fur- ther resolved by subclustering analysis. We identified 4 distinct fibroblast subpopulations dif- ferentially present in the biopsies and made the surprising finding that one fibroblast subtype in the transplant biopsies was kidney-recipient rather donor-derived. We also identified tubu- lar progenitor cells with profibrotic gene signature. Finally, the transcriptomes of endothelial cell subtypes provided additional insights into the anti-allograft response. Editor: Alain Haziot, INSERM, FRANCE Editor: Alain Haziot, INSERM, FRANCE Received: May 30, 2021 Accepted: April 13, 2022 Published: June 3, 2022 Copyright: © 2022 Suryawanshi et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: The RNA sequencing data files have been deposited at NCBI’s Gene Expression Omnibus under the accession number GSE151671. Funding: Supported in part by awards from the: National Institutes of Health -NIH MERIT Award, R37-AI051652 to M. Suthanthiran, -K08- DK087824 and R03-DK105270 to T. Muthukumar, -NCATS Clinical and Translational Science Award, UL1TR000457 to Weill Cornell Medical College) Mendez National Institute of Transplantation Foundation to M. Suthanthiran. 1 / 30 PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 PLOS ONE Infiltrating fibroblasts in human kidney allografts Competing interests: The authors have declared that no competing interests exist. Competing interests: The authors have declared that no competing interests exist. PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 Competing interests: The authors have declared that no competing interests exist. Tissue collection, dissociation, and single-cell preparation We followed a standard operating procedure for performing kidney allograft biopsies to obtain samples for scRNA-seq. Tissue samples were collected under local anesthesia by real-time ultrasound guidance using an 18g Bard Monopty automated spring-loaded biopsy gun, and a Civco Ultra-Pro II in-plane needle guide attached to the ultrasound probe to prevent any con- tamination by tissues other than kidney. The presence of kidney cortical parenchyma without the presence of kidney medulla, kidney capsule, or any extra-renal tissue was verified by exam- ination of the biopsy tissue under the microscope. Native kidney needle biopsy was obtained from a kidney donor in the operating room during the back-table preparation of the kidney prior to its implantation in the recipient. The biopsies were transported in phosphate-buffered saline on ice to our Gene Expression Monitoring laboratory and immediately dissociated for single-cell capture. We developed and used an in-house protocol for single-cell suspension preparation. In brief, the sample was placed in 400 μl of freshly prepared tissue dissociation PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 2 / 30 PLOS ONE Infiltrating fibroblasts in human kidney allografts solution comprised of 100 μl Liberase TL solution (2 mg/ml, Sigma-Aldrich), 500 μl Tyrode’s solution-HEPES-based (Boston BioProducts), and 200 μl DNase I solution (1 mg/ml, Stemcell technologies) and incubated at 37˚C water bath for 15 min. The cell suspension was passed through a 40 μm Falcon™cell strainer (ThermoFisher Scientific) into a 50 ml centrifuge tube filled with 5 ml fetal bovine serum (ThermoFisher Scientific), washed through with Dulbecco’s phosphate-buffered saline (ThermoFisher Scientific), and centrifuged for 5 min at 300g. Cells were resuspended in 2% bovine serum albumin (New England BioLabs) and were transferred immediately in ice to the genomics core laboratory. Single-cell capture, library preparation, sequencing, data processing, and generation of gene expression matrix Single-cell suspension on ice was immediately transferred to the Weill Cornell Medicine geno- mics core facility. scRNA-seq libraries were prepared using the Chromium Single Cell 3’ Reagent Kit V2 (10x Genomics) according to the manufacturer’s instructions. The library was sequenced on Illumina HiSeq 2500 platform as follows: 26 bp (Read1) and 98 bp (Read2). The sequencing was performed to obtain 150–200 million reads (each for Read1 and Read2). The 10x raw data were processed individually for 3 kidney samples using previously described Drop-seq pipeline (Drop-seq core computational protocol V1.2, http://mccarrolllab.com/ dropseq/) with the following parameters. The Read1 bases 1–16 was tagged with cell barcode ‘XC’ and bases 17–26 was tagged with a unique molecular identifier (UMI) ‘XM’. Low-quality bases containing reads were removed and the 3’-end of Read2 was trimmed to remove poly(A) sequences of six or more bases and were aligned to human genome (hg38) reference using STAR aligner (STAR_2.5.1a), allowing no more than three mismatches. The gene expression matrix was then generated using the ‘MIN_BC_READ_THRESHOLD = 2’ option to retain UMIs with read evidence of two or more. PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 scRNA-seq data analysis Cell clustering analysis was performed collectively for all 3 samples using ‘Seurat V3.1’, an R package for exploration of single-cell data. Briefly, only those genes that were expressed in more than three cells and cells that expressed more than 200 but less than 5000 genes were retained. Ubiquitously expressed genes such as ribosomal protein-coding (RPS and RPL) and non-coding RNA (MALAT1) genes were removed. We also removed miRNA and snoRNA genes from clustering analysis. The clustering analysis was performed in an iterative manner where in round 1 the cells with >25% mitochondrial content was identified and removed from round 2 clustering analysis. In round 2 analysis, mitochondrially coded genes (MT-) were also dropped from clustering analysis. We first generated three separate objects for AK1, AK2, and HK samples using ‘CreateSeuratObject’ function. Next, we merged these objects into one object using ‘merge’ function. We separated endothelial, epithelial, immune and stromal cells into their individual Seurat objects and conducted at least two rounds of analysis on each to identify and remove doublet captures. Typically, doublet capture show expression of mark- ers of two different lineages (e.g., T cells and epithelial cells). Normalization was performed applying ‘logNormalize’ function of Seurat. Using function ‘FindVariableFeatures’ ~2000 genes were identified, followed by ‘ScaleData’. Next, ‘RunPCA’, ‘FindNeighbors’, ‘FindClus- ters’, and ‘RunUMAP’ was used with different dimensions 1:10 and resolution of 0.5 wherever these options were needed. Reduction method “umap” was used and clusters were plotted using ‘DimPlot’ option. To identify differentially expressed genes by each cluster, Wilcoxon rank sum test inbuilt in the Seurat package was used with parameters ‘min.pct = 0.25’ and ‘thresh.use = 0.25’. 3 / 30 PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 PLOS ONE Infiltrating fibroblasts in human kidney allografts Subclustering analyses of the cell groups described in this manuscript were performed using similar strategy as described above. The expression of established lineage marker genes was used to assign cell types. Once the cell types were identified, average expression was calcu- lated for each cell type followed by normalization to 10,000 to create a TPM (transcript per million)-like value. For donor/recipient origin of cell analysis, we quantified the expression of female-specific gene (XIST) and male-specific Y chromosome autosomal genes (RPS4Y, EIF1AY, DDX3Y) as previously described [12–16]. We created separate Seurat objects for HK, AK1, and AK2 cells maintaining the original cell type identity. scRNA-seq data analysis Using ‘DotPlot’ function, we plotted the expression of female-specific gene (XIST) and male-specific Y chromosome autosomal genes (RPS4Y, EIF1AY, DDX3Y) [13–19]. We verified the results of the donor/recipient origin of cell analysis by also clustering cells based on genotype. Raw sequencing data from each sample was individ- ually aligned using 10x Genomics Cell Ranger 6.1.1 to human genome reference (GRCh38) downloaded from 10x Genomics (https://cf.10xgenomics.com/supp/cell-exp/refdata-gex- GRCh38-2020-A.tar.gz) using default parameters [20]. The output BAM and barcode file were used as inputs for ‘Singularity’ [21] version of ‘Souporcell’ pipeline [22]. The samples were individually processed with default parameters of pipeline with k = 2. The cluster file was fur- ther filtered to include only singlet cell cluster assignment. The cluster data was merged into Seurat object [23] for clustered fibroblast cells in R version 4.1.0 [24]. For cells of the peritubular capillaries (PTC), in order to perform differential gene expres- sion analysis by samples, we could not employ the conventional differential gene expression analysis methods such as DESeq2 or edgeR, due to the small number of samples. For this anal- ysis, we only allowed genes that had >2 TPM expression in at least one of the samples (HK, AK1, or AK2). Next, for each gene, we calculated the largest (maximal) and second largest TPM expression values across all three samples. Finally, only the genes with at least two-fold difference between maximal and second largest TPM were reported. For the assessment of immunoglobulin expression in plasmablast cells, we mapped all reads to the Ensembl transcriptome with STAR and subjected the top 10 immunoglobulin genes to unsupervised clustering using R packages stats v.1.0.3 and gplots v.3.1.1. Normalized expression profiles of PT, PG1 and PG2 were generated and used for matri- some analysis. We utilized a previously described list of matrisome genes and then subset the list for each category of matrisomes (collagen, proteoglycan, and secreted factor) from the TPM expression profiles of PT, PG1, and PG2 cell types. Log2(TPM+1) values for top expressed genes in each of the matrisome category were represented in the heatmap. To authenticate our annotation of cells, we used ‘Azimuth’ to map our data to an annotated reference dataset [23]. We mapped our data of 7,217 high quality cells from the three samples to the reference consisting of 64,693 kidney cells generated in the Human Biomolecular Atlas Program (HuBMAP) and the Kidney Precision Medicine Project (KPMP) [25]. Clinical characteristics and kidney cortical biopsy specimens A summary of the clinical and histopathological characteristics of the healthy kidney donor and the two kidney transplant recipients is provided in Table 1. The two allograft biopsies included transplant glomerulopathy with graft dysfunction and follow up after active anti- body-mediated rejection with normal graft function. Biopsy HK was obtained from a healthy 40-year-old living kidney donor, in the operating room at the time of kidney donation. The recipient who received kidney from this healthy living donor had immediate graft function and normal serum creatinine at 12 months after transplantation with no major infections or acute rejection. Biopsy AK1 was obtained by real-time ultrasound guidance from a 51-year-old woman. She previously developed end-stage kidney disease of her native kidneys due to lupus nephritis and received a living-donor kidney transplant. She had worsening of proteinuria after the liv- ing-donor kidney transplantation and an allograft biopsy 26 months after transplantation Table 1. Clinical characteristics of the three subjects whose kidney biopsies were used for scRNA-seq. Study approval The patients that we describe herein underwent living donor kidney transplantation and were followed up at our center—New York Presbyterian Hospital/Weill Cornell Medicine. Donor nephrectomies and recipient transplant surgeries were both done at our center and no organs or tissues were procured from individuals who were incarcerated. Patients provided written informed consent to participate in the study and the informed consent was obtained prior to their inclusion in the study. The research protocol was approved by the Weill Cornell Medi- cine Institutional Review Board (protocol number: 1404015008). The clinical and research activities that we report here are consistent with the principles of the “Declaration of Istanbul PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 4 / 30 PLOS ONE Infiltrating fibroblasts in human kidney allografts on Organ Trafficking and Transplant Tourism”[26]. This study did not generate new unique reagents. We have deposited our scRNA-seq data files, compliant with the single cell version of MIAME/MINSEQE standards [27], at NCBI’s Gene Expression Omnibus under the accession number GSE151671. (Continued) PLOS ONE Infiltrating fibroblasts in human kidney allografts Table 1. (Continued) Clinical information Kidney biopsiesa Native kidney [HK]b Allograft kidney #1 [AK1 (TG w/ graft dysfunction)]c Allograft kidney #2 [AK2 (aABMR f/u)]d Index biopsy used for scRNA-seqh Time, transplant to index biopsy Maintenance immunosuppression Time, prior biopsy to index biopsy Serum creatinine, mg/dl Serum tacrolimus trough, ng/dl Urine albumin:creatinine ratio, mg/g Index biopsy findingsa Index biopsy Banff lesion scoresk Donor-specific anti-HLA antibodies at index biopsy Mean fluorescence intensity value 0 months (at donation) No inflammation No tubulointerstitial fibrosis g0, ptc0, i0, t0, v0, cg0, cptc0, ci0, ct0, C4d- 42 months Tacrolimus/Mycophenolate/ Prednisone 16 months 2.63 4.6 4.8 Severe microvascular inflammation Severe Transplant glomerulopathy Moderate tubulointerstitial fibrosisl g3, ptc3, i1, t0, v0, cg3, cptc0, ci2, ct2, C4d- Not detected Class I—0 Class II—0 84 months Tacrolimus/Mycophenolate/ Prednisone 8 months 1.69 7.8 0.22 Mild microvascular inflammation No Transplant glomerulopathy No tubulointerstitial fibrosism g0, ptc1, i1, t0, v0, cg0, cptc0, ci0, ct0, C4d- Detected Class I—1839 against Cw5 Class II—10512 against DQ4 From index biopsy to last follow uph Time, index biopsy to last follow up Serum creatinine, mg/dl Urine albumin:creatinine ratio, mg/g Donor-specific anti-HLA antibodies Mean fluorescence intensity value Maintenance immunosuppression 12 months 1.14 21 Not detected Class I—0 Class II—0 Tacrolimus/Mycophenolate 12 months Graft failure, Initiated on dialysis - Not detected Class I—0 Class II—0 None 12 months 1.64 0.09 Detected Class I—2228 against Cw5 Class II—5361 against DQ4 Tacrolimus/Mycophenolate/Prednisone uorescence, and electron microscopy. Sections for light microscopy were stained with hematoxylin and eosin, periodic acid– a:Biopsies were evaluated by light, immunofluorescence, and electron microscopy. Sections for light microscopy were stained with hematoxylin and eosin, periodic acid– Schiff, and Masson’s trichrome. Immunohistochemistry of SV40 large T antigen was done to identify nuclear inclusion bodies of polyoma virus. Presence of immunoglobulins and complement proteins including complement factor 4 degradation product d (C4d) were evaluated by immunofluorescence microscopy. Presence of glomerular capillary basement membrane double contours and peritubular capillary basement multilayering were assessed by electron microscopy. Each patient provided a single biopsy sample for this study. All three biopsies were done using an 18g size Bard1 Monopty1 disposable core biopsy instrument (Bard Biopsy, Tempe, AZ). K) was obtained from the healthy kidney donor at the time of transplantation. b:Native kidney biopsy tissue (HK) was obtained from the healthy kidney donor at the time of transplantation. PLOS ONE c:Allograft kidney biopsy tissue #1 (AK1) was obtained at the time of core needle biopsy of the allograft. In an earlier biopsy, chronic tissue injury and remodeling characterized by glomerular capillary basement membrane duplication (transplant glomerulopathy [TG]) was observed. Transplant glomerulopathy represents a form of chronic immune rejection likely mediated by circulating antibodies predominantly directed against the donor HLA. However, circulating IgG antibodies directed against the donor HLA prior to transplant or at any time in the post-transplant period were not detected using the highly sensitive single antigen bead assay. We did not test for non-HLA antibodies. Biopsy AK1 was done for worsening serum creatinine and proteinuria. d:Allograft kidney biopsy tissue #2 (AK2) was obtained at the time of core needle biopsy of the allograft. In an earlier biopsy, active antibody-mediated rejection (aABMR), characterized histologically by inflammation of the microvasculature in the kidney and likely mediated by circulating antibodies predominantly directed against the donor HLA was observed. Kidney allograft dysfunction associated with active AMR was successfully reversed and Biopsy AK2 was done for surveillance purpose. e:Complement-dependent cytotoxicity. f e:Complement-dependent cytotoxicity. f:Flow cytometric cross match was positive for both recipient B cells (auto flow cytometry crossmatch) and kidney donor B cells (donor flow cytometry crossmatch). g:Circulating immunoglobulin G antibodies in the transplant recipient directed against one or more donor HLA were measured using the highly sensitive single antigen bead assay on a Luminex platform. Mean fluorescence intensity is a measure of the degree of saturation of target antigens present on a single bead by antibodies and is used as a surrogate for the level of antibody titers. A mean fluorescence intensity value of 2000 is considered positive for the presence of antibodies. h:The index biopsy was the kidney allograft biopsy tissue sample used for scRNA-seq. i:The prior biopsy in the 51 years old female recipient had transplant glomerulopathy (chronic glomerulopathy [cg] score >0) and microvascular inflammation but did not fulfil the Banff criteria for antibody-mediated rejection. The prior biopsy in the 29 years old male recipient was categorized as active antibody-mediated rejection, based on microvascular inflammation (MVI), positive peritubular capillary staining for complement split product 4d (C4d), and the presence of circulating antibodies directed against the donor HLA (DSA). Clinical characteristics and kidney cortical biopsy specimens Clinical information Kidney biopsiesa Native kidney [HK]b Allograft kidney #1 [AK1 (TG w/ graft dysfunction)]c Allograft kidney #2 [AK2 (aABMR f/u)]d Donor characteristics Living/Deceased Relation of the donor to recipient Age at donation Gender Race/Ethnicity Living Sister 40 years Female White Living Cousin 49 years Male Hispanic Living Brother 28 years Female African American Recipient characteristics Age at Transplant Gender Race/Ethnicity Native kidney disease 50 Male White Diabetic nephropathy 51 years Female Hispanic Lupus nephritis 29 years Male African American Focal segmental glomerulosclerosis At the time of transplantation HLA-ABDR matching CDC cross matche Flowcytometry cross match Donor-specific anti-HLA antibodiesg Mean fluorescence intensity value Induction immunosuppression Maintenance immunosuppression 1-Haploidentical Negative Not done Not Detected Class I—0 Class II—0 Basiliximab Tacrolimus/Mycophenolate 1-Antigen mismatch Negative Donor and Auto B cell Positivef Not Detected Class I—0 Class II—0 Thymoglobulin Tacrolimus/Mycophenolate/Prednisone 1-Haploidentical Negative Negative Detected HLA Class I—2573 against Cw5 HLA Class II MFI—0 Thymoglobulin Tacrolimus/Mycophenolate From transplantation to index biopsyh Delayed graft function BK virus/Cytomegalovirus viremia Number of biopsies prior to index biopsy Time, transplant to prior biopsy Serum creatinine at prior biopsy, mg/dl Urine albumin:creatinine ratio at prior biopsy, mg/g Prior biopsy findingsa Prior biopsy Banff lesion scoresk Donor-specific anti-HLA antibodies at prior biopsy Mean fluorescence intensity value No No 1 26 months 1.11 1.2 Severe microvascular inflammation Severe transplant glomerulopathy No tubulointerstitial fibrosisi g3, ptc2, i0, t0, cg3, cptc0, ci0, ct0, C4d- Not Detected Class I—0 Class II—0 No No 1 76 months 2.16 0.97 Active antibody-mediated rejection No transplant glomerulopathy No tubulointerstitial fibrosisj g3, ptc3, i1, t0, cg0, cptc0, ci0, ct0, C4d+ Detected Class I—2760 against Cw5 Class II—17672 against DQ4 (Continued) al characteristics of the three subjects whose kidney biopsies were used for scRNA-seq. Table 1. Clinical characteristics of the three subjects whose kidney biopsies were used for scRNA-seq. (Continued) PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 5 / 30 PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 https://doi.org/10.1371/journal.pone.0267704.t001 PLOS ONE Banff chronic lesions include chronic glomerulopathy (cg), peritubular capillary basement membrane multilayering (cptc), interstitial fibrosis (ci), tubular atrophy (ct), chronic vascular lesions (cv), and arteriolar hyaline thickening (ah). Peritubular capillary staining for complement split product 4d (C4d) by immunofluorescence was negative. h //d i /10 13 1/j l 026 04 001 Fig 1. Histopathological characteristics of kidney biopsies. Figure depicts the histopathological characteristics of the kidney biopsies used for scRNA-seq, HK, AK1 and AK2. Allograft biopsies, AK1 and AK2, were evaluated by light, immunofluorescence, and electron microscopy. Colors represent semi-quantitative Banff lesion scores from 0 to 3. Banff acute lesions include glomerular inflammation (g), peritubular capillary inflammation (ptc), tubular inflammation (t), interstitial inflammation (i), and vascular inflammation (v). Banff chronic lesions include chronic glomerulopathy (cg), peritubular capillary basement membrane multilayering (cptc), interstitial fibrosis (ci), tubular atrophy (ct), chronic vascular lesions (cv), and arteriolar hyaline thickening (ah). Peritubular capillary staining for complement split product 4d (C4d) by immunofluorescence was negative. https://doi.org/10.1371/journal.pone.0267704.g001 https://doi.org/10.1371/journal.pone.0267704.g001 revealed interstitial fibrosis and tubular atrophy, microvascular inflammation, and transplant glomerulopathy. The biopsy, however, did not fulfill the Banff criteria (an international stan- dardized criteria for reporting allograft biopsies) for antibody-mediated rejection and there was no evidence for recurrence of lupus nephritis. Circulating immunoglobulin G antibodies directed against the donor HLA prior to transplant or at any time in the post-transplant period were not detected using the highly sensitive single-antigen bead assay. We did not test for non- HLA antibodies. Before transplant, she had a negative T and B cell complement-dependent cytotoxicity cross match and a negative T cell flowcytometry crossmatch. However, she had a positive flowcytometry crossmatch to the B cells of her kidney donor (donor flow cytometry B cell crossmatch) and to her own B cells (auto flow cytometry B cell crossmatch). The index biopsy that was used for scRNA-seq was done for worsening proteinuria and kidney function 16 months after the initial biopsy (Fig 1 and S1 Fig in S1 File). Subject AK2 had end stage kidney disease due to focal and segmental glomerulosclerosis. He developed acute elevation of serum creatinine, 76 months after a living-donor kidney transplantation and allograft biopsy revealed active antibody-mediated rejection (Banff cate- gory 2) with no chronic glomerular or tubulointerstitial changes. He was treated with our transplant center protocol comprised of methylprednisolone, plasmapheresis, intravenous immunoglobulin, and bortezomib and had resolution of graft dysfunction. PLOS ONE The index biopsy that was used for scRNA-seq was done by the treating physician for surveillance purpose 8 months after the acute rejection episode (Fig 1 and S2 Fig in S1 File). He had normal graft function (serum creatinine <2 mg/dl and albuminuria <500 mg/day) at the time of the index biopsy but had persistent circulating IgG antibodies directed against the donor HLA. PLOS ONE j:The prior biopsy in the 29 years old male recipient was categorized as active antibody-mediated rejection, based on microvascular inflammation (MVI), positive peritubular capillary staining for complement split product 4d (C4d), and the presence of circulating antibodies directed against the donor HLA (DSA). k:Biopsies were reported—based on the Banff 2017 update of the Banff ‘97 classification of allograft pathology[2]—independently by two transplant pathologists at our center who were blinded to the sequencing data. k:Biopsies were reported—based on the Banff 2017 update of the Banff ‘97 classification of allograft pathology[2]—independently by two transplant pathologists at our center who were blinded to the sequencing data. recipient had severe transplant glomerulopathy (chronic glomerulopathy [cg] score >0) and severe microvascular eria for chronic active antibody-mediated rejection. l:The index biopsy in the 51 years old female recipient had severe transplant glomerulopathy (chronic glomerulopathy [cg] s inflammation but did not fulfil the Banff criteria for chronic active antibody-mediated rejection. rs old male recipient had mild microvascular inflammation and did not fulfil the criteria for active antibody-mediated rejecti ndex biopsy in the 29 years old male recipient had mild microvascular inflammation and did not fulfil the criteria for active a https://doi.org/10.1371/journal.pone.0267704.t001 6 / 30 PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 PLOS ONE Infiltrating fibroblasts in human kidney allografts Fig 1. Histopathological characteristics of kidney biopsies. Figure depicts the histopathological characteristics of the kidney biopsies used for scRNA-seq, HK, AK1 and AK2. Allograft biopsies, AK1 and AK2, were evaluated by light, immunofluorescence, and electron microscopy. Colors represent semi-quantitative Banff lesion scores from 0 to 3. Banff acute lesions include glomerular inflammation (g), peritubular capillary inflammation (ptc), tubular inflammation (t), interstitial inflammation (i), and vascular inflammation (v). Banff chronic lesions include chronic glomerulopathy (cg), peritubular capillary basement membrane multilayering (cptc), interstitial fibrosis (ci), tubular atrophy (ct), chronic vascular lesions (cv), and arteriolar hyaline thickening (ah). Peritubular capillary staining for complement split product 4d (C4d) by immunofluorescence was negative. https://doi.org/10.1371/journal.pone.0267704.g001 Fig 1. Histopathological characteristics of kidney biopsies. Figure depicts the histopathological characteristics of the kidney biopsies used for scRNA-seq, HK, AK1 and AK2. Allograft biopsies, AK1 and AK2, were evaluated by light, immunofluorescence, and electron microscopy. Colors represent semi-quantitative Banff lesion scores from 0 to 3. Banff acute lesions include glomerular inflammation (g), peritubular capillary inflammation (ptc), tubular inflammation (t), interstitial inflammation (i), and vascular inflammation (v). Identification of distinct cell types in healthy and allograft kidneys We conducted iterative cell clustering analysis using Seurat V3.1, an R package for exploration of single-cell data [28]. We obtained 9762 cells; 2545 cells with >25% mitochondrial content was removed from subsequent analysis. The final single-cell gene expression matrices for the three kidney biopsies were comprised of 7217 high-quality cells and separated into 12 cell clus- ters by gene expression (Fig 2A, left) with no contributions from batch processing (Fig 2A, 7 / 30 PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 PLOS ONE Infiltrating fibroblasts in human kidney allografts Fig 2. scRNA-seq applied to kidney allografts differentiated cell types and resolved recipient and donor origin based on sex-specific gene expression. (A). Uniform manifold approximation and projection (UMAP)-based visualization of 7217 individual cells obtained from the three kidney biopsy tissues. Left Panel: UMAP-based visualization in which different clusters represent different cell types. PT1 and PT2, proximal tubular cells 1 and 2; PG, progenitor cells; LH.CD.IC, cluster of loop of Henle, collecting duct, and intercalated cells; FB, fibroblasts; EC, endothelial cells; PC.vSMC, cluster of pericytes and vascular smooth muscle cells; TC, T lymphocytes; NK, natural killer cells; BC.PLASMA, cluster of B lymphocytes and plasmablast cells; MAC.DC, cluster of macrophages and dendritic cells; and MONO, monocytes. Right Panel: UMAP-based visualization of the same cell clusters shown in the left panel in which the cells are colored by the samples. HK-healthy kidney biopsy tissue, AK1 and AK2-allograft kidney biopsy tissues. (B). Dot-plot showing expression of known lineage markers. The size of the dot represents the proportion of cells within each cluster expressing the marker. The intensity of the color represents the standard score for each marker across different cell clusters. (C). Dot-plot showing the annotation of donor/recipient origin of cells in each sample based on female (XIST) and Y chromosome (RPS4Y1, EIF1AY, and DDX3Y)-specific gene expression patterns. HK is a female donor kidney; AK1 is a female recipient of kid f l d AK2 i l i i t f kid f f l d XIST i f l ifi ll d XIST t i ti Fig 2. scRNA-seq applied to kidney allografts differentiated cell types and resolved recipient and donor origin based on sex-specific gene expression. (A). Uniform manifold approximation and projection (UMAP)-based visualization of 7217 individual cells obtained from the three kidney biopsy tissues. Left Panel: UMAP-based visualization in which different clusters represent different cell types. https://doi.org/10.1371/journal.pone.0267704.g002 Sex differences between kidney recipient and donor reveal migratory graft- infiltrating cells AK1 was a female recipient of a kidney from a male donor and AK2 was a male recipient of a kidney from a female donor. We took advantage of the sex mismatch between kidney recipi- ents and their donors and monitored the expression of male-specific Y chromosome-encoded RPS4Y, EIF1AY, and DDX3Y, and, and female-specific XIST, involved in female X chromo- some inactivation, to determine recipient or donor origin of the cells in the allograft [13]. We first separated the cell clusters by individual samples and then assigned each cluster to either female or male origin based on the expression pattern of XIST, RPS4Y1, EIF1AY, and DDX3Y, and genes. HK biopsy was obtained from a female kidney donor and as expected, the cell clus- ters expressed XIST and were devoid for expression of RPS4Y1, EIF1AY, and DDX3Y indicat- ing their female origin (Fig 2C). In accord with AK1 being a female recipient of male kidney, all graft infiltrating immune cell groups, TC, NK, BC.PLASMA, and MAC.DC were of female origin and matched the sex of the allograft recipient. In accord with AK2 being a male recipient of female kidney, the graft infiltrating immune cell types were of male origin matching the sex of the recipient. In contrast to the infiltrating cells, the AK1 kidney parenchymal cells were male and matched the sex of the organ donor, while AK2 kidney parenchymal cells were female but also matched the donor. Importantly, XIST RNA expression was not universal across all female parenchymal cells as exemplified by absence in PT1 and PT2 in AK2 and HK female kidneys. The absence or dramatically reduced abundance of XIST in female PT cells is a surprising biological phe- nomenon, given that XIST RNA was captured across all other female origin cell types. g p g yp Unexpectedly, most of the FBs identified in male kidney AK1 were of female recipient ori- gin (like the infiltrated immune cells), indicating that these FBs are migratory in nature and their presence in the allograft is by infiltration of cells from the recipient (S2 Table in S1 File). The AK1 kidney biopsy had a Banff chronic lesion score of 2 for interstitial fibrosis (ci score), defined by interstitial fibrosis involving 26–50% of cortical area. Identification of distinct cell types in healthy and allograft kidneys PT1 and PT2, proximal tubular cells 1 and 2; PG, progenitor cells; LH.CD.IC, cluster of loop of Henle, collecting duct, and intercalated cells; FB, fibroblasts; EC, endothelial cells; PC.vSMC, cluster of pericytes and vascular smooth muscle cells; TC, T lymphocytes; NK, natural killer cells; BC.PLASMA, cluster of B lymphocytes and plasmablast cells; MAC.DC, cluster of macrophages and dendritic cells; and MONO, monocytes. Right Panel: UMAP-based visualization of the same cell clusters shown in the left panel in which the cells are colored by the samples. HK-healthy kidney biopsy tissue, AK1 and AK2-allograft kidney biopsy tissues. (B). Dot-plot showing expression of known lineage markers. The size of the dot represents the proportion of cells within each cluster expressing the marker. The intensity of the color represents the standard score for each marker across different cell clusters. (C). Dot-plot showing the annotation of donor/recipient origin of cells in each sample based on female (XIST) and Y chromosome (RPS4Y1, EIF1AY, and DDX3Y)-specific gene expression patterns. HK is a female donor kidney; AK1 is a female recipient of a kidney from a male donor; AK2 is a male recipient of a kidney from a female donor. XIST is a female-specifically expressed gene. XIST gene transcription produces X-inactive specific transcript (Xist) RNA, a non-coding RNA, which is a major effector of the X chromosome inactivation. The Xist RNA is expressed only on the inactive chromosome and not on the active chromosome. Males (XY), who have only one X chromosome that is active, do not express it. Females (XX), who have one active and one inactive X chromosome, express it. In HK biopsy (female kidney), all the cells in the kidney express XIST and none express the Y chromosome markers. In AK1 biopsy (male donor and female recipient), all the kidney parenchymal cells express Y chromosome markers whereas all the recipient-derived immune infiltrating cells express XIST. In AK2 biopsy (female donor and male recipient), all the kidney parenchymal cells express XIST whereas all the recipient-derived immune infiltrating cells express the Y chromosome markers. https://doi.org/10.1371/journal.pone.0267704.g002 https://doi.org/10.1371/journal.pone.0267704.g002 8 / 30 PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 PLOS ONE Infiltrating fibroblasts in human kidney allografts right). Sex differences between kidney recipient and donor reveal migratory graft- infiltrating cells This finding is particularly striking considering that FBs in AK2, where the biopsy had a Banff ci score of 0, defined by interstitial fibrosis involving 5% of cortical area, largely matched the donor (like the kidney parenchymal cells) and not the recipient. Identification of distinct cell types in healthy and allograft kidneys Based on differential gene expression and previously established markers of cell types or states we designated these clusters as proximal tubular cells (PT1 and PT2), tubular progenitor cells (PG), loop of Henle cells, collecting duct cells, and intercalated cells (LH.CD.IC), fibro- blasts (FB), endothelial cells (EC), pericytes and vascular smooth muscle cells (PC.vSMC), T cells (TC), natural killer cells (NK), B cells and plasmablast cells (BC.PLASMA), macrophages and dendritic cells (MAC.DC), and monocytes (MONO) (Fig 2B and S1 Table in S1 File). Dur- ing subclustering analysis, we removed 113 endothelial cell and pericyte doublets as well as 26 epithelial and T cell doublet. Comparative analysis of fibroblast-specific gene expression in healthy and allograft kidneys To further resolve FB subtypes, we performed FB subclustering analysis and identified FB1 to FB4 (Fig 3A, top), all of which expressed the canonical fibroblast marker DCN (Fig 3B). The subtypes distinguish disease (FB1 and FB4) and healthy samples (FB2 and FB3) (Fig 3A, bot- tom). FB1 was comprised of all sex-mismatched (recipient-derived) ‘migratory’ FBs (migrating from the recipient/host to the donor/kidney allograft) (Fig 4A–4C) whereas FB4 was com- prised of sex-matched (donor-derived) ‘transplant-organ-originating’ fibroblasts of AK1 and PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 9 / 30 PLOS ONE Infiltrating fibroblasts in human kidney allografts Fig 3. Sub-clustering of fibroblasts revealed four distinct subtypes. (A). UMAP-based visualization of subpopulations of fibroblasts. In the top panel the fibroblasts are colored by different sub-populations. In the bottom panel the cells are colored by the biopsies HK, AK1 and AK2. (B). Violin plots depict the expression of the lineage gene markers. Fig 3. Sub-clustering of fibroblasts revealed four distinct subtypes. (A). UMAP-based visualization of subpopulations of fibroblasts. In the top panel the fibroblasts are colored by different sub-populations. In the bottom panel the cells are colored by the biopsies HK, AK1 and AK2. (B). Violin plots depict the expression of the lineage gene markers. https://doi.org/10.1371/journal.pone.0267704.g003 AK2 biopsies. FB3 and FB4 were defined by expression of GGT5 and EMILIN1, markers recently reported in healthy kidney biopsy to represent interstitial fibroblasts [29]. In addition, these FBs expressed ACTA2, indicative of myofibroblast (mFB)-like characteristics [29]. FB4 uniquely expressed POSTN, another mFB marker recently defined in a cell subpopulation in human kidneys expressing the most ECM genes [30]. Furthermore, FB4 cells uniquely expressed TNC, COL4A1, COL18A1, and TGM2, genes involved in beta-1 integrin cell surface interactions and cell adhesion to the extracellular matrix. The fibroblast subpopulations FB2 and FB3 are dominated by cells from HK and abundantly express cellular stress response genes, including JUN and FOSB, commonly observed in scRNA-seq due to stress from mechanical cell dissociation and refrigeration [31]. The stress signature in HK may be more pronounced as the biopsy was collected during the back-table preparation of the kidney when it was on ice and perfused with ice cold solution prior to implantation. FB1 and FB2 expressed the secretory factors CFD, SFRP2 and SFRP4 in addition to genes implicated in promoting cell migration such as MFAP5, S100A4 and S100A10 [32]. The infil- trating FB1 cells uniquely expressed FBN1, IFI27, WISP2 and PLA2G2A. PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 Comparative analysis of fibroblast-specific gene expression in healthy and allograft kidneys The migratory FB1 expression signature was devoid of lymphoid or myeloid cell lineage markers suggesting that these cells are either not derived from the hematopoietic lineage or their expression is lost in the target (donor) kidney. The absence of markers of other cell lineages also indicates these are not rare cell doublets in incompletely dissociated cells (Fig 5). In addition, using an R package ‘DoubletFinder’ [33]—a computational doublet detection tool, we confirmed that there were no cell doublets in this migratory FBs. The migratory FB1 cells were comprised predominantly of female recipient-derived FB from AK1 biopsy. We also detected a single male recipient- derived migratory FB in the AK2 biopsy (Fig 4B and 4C), indicating that fibroblast migration is more universal. Clustering of the FB cells based on their genotype matched their clustering based on donor/recipient sex and confirmed the presence of recipient-derived migratory fibro- blasts (Fig 4D). As expected, our FB cells mapped with a high prediction score to fibroblasts in PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 10 / 30 PLOS ONE Infiltrating fibroblasts in human kidney allografts A C B D Fig 4. Visualization of fibroblast subpopulations clustered based on donor/recipient sex and clustered based on their genotype. (A). Fibroblast subpopulations. (B). Fibroblast subpopulations visualized based on their expression of female-specific gene XIST. (C). Fibroblast subpopulations visualized based on their expression of male-specific Y chromosome autosomal gene RPS4Y1. (D). Fibroblast subpopulations visualized based on their genotype clusters. https://doi.org/10.1371/journal.pone.0267704.g004 A C C B D D B Fig 4. Visualization of fibroblast subpopulations clustered based on donor/recipient sex and clustered based on their genotype. (A). Fibroblast subpopulations. (B). Fibroblast subpopulations visualized based on their expression of female-specific gene XIST. (C). Fibroblast subpopulations visualized based on their expression of male-specific Y chromosome autosomal gene RPS4Y1. (D). Fibroblast subpopulations visualized based on their genotype clusters. https://doi.org/10.1371/journal.pone.0267704.g004 https://doi.org/10.1371/journal.pone.0267704.g004 the kidney reference dataset generated in the HuBMAP and KPMP projects (S3 Fig in S1 File). Also, the expression of our FB lineage markers as well as the uniquely expressed genes of our migratory FB cells were largely restricted in the kidney reference dataset to FB cells (S4 Fig in S1 File). Importantly, they did not map to any immune cells in the reference dataset, confirm- ing the authenticity of our FB annotation. Both AK1 and AK2 biopsies contained exclusively kidney cortex and were not contaminated by kidney capsule or other extra-kidney tissues. PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 Validation of the four subpopulations of fibroblasts Having defined four subpopulations of FBs including the migratory FB1, we compared our fibroblast subtype expression pattern with those of recently published kidney scRNA-seq stud- ies (Fig 6). Subpopulation F3 and F4 expressing GGT5 and EMILIN1 (Fig 6A), were similar to kidney FBs (Fig 6B) observed in the study by Wu et al. [29] comprising a kidney allograft biopsy with histologic diagnosis of acute T-cell–mediated rejection with plasma cells and acute C4d-negative ABMR as well as a healthy kidney tissue biopsy from a discarded human donor kidney. Kidney allograft tissue was processed as single-cell and healthy kidney tissue was PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 11 / 30 PLOS ONE Infiltrating fibroblasts in human kidney allografts Fig 5. Expression of cell lineage markers in the fibroblast subpopulations. Figure depicts the violin plots showing expression of cell lineage markers in the fibroblast subpopulations. No lymphoid or myeloid cell lineage markers were expressed in the fibroblasts. https://doi.org/10.1371/journal.pone.0267704.g005 Fi 5 E i f ll li k i th fib bl t b l ti Fi d i t th i li l t h i i f ll li k i th Fig 5. Expression of cell lineage markers in the fibroblast subpopulations. Figure depicts the violin plots showing expression of cell lineage markers in the fibroblast subpopulations. No lymphoid or myeloid cell lineage markers were expressed in the fibroblasts. https://doi.org/10.1371/journal.pone.0267704.g005 Fig 5. Expression of cell lineage markers in the fibroblast subpopulations. Figure depicts the violin plots showing expression of cell lineage markers in the fibroblast subpopulations. No lymphoid or myeloid cell lineage markers were expressed in the fibroblasts. https://doi.org/10.1371/journal.pone.0267704.g005 processed as single-nucleus RNA-seq. In addition, our FB4 subtype expressed PDGFRA, PDGFRB and POSTN noted in ‘high-ECM expressing’ transitioning FB to mFB (Fig 6C) reported by Kuppe et al. [30]. The latter study included human kidney tissue from patients with normal kidney function or chronic kidney disease (CKD) due to hypertensive nephro- sclerosis undergoing partial nephrectomy because of kidney cancer. In a recent study by Valenzi et al. [34] of human lung tissue from healthy controls and from patients with systemic sclerosis-associated interstitial lung disease, two major FB subpopulations were identified, including an MFAP5hi population in the lung with systemic sclerosis-associated interstitial lung disease (Fig 6D). PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 Validation of the four subpopulations of fibroblasts The expression profile of our migratory FB1 fibroblasts strikingly matches with this MFAP5hi FBs sharing expression of MFAP5, PLA2G2A, SLPI, CD34, and THY1 genes (Fig 6A and 6D). We next compared the expression profile of FB1 to FB4 to FBs detected in skin biopsies of patients with atopic dermatitis and normal subjects [35]. All four subpopulations in the kidney, including the migratory FB1 subpopulation, were distinctly dif- ferent from skin FBs (S5 Fig in S1 File). processed as single-nucleus RNA-seq. In addition, our FB4 subtype expressed PDGFRA, PDGFRB and POSTN noted in ‘high-ECM expressing’ transitioning FB to mFB (Fig 6C) reported by Kuppe et al. [30]. The latter study included human kidney tissue from patients with normal kidney function or chronic kidney disease (CKD) due to hypertensive nephro- sclerosis undergoing partial nephrectomy because of kidney cancer. In a recent study by Valenzi et al. [34] of human lung tissue from healthy controls and from patients with systemic sclerosis-associated interstitial lung disease, two major FB subpopulations were identified, including an MFAP5hi population in the lung with systemic sclerosis-associated interstitial lung disease (Fig 6D). The expression profile of our migratory FB1 fibroblasts strikingly matches with this MFAP5hi FBs sharing expression of MFAP5, PLA2G2A, SLPI, CD34, and THY1 genes (Fig 6A and 6D). We next compared the expression profile of FB1 to FB4 to FBs detected in skin biopsies of patients with atopic dermatitis and normal subjects [35]. All four subpopulations in the kidney, including the migratory FB1 subpopulation, were distinctly dif- ferent from skin FBs (S5 Fig in S1 File). 12 / 30 PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 PLOS ONE Infiltrating fibroblasts in human kidney allografts Fig 6. FB1-FB4 subpopulations compared with FBs described in recently published human scRNA-seq studies. (A). Violin plots depicting the expression of marker genes for FB1 to FB4 subpopulations. (B). FBs in a kidney allograft with rejection [29]. (C). FBs in kidneys with CKD due to hypertensive nephrosclerosis [30]. (D). FBs in lungs with systemic sclerosis-associated interstitial lung disease [34]. https://doi.org/10.1371/journal.pone.0267704.g006 studies. (A). Violin plots depicting the expression Fig 6. FB1-FB4 subpopulations compared with FBs described in recently published human scRNA-seq studies. (A). Violin plots depicting the expression of marker genes for FB1 to FB4 subpopulations. (B). FBs in a kidney allograft with rejection [29]. (C). FBs in kidneys with CKD due to hypertensive nephrosclerosis [30]. (D). Validation of the four subpopulations of fibroblasts FBs in lungs with systemic sclerosis-associated interstitial lung disease [34]. https://doi.org/10.1371/journal.pone.0267704.g006 Pericytes and vascular smooth muscle subtypes in the kidney The vascular smooth muscle cell and pericyte cell types were further resolved into six subpopu- lations: vSMC1 to 4 and PC1 and 2 (Figs 7 and 8, S3 Table in S1 File). The pericytes were char- acterized by PDGFRA-negative, PDGFRB-high, and RGS5-high signature (Fig 7B). vSMC1 and 2, were dominated by the HK biopsy, and showed abundant expression of stress-induced genes (relatively more in vSMC1 than vSMC2) such as JUNB and FOSB (Fig 7B). Interestingly, vSMC3 and 4, contributed by AK1 and AK2 biopsies, respectively, differentiated from the PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 13 / 30 PLOS ONE Infiltrating fibroblasts in human kidney allografts Fig 7. Sub-clustering of pericytes and vascular smooth cells. (A). UMAP-based visualization of subpopulations of pericytes and vascular smooth muscle cells. In the left panels the pericytes and vascular smooth muscle cells are colored by different sub-populations. In the right panel the cells are colored by the biopsies HK, AK1 and AK2. (B). Violin plots of the expression of lineage gene markers of pericytes and vascular smooth muscle cells. https://doi.org/10.1371/journal.pone.0267704.g007 Fig 7. Sub-clustering of pericytes and vascular smooth cells. (A). UMAP-based visualization of subpopulations of pericytes and vascular smooth muscle cells. In the left panels the pericytes and vascular smooth muscle cells are colored by different sub-populations. In the right panel the cells are colored by the biopsies HK, AK1 and AK2. (B). Violin plots of the expression of lineage gene markers of pericytes and vascular smooth muscle cells. https://doi.org/10.1371/journal.pone.0267704.g007 https://doi.org/10.1371/journal.pone.0267704.g007 healthy vSMC1 and 2 by additional expression of cardiac muscle alpha actin ACTC1 likely induced due to inflammation. vSMC3 was further characterized by induction of NNMT, con- sidered as master metabolic regulator and contributor to secretion of cytokines and oncogenic extracellular matrix of cancer-associated fibroblasts [36]. The expression of NNMT also in PTC cells was unique to the AK1 allograft with high interstitial fibrosis. The only additional difference between vSMC3 and vSMC4 was the expression of male specific EIF1AY. A higher level of stress response genes including JUNB and FOSB separated PC1 from PC2. In addition, PC2 cells expressed THY1, S100A4, and CCL2 and exclusively originated from HK cells. PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 Tubular progenitor cell populations are increased in allograft kidneys Reclustering analysis of the epithelial cell types PT1, PT2, PG, and LH.CD.IC yielded 6 sub- populations (Fig 9A and 9B). A single proximal tubular (PT) cell cluster was defined by expres- sion of MIOX, ANPEP, and SLC13A1 genes (Fig 9C). Tubular progenitor (PG) cells expressing PROM1, CD24, and VIM, however, separated into a major (PG1) and a minor (PG2) popula- tion. Fibrotic AK1 biopsy contributed the most to PGs (PG1 79.3% and PG2 55%) (Fig 9D). Heterogeneity in the proportion of kidney parenchymal cells among the three samples could be due sampling variability and should be interpreted with caution. PGs lack brush borders, are scattered throughout proximal tubules in the normal kidney and become more numerous and participate in tubular regeneration after acute tubular injury [37, 38]. We note that PGs also expressed CDH2 (N-cadherin), a known marker for epithelial-mesenchymal transition (EMT) [39]. The co-expression of proximal tubular cell marker genes in PG1 suggests a more differentiated state compared to PG2. Furthermore, PG1 and PG2 are characterized by expres- sion of injury markers HAVCR1, LCN2, and MYC, which are absent in PT (Fig 10). The key histological feature of AK1 kidney was interstitial fibrosis. To examine whether PG abundance and gene expression contributed to the fibrosis, we performed matrisome (the ensemble of extracellular matrix [ECM] and ECM-associated proteins) enrichment analysis [40, 41] of the PT and PG subtypes (Fig 9E) and also compared it to those of FB cells from our study (Fig 11). Most of the expressed collagen genes including COL4A1, COL4A2, and COL1A1, were selectively enriched in PGs and not or minimally expressed in PT cells. Simi- larly, the second category of matrisome genes ‘proteoglycans’ also showed enrichment in PGs and not in PT. In addition, secreted factors including the S100A family and cytokines such as CCL2, CXCL1, and CXCL6 were abundant in PGs. The other epithelial cells of the kidney such as LH, CD, and IC-A were identified by unique expression of marker genes UMOD, AQP2, and ATP6V0D2, respectively. 14 / 30 PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 PLOS ONE Infiltrating fibroblasts in human kidney allografts Fig 8. Differential gene expression in pericytes and vascular smooth muscle cells. Figure depicts heatmap of differentially expressed genes in pericytes and vascular smooth muscle cells. https://doi org/10 1371/journal pone 0267704 g008 Fig 8. Differential gene expression in pericytes and vascular smooth muscle cells. Tubular progenitor cell populations are increased in allograft kidneys Figure depicts heatmap of differentially Fig 8. Differential gene expression in pericytes and vascular smooth muscle cells. Figure depicts heatmap of differential expressed genes in pericytes and vascular smooth muscle cells. https://doi.org/10.1371/journal.pone.0267704.g008 15 / 30 PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 PLOS ONE Infiltrating fibroblasts in human kidney allografts Fig 9. Epithelial cell sub-clustering revealed collagen-producing tubular progenitor cells. (A). UMAP-based visualization of epithelial cells colored by different cell types. PT, proximal tubular cells; IC-A, intercalated cells type A; CD, collecting duct cells; LH, loop of Henle cells; PG, progenitor cells. (B). UMAP-based visualization of epithelial cells colored by the biopsies HK, AK1 and AK2. (C). Violin plot showing expression of the lineage gene markers. (D). Stacked bar plots show the proportion of epithelial cells in each sample. The number on the right is the total number of cells. (E). Heatmap showing top expressed genes belonging to categories of matrisome groups. https://doi.org/10.1371/journal.pone.0267704.g009 Fig 9. Epithelial cell sub-clustering revealed collagen-producing tubular progenitor cells. (A). UMAP-based visualization of epithelial cells colored by different cell types. PT, proximal tubular cells; IC-A, intercalated cells type A; CD, collecting duct cells; LH, loop of Henle cells; PG, progenitor cells. (B). UMAP-based visualization of epithelial cells colored by the biopsies HK, AK1 and AK2. (C). Violin plot showing expression of the lineage gene markers. (D). Stacked bar plots show the proportion of epithelial cells in each sample. The number on the right is the total number of cells. (E). Heatmap showing top expressed genes belonging to categories of matrisome groups. https://doi.org/10.1371/journal.pone.0267704.g009 PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 Endothelial cell types in disease biopsies show a chemoattractant cytokine signature Further investigation of endothelial cell types revealed four distinct subtypes; peritubular capillaries (PTC1-3) and descending vasa recta (DVR), primarily based on the distinct expres- sion for PLVAP, AQP1, and SLC14A (Fig 12A and 12B and S3 Table in S1 File). Glomerular endothelial cells did not separate out by this analysis although some ECs co-expressed glomer- ular endothelial markers. All EC subclusters shared expression of canonical markers such as PECAM1 and CDH5. The PTC2 subpopulation, mostly composed of AK2 cells, was character- ized by the unique expression of the structurally and functionally related cytokines CXCL9, CXCL10, and CXCL11 (Fig 12A–12C). These cytokines act as chemoattractants during inflam- mation through binding to the receptor CXCR3 mostly expressed by activated T cells [42]. AK1 and AK2 PTCs showed a >6-fold upregulation of the cell-surface-glycoprotein-encoding SELE gene compared to HK (SELE TPM, HK: 0.6, AK1: 2.6, AK2: 4.1). Under inflammatory conditions, endothelial cells induce expression of SELE in order to facilitate trans-endothelial passage of leukocytes through adhesion to the vascular lining [43]. The PTC3 subpopulation dominated by cells from HK showed higher expression of JUN and FOS gene family (Fig 13), indicative of cold shock response in the biopsy collected of the chilled and perfused donor organ prior to implantation. PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 16 / 30 PLOS ONE Infiltrating fibroblasts in human kidney allografts Fig 10. Analysis of kidney epithelial cell subpopulations. Figure depicts UMAP-based visualization of epithelial cell groups. (A). UMAP-based visualization of epithelial cells. (B). Feature plot showing co-expression of kidney injury markers HAVCR1, LCN2, and MYC. Fig 10. Analysis of kidney epithelial cell subpopulations. Figure depicts UMAP-based visualization of epithelial cell groups. (A). UMAP-based visualization of epithelial cells. (B). Feature plot showing co-expression of kidney injury markers HAVCR1, LCN2, and MYC. https://doi.org/10.1371/journal.pone.0267704.g010 https://doi.org/10.1371/journal.pone.0267704.g010 PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 Immune cell heterogeneity in the healthy and allograft kidneys Subclustering analysis of the immune cell populations identified TC, cytotoxic TC, NK, MONO, MAC, DC, BC, PLASMA, and MAST (mast cells) cell types (Fig 14A–14C). The transplant AK1 and AK2 samples showed higher proportions of immune cell infiltrates (52% and 66%, respectively, vs. 16% in HK of all cells) (Fig 14D). The T cells in HK were dominated by granzyme-K- (GZMK-) producing CD8+ T cells (Fig 14E). Furthermore, AK2 showed a large subpopulation of GZMK-expressing CD8+ cytotoxic T cells, and a smaller subgroup of CD8+ cytotoxic T cells, defined by high expression of granzyme B (GZMB) and perforin (PRF1). We also identified a minor population of central memory T cells in AK2 characterized by expression of CCR7, SELL, and TCF7. All three samples had a small subset of interferon- stimulated-gene- (ISG-) high CD4+ T cells with increased expression of ISG15, MX1, RSAD2, IFIT1, and IFIT2. MACs expressed MS4A4A, STAB1 and SEPP1 typically considered as gene signatures of “alternatively activated” M2 macrophages. The content of NK and MONO in AK1 (0.38% and 0.44%, respectively) compared to AK2 (9.8% and 7.8%, respectively) was remarkably reduced. IRF1 was 10-fold more abundant in MACs of AK2 (TPM 3.1) com- pared to AK1 (TPM 0.3) and HK (TPM 0.1). IRF1 was also abundant in cytotoxic TCs of AK2 (TPM 3.0) compared to AK1 (TPM 1.7) and HK (TPM 1.4). The accumulation of PLASMA (plasmablast cells) of the AK2 sample was striking, reaching about 1.5% of total cells, but were also identified in lower numbers in AK1 and HK. Mapping the reads of PLASMA to the con- stant regions of the light and heavy chain immunoglobulins (IG), we identified many distinct subtypes defined by the strong expression of light chains of IGK, IGL, or IGLL5, in combina- tion with heavy chain IGHG, IGHA, or IGHD. These IG genes were more than an order of magnitude more expressed than abundant housekeeping genes (S4 Table in S1 File). IGHG and IGHA expressing PLASMA were about equal in abundance (Fig 15). Secreted IgA are rec- ognized by epithelial cells expressing PIGR at their basal side and transported and secreted at PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 17 / 30 PLOS ONE Infiltrating fibroblasts in human kidney allografts Fig 11. Top expressed genes of matrisome groups in fibroblasts, proximal tubular, and tubular progenitor cells. Immune cell heterogeneity in the healthy and allograft kidneys Figure depicts heatmap of the top expressed matrisome genes in the proximal tubular cells, tubular progenitor cells, and fibroblasts. https //doi org/10 1371/journal pone 0267704 g011 Fig 11. Top expressed genes of matrisome groups in fibroblasts, proximal tubular, and tubular progenitor cells. Figure depicts heatmap of the top expressed matrisome genes in the proximal tubular cells, tubular progenitor cells, and fibroblasts. Fig 11. Top expressed genes of matrisome groups in fibroblasts, proximal tubular, and tubular progenitor cells. Figure depicts heatmap of the top expressed matrisome genes in the proximal tubular cells, tubular progenitor cells, and fibroblasts. https://doi.org/10.1371/journal.pone.0267704.g011 https://doi.org/10.1371/journal.pone.0267704.g011 https://doi.org/10.1371/journal.pone.0267704.g011 the apical ciliated surface. PIGR expression was detected across most epithelial cell types including CD and PROG1 in AK1, AK2, and HK. the apical ciliated surface. PIGR expression was detected across most epithelial cell types including CD and PROG1 in AK1, AK2, and HK. Discussion The key findings of our single-cell transcriptome analysis include: (i) allograft kidney biopsy AK1, with ongoing tubulointerstitial fibrosis, contained more recipient-derived fibroblasts in contrast to the allograft kidney AK2 biopsy with no fibrosis and almost exclusively donor- derived FBs; (ii) allograft kidney AK1 biopsy also contained most proximal tubular progenitor cells that were enriched in the expression of ECM glycoproteins, collagens, and proteoglycans, PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 18 / 30 PLOS ONE Infiltrating fibroblasts in human kidney allografts Fig 12. Sub-clustering of endothelial cells. (A). UMAP-based visualization of subpopulations of endothelial cells. In the left panel the endothelial cells, are colored by different sub-populations. PTC, peritubular capillaries; DVR, descending vasa recta. In the right panel the cells are colored by the biopsies HK, AK1 and AK2. (B). Violin plots of the expression of lineage markers of endothelial cells. (C). Heatmap shows the differentially expressed genes among the three biopsies in PTC cells (PTC1, PTC2 and PTC3 combined). https://doi.org/10.1371/journal.pone.0267704.g012 Fig 12. Sub-clustering of endothelial cells. (A). UMAP-based visualization of subpopulations of endothelial cells. In the left panel the endothelial cells, are colored by different sub-populations. PTC, peritubular capillaries; DVR, descending vasa recta. In the right panel the cells are colored by the biopsies HK, AK1 and AK2. (B). Violin plots of the expression of lineage markers of endothelial cells. (C). Heatmap shows the differentially expressed genes among the three biopsies in PTC cells (PTC1, PTC2 and PTC3 combined). https://doi.org/10.1371/journal.pone.0267704.g012 https://doi.org/10.1371/journal.pone.0267704.g012 and (iii) allograft kidney AK2 biopsy, eight months after successful treatment of antibody- mediated rejection as defined by clinical and histological criteria, contained endothelial cells expressing T cell chemoattractant cytokines. Determining the frequency of these observations is essential and will require scRNA-seq studies on larger cohorts of similar phenotype. The discovery of migratory recipient-derived fibroblast subtypes in human allograft kidney biopsies is prompting for the investigation of molecular clues leading into recruitment of such migratory FBs to the kidney, the answer of which may hold immense therapeutic implications. In humans and in experimental kidney disease, macrophages can directly transdifferentiate into collagen-producing myofibroblasts [44]. However, the absence of immune cell or other cell lineage markers in our migratory FB1 subpopulation suggests that cells are either not derived from hematopoietic lineage or their expression is lost in the target (donor) kidney. PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 Discussion The scRNA-seq study of human lung tissue from healthy controls and from patients with sys- temic sclerosis-associated interstitial lung disease identified two distinct major subpopulations of FB in the lungs [34]. One of the two subpopulation was called MFAP5hi FB and was charac- terized by high expression of MFAP5, CD34, THY1, SLPI, and PLA2G2A in both healthy and diseased lungs [34]. Our migratory FB1 subpopulation showed high expression of MFAP5, CD34, THY1, SLPI, and PLA2G2A, qualifying as MFAP5hi FBs found in lungs. This migratory FB1 subpopulation expressed fibronectin (FBN1), interferon-alpha-inducible protein 27 (IFI27), WNT1-inducible signaling pathway protein 2 (WISP2), and complement-decay-accel- erating factor (CD55) uniquely. All these genes have been implicated in cancer cell migration and metastasis [45–48]. Increased MFAP5 expression has been reported to stimulate cancer cell motility and invasion and predicted poor survival, and MFAP5 in vivo silencing reduced tumor progression [32]. In addition to MFAP5, migratory FB1 identified in the AK1 biopsy PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 19 / 30 PLOS ONE Infiltrating fibroblasts in human kidney allografts Fig 13. Differential gene expression in endothelial cell subtypes. Figure depicts heatmap of the differentially expressed genes in the endothelial cell subtypes. Fig 13. Differential gene expression in endothelial cell subtypes. Figure depicts heatmap of the differentially expressed genes in the endothelial cell subtypes. https://doi.org/10.1371/journal.pone.0267704.g013 Differential gene expression in endothelial cell subtypes. Figure depicts heatmap of the differentially expressed genes in the elial cell subtypes. Fig 13. Differential gene expression in endothelial cell subtypes. Figure depicts heatmap of the differentially expressed genes in the endothelial cell subtypes. https://doi.org/10.1371/journal.pone.0267704.g013 https://doi.org/10.1371/journal.pone.0267704.g013 https://doi.org/10.1371/journal.pone.0267704.g013 were differentiated from the non-migratory FB4 subpopulation by S100A4 expression in the migratory FBs. Interstitial fibroblasts appear after gastrulation because of EMT from secondary epithelium [49]. Calcium-binding protein S100A4—also called FSP1—likely plays an impor- tant role in facilitating and maintaining EMT phenotype [50]. S100A4 is also expressed in some endosteal lining cells and marrow stromal cells and may be an EMT niche for osteogenic precursor cells, indifferent endosteum, fibroblasts, and marrow stromal cells [50]. Addition- ally, S100A4 promotes metastasis and is associated with intestinal fibroblast migration in PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 20 / 30 PLOS ONE Infiltrating fibroblasts in human kidney allografts Fig 14. Immune cell heterogeneity in the healthy and allograft kidneys. (A). UMAP-based visualization of immune cells colored by different cell types. Discussion TC, T lymphocytes; CTC, cytotoxic T lymphocytes; NK, natural killer cells; MONO, monocytes; MAC, macrophages; DC, dendritic cells; BC, B lymphocytes; PLASMA, plasmablast cells. (B). UMAP-based visualization of immune cells colored by the biopsies HK, AK1 and AK2. (C). Dot-plot showing expression of known lineage gene markers. (D). Pie chart depicting the proportion of cell types in each biopsy sample. The immune cells are labelled in red. (E). Bar graphs showing the different types of T cells in HK, AK1, and AK2. Central memory CD4+ T cells: CCR7+SELL+TCF7+; ISG-high CD4+ T cells: CD4+ISG15+; Cytotoxic CD8+ T cells: CD8A+GZMB+. Fig 14. Immune cell heterogeneity in the healthy and allograft kidneys. (A). UMAP-based visualization of immune cells colored by different cell types. TC, T lymphocytes; CTC, cytotoxic T lymphocytes; NK, natural killer cells; MONO, monocytes; MAC, macrophages; DC, dendritic cells; BC, B lymphocytes; PLASMA, plasmablast cells. (B). UMAP-based visualization of immune cells colored by the biopsies HK, AK1 and AK2. (C). Dot-plot showing expression of known lineage gene markers. (D). Pie chart depicting the proportion of cell types in each biopsy sample. The immune cells are labelled in red. (E). Bar graphs showing the different types of T cells in HK, AK1, and AK2. Central memory CD4+ T cells: CCR7+SELL+TCF7+; ISG-high CD4+ T cells: CD4+ISG15+; Cytotoxic CD8+ T cells: CD8A+GZMB+. https://doi.org/10.1371/journal.pone.0267704.g014 patients with Crohn’s disease [51, 52]. Our observation suggests that the migratory and tissue- invasive nature of the FBs may play a crucial role in fibrosis and complements the prior obser- vation of mesenchymal cells of host origin in the vascular and interstitial compartments of kid- ney allografts undergoing chronic rejection [53]. The discovery of migratory FBs expressing unique genes associated with cell mobility and migration and may provide opportunities for targeted intervention to ameliorate interstitial fibrosis. A recent scRNA-seq analysis of study of human kidney tissue collected from patients with normal kidney function or CKD due to hypertensive nephrosclerosis undergoing partial nephrectomy because of kidney cancer sought to identify the cell types that contribute to the production of extracellular matrix (ECM) in kidney fibrosis [30]. Among the four major cell types in the kidney—mesenchymal, epithelial, endothelial, and immune—mesenchymal cells exhibited the highest extracellular matrix expression in the kidney with minor contribution from epithelial cells. Within the mesenchymal cell type, 3 fibroblasts 2 myofibroblasts, 2 peri- cytes, and 1 vascular smooth muscle cells were identified. PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 Discussion myofibroblasts (mFB) were defined as cells that expressed the most ECM genes and expressed the marker POSTN. Three main sources of mFB were identified—NOTCH3+ RGS5+PDGFRA- pericytes, MEG3+ PDGFRA+ fibroblasts, and COLEC11+CXCL12+ fibroblasts [30]. Using such a definition, our FB4 subpop- ulation qualified as mFB. In both AK1 and AK2 biopsies, FB4 subpopulation expressed mark- ers that suggests contribution from all three sources of mFB. In our native kidney HK biopsy with no fibrosis, both FB2 and FB3 did not express POSTN and were likely not derived from pericytes. 21 / 30 PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 PLOS ONE Infiltrating fibroblasts in human kidney allografts Fig 15. Immunoglobulin gene expression in plasmablast cells. Heatmap depicts the expression of immunoglobulin gene constant regions in plasmablast cells in HK, AK1, and AK2 samples. The X axis shows barcodes of individual plasmablast cells. Fig 15. Immunoglobulin gene expression in plasmablast cells. Heatmap depicts the expression of immunoglobulin gene constant regions in plasmablast cells in HK, AK1, and AK2 samples. The X axis shows barcodes of individual plasmablast cells. Fig 15. Immunoglobulin gene expression in plasmablast cells. Heatmap depicts the expression of immunoglobulin gene constant regions in plasmablast cells in HK, AK1, and AK2 samples. The X axis shows barcodes of individual plasmablast cells. https://doi.org/10.1371/journal.pone.0267704.g015 If mFB may be defined as a cell population that express the most ECM genes and express the marker POSTN, then the source(s) of such cells in the kidney continues to remain contro- versial [54]. These cells were traditionally thought to arise from kidney resident interstitial FB —residual embryonic mesenchymal cells left over from organogenesis [55]. Data from more recent preclinical models suggests that they may originate from EMT [50], endothelial-to-mes- enchymal transition [56],https://www.frontiersin.org/articles/10.3389/fphys.2016.00061/full - B36 pericytes [57], and bone marrow-derived fibroblast precursor fibrocytes [58]. However, their relative contribution is not known and may depend on the nature of injury resulting in fibrosis [59]. Based on our findings, we speculate that kidney resident interstitial FB are the dominant cells that produce ECM in normal healthy kidney. In kidney allografts, in the con- text of alloimmune injury and toxicity due to drugs such as calcineurin inhibitors, there is additional contribution of pericyte-derived mFB to the pool of ECM-producing cells. As allo- graft fibrosis worsens, there is further recruitment of FBs to the pool from the recipient. Discussion However, to our knowledge, endothelial cells as a source of CXCL9,10,11 has not been previously described in human kidney allografts. Such communication of endothe- lial cells through T cell chemoattractant, eight months after successful treatment of antibody- mediated rejection, is striking and highlights the role of donor-derived endothelium in perpetu- ating tissue injury in the presence of circulating antibodies directed against the donor HLA. Importantly, in this recipient, even though there was clinical and histological improvement, anti- bodies directed against the donor HLA persisted in the circulation. Unlike donor and recipient FB chimerism, we did not find donor and recipient immune cell chimerism in the kidney allograft. Our findings of the absence of donor-derived T-lym- phocytes in the kidney allograft agrees with the observation that donor T-lymphocyte number in the allograft decreased as a function of time, in a recent study that utilized whole exome sequencing of donor and recipient DNA, and scRNA-seq of the kidney allograft to identify the origin of each cell [66]. This study also found that recipient-derived macrophages dominate during rejection, and in a single patient who had a biopsy after treatment of rejection—33 months after transplantation, the proportion of donor-derived macrophages increased. In con- tradistinction, the lack of donor macrophages in our post-rejection treatment AK2 biopsy— performed at 84 months after transplantation—suggests the possibility that donor macro- phages in the allograft also decreased as a function of time after transplantation. Persistent immune activity as noted in the AK2 biopsy is frequently observed in follow-up biopsies of kidney transplant recipients treated for active AMR [67]. Despite clinical and biochemical reversal of active AMR in subject AK2, the continued presence of plasmablast cells within the allograft implicated in secreting antibodies directed against donor HLA and the resultant endothelial cell elaboration of T cell chemoattractant cytokines CXCL9,10,11 could explain the persistent presence of cytotoxic T cells. Generally, long-lived plasma cells in the bone marrow are the major source of persistent antibody production [68, 69]. Such long-lived plasma cells— resistant to conventional immunosuppressive therapy—have been shown to localize in inflamed kidneys in lupus mice and patients with lupus [68]. Acute rejection in humans is characterized by intragraft B cells and plasma cells [70, 71], which could be functionally active [72]. Discussion The striking similarity between expression profile of our migratory FB1 and the MFAP5hi FB in the lungs suggests a common origin of these fibroblasts with potential role to affect multiple organs. Confirming our findings is essential and will require scRNA-seq studies on larger cohorts of similar phenotype. In this context, it will be interesting to study allografts with chronic active T-cell-mediated rejection or interstitial fibrosis and tubular atrophy not other- wise specified, and well-functioning allografts over the long term that never had acute rejection to assess whether recipient-derived migratory FBs are operational in those conditions as well. The detection of PROM1- (CD133-) positive tubular progenitor (PG) cells expressing ECM glycoproteins, collagens, and proteoglycans in AK1 biopsy with interstitial fibrosis and tubular atrophy is a novel finding. While the contribution of EMT to kidney fibrosis is a subject of considerable controversy, recent evidence suggests that partial EMT is sufficient to induce the release of fibrogenic cytokines [60, 61]. The expression of CDH2 (N-cadherin) in PGs and not PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 22 / 30 PLOS ONE Infiltrating fibroblasts in human kidney allografts any other tubular epithelial cell type, suggests PGs undergo partial EMT. The expression of injury markers seen in PGs suggest that chronic tubular injury likely induces EMT transition leading to production of extracellular matrix proteins and contribute to kidney fibrosis. Fur- thermore, although PGs express tubular cell markers, they could not be matched to previously established tubular cell subsets. PGs express several S100 proteins, a family of calcium-binding proteins involved in cell apoptosis, migration, proliferation, differentiation, energy metabo- lism, and inflammation. PROM1+ cells are known to be distributed throughout the kidney and capable of expansion and limited self-renewal [62]. Identifying partial EMT in tubular progen- itor cells rekindles the role of tubular cells in perpetuating fibrosis [63]. A subpopulation of endothelial cells in AK2 biopsy expressed mRNA for cytokines CXCL9, CXCL10 and CXCL11 while T cells expressed the cognate receptor CXCR3. Our finding of endo- thelium-derived T-cell chemoattractant cytokines provide a mechanistic basis for the presence of T cells in the AK2 biopsy. Donor endothelium derived CXCL10 as initiator of alloresponse has been previously noted in a cardiac allograft model [64]. In kidney transplant recipients, urinary CXCL10 levels increase during antibody-mediated rejection that is accompanied by microvascu- lar inflammation [65]. PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 Discussion The increased accumulation of plasmablast cells with high expression of immunoglobulins in the AK2 biopsy, 8 months after successful treatment of active AMR, may also contribute to the ongoing production of IgG antibodies directed against donor class I and class II HLA we noted in this patient after successful treatment of active AMR (Table 1). None of our biopsies showed histological features of tertiary lymphoid tissues [73]. To our knowledge, the presence of func- tionally active long-lived plasma cells in successfully treated active AMR of kidney allograft have PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 23 / 30 PLOS ONE Infiltrating fibroblasts in human kidney allografts not been reported. Also, the significance of IgG expressing plasmablast cells in healthy kidney (HK biopsy sample) is unclear. Interestingly, the presence of IgA producing plasmablast cells in all three biopsy samples and the epithelial cell expression of PIGR, that is involved in the dimeric IgA transcytosis, rekindles the role of secretory IgA in the defense against bacterial infections in the kidney [74, 75]. Given the small number of sample and their limited number of infiltrating immune cells further studies are needed to substantiate these early observations. A strength of our study is the rapid processing of freshly collected biopsies without cryo- preservation, albeit on different days. We minimized the time for transfer of the samples from the ultrasound suite or the operating room, where the biopsies were performed, to the research laboratory for generating single-cell suspensions, and from the research laboratory to the genomics core laboratory for library preparation. We used the 10x Chromium platform (10x Genomics) with high cell-capture efficiency and permitting the use of human kidney allograft biopsies with limiting amount of tissue [4]. Earlier reports on scRNA-seq of healthy and dis- eased human kidney tissue, both native and allograft kidney, have included a combination of single-cell and single-nucleus sequencing [29], fresh and frozen specimens [76], multiple plat- forms to capture the single cells [29, 77] and analyzing the transcriptome after combining cells obtained from biopsies from tumor nephrectomies and discarded donor kidneys [78]. Another strength is the use of normal kidney tissue from a living kidney donor, instead of unaffected areas of tumor nephrectomies or kidneys rejected for clinical transplantation. Our study has limitations. The number of study subjects was sparse, and we analyzed only three biopsies. Our findings need to be validated in more biopsies. Discussion Extrapolation from these results is challenging because of the heterogeneity in allograft pathology and tissue sampling depth by core needle biopsy. We did not study chronically active T-cell-mediated rejection or allografts with no acute rejection and stable function several years after transplantation. How- ever, having implemented a standardized protocol for sample accusation, sample preparation, single-cell capture, RNA-seq, and data analysis, we believe that our results are robust and have provided rich mechanistic information. Another limitation is that not every cell type is cap- tured by our whole cell dissociation protocol, in particular podocytes. Nevertheless, we were able to capture and analyze several nephron cell types; continued refinement of tissue process- ing techniques is expected to further improve the types of cells captured. Finally, there are inherent limitations to the scRNA-seq technique, for example, only a fraction of mRNAs pres- ent in cells are captured and converted to cDNA and the tissue dissociation process disrupts tissue architecture and loses relative spatial positioning information of cells. Conclusions We have demonstrated the utility of scRNA-seq in interrogating intragraft events in kidney allografts. Our analysis has revealed unique cell types and cell states in kidney allograft biopsies and confirmed our hypothesis that applying scRNA-seq furthers precision transplantation medicine approaches by providing mechanistic insights and opportunities for drug target and pathway identification at hitherto unknown and unavailable resolution. With improvement in the technology, refinement in computational approaches, and decreasing operational costs, it is possible in the future to apply single-cell transcriptomics to complement conventional histo- pathology, in the clinic, for the idealized care of transplant recipients. Supporting information S1 File. 24 / 30 PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 PLOS ONE Infiltrating fibroblasts in human kidney allografts Author Contributions Conceptualization: Hemant Suryawanshi, Mila Lagman, Manikkam Suthanthiran, Thomas Tuschl, Thangamani Muthukumar. Conceptualization: Hemant Suryawanshi, Mila Lagman, Manikkam Suthanthiran, Thomas Tuschl, Thangamani Muthukumar. Data curation: Hemant Suryawanshi, Pavel Morozov, Alicia Alonso, Thomas Tuschl, Thanga- mani Muthukumar. Formal analysis: Hemant Suryawanshi, Gaurav Thareja, Alicia Alonso, Aziz Belkadi, Steven P. Salvatore, Surya V. Seshan, Karsten Suhre, Thangamani Muthukumar. Funding acquisition: Manikkam Suthanthiran, Thomas Tuschl. Funding acquisition: Manikkam Suthanthiran, Thomas Tuschl. Investigation: Hemant Suryawanshi, Hua Yang, Michelle Lubetzky, Mila Lagman, Alicia Alonso, Carol Li, Catherine Snopkowski, Franco B. Mueller, John R. Lee, Darshana M. Dadhania, Vijay K. Sharma, Thomas Tuschl, Thangamani Muthukumar. Methodology: Hemant Suryawanshi, Hua Yang, Alicia Alonso, Thangamani Muthukumar. Methodology: Hemant Suryawanshi, Hua Yang, Alicia Alonso, Thangamani Muthukumar. Project administration: Manikkam Suthanthiran, Thomas Tuschl. Project administration: Manikkam Suthanthiran, Thomas Tuschl. Resources: Alicia Alonso, Karsten Suhre, Manikkam Suthanthiran, Thomas Tuschl, Thanga- mani Muthukumar. Software: Hemant Suryawanshi, Pavel Morozov, Thomas Tuschl. Software: Hemant Suryawanshi, Pavel Morozov, Thomas Tuschl. Supervision: Thomas Tuschl, Thangamani Muthukumar. Validation: Hemant Suryawanshi, Pavel Morozov, Gaurav Thareja, Aziz Belkadi, Karsten Suhre. Visualization: Hemant Suryawanshi, Pavel Morozov, Gaurav Thareja, Aziz Belkadi, Karsten Suhre. Writing – original draft: Hemant Suryawanshi, Thomas Tuschl, Thangamani Muthukumar. Writing – review & editing: Hemant Suryawanshi, Manikkam Suthanthiran, Thomas Tuschl, Thangamani Muthukumar. Acknowledgments We thank ThuTrang Du, Division Administrator, Division of Nephrology, Weill Cornell Med- ical College, for her help with the execution of this project. We thank A. Hurley and the Research Facilitation Office staff at Rockefeller University for their regulatory and administra- tive assistance. We thank ThuTrang Du, Division Administrator, Division of Nephrology, Weill Cornell Med- ical College, for her help with the execution of this project. We thank A. Hurley and the ical College, for her help with the execution of this project. We thank A. Hurley and the Research Facilitation Office staff at Rockefeller University for their regulatory and administra- tive assistance. PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 3. Naesens M, Anglicheau D. Precision Transplant Medicine: Biomarkers to the Rescue. J Am Soc Nephrol. 2018; 29(1):24–34. Epub 2017/10/11. https://doi.org/10.1681/ASN.2017010004 PMID: 28993504; PubMed Central PMCID: PMC5748900. References Hart A, Smith JM, Skeans MA, Gustafson SK, Wilk AR, Castro S, et al. OPTN/SRTR 2017 Annual Data Report: Kidney. Am J Transplant. 2019; 19 Suppl 2:19–123. Epub 2019/02/28. https://doi.org/10.1111/ ajt.15274 PMID: 30811893. 11. El Ters M, Grande JP, Keddis MT, Rodrigo E, Chopra B, Dean PG, et al. Kidney allograft survival after acute rejection, the value of follow-up biopsies. Am J Transplant. 2013; 13(9):2334–41. Epub 2013/07/ 20. https://doi.org/10.1111/ajt.12370 PMID: 23865852. 12. Skelly DA, Squiers GT, McLellan MA, Bolisetty MT, Robson P, Rosenthal NA, et al. Single-Cell Tran- scriptional Profiling Reveals Cellular Diversity and Intercommunication in the Mouse Heart. Cell Rep. 2018; 22(3):600–10. Epub 2018/01/19. https://doi.org/10.1016/j.celrep.2017.12.072 PMID: 29346760. 13. Suryawanshi H, Morozov P, Straus A, Sahasrabudhe N, Max KEA, Garzia A, et al. A single-cell survey of the human first-trimester placenta and decidua. Sci Adv. 2018; 4(10):eaau4788. Epub 2018/11/08. https://doi.org/10.1126/sciadv.aau4788 PMID: 30402542; PubMed Central PMCID: PMC6209386. 14. Stevant I, Kuhne F, Greenfield A, Chaboissier MC, Dermitzakis ET, Nef S. Dissecting Cell Lineage Specification and Sex Fate Determination in Gonadal Somatic Cells Using Single-Cell Transcriptomics. Cell Rep. 2019; 26(12):3272–83 e3. Epub 2019/03/21. https://doi.org/10.1016/j.celrep.2019.02.069 PMID: 30893600. 15. Zhou L, Adrianto I, Wang J, Wu X, Datta I, Mi QS. Single-Cell RNA-Seq Analysis Uncovers Distinct Functional Human NKT Cell Sub-Populations in Peripheral Blood. Front Cell Dev Biol. 2020; 8:384. Epub 2020/06/13. https://doi.org/10.3389/fcell.2020.00384 PMID: 32528956; PubMed Central PMCID: PMC7264113. 16. Thomas JR, Appios A, Zhao X, Dutkiewicz R, Donde M, Lee CYC, et al. Phenotypic and functional char- acterization of first-trimester human placental macrophages, Hofbauer cells. J Exp Med. 2021; 218(1). Epub 2020/10/20. https://doi.org/10.1084/jem.20200891 PMID: 33075123; PubMed Central PMCID: PMC7579740. 17. Staedtler F, Hartmann N, Letzkus M, Bongiovanni S, Scherer A, Marc P, et al. Robust and tissue-inde- pendent gender-specific transcript biomarkers. Biomarkers. 2013; 18(5):436–45. Epub 2013/07/09. https://doi.org/10.3109/1354750X.2013.811538 PMID: 23829492. 18. Gershoni M, Pietrokovski S. The landscape of sex-differential transcriptome and its consequent selec- tion in human adults. BMC Biol. 2017; 15(1):7. Epub 2017/02/09. https://doi.org/10.1186/s12915-017- 0352-z PMID: 28173793; PubMed Central PMCID: PMC5297171. 19. Zhou Q, Wang T, Leng L, Zheng W, Huang J, Fang F, et al. Single-cell RNA-seq reveals distinct dynamic behavior of sex chromosomes during early human embryogenesis. Mol Reprod Dev. 2019; 86 (7):871–82. Epub 2019/05/17. https://doi.org/10.1002/mrd.23162 PMID: 31094050. 20. Zheng GX, Terry JM, Belgrader P, Ryvkin P, Bent ZW, Wilson R, et al. Massively parallel digital tran- scriptional profiling of single cells. Nat Commun. 2017; 8:14049. Epub 2017/01/17. https://doi.org/10. References 1. Loupy A, Haas M, Solez K, Racusen L, Glotz D, Seron D, et al. The Banff 2015 Kidney Meeting Report: Current Challenges in Rejection Classification and Prospects for Adopting Molecular Pathology. Am J Transplant. 2017; 17(1):28–41. Epub 2016/11/20. https://doi.org/10.1111/ajt.14107 PMID: 27862883; PubMed Central PMCID: PMC5363228. 2. Haas M, Loupy A, Lefaucheur C, Roufosse C, Glotz D, Seron D, et al. The Banff 2017 Kidney Meeting Report: Revised diagnostic criteria for chronic active T cell-mediated rejection, antibody-mediated rejec- tion, and prospects for integrative endpoints for next-generation clinical trials. Am J Transplant. 2018; 18(2):293–307. Epub 2017/12/16. https://doi.org/10.1111/ajt.14625 PMID: 29243394; PubMed Central PMCID: PMC5817248. 2. Haas M, Loupy A, Lefaucheur C, Roufosse C, Glotz D, Seron D, et al. The Banff 2017 Kidney Meeting Report: Revised diagnostic criteria for chronic active T cell-mediated rejection, antibody-mediated rejec- tion, and prospects for integrative endpoints for next-generation clinical trials. Am J Transplant. 2018; 18(2):293–307. Epub 2017/12/16. https://doi.org/10.1111/ajt.14625 PMID: 29243394; PubMed Central PMCID: PMC5817248. 3. Naesens M, Anglicheau D. Precision Transplant Medicine: Biomarkers to the Rescue. J Am Soc Nephrol. 2018; 29(1):24–34. Epub 2017/10/11. https://doi.org/10.1681/ASN.2017010004 PMID: 28993504; PubMed Central PMCID: PMC5748900. 25 / 30 PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 PLOS ONE Infiltrating fibroblasts in human kidney allografts 4. Malone AF, Wu H, Humphreys BD. Bringing Renal Biopsy Interpretation Into the Molecular Age With Single-Cell RNA Sequencing. Semin Nephrol. 2018; 38(1):31–9. Epub 2018/01/03. https://doi.org/10. 1016/j.semnephrol.2017.09.005 PMID: 29291760; PubMed Central PMCID: PMC5753432. 5. Malone AF, Humphreys BD. Single-cell Transcriptomics and Solid Organ Transplantation. Transplanta- tion. 2019; 103(9):1776–82. Epub 2019/04/05. https://doi.org/10.1097/TP.0000000000002725 PMID: 30946217; PubMed Central PMCID: PMC6713581. 6. Varma E, Luo X, Muthukumar T. Dissecting the human kidney allograft transcriptome: single-cell RNA sequencing. Curr Opin Organ Transplant. 2020. Epub 2020/12/15. https://doi.org/10.1097/MOT. 0000000000000840 PMID: 33315769. 7. Cole EH, Johnston O, Rose CL, Gill JS. Impact of acute rejection and new-onset diabetes on long-term transplant graft and patient survival. Clin J Am Soc Nephrol. 2008; 3(3):814–21. Epub 2008/03/07. https://doi.org/10.2215/CJN.04681107 PMID: 18322046; PubMed Central PMCID: PMC2386715. 8. Alkadi MM, Kim J, Aull MJ, Schwartz JE, Lee JR, Watkins A, et al. Kidney allograft failure in the steroid- free immunosuppression era: A matched case-control study. Clin Transplant. 2017; 31(11). Epub 2017/ 09/19. https://doi.org/10.1111/ctr.13117 PMID: 28921709. 9. Clayton PA, McDonald SP, Russ GR, Chadban SJ. Long-Term Outcomes after Acute Rejection in Kid- ney Transplant Recipients: An ANZDATA Analysis. J Am Soc Nephrol. 2019. Epub 2019/07/17. https:// doi.org/10.1681/ASN.2018111101 PMID: 31308074. 10. References 1038/ncomms14049 PMID: 28091601; PubMed Central PMCID: PMC5241818. 21. Kurtzer GM, Sochat V, Bauer MW. Singularity: Scientific containers for mobility of compute. PLoS One. 2017; 12(5):e0177459. Epub 2017/05/12. https://doi.org/10.1371/journal.pone.0177459 PMID: 28494014; PubMed Central PMCID: PMC5426675. 22. Heaton H, Talman AM, Knights A, Imaz M, Gaffney DJ, Durbin R, et al. Souporcell: robust clustering of single-cell RNA-seq data by genotype without reference genotypes. Nat Methods. 2020; 17(6):615–20. Epub 2020/05/06. https://doi.org/10.1038/s41592-020-0820-1 PMID: 32366989. PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 26 / 30 PLOS ONE Infiltrating fibroblasts in human kidney allografts 23. Hao Y, Hao S, Andersen-Nissen E, Mauck WM 3rd, Zheng S, Butler A, et al. Integrated analysis of mul- timodal single-cell data. Cell. 2021; 184(13):3573–87 e29. Epub 2021/06/02. https://doi.org/10.1016/j. cell.2021.04.048 PMID: 34062119; PubMed Central PMCID: PMC8238499. 24. Team RC. R: A language and environment for statistical computing. 2021. 25. Lake BB, Menon R, Winfree S, Hu Q, Ferreira RM, Kalhor K, et al. An atlas of healthy and injured cell states and niches in the human kidney. bioRxiv. 2021:2021.07.28.454201. https://doi.org/10.1101/ 2021.07.28.454201 26. International Summit on Transplant T, Organ T. The Declaration of Istanbul on Organ Trafficking and Transplant Tourism. Clin J Am Soc Nephrol. 2008; 3(5):1227–31. Epub 2008/08/15. https://doi.org/10. 2215/CJN.03320708 PMID: 18701611; PubMed Central PMCID: PMC4571160. 27. Fullgrabe A, George N, Green M, Nejad P, Aronow B, Fexova SK, et al. Guidelines for reporting single- cell RNA-seq experiments. Nat Biotechnol. 2020; 38(12):1384–6. Epub 2020/11/15. https://doi.org/10. 1038/s41587-020-00744-z PMID: 33188371. 28. Stuart T, Butler A, Hoffman P, Hafemeister C, Papalexi E, Mauck WM 3rd, et al. Comprehensive Inte- gration of Single-Cell Data. Cell. 2019; 177(7):1888–902 e21. Epub 2019/06/11. https://doi.org/10. 1016/j.cell.2019.05.031 PMID: 31178118; PubMed Central PMCID: PMC6687398. 29. Wu H, Malone AF, Donnelly EL, Kirita Y, Uchimura K, Ramakrishnan SM, et al. Single-Cell Transcrip- tomics of a Human Kidney Allograft Biopsy Specimen Defines a Diverse Inflammatory Response. J Am Soc Nephrol. 2018; 29(8):2069–80. Epub 2018/07/08. https://doi.org/10.1681/ASN.2018020125 PMID: 29980650; PubMed Central PMCID: PMC6065085. 30. Kuppe C, Ibrahim MM, Kranz J, Zhang X, Ziegler S, Perales-Paton J, et al. Decoding myofibroblast ori- gins in human kidney fibrosis. Nature. 2021; 589(7841):281–6. Epub 2020/11/12. https://doi.org/10. 1038/s41586-020-2941-1 PMID: 33176333. 31. van den Brink SC, Sage F, Vertesy A, Spanjaard B, Peterson-Maduro J, Baron CS, et al. Single-cell sequencing reveals dissociation-induced gene expression in tissue subpopulations. Nat Methods. 2017; 14(10):935–6. Epub 2017/09/30. https://doi.org/10.1038/nmeth.4437 PMID: 28960196. 32. Leung CS, Yeung TL, Yip KP, Pradeep S, Balasubramanian L, Liu J, et al. PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 References Panzer U, Steinmetz OM, Reinking RR, Meyer TN, Fehr S, Schneider A, et al. Compartment-specific expression and function of the chemokine IP-10/CXCL10 in a model of renal endothelial microvascular injury. J Am Soc Nephrol. 2006; 17(2):454–64. Epub 2005/12/31. https://doi.org/10.1681/ASN. 2005040364 PMID: 16382019. 43. Silva M, Videira PA, Sackstein R. E-Selectin Ligands in the Human Mononuclear Phagocyte System: Implications for Infection, Inflammation, and Immunotherapy. Front Immunol. 2017; 8:1878. Epub 2018/ 02/07. https://doi.org/10.3389/fimmu.2017.01878 PMID: 29403469; PubMed Central PMCID: PMC5780348. 44. Wang YY, Jiang H, Pan J, Huang XR, Wang YC, Huang HF, et al. Macrophage-to-Myofibroblast Transi- tion Contributes to Interstitial Fibrosis in Chronic Renal Allograft Injury. J Am Soc Nephrol. 2017; 28 (7):2053–67. Epub 2017/02/18. https://doi.org/10.1681/ASN.2016050573 PMID: 28209809; PubMed Central PMCID: PMC5491278. 45. Gopal S, Veracini L, Grall D, Butori C, Schaub S, Audebert S, et al. Fibronectin-guided migration of car- cinoma collectives. Nat Commun. 2017; 8:14105. Epub 2017/01/19. https://doi.org/10.1038/ ncomms14105 PMID: 28102238; PubMed Central PMCID: PMC5253696. 46. Dho SH, Lim JC, Kim LK. Beyond the Role of CD55 as a Complement Component. Immune Netw. 2018; 18(1):e11. Epub 2018/02/20. https://doi.org/10.4110/in.2018.18.e11 PMID: 29503741; PubMed Central PMCID: PMC5833118. 47. Liu Y, Song Y, Ye M, Hu X, Wang ZP, Zhu X. The emerging role of WISP proteins in tumorigenesis and cancer therapy. J Transl Med. 2019; 17(1):28. Epub 2019/01/16. https://doi.org/10.1186/s12967-019- 1769-7 PMID: 30651114; PubMed Central PMCID: PMC6335850. 48. Xu L, Zu T, Li T, Li M, Mi J, Bai F, et al. ATF3 downmodulates its new targets IFI6 and IFI27 to suppress the growth and migration of tongue squamous cell carcinoma cells. PLoS Genet. 2021; 17(2): e1009283. Epub 2021/02/04. https://doi.org/10.1371/journal.pgen.1009283 PMID: 33539340; PubMed Central PMCID: PMC7888615. 49. Iwano M, Plieth D, Danoff TM, Xue C, Okada H, Neilson EG. Evidence that fibroblasts derive from epi- thelium during tissue fibrosis. J Clin Invest. 2002; 110(3):341–50. Epub 2002/08/07. https://doi.org/10. 1172/JCI15518 PMID: 12163453; PubMed Central PMCID: PMC151091. 50. Kalluri R, Neilson EG. Epithelial-mesenchymal transition and its implications for fibrosis. J Clin Invest. 2003; 112(12):1776–84. Epub 2003/12/18. https://doi.org/10.1172/JCI20530 PMID: 14679171; PubMed Central PMCID: PMC297008. 51. Boye K, Maelandsmo GM. S100A4 and metastasis: a small actor playing many roles. Am J Pathol. 2010; 176(2):528–35. Epub 2009/12/19. https://doi.org/10.2353/ajpath.2010.090526 PMID: 20019188; PubMed Central PMCID: PMC2808059. 52. Cunningham MF, Docherty NG, Burke JP, O’Connell PR. S100A4 expression is increased in stricture fibroblasts from patients with fibrostenosing Crohn’s disease and promotes intestinal fibroblast migra- tion. Am J Physiol Gastrointest Liver Physiol. 2010; 299(2):G457–66. References Calcium-dependent FAK/ CREB/TNNC1 signalling mediates the effect of stromal MFAP5 on ovarian cancer metastatic potential. Nat Commun. 2014; 5:5092. Epub 2014/10/04. https://doi.org/10.1038/ncomms6092 PMID: 25277212; PubMed Central PMCID: PMC4185407. 33. McGinnis CS, Murrow LM, Gartner ZJ. DoubletFinder: Doublet Detection in Single-Cell RNA Sequenc- ing Data Using Artificial Nearest Neighbors. Cell Syst. 2019; 8(4):329–37 e4. Epub 2019/04/08. https:// doi.org/10.1016/j.cels.2019.03.003 PMID: 30954475; PubMed Central PMCID: PMC6853612. 34. Valenzi E, Bulik M, Tabib T, Morse C, Sembrat J, Trejo Bittar H, et al. Single-cell analysis reveals fibro- blast heterogeneity and myofibroblasts in systemic sclerosis-associated interstitial lung disease. Ann Rheum Dis. 2019; 78(10):1379–87. Epub 2019/08/14. https://doi.org/10.1136/annrheumdis-2018- 214865 PMID: 31405848; PubMed Central PMCID: PMC7255436. 35. He H, Suryawanshi H, Morozov P, Gay-Mimbrera J, Del Duca E, Kim HJ, et al. Single-cell transcriptome analysis of human skin identifies novel fibroblast subpopulation and enrichment of immune subsets in atopic dermatitis. J Allergy Clin Immunol. 2020; 145(6):1615–28. Epub 2020/02/10. https://doi.org/10. 1016/j.jaci.2020.01.042 PMID: 32035984. 36. Eckert MA, Coscia F, Chryplewicz A, Chang JW, Hernandez KM, Pan S, et al. Proteomics reveals NNMT as a master metabolic regulator of cancer-associated fibroblasts. Nature. 2019; 569(7758):723– 8. Epub 2019/05/03. https://doi.org/10.1038/s41586-019-1173-8 PMID: 31043742; PubMed Central PMCID: PMC6690743. 37. Lindgren D, Bostrom AK, Nilsson K, Hansson J, Sjolund J, Moller C, et al. Isolation and characterization of progenitor-like cells from human renal proximal tubules. Am J Pathol. 2011; 178(2):828–37. Epub 2011/02/02. https://doi.org/10.1016/j.ajpath.2010.10.026 PMID: 21281815; PubMed Central PMCID: PMC3070548. 38. Smeets B, Boor P, Dijkman H, Sharma SV, Jirak P, Mooren F, et al. Proximal tubular cells contain a phenotypically distinct, scattered cell population involved in tubular regeneration. J Pathol. 2013; 229 (5):645–59. Epub 2012/11/06. https://doi.org/10.1002/path.4125 PMID: 23124355; PubMed Central PMCID: PMC3951144. 39. Wheelock MJ, Shintani Y, Maeda M, Fukumoto Y, Johnson KR. Cadherin switching. J Cell Sci. 2008; 121(Pt 6):727–35. Epub 2008/03/07. https://doi.org/10.1242/jcs.000455 PMID: 18322269. 40. Hynes RO, Naba A. Overview of the matrisome—an inventory of extracellular matrix constituents and functions. Cold Spring Harb Perspect Biol. 2012; 4(1):a004903. Epub 2011/09/23. https://doi.org/10. 1101/cshperspect.a004903 PMID: 21937732; PubMed Central PMCID: PMC3249625. 27 / 30 PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 PLOS ONE Infiltrating fibroblasts in human kidney allografts 41. Suryawanshi H, Clancy R, Morozov P, Halushka MK, Buyon JP, Tuschl T. Cell atlas of the fetal human heart and implications for autoimmune-mediated congenital heart block. Cardiovasc Res. 2019. Epub 2019/10/08. https://doi.org/10.1093/cvr/cvz257 PMID: 31589297. 42. PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 References Epub 2010/05/22. https://doi.org/ 10.1152/ajpgi.00351.2009 PMID: 20489045. 53. Grimm PC, Nickerson P, Jeffery J, Savani RC, Gough J, McKenna RM, et al. Neointimal and tubuloin- terstitial infiltration by recipient mesenchymal cells in chronic renal-allograft rejection. N Engl J Med. 2001; 345(2):93–7. Epub 2001/07/14. https://doi.org/10.1056/NEJM200107123450203 PMID: 11450677. 54. Yan J, Zhang Z, Jia L, Wang Y. Role of Bone Marrow-Derived Fibroblasts in Renal Fibrosis. Front Phy- siol. 2016; 7:61. Epub 2016/03/05. https://doi.org/10.3389/fphys.2016.00061 PMID: 26941655; PubMed Central PMCID: PMC4766307. 55. Picard N, Baum O, Vogetseder A, Kaissling B, Le Hir M. Origin of renal myofibroblasts in the model of unilateral ureter obstruction in the rat. Histochem Cell Biol. 2008; 130(1):141–55. Epub 2008/05/02. https://doi.org/10.1007/s00418-008-0433-8 PMID: 18449560; PubMed Central PMCID: PMC2413112. 56. Zeisberg EM, Potenta SE, Sugimoto H, Zeisberg M, Kalluri R. Fibroblasts in kidney fibrosis emerge via endothelial-to-mesenchymal transition. J Am Soc Nephrol. 2008; 19(12):2282–7. Epub 2008/11/07. https://doi.org/10.1681/ASN.2008050513 PMID: 18987304; PubMed Central PMCID: PMC2588112. 57. Wu CF, Chiang WC, Lai CF, Chang FC, Chen YT, Chou YH, et al. Transforming growth factor beta-1 stimulates profibrotic epithelial signaling to activate pericyte-myofibroblast transition in obstructive kid- ney fibrosis. Am J Pathol. 2013; 182(1):118–31. Epub 2012/11/13. https://doi.org/10.1016/j.ajpath. 2012.09.009 PMID: 23142380; PubMed Central PMCID: PMC3538028. 28 / 30 PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 PLOS ONE Infiltrating fibroblasts in human kidney allografts 58. Chen G, Lin SC, Chen J, He L, Dong F, Xu J, et al. CXCL16 recruits bone marrow-derived fibroblast pre- cursors in renal fibrosis. J Am Soc Nephrol. 2011; 22(10):1876–86. Epub 2011/08/06. https://doi.org/ 10.1681/ASN.2010080881 PMID: 21816936; PubMed Central PMCID: PMC3187185. 59. Sun YB, Qu X, Caruana G, Li J. The origin of renal fibroblasts/myofibroblasts and the signals that trigger fibrosis. Differentiation. 2016; 92(3):102–7. Epub 2016/10/26. https://doi.org/10.1016/j.diff.2016.05.008 PMID: 27262400. 60. Lovisa S, Zeisberg M, Kalluri R. Partial Epithelial-to-Mesenchymal Transition and Other New Mecha- nisms of Kidney Fibrosis. Trends Endocrinol Metab. 2016; 27(10):681–95. Epub 2016/07/04. https:// doi.org/10.1016/j.tem.2016.06.004 PMID: 27372267. 61. Liu BC, Tang TT, Lv LL, Lan HY. Renal tubule injury: a driving force toward chronic kidney disease. Kid- ney Int. 2018; 93(3):568–79. Epub 2018/01/24. https://doi.org/10.1016/j.kint.2017.09.033 PMID: 29361307. 62. Bussolati B, Bruno S, Grange C, Buttiglieri S, Deregibus MC, Cantino D, et al. Isolation of renal progeni- tor cells from adult human kidney. Am J Pathol. 2005; 166(2):545–55. Epub 2005/02/01. https://doi.org/ 10.1016/S0002-9440(10)62276-6 PMID: 15681837; PubMed Central PMCID: PMC1602314. 63. Sheng L, Zhuang S. References New Insights Into the Role and Mechanism of Partial Epithelial-Mesenchymal Tran- sition in Kidney Fibrosis. Front Physiol. 2020; 11:569322. Epub 2020/10/13. https://doi.org/10.3389/ fphys.2020.569322 PMID: 33041867; PubMed Central PMCID: PMC7522479. 64. Hancock WW, Gao W, Csizmadia V, Faia KL, Shemmeri N, Luster AD. Donor-derived IP-10 initiates development of acute allograft rejection. J Exp Med. 2001; 193(8):975–80. Epub 2001/04/17. https:// doi.org/10.1084/jem.193.8.975 PMID: 11304558; PubMed Central PMCID: PMC2193411. 65. Rabant M, Amrouche L, Lebreton X, Aulagnon F, Benon A, Sauvaget V, et al. Urinary C-X-C Motif Che- mokine 10 Independently Improves the Noninvasive Diagnosis of Antibody-Mediated Kidney Allograft Rejection. J Am Soc Nephrol. 2015; 26(11):2840–51. Epub 2015/05/08. https://doi.org/10.1681/ASN. 2014080797 PMID: 25948873; PubMed Central PMCID: PMC4625672. 66. Malone AF, Wu H, Fronick C, Fulton R, Gaut JP, Humphreys BD. Harnessing Expressed Single Nucleo- tide Variation and Single Cell RNA Sequencing To Define Immune Cell Chimerism in the Rejecting Kid- ney Transplant. J Am Soc Nephrol. 2020; 31(9):1977–86. Epub 2020/07/17. https://doi.org/10.1681/ ASN.2020030326 PMID: 32669324; PubMed Central PMCID: PMC7461682. 67. Pineiro GJ, Montagud-Marrahi E, Rios J, Ventura-Aguiar P, Cucchiari D, Revuelta I, et al. Influence of Persistent Inflammation in Follow-Up Biopsies After Antibody-Mediated Rejection in Kidney Transplan- tation. Front Med (Lausanne). 2021; 8:761919. Epub 2021/12/07. https://doi.org/10.3389/fmed.2021. 761919 PMID: 34869466; PubMed Central PMCID: PMC8632955. 68. Lindquist RL, Niesner RA, Hauser AE. In the Right Place, at the Right Time: Spatiotemporal Conditions Determining Plasma Cell Survival and Function. Front Immunol. 2019; 10:788. Epub 2019/05/10. https://doi.org/10.3389/fimmu.2019.00788 PMID: 31068930; PubMed Central PMCID: PMC6491733. 69. Su H, Zhang CY, Lin JH, Hammes HP, Zhang C. The Role of Long-Lived Plasma Cells in Antibody- Mediated Rejection of Kidney Transplantation: An Update. Kidney Dis (Basel). 2019; 5(4):211–9. Epub 2019/11/27. https://doi.org/10.1159/000501460 PMID: 31768378; PubMed Central PMCID: PMC6873015. 70. Zarkhin V, Kambham N, Li L, Kwok S, Hsieh SC, Salvatierra O, et al. Characterization of intra-graft B cells during renal allograft rejection. Kidney Int. 2008; 74(5):664–73. Epub 2008/06/13. https://doi.org/ 10.1038/ki.2008.249 PMID: 18547992. 71. Heidt S, Vergunst M, Anholts JDH, Swings G, Gielis EMJ, Groeneweg KE, et al. Presence of intragraft B cells during acute renal allograft rejection is accompanied by changes in peripheral blood B cell sub- sets. Clin Exp Immunol. 2019; 196(3):403–14. Epub 2019/02/04. https://doi.org/10.1111/cei.13269 PMID: 30712266; PubMed Central PMCID: PMC6514375. 72. Bhat ZY, Bostwick DG, Hossain D, Zeng X. Participation of functionally active plasma cells in acute rejection and response to therapy in renal allografts. DNA Cell Biol. 2014; 33(7):448–54. Epub 2014/04/ 02. PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 78. Stewart BJ, Ferdinand JR, Young MD, Mitchell TJ, Louden KW, Riding AM, et al. Spatiotemporal immune zonation of the human kidney. Science. 2019; 365:1461–6. https://doi.org/10.1126/science. aat5031 PMID: 31604275 References https://doi.org/10.1089/dna.2014.2371 PMID: 24684655. 73. Lee YH, Sato Y, Saito M, Fukuma S, Saito M, Yamamoto S, et al. Advanced Tertiary Lymphoid Tissues in Protocol Biopsies are Associated with Progressive Graft Dysfunction in Kidney Transplant Recipi- ents. J Am Soc Nephrol. 2022; 33(1):186–200. Epub 2021/11/03. https://doi.org/10.1681/ASN. 2021050715 PMID: 34725107; PubMed Central PMCID: PMC8763171. 74. Tourville DR, Adler RH, Bienenstock J, Tomasi TB Jr. The human secretory immunoglobulin system: immunohistoligical localization of gamma A, secretory "piece," and lactoferrin in normal human tissues. J Exp Med. 1969; 129(2):411–29. Epub 1969/02/01. https://doi.org/10.1084/jem.129.2.411 PMID: 4178353; PubMed Central PMCID: PMC2138606. 29 / 30 PLOS ONE \| https://doi.org/10.1371/journal.pone.0267704 June 3, 2022 PLOS ONE Infiltrating fibroblasts in human kidney allografts 75. Krawczyk KM, Nilsson H, Nystrom J, Lindgren D, Leandersson K, Sward K, et al. Localization and Reg- ulation of Polymeric Ig Receptor in Healthy and Diseased Human Kidney. Am J Pathol. 2019; 189 (10):1933–44. Epub 2019/08/14. https://doi.org/10.1016/j.ajpath.2019.06.015 PMID: 31404540. 76. Der E, Suryawanshi H, Morozov P, Kustagi M, Goilav B, Ranabathou S, et al. Tubular cell and keratino- cyte single-cell transcriptomics applied to lupus nephritis reveal type I IFN and fibrosis relevant path- ways. Nat Immunol. 2019; 20(7):915–27. Epub 2019/05/22. https://doi.org/10.1038/s41590-019-0386- 1 PMID: 31110316; PubMed Central PMCID: PMC6584054. 77. Arazi A, Rao DA, Berthier CC, Davidson A, Liu Y, Hoover PJ, et al. The immune cell landscape in kid- neys of patients with lupus nephritis. Nat Immunol. 2019; 20(7):902–14. Epub 2019/06/19. https://doi. org/10.1038/s41590-019-0398-x PMID: 31209404. 78. Stewart BJ, Ferdinand JR, Young MD, Mitchell TJ, Louden KW, Riding AM, et al. Spatiotemporal immune zonation of the human kidney. Science. 2019; 365:1461–6. https://doi.org/10.1126/science. aat5031 PMID: 31604275 30 / 30
https://openalex.org/W3004206645	https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0227884&type=printable	English	null	Improving measurement of child abuse and neglect: A systematic review and analysis of national prevalence studies	PloS one	2,020	cc-by	13,785	RESEARCH ARTICLE Editor: Abraham Salinas-Miranda, University of South Florida, UNITED STATES Editor: Abraham Salinas-Miranda, University of South Florida, UNITED STATES Received: September 10, 2019 Accepted: December 31, 2019 Published: January 28, 2020 Received: September 10, 2019 Accepted: December 31, 2019 Published: January 28, 2020 Methods Copyright: © 2020 Mathews et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Guided by PRISMA and following a published protocol, we searched 22 databases from inception to 31 May 2019 to identify nationwide studies of the prevalence of either all five or at least four forms of child maltreatment. We conducted a formal quality assessment and critical analysis of study design. Data Availability Statement: All relevant data are within the manuscript and its Supporting Information files. Ben MathewsID1,2, Rosana PacellaID3, Michael P. Dunne4, Marko Simunovic5, Cicely Marston6 1 Director, Childhood Adversity Research Program, Faculty of Health, Queensland University of Technology, Brisbane, Queensland, Australia, 2 Adjunct Professor, Johns Hopkins University, Bloomberg School of Public Health, Baltimore MD, United States of America, 3 Institute for Lifecourse Development, University of Greenwich, Greenwich, London, United Kingdom, 4 School of Public Health, Faculty of Health, Queensland University of Technology, Brisbane, Queensland, Australia, 5 Institute for Health and Biomedical Innovation, Queensland University of Technology, Brisbane, Queensland, Australia, 6 Faculty of Public Health and Policy, London School of Hygiene and Tropical Medicine, London, United Kingdom a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 b.mathews@qut.edu.au * b.mathews@qut.edu.au OPEN ACCESS Child maltreatment through physical abuse, sexual abuse, emotional abuse, neglect, and exposure to domestic violence, causes substantial adverse health, educational and beha- vioural consequences through the lifespan. The generation of reliable data on the preva- lence and characteristics of child maltreatment in nationwide populations is essential to plan and evaluate public health interventions to reduce maltreatment. Measurement of child mal- treatment must overcome numerous methodological challenges. Little is known to date about the extent, nature and methodological quality of these national studies. This study aimed to systematically review the most comprehensive national studies of the prevalence of child maltreatment, and critically appraise their methodologies to help inform the design of future studies. Citation: Mathews B, Pacella R, Dunne MP, Simunovic M, Marston C (2020) Improving measurement of child abuse and neglect: A systematic review and analysis of national prevalence studies. PLoS ONE 15(1): e0227884. https://doi.org/10.1371/journal.pone.0227884 Improving measurement of child abuse and neglect: A systematic review and analysis of national prevalence studies Ben MathewsID1,2*, Rosana PacellaID3, Michael P. Dunne4, Marko Simunovic5, Cicely Marston6 Introduction Child maltreatment is common and causes substantial adverse health, educational and beha- vioural consequences [1]. Understanding its prevalence and characteristics in nationwide pop- ulations is essential to plan and evaluate interventions to reduce maltreatment. However, measurement of child maltreatment is known to be far from universal, and when performed must confront methodological challenges. This study systematically reviews the most compre- hensive national studies of the prevalence of child maltreatment, and critically appraises their methodologies to help inform future measurement. Child maltreatment in its five recognised forms is a major public health issue [2]. Physical and mental diseases are caused through proximal and distal pathways. Immediate physical injuries and conditions include brain injury and failure to thrive, and a panoply of psychologi- cal disorders include anxiety, depression, and suicidality. Studies have found serious effects of physical abuse [3,4,5], sexual abuse [6,7], emotional abuse [5,8–10], neglect [5,9,11], and expo- sure to domestic violence [12–14]. Experiencing multiple forms of maltreatment is common [12], and is associated with more severe outcomes [14,15], including alcohol and drug abuse, mental illness, interpersonal violence, and sexual risk taking [16]. The adoption of coping mechanisms such as smoking, alcohol and drug use can compound the damage by causing diseases in the medium to long term; and chronic stress can cause coro- nary artery disease, pulmonary fibrosis, and inflammation [17–22]. Potent mediators include prolonged psychological distress, risky behaviours, social withdrawal and dysfunction, The adoption of coping mechanisms such as smoking, alcohol and drug use can compound the damage by causing diseases in the medium to long term; and chronic stress can cause coro- nary artery disease, pulmonary fibrosis, and inflammation [17–22]. Potent mediators include prolonged psychological distress, risky behaviours, social withdrawal and dysfunction, impaired cognitive development, low educational and occupational attainment, and interper- sonal relationship difficulties. A growing body of evidence is showing child maltreatment affects brain development, shortens telomeres, and produces epigenetic neurobiological changes [23–26]. The disease and economic burdens are substantial: a recent estimate of the cost of disability-adjusted life years (DALYs) lost across East Asia and the Pacific was 1.88% of the region’s GDP, equating to $194 billion in 2012 dollars [27]. As a global policy imperative, the United Nations recognises the gravity of child maltreat- ment and its consequences. The United Nations Agenda for Sustainable Development includes a target of ending abuse of children [28]. Results This review identified 30 national prevalence studies of all five or at least four forms of child maltreatment, in 22 countries. While sound approaches are available for different settings, methodologies varied widely in nature and robustness. Some instruments are more reliable and obtain more detailed and useful information about the characteristics of the maltreat- ment, including its nature, frequency, and the relationship between the child and the person Funding: The authors received no specific funding for this work. Competing interests: The authors have declared that no competing interests exist. PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 1 / 22 Improving measurement of child abuse and neglect: A systematic review who inflicted the maltreatment. Almost all studies had limitations, especially in the level of detail captured about maltreatment, and the adequacy of constructs of maltreatment types. Conclusions Countries must invest in rigorous national studies of the prevalence of child maltreatment. Studies should use a sound instrument containing appropriate maltreatment constructs, and obtain nuanced information about its nature. PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 Introduction Reliable scientific data on national prevalence is essential to measure progress against this goal, and to inform policy efforts aimed at preven- tion, early identification and response [29–30]. Nationwide studies of the experience of childhood maltreatment can identify baseline prev- alence stratified by maltreatment type, as well as important contextual features including the child’s sex, age, and relationship with the abusive person. Without good measurement tech- niques and repeated measures over time, we lack understandings of baseline measures, of whether maltreatment is increasing or declining, of changes in maltreatment types over time, and of the efficacy of policy and practice interventions designed to reduce child maltreatment for the whole population and for key sub-populations. 2 / 22 PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 Improving measurement of child abuse and neglect: A systematic review Despite the necessity for good data in public health generally and in child maltreatment in particular, approximately half of all countries have failed to report any kind of robust preva- lence estimates [2], and extant studies are often limited to measuring one or few maltreatment types [31]. Accordingly, prevalence estimates are often inadequately specified, and are almost certainly underestimated. In addition, existing studies vary widely in design, sample and meth- ods, and often use non-standardized instruments [5,32]. Where an instrument is non-stan- dardized and untested the risk may be heightened that the study will fail to capture Despite the necessity for good data in public health generally and in child maltreatment in particular, approximately half of all countries have failed to report any kind of robust preva- lence estimates [2], and extant studies are often limited to measuring one or few maltreatment types [31]. Accordingly, prevalence estimates are often inadequately specified, and are almost certainly underestimated. In addition, existing studies vary widely in design, sample and meth- ods, and often use non-standardized instruments [5,32]. Where an instrument is non-stan- dardized and untested, the risk may be heightened that the study will fail to capture experiences that constitute maltreatment, and may capture experiences that do not constitute maltreatment, hence producing unreliable results. Importantly, the use of unsound maltreat- ment constructs and operational definitions also compromises the reliability of recorded mea- sures [33–34]. As an example of this, studies of sexual abuse that do not include non-contact sexual abuse will underestimate prevalence; conversely, studies that include as sexual abuse genuinely consensual acts between peers will overestimate prevalence. Introduction These factors influence health outcomes and provide evidence about specific risk and protective factors and how these may best be targeted. Rigor- ous measurement of child maltreatment is complex, but is essential to inform prevention efforts and drive nationwide social change [2,14,29,36,37]. Recent research has reviewed global prevalence estimates [2,31], the nature of population health surveys exploring consequences of child maltreatment [37], and approaches in studies of youth [38]. However, to date, there has not been a systematic review and methodological appraisal of high quality national population prevalence studies of child maltreatment to pro- vide a baseline for future measurement efforts. This study aimed to investigate three questions. First, what national studies have been con- ducted of the prevalence and nature of all five, or at least four, major forms of child maltreat- ment? Second, what methodologies were used in these studies? Third, what does a critical analysis of these studies indicate about the methodological rigour, quality, and practical viabil- ity of different approaches? The results of our investigation can inform future efforts to gener- ate baseline prevalence estimates, design policy responses, and chart trends over time, as more societies confront the challenge of childhood maltreatment. Introduction Similarly, studies of neglect that do not consider medical neglect will underestimate prevalence. Studies of emo- tional abuse that include non-abusive yelling will overestimate prevalence. Optimal methodologies for measuring population characteristics of child maltreatment can ensure adequate detail is captured to yield reliable, detailed, useful data. For best quality esti- mates, prevalence studies should adopt robust conceptual understandings of maltreatment types and their operational definitions [33]. In addition, prevalence studies need to ask a series of items to obtain accurate data, rather than a single question which will tend to underestimate prevalence [35]. Similarly, to avoid underestimates, items should be behaviourally specific, rather than vague, ambiguous or non-specific [36]. All national prevalence studies face meth- odological and practical challenges, and studies take different approaches [2,12,14,30]. Ideally, however, all five forms of child maltreatment should be measured simultaneously, since many children experience such poly-victimization and its heightened consequences [1,14,16]. To provide nuanced, useful information, studies should ask about prevalence, and about the spe- cific nature of the acts, their severity, frequency, and timing, and the relationship of the child to the person inflicting the abuse [33]. These factors influence health outcomes and provide evidence about specific risk and protective factors and how these may best be targeted. Rigor- ous measurement of child maltreatment is complex, but is essential to inform prevention efforts and drive nationwide social change [2,14,29,36,37]. Optimal methodologies for measuring population characteristics of child maltreatment can ensure adequate detail is captured to yield reliable, detailed, useful data. For best quality esti- mates, prevalence studies should adopt robust conceptual understandings of maltreatment types and their operational definitions [33]. In addition, prevalence studies need to ask a series of items to obtain accurate data, rather than a single question which will tend to underestimate prevalence [35]. Similarly, to avoid underestimates, items should be behaviourally specific, p y y p rather than vague, ambiguous or non-specific [36]. All national prevalence studies face meth- odological and practical challenges, and studies take different approaches [2,12,14,30]. Ideally, however, all five forms of child maltreatment should be measured simultaneously, since many children experience such poly-victimization and its heightened consequences [1,14,16]. To provide nuanced, useful information, studies should ask about prevalence, and about the spe- cific nature of the acts, their severity, frequency, and timing, and the relationship of the child to the person inflicting the abuse [33]. Our systematic review was guided by PRISMA [39] (S1 Fig). We developed a protocol, regis- tered at PROSPERO [40]. #CRD42017068120, https://www.crd.york.ac.uk/PROSPERO). Search strategy Our systematic review was guided by PRISMA [39] (S1 Fig). We developed a protocol, regis- tered at PROSPERO [40]. #CRD42017068120, https://www.crd.york.ac.uk/PROSPERO). 3 / 22 PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 Improving measurement of child abuse and neglect: A systematic review Adopting our search strategy (S1 File), we searched 22 databases from their inception to 31 May 2019. Adopting our search strategy (S1 File), we searched 22 databases from their inception to 31 May 2019. Screening As detailed in our search strategy (S1 File), in Phase 1, MS, JD and ED screened records by title. We removed duplicates using electronic software (Endnote), and removed remaining duplicates about the same study, selecting the publication providing the most detailed account. In Phase 2, BM and RP independently screened records by title and abstract. Disagreements were discussed between BM and RP to achieve consensus. To identify any further potential eli- gible studies at this stage that may not have been captured in the search, all co-authors consid- ered if there were any further known studies requiring inclusion that were not in the Phase 2 shortlist. In Phase 3, BM and RP independently assessed full text of screened in articles. Dis- agreements were discussed between BM and RP to achieve consensus, with reasons recorded. We screened reference lists of included studies to identify any further potential eligible studies. We used a translator to assist in screening non-English studies. This process resulted in 23 eli- gible studies (Fig 1). Eligibility criteria We searched for quantitative studies of the prevalence of child maltreatment. Included studies met four criteria: (1) primary empirical studies of the prevalence of four or five types of child maltreatment: ((i) physical abuse; (ii) emotional or psychological abuse; (iii) neglect; (iv) expo- sure to domestic violence; and (v) sexual abuse; (2) studies conducted nationwide using a rep- resentative sample of the population; (3) studies involving adult or child participants providing self-reported information about their experience, or studies where adults provided information about their child’s experience; (4) peer-reviewed studies or substantial grey literature. PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 Data extraction and analysis We developed a template to extract 60 data items from each study considering design, proce- dure, sample, instrument, ethics, and subpopulation analysis (S2 File). We extracted 45 items about the instrument, including: name, psychometric data, definitions of maltreatment con- structs, number of questions asked about each type, and whether questions explored: (a) the relationship between the child and the person inflicting maltreatment; (b) nature of the acts; (c) severity (e.g., if they caused injuries); (d) frequency. MS and BM extracted these data. We separately tabulated the extracted items each study asked about maltreatment, with BM con- ducting a final double-check regarding these (S3 File). Our critical analysis included an appraisal of the construct validity of study items and the soundness of their operational definition. To inform this analysis, we identified robust concep- tual understandings of each maltreatment type as established in the scholarly literature, and adopted these as an evaluative standard. Physical abuse involves intentional acts of physical force by a parent or caregiver, excluding lawful corporal punishment [41]. Sexual abuse involves contact and non-contact sexual acts, inflicted by any adult or child in a position of power over the victim, to seek or obtain physical or mental sexual gratification, when the child does not have capacity to provide consent, or has capacity but does not provide consent [42]. Emotional or psychological abuse is inflicted by a parent or caregiver, and includes emotional unavailability, hostile interaction, developmentally inappropriate interaction, failure to acknowledge the child’s individuality, and failure to integrate the child into the social world [43]. Neglect involves parental or caregiver omissions to provide the basic necessities of life PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 4 / 22 Improving measurement of child abuse and neglect: A systematic review Fig 1. PRISMA flow diagram. https://doi.org/10.1371/journal.pone.0227884.g001 Fig 1. PRISMA flow diagram. https://doi.org/10.1371/journal.pone.0227884.g001 suited to the child’s developmental stage, as recognised by the child’s cultural context, includ- ing physical, emotional, medical, environmental, supervisory, and educational neglect [44]. Exposure to domestic violence involves the child witnessing a parent or other family member being subjected to assaults, threats or property damage by another adult or teenager normally resident in the household [12]. Our critical analysis was also informed by an understanding that prevalence studies must be conducted with low risk of bias to obtain reliable findings. Systematic review This review identified 23 articles reporting the results of national studies of the prevalence of all five or four of the recognized forms of child maltreatment. One of these articles reported the results of a study conducted simultaneously in nine countries in the Balkan Peninsula, and eight of these national studies met our eligibility criteria [46]. Accordingly, in total, our review identified 30 national studies, conducted in 22 countries. Studies were published between 2005 and 2019. Extracted data revealed study location, scope, participants, data collection method, and instrument. Table 1 presents the extracted information from included studies. The sup- porting information details the prevalence rates reported by each study (S5 File). There were four studies in the USA [47–50], three in the UK [51–53], two in Hong Kong [54–55] two in Taiwan [56–57], and two in Germany [58,59]. There was one study in Den- mark [60], the Netherlands [61], Switzerland [62], Japan [63], Suriname [64], Saudi Arabia [65], Israel [66], South Africa [67], and Hungary [68]. In the Balkans study [46], eight met eli- gibility criteria based on the number of types of maltreatment studied: Albania, Bosnia & Her- zegovina, Bulgaria, Croatia, the Former Yugoslavian Republic of Macedonia, Greece, Romania, and Serbia; in general for our purposes, we treat these as one study. The Turkish study involved three forms of maltreatment, so was excluded from our analyses. Fourteen studies measured all five maltreatment types [47–51,53,56–57,61,64–68]. Of nine studies measuring four maltreatment types, seven omitted EDV [46,52,58–60,62–63], and two omitted sexual abuse [54–55]. Eleven studies measured prevalence throughout childhood and in the past year; nine measured prevalence through childhood only, and three measured past year incidence only. Only nine studies explored all five types of maltreatment across childhood, defined as aged under 18 [48–50,53,64–68]. These studies occurred in seven countries (USA, UK, Suriname, Saudi Arabia, Israel, South Africa and Hungary), and only three involved a sample of adults providing data about experiences over their entire childhood [53,65,68]. Four studies in Ger- many, the UK and Japan obtained information from adults about all maltreatment across childhood except EDV [52,58–59,63]. Eight studies involved only child participants aged under 18 providing self-report data. Three studies included child and adult participants each providing self-report data. Data extraction and analysis In our analysis, we assessed study rigour, quality and practicability, and used a quality assessment tool designed to assess risk of bias in population-based prevalence studies [45, S4 File]. Using our quality assessment 5 / 22 PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 Improving measurement of child abuse and neglect: A systematic review tool, we created an overall risk of bias score for each study which summed scores for individual items (maximum score 10). RP and CM independently assessed each study considering four external validity items and five internal validity items. Disputes were resolved through an inde- pendent third assessor (MD, BM). Our critical analysis further considered suitability of approach, considering: methodology to recruit the sample and accommodate high-risk sub- samples; administration method; instrument; soundness of conceptual constructs; ethics; and practical viability. PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 Improving measurement of child abuse and neglect: A systematic review Table 1. Summary of study characteristics. STUDY DETAILS DESIGN SAMPLE PROCEDURE Citation Nation Prev/ Incid/ Both Period studied Number of types of maltreatment Participant age (yrs) Sampling frame Sampling strategy Sample size (n) Response rate (%) Recruitment How administered1 Al Muneef 2017 Saudi Arabia Prev Childhood < 18 5 18 and over Cross-sectional national sample of citizens raised and resident in S. A. Random selection of large and small cities in all 13 regions of SA; participants selected from 182 locations 10,156 Not reported NR HH, hard copy (6 mths) Chan 2011 Hong Kong Both Lifetime and past year 4: PEN EDV 12–17 Chinese children aged 12–17 Randomly sampled HHs from registry of Quarters (fixed sampling intervals); N = 4347 households 1094 (3049 HHs) 70% NR HH, FTF interviews Chan 2011 Hong Kong Both Lifetime and past year 4: PEN EDV 12–17 Chinese children aged 12–17 HHs from registry of Quarters, stratified random sampling (N not reported) 2062 70% NR HH, FTF interviews Christoffersen 2013 Denmark Prev PA, EA, Neg < 12; SA < 24 4: SPEN 24 yrs only All children born 1984 Stratified random sample of 24 year olds N = 4718 2980 63% Letter + phone CATI + HH (2 yrs) Denholm 2013 UK Prev Childhood < 16 yrs 5 Age 7, 11,16, 23, 33, 42, 45, 50 All born March 1958 England, Scotland, Wales 1958 British cohort, born during one week in March (prospective cohort study). Aged 45, N = 11971 9310 78% Birth registry HH FTF (parent proxy for child; then direct (1 yr) Euser 2013 Netherlands Incid Past year 5 12–17 Students aged 12–17 Random selection of 42 high schools from database of all schools nationally, each with 4 randomly selected classes 1920 (from 29/ 42 schools) Not reported; 29/42 schools participated NR In school, hard copy Feng 2015 Taiwan Both Lifetime and past year 5 12–18 Taiwanese adolescents aged 12–18 35 schools out of 44 invited schools, across 17 cities and townships 5236 (in 35 participating schools) Not reported; 994% of participating schools Phone, through school In school Finkelhor 2005 USA Incid Past 12 months 5 2–17 (parents for children 2–9; children 10–17) Nationally representative sample children aged 2–17 Random digit dialling 2030 (1000 children 10–17; 1030 parents of children 2–9) 795% of eligible persons contacted Phone CATI (3 mths) Finkelhor 2009 USA Both Lifetime and past 12 months 5 0–17 (parents for children 0–9; children aged 10–17) Cross-sectional national sample of children aged 0–17 National landline residential telephone survey 4549 (3053 national cross- section; 1496 oversample) 71% cross- section; 63% oversample Phone CATI (5 mths) Finkelhor 2014 USA Both Lifetime and past 12 months 5 1 mth-17 yrs (parents for those 1 mth-9 yrs; children 10–17) Nationwide sampling frame of residential phone numbers Random digit dialling + two samples to capture those without landlines: cell phone sample (n = 31; abandoned due to low yield) and address-based sample (n = 750) 4503 60% of eligible respondents Phone CATI (12 mths) Finkelhor 2015 USA Both Lifetime and past year 5 0–17 (parents for children 0–9; children 10–17) National sample Nationally representative sample of phone numbers via 4 methods (address-based sample (ABS) of HH phone numbers; prescreened sample (PSS) of HH with children; landline sample (LLS); cellphone sample (CS) 4000 (1011 ABS; 520 PSS; 2443 LLS; 26 CS) Differed across 4 sample frames (142%-67%) Phone CATI (9 mths) Hauser 2011 Germany Prev Childhood < 18 4: SPEN 14–90 14–90 yr olds understanding written German Cross sectional randomly generated sample of the population N = 4455 2504 56% HH in person HH, hard copy (1 mth) Lev-Wiesel 2018 Israel Both Lifetime and past year 5 12–17 Students in grades 6, 8 and 10 in the national public school system Two-stage random sample by school level (primary, junior high, high), school district (Northern, Central, Southern Israel, Jerusalem); and school SES indicator; and random selection of two classes from each grade within each school 12,035 Not reported School In school, either hard copy or CASI using iPod (Continued) Table 1. Systematic review Five stud- ies involved a household’s child participant aged under 18 providing self-report data (four involved children aged 10–17 and one involved children aged 11–17) and the household’s parents providing proxy data about a child aged under the cut-off. Five studies involved only adults providing self-report data (24 year olds; 18–24 year olds; 20–49 year olds; 18 and over). Sample sizes ranged from 1094 to 12,035 participants. Five studies adopted measures to recruit high-risk sub-populations [48,56,60,62,64]. Seven studies were conducted in schools: Taiwan [56–57], the Netherlands [61]. Switzer- land [62], Suriname [64], and the Balkans study [46]. Eleven studies were conducted in house- holds by interviews, in Hong Kong [54–55], Hungary [68], the UK [51–53], Germany [58–59], 6 / 22 PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 Improving measurement of child abuse and neglect: A systematic review Table 1. (Continued) STUDY DETAILS DESIGN SAMPLE PROCEDURE Citation Nation Prev/ Incid/ Both Period studied Number of types of maltreatment Participant age (yrs) Sampling frame Sampling strategy Sample size (n) Response rate (%) Recruitment How administered1 May-Chahal 2005 UK Prev Childhood < 18 4: SPEN 18–24 18–24 year olds in the UK Random sample using postcode address file: 633 postcode sectors with probability proportional to population of 18–24 year olds after stratification; 90 addresses in each postcode; N = 56,979 addresses 2869 69% Direct to individual HH, CAPI (5 mths) Nagy 2019 Hungary Prev Childhood < 18 5 18–112 Hungarian adults aged 18 or older Multi-stage stratified cluster sampling using 120 census sampling units with randomised selection of 10 addresses within each unit 1174 74.6% HH in person HH, hard copy (1 mth) Nikolaidis 2018 Balkans Both Lifetime and past year 4: SPEN 11, 13, and 16 year olds 11, 13, and 16 year olds in nine Balkan nations Random sample of schools, derived from number of schools per region; 63,250 students 42,194 66.7% (students); 45.8%-82.7% (nations) School In school, hard copy Radford 2013 UK Both Childhood < 18, and past year 5 Parents of children 2 mths-10 yrs; Children 11– 17; Adults 18– 24 Children and young people in the UK aged under 25 Random probability sampling of HHs from UK Postcode Address File (50,000 by mail), and eligibility determined by visits 2160 parents of children 2 mths- 10 yrs; 2275 children 11–17; 1761 adults 18– 24 60.4% (number of interviews completed as a % of HHs approached) Direct to individual by mail, door- knock HH, CASI + option of headphones, hard copy for parent (10 mths) Schick 2016 Switzerland Prev Childhood 4: SPEN 9th grade students 9th grade adolescents in representative population based sample Probability proportional to cluster size, stratified by 7 regions & 26 cantons; then N = 228 randomly selected schools with 560 classes 6787 (177 participating schools with 449 classes) RR in participating classrooms was 92% School In school, CASI (9 mths) Shen 2016 Taiwan Incid Past 12 months 5 10–11 Fourth grade Taiwanese primary school children Proportionally stratified according to county and randomly selected. N = 25% of all primary schools in Taiwan 49% of invited schools (314 schools; 6233 children) Not reported. Summary of study characteristics. PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 7 / 22 999% of consenting parents’ children participated Phone In school, hard copy (spring semester) Tsuboi 2015 Japan Prev Less than 18 yrs 4: SPEN 20–49 Japanese adults aged 16–49 Multi-stage randomised cluster sampling; 44 clusters from 11 geographical units; N = 2693 1540 5720% Door-knock Hard copy to home (1 mth) van der Kooij 2015 Suriname Both 12–17: childhood and past 12 mths; adults: in childhood 5 12–17; and 18–22 Suriname— national sample of students Stratified national sample of students from high schools and vocational education classes. Random probability sampling. 1391 (57 schools); 1072 children 12–17; 239 adults 18– 22 Not reported School In school, hard copy (4 mths) Ward 2018 South Africa Both Lifetime and past year 5 15–17 South Africa– nationwide sample of 15–17 year olds Multi-stage stratified random sample using 725/80,787 randomised census enumerator areas, with randomised selection of 5–10 HHs in each 5631 94.8% participation rate HH in person HH, hard copy interview (1 yr 5 mths) Witt 2017 Germany Prev Childhood < 18 4: SPEN 14–94 14–94 yr olds understanding German Randomly generated representative sample obtained by households in every third street 2487 51.20% 2510/ 4902 HHs HH in person HH, hard copy (3 mths) CAPI: Computer assisted personal interview. CASI: Computer assisted self-interview. CATI: Computer assisted telephone interview. EDV: Exposure to domestic violence. E: Emotional or psychological abuse. HH: Household. FTF: Face to face. N: Neglect. P: Physical abuse. S: Sexual abuse. Each wave did not measure all five types. 1 Instrument administration time generally ranged from 30 55 minutes Table 1. (Continued) CAPI: Computer assisted personal interview. CASI: Computer assisted self-interview. CATI: Computer assisted telephone interview. EDV: Exposure to domestic violence. E: Emotional or psychological abuse. HH: Household. FTF: Face to face. N: Neglect. P: Physical abuse. S: Sexual abuse. Each wave did not measure all five types. 1 PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 8 / 22 Improving measurement of child abuse and neglect: A systematic review Japan [63], Saudi Arabia [65], and South Africa [67]. Five studies used remote computer assis- ted telephone interviews (CATI), with four in the USA [47–50], and one in Denmark [60]. Data collection time ranged from 1 month to 2 years. Methodologies to recruit the sample and accommodate high-risk subpopulations also var- ied. In most studies, the target population was a close representation of the national popula- tion. Studies in schools were done in countries with high school attendance. All studies used random selection. However, few studies used strategies to capture participants from culturally and linguistically diverse groups, or from high-risk groups such as those in out of home care. Response rates for household studies generally ranged from 56% to 78%, with one reporting a participation rate of 94.8% [67]. Rates in school-based studies showed schools’ participation rate ranging from 49%-79%, and then with almost 100% response rates from children in par- ticipating schools. Response rates in CATI studies ranged from 60% to 79.5%, with more recent studies having lower rates [47–49]. Regarding consent to participate, 18 of the studies involved child participants exclusively or with adult participants. Nine studies involved only child participants; in these, two required only the child’s consent [56,62], one required the child’s consent and parental passive consent [64], one required the child’s consent and either passive or active parental consent [46], and five required parental active consent and the child’s consent [54–55,57,61,66–67]. Of the studies involving child participants, seven reported the measures used by research teams when a child was suspected to have been harmed or at risk [46–50,53,67]. Nine studies reported other measures to assist any distressed participants [46,48,52,54,56,60,62,64]. Studies used a range of instruments and approaches to measuring each maltreatment type. Table 2 presents key data extracted from the instrument used in each study. Comprehensive details about the maltreatment items are detailed in the supporting information (S3 File). Eight studies used the Juvenile Victimization Questionnaire (JVQ). These studies used dif- ferent versions of the JVQ, either using its original form [72], an enhanced form [48–50], or an adapted version [53,62,66–67]. Two studies used the Conflict Tactics Scale Parent-Child ver- sion (measuring physical abuse, emotional abuse, and neglect), and the CTS2 (EDV) [54–55]. Two studies used the ICAST-C, in either its original form [56] or an adapted version [46]. Two studies used the Childhood Trauma Questionnaire [58–59]. Single studies used the Adverse Childhood Experiences International Questionnaire [65], the Adverse Childhood Experiences questionnaire [68], and the Lifestyle and Attitudes Towards Sexual Behavior instrument [63]. Four studies used a blend of instruments [51,57,61,64]. Two studies used self-developed instruments [52,60]. Six studies did not report psychometric data on instrument validity and reliability. Six stud- ies reported psychometric data on the instrument as used [46,54–56,58,72]. PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 Improving measurement of child abuse and neglect: A systematic review Table 2. Key features of instruments used in prevalence studies. Study INSTRUMENT SEXUAL ABUSE PHYSICAL ABUSE EMOTIONAL ABUSE NEGLECT EDV Instrument Psycho- metrics reported Approach to constructs How many items Identity, nature, severity, frequency How many items Identity, nature, severity, frequency How many items Identity, nature, severity, frequency How many items Identity, nature, severity, frequency How many items Identity, nature, severity, frequency Al Muneef 2017 ACE-IQ No S3 File 4 N N N Y 2 N N N Y 2 N N N Y 4 N N N Y 3 N N N Y Chan 2011 CTSPC (P, E, N); CTS2 (EDV) Yes [69– 70]1 p. 536 [69– 70]1 n.a n.a 13 Y Y N Y 5 Y Y N Y 5 Y Y N Y 39 Y Y N Y Chan 2011 CTSPC (P, E, N); CTS2 (EDV) Yes [69– 70]1 p. 6–8 [69–70]1 n.a n.a 13 Y Y N Y 5 Y Y N Y 5 Y Y N Y 39 Y Y N Y Christoffersen 2013 Self- developed No p. 152–3 4 Y Y N N 7 Y Y Y N 6 Y Y N N 7 Y Y N N n.a n.a Denholm 2013 Blended tools No p. 342, 346 1 Y Y N N 1 Y Y N N 2 Y Y N N 11 Y Y Y N 1 Y Y N N Euser 2013 Blended tools No p. 844; EDV not reported 8 Y Y N N 8 Y Y N N 1 Y Y N N 8 Y Y N N 7 Y Y N N Feng 2015 ICAST-CH Yes [71] p. 12, 15 6 Y Y N Y 9 Y Y N Y 8 Y Y N Y 6 Y Y N Y 7 Y Y N Y Finkelhor 2005 JVQ Yes [72]2 p. 21–23 7 + followups Y Y Y Y 1 + followups Y Y Y Y 1 + followups Y Y Y Y 1 + followups Y Y Y Y 2 + followups Y Y Y Y Finkelhor 2009 JVQ (1st enhanced)3 Yes [72]2 p. 1418–22 7 + followups Y Y Y Y 1 + followups Y Y Y Y 1 + followups Y Y Y Y 1 + followups Y Y Y Y 8 + followups Y Y Y Y Finkelhor 2014 JVQ (2nd enhanced)3 Yes [72]2 p. 1433–35 7 + followups Y Y Y Y 1 + followups Y Y Y Y 1 + followups Y Y Y Y 5 + followups Y Y Y Y 8 + followups Y Y Y Y Finkelhor 2015 JVQ (3rd enhanced)3 Yes [72]2 eApp. 2 8 + followups Y Y Y Y 1 + followups Y Y Y Y 1 + followups Y Y Y Y 5 + followups Y Y Y Y 8 + followups Y Y Y Y Hauser 2011 CTQ Short- form Yes [73] Referred to [73] 5 N Y N Y 5 N Y Y Y 5 N Y Y Y 10 Y Y Y Y n.a n.a Lev-Wiesel 2018 JVQ (modified) + CTQ (modified) Yes [72– 73]2 Referred to [72– 73]2 11 Y Y Y Y 6 Y Y Y Y 5 Y Y Y Y 12 Y Y Y Y 2 Y Y Y Y May-Chahal 2005 Self- developed No p. 972–976 14 Y Y Y Y 9 Y Y Y Y 7 Y Y Y Y 8 Y Y Y Y n.a n.a Nagy 2019 ACE No [37] p. 14 1 N N N N 1 N N N N 1 N N N N 2 N N N N 1 N N N N Nikolaidis 2018 ICAST-CH (modified) Yes, p. 5 S3 File 5 (11 yrs), 6 (13, 16 yrs) Y N N Y 15 (11 yrs), 16 (13, 16 yrs) Y N N Y 17 (11 yrs),19 (13, 16 yrs) Y N N Y 4 Y N N Y n.a n.a Radford 2013 JVQ (modified) Yes [72]2 p. 812–3 7 + followups Y Y N Y 2 + followups Y Y N Y 1 + followups Y Y N Y 14 + followups Y Y N Y 6 + followups Y Y N Y Schick 2016 JVQ (modified) Yes [72]2 p. 5 1 Y Y N N 1 Y Y N N 1 Y Y N N 1 Y Y N N n.a n.a Shen 2016 Blended tools Yes p. CTSPC: Conflict Tactics Scale Parent-Child version. CTS2: Conflict Tactics Scale 2. CTQ: Childhood Trauma. ICAST-CH: International Child Abuse Screening Tool: Children’s Home version. JVQ: Juvenile Victimization Questionnaire. 1 Studies using enhanced or adapted versions of instruments generally cited the original instrument’s data but did not report further psychometric tests. Most studies did not define overarching concepts of each form of maltreatment, instead operationalising these concepts into questions about the participant’s experiences. Approaches to some but not all forms of maltreatment broadly aligned with the nature of maltreatment concepts as established by the scholarly literature. Approaches to physical abuse and sexual abuse were generally sound. Approaches to the construct and operationalisation of emotional abuse were generally sub-optimal, with some exceptions (e.g., [46,52]). Neglect was also rarely well-operationalised, with some exceptions (e.g., [49,52–53,58–59,66]. Studies explored maltreatment experiences in varying depth, reflected by the number and nature of questions asked (Table 2). For sexual abuse, 12 studies asked between five and eight questions. Most studies asked about the relationship with the person inflicting the abuse, and the nature of the acts; more than half asked about frequency; but few asked about severity. PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 9 / 22 https://doi.org/10.1371/journal.pone.0227884.t002 For further details see S3 File. 2 For full details on the original and subsequent enhanced JVQ, see S3 File. 3 For full details see S3 File see S3 File. he original and subsequent enhanced JVQ, see S3 File. 2 For full details on the original and subsequent enhanced JVQ, see S3 File. 3 For full details see S3 File. 1 For further details see S3 File. 2 For full details on the original and subsequent enhanced JVQ, see S3 File. ails see S3 File. h l d b h d Q S l r further details see S3 File. tails on the original and subsequent enhanced JVQ, see S3 File. tails see S3 File. equent enhanced JVQ, see S3 File. g q J Q, 3 For full details see S3 File. CTSPC: Conflict Tactics Scale Parent-Child version. CTS2: Conflict Tactics Scale 2. CTQ: Childhood Trauma. ICAST-CH: I Children’s Home version. JVQ: Juvenile Victimization Questionnaire. 8–9 2 N Y N Y 7 Y Y N Y 4 Y Y N Y 4 Y Y N Y 2 Y Y N Y Tsuboi 2015 Lifestyle & Attitudes No p. 2580 1 N Y N N 1 N Y N N 1 N Y N N 1 N Y N N n.a n.a van der Kooij 2015 Blended tools Yes3 [74–75] p. 155 + S3 File 7 N Y N Y 8 Y Y N Y 1 Y Y N Y 8 Y Y N Y 7 Y Y N Y Ward 2018 JVQ (modified) Yes [72]2 S3 File 7 Y Y Y Y 1 Y Y Y Y 1 Y Y Y Y 4 Y Y Y Y 7 Y Y Y Y Witt 2017 CTQ Short- form Yes [73] Referred to [73] 5 N Y N Y 5 N Y Y Y 5 N Y Y Y 10 Y Y Y Y n.a n.a CTSPC C fli t T ti S l P t Child i CTS2 C fli t T ti S l 2 CTQ Childh d T ICAST CH I t ti l Child Ab S i T l Table 2. Key features of instruments used in prevalence studies. PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 10 / 22 Improving measurement of child abuse and neglect: A systematic review Other notable differences included: two studies being limited to sexual abuse by a parent/ guardian [51,60]; most studies including contact and non-contact acts, but three studies included contact abuse only [62,65,68]; four studies asking only one question [51,62–63,68]. For physical abuse, eight studies asked only one question, although these included multiple distinct concepts [47–51,62,63,68]. Six studies asked between five and nine questions. Most asked about relationship and nature; more than half asked about frequency; but few asked about severity. A notable difference was in the treatment of spanking on a child’s bottom: seven studies excluded “spanking on your bottom” from the definition of physical abuse [47– 50,53,62,66]; four studies included spanking with a bare hand as physical abuse [46,54–56]; and four studies included as physical abuse being hit or spanked on the bottom but only when done with an implement or hard object [51,52,57,64]. For emotional or psychological abuse, eight studies asked between five and eight questions. Most asked about relationship and nature; more than half asked about frequency; but few asked about severity. Other notable differences included: three studies being limited to a single generic question [51,61,64]; seven studies using a single compound question [47–51,62,67]; and only two studies using a detailed scale of items closely aligned with a sound conceptual model [46,52]. For neglect, 12 studies asked between five and 11 questions. Five studies asked one question [47–48,62,63,68]. Most asked about relationship and nature; more than half asked about fre- quency; but few asked about severity. Six studies asked detailed questions about multiple dimensions of neglect, and their severity [49–50,52,58–59,66]. Other notable differences included: some studies operationalising neglect very broadly, including a parent having low aspirations [51], or not helping with homework [64]; only one study asking about educational neglect [64]; and one study omitting physical and nutritional neglect [46]. For exposure to domestic violence, six studies asked between six and eight questions. Most asked about relationship and nature; more than half asked about frequency; but few asked about severity. Notable differences were: two studies used the comprehensive CTS2 scale of 39 items originally devised for use with adult couples [54–55]; and the original JVQ had two phys- ical assault items [72], and later added six items about threats or property damage by other family members [48–50]. Risk of bias Table 3 sets out the quality assessment and scoring results for each study. Scores ranged from 6 to 10. Most studies had relatively high internal and external validity. We concluded that stud- ies scoring 9.5 or 10 had minimal bias. Five studies achieved scores of 10: two in Hong Kong [54–55], and one each in Taiwan [56], Israel [66] and South Africa [67]. Five studies achieved scores of 9.5: three in the USA [48–50], one in the UK [53], and the Balkans study [46]. Five other studies achieved scores of 9, from Saudi Arabia [65], the UK [52], Germany [62], Hun- gary [68], and Taiwan [57]. Four studies scored 7, and two scored 6; here we concluded risk of bias was high, particularly regarding selection bias and non-response bias. Discussion This systematic review identified 30 studies of the prevalence of either four or five forms of child maltreatment, conducted in 22 nations. In addition, many other studies have been con- ducted of three or fewer maltreatment types, such as studies of sexual, physical and emotional abuse. These have been conducted on a stand-alone basis [76], or as part of a systematic cam- paign supported by a global public private partnership [77]. By 2019, the Violence Against Children Surveys (VACS), which also measure the prevalence of physical, sexual and PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 11 / 22 Improving measurement of child abuse and neglect: A systematic review Table 3. Quality assessment of included studies. Discussion Quality assessment criteria Study Location Target population nationally representative (age, sex) Representative sampling frame Random selection Non- response bias Data collected from participant Questions congruent with conceptual understandings of maltreatment Instrument reliability and validity Consistent data collection Appropriate numerators, denominators Quality score Al Muneef 2017 Saudi Arabia 2 1 1 1 1 1 0 1 1 9 Chan 2011 Hong Kong 2 1 1 1 1 1 1 1 1 10 Chan 2011 Hong Kong 2 1 1 1 1 1 1 1 1 10 Christoffersen 2013 Denmark 1 1 1 1 1 0 0 1 1 7 Denholm 2013 UK 1 0 1 1 1 0 0 1 1 6 Euser 2013 Netherlands 1 0 1 1 1 1 0 1 1 7 Feng 2015 Taiwan 2 1 1 1 1 1 1 1 1 10 Finkelhor 2005 USA 2 1 1 1 05 0 1 1 1 85 Finkelhor 2009 USA 2 1 1 1 05 1 1 1 1 95 Finkelhor 2014 USA 2 1 1 1 05 1 1 1 1 95 Finkelhor 2015 USA 2 1 1 1 05 1 1 1 1 95 Hauser 2011 Germany 1 1 1 0 1 0 1 1 1 7 Lev-Wiesel 2018 Israel 2 1 1 1 1 1 1 1 1 10 May-Chahal 2005 UK 2 1 1 1 1 1 0 1 1 9 Nagy 2019 Hungary 2 1 1 1 1 0 1 1 1 9 Nikolaidis 2018 Balkans 2 1 1 0.5 1 1 1 1 1 9.5 Radford 2013 UK 2 1 1 1 05 1 1 1 1 95 Schick 2016 Switzerland 2 1 1 1 1 0 1 1 1 9 Shen 2016 Taiwan 2 1 1 0 1 1 1 1 1 9 Tsuboi 2015 Japan 2 1 1 0 1 0 0 1 1 7 van der Kooij 2015 Suriname 1 0 1 0 1 1 0 1 1 6 Ward 2018 South Africa 2 1 1 1 1 1 1 1 1 10 Witt 2017 Germany 2 1 1 0 1 0 1 1 1 8 https://doi.org/10.1371/journal.pone.0227884.t003 Table 3. Quality assessment of included studies. PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 12 / 22 Improving measurement of child abuse and neglect: A systematic review emotional abuse, had been conducted in 16 countries and were being planned in a further eight countries in Africa, Asia and the Caribbean [30,77–78]. Discussion Other studies have considered the prevalence of a mixture of peer violence and maltreatment by parents or caregivers [79– 80]. Accordingly, a good deal of evidence has been generated about the prevalence of child maltreatment in several dozen nations, representing substantial progress in the international understanding of the epidemiology of child maltreatment. However, this review has highlighted the fact that the vast majority of nations lack reliable benchmark national preva- lence data on a comprehensive assessment of maltreatment, including measurement of four or five of the recognised five types of maltreatment, and almost all lack follow-up studies to estab- lish trends over time. This study demonstrates the urgent need to conduct more rigorous prev- alence studies, particularly those by measuring all relevant types of maltreatment, to generate more accurate understandings of the extent of maltreatment, and to enable progress in reduc- ing child maltreatment against the SDG target. Our review also shows that there is substantial variation in study participants across the dif- ferent studies, limiting comparability and introducing certain strengths and limitations which are important to consider in designing future work. Several studies obtained data using parents as proxies for children under 10, and reported reliable responses. This approach may capture data about very young children’s experiences that is otherwise unattainable, although accurate estimates rely on parents being both knowledgeable and truthful in their responses [47]. Yet, the literature reports no evidence of reporter bias in comparisons of adult proxy and youth self-report data [47,48]. Arguably, from a public health perspective, studies provide most comprehensive and reli- able estimates when capturing prevalence data over the entire span of childhood up to age 18. Furthermore, where a study’s participants are children and or young adolescents, past year incidence data is useful. Over half of the studies in this review included children as respon- dents. In these studies, responses benefitted from being direct and proximate to the experience as well as capturing useful stratified data about single year incidence in a closely contempora- neous time period. Developmental evidence suggests children’s and adolescents’ participation is entirely appropriate. While adolescents may generally differ from adults in the attainment of psychosocial capacities to understand long-term consequences, regulate conduct, and with- stand social and emotional pressures, their cognitive capacity is not substantially different from that of adults [81–84]. PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 Discussion One limitation of such studies is that they will not obtain recent proximal data of single year incidence. An additional potential limitation, yet to be fully analysed, may be that retrospective accounts are affected by various kinds of recall bias. We acknowledge that some have argued that retrospective studies do not provide data about child abuse experiences that is as accurate as prospective studies [92–93] and have cau- tioned against sole reliance on retrospective accounts, especially where prevalence estimates are used to draw causal inferences about the effect of maltreatment on biomedical diseases. A recent systematic review and meta-analysis concluded that prospective and retrospective mea- sures of childhood maltreatment identify different groups of individuals [94]. However, it was also recognised that prospective measures may have lower sensitivity than retrospective mea- sures of the experience of maltreatment, and concluded that “the low agreement between pro- spective and retrospective measures cannot be interpreted to directly indicate poor validity of retrospective measures” and that retrospective measures could have greater ability to identify true cases [94]. The well-known discrepancies between true maltreatment rates and those recorded in many data sources used for prospective studies is attributable to the low correla- tion between actual experiences and their representation in official data such as crime statistics and child protection service records. Few maltreatment experiences are ever brought to the attention of criminal justice agencies or child protection services. The caution urged regarding retrospective reports appropriately appears more directed towards studies considering causa- tion of disease than estimation of population prevalence. It is also accepted that lack of validity tends to underreport the experience of abuse [95–97], and studies of test-retest reliability regarding retrospective accounts have indicated general stability over time [98]. We acknowl- edge that retrospective reports may have compromised validity for various reasons, including motivational factors and memory biases, and measurement features including poorly worded questions [92,94]. Overall, however, our view is that retrospective studies of child maltreat- ment, especially when well-designed with behaviourally-specific questions grounded in sound constructs of maltreatment, with representative samples of the population, offer the opportu- nity to obtain sufficiently accurate estimates of the prevalence of child maltreatment experiences. The fourth finding is that while considerable investment is required for all kinds of approach, viable approaches to survey administration are available for diverse geographical settings to accommodate large and small nations, and attain sufficient participation. Discussion Similarly, apart from those still in early developmental stages, chil- dren’s cognition and reliable episodic memory is sufficiently developed to enable participation in survey research [85–86]. This justifies the design of instruments for child and adolescent participants, including the careful approach of the developers of the Juvenile Victimization Questionnaire in designing an instrument suitable for participants as young as eight [72]. Ethically, there is no impediment to involving child and adolescent participants [87]. Ado- lescents and children are cognitively capable of providing their own consent, and are ethically entitled to do so as autonomous individuals. Moreover, adolescents and children have rights to freedom of expression, and bear the right of participation in matters affecting them. While there remains no consensus on the most justifiable approach to confidentiality and welfare [87–90], we assert that studies can adopt robust measures to balance imperatives of attaining sufficient study participation, while ensuring participant welfare and confidentiality. While confidentiality is a foundational principle in these studies, the exception to this, conveyed to youth participants at the outset, that cases of current or imminent significant risk of danger may be referred to welfare authorities, has been found not to affect response rates [38,53]. Alongside this, studies can adopt stepwise approaches drawing on multiple psychological and legal resources to support participants who disclose severe incidents or who experience distress [87]. However, it is important not overstate the frequency of distress. Several studies have 13 / 22 PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 Improving measurement of child abuse and neglect: A systematic review found low rates of distress among youth participants in studies of maltreatment, and the level of youth distress does not differ significantly from that of adults. Furthermore, even distressed participants mostly maintain their involvement was worthwhile [38,91]. A recent US study, for example, found only 0.8% of participants aged 10–17 reported being “pretty or a lot” upset by answering the questions, and even this did not unduly affect their reported willingness to par- ticipate [91]. An associated finding is that children in high-risk sub-populations, such as those in out-of-home care, have not been well represented, leading to likely underestimates of preva- lence and scarce evidence about specific risk profiles. Studies that rely on adults’ retrospective accounts offer the substantial benefit of capturing data about experiences across childhood. PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 Discussion p g p Our sixth finding is that instruments must soundly operationalise constructs of each mal- treatment type by describing them in a way congruent with sound conceptual understandings. This review and critical appraisal found that instruments most often adopted unsound con- structs and operationalisation of neglect, and emotional abuse. In particular, many studies did not consider sufficient operational categories of these maltreatment types as required by sound conceptual models, which will lead to under-estimates of prevalence, and will miss the oppor- tunity to capture important information about the nature of specific experiences. Other studies used broad or vague conceptual expressions, which will have the opposite effect of over-esti- mating prevalence. This finding provides a contextual demonstration of the problem of unsound constructs compromising reliability and validity in general [33,34], and of the ongo- ing challenge to this field to adopt sound constructs of maltreatment and sound behaviourally- specific examples of these constructs [99]. Additionally in this regard, many studies asked only one question about a maltreatment type, which leads to underestimates of prevalence [36]. Sin- gle-item assessment, even through a compound question involving multiple elements of a con- struct, cannot capture accurate or nuanced data and should be avoided wherever possible. Finally, we found few questions about educational neglect. Arguably, since education is a human right recognised by the United Nations Convention on the Rights of the Child article 28, and is a condition for human flourishing [100] and a protective factor against multiple adversities such as child marriage [101], this is a significant dimension of neglect warranting greater priority. We recommend particularly close attention to how future studies conceptual- ise and operationalise these forms of maltreatment. A seventh finding is that few studies asked detailed follow-up questions about the child’s relationship with the person inflicting the acts, and the severity and frequency of the acts. Gen- erally, studies using the JVQ asked the most detailed follow-up questions. Obtaining informa- tion about the severity, frequency, timing, and relational setting of abuse and neglect is important, since the closeness of the relationship between the person maltreating the child and the child can have significant effects [102–103], and the timing of maltreatment is also impor- tant, with studies finding effects for both sex and age [104]. Discussion The implications of this are clear for future study design. School-based studies appeared most often in small nations, which may more readily facilitate centralised educational sector endorsement for the research, or which may have a high commitment to social research. When school lead- ers agree for their school to participate, children generally participate at a very high rate. Simi- larly, household studies identified in this review generally occurred in small nations. Both school-based and household studies require substantial numbers of staff, but may be most fea- sible where labour costs are manageable and where the social ecology is of sufficient strength to support and perhaps even require direct personal involvement in such research. In larger PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 14 / 22 Improving measurement of child abuse and neglect: A systematic review nations, for reasons of practicability and cost, studies used CATI and achieved satisfactory response rates. Perhaps for reasons of cost, and practical difficulty, a challenge remains to cap- ture the experience of culturally and linguistically diverse sub-populations, and hard to reach groups such as children who are not in school, or who are in out of home care. Future research could consider optimal local strategies to respond to this challenge. Our fifth finding is that selection, design and testing of an appropriate instrument appears an enduring challenge. In this regard, two coexisting needs must be balanced by any study: first, to be practicable in terms of the time and cost required to design, test and administer an instrument and minimise missing data; and second, to achieve sufficient comprehensiveness and ensure construct validity by describing maltreatment types in a way congruent with con- ceptual understandings [33]. Our review showed that a wide variety of instruments have been used, with psychometric data often not reported. The JVQ was the instrument most often used in either full-form or short-form; moreover, several studies adapted the original JVQ, some- times adding a considerable number of items. These adapted versions did not appear to have been subjected to testing. While inconsequential modification of a proven instrument obviates the need for re-testing, substantial modification may be further supported by cognitive testing and test-retest reliability. What is relatively clear is that a proven, sound instrument offers both practicable and methodological benefits over a blended tool, or a new unproven instrument. PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 Discussion From a public health perspective, the measurement of maltreatment should ideally move beyond raw prevalence, and yield suffi- ciently sensitive and nuanced information about these key contextual features of the maltreat- ment to inform future public health policy and prevention efforts, including the indication of PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 15 / 22 Improving measurement of child abuse and neglect: A systematic review priority areas for responses. The addition of such questions presents challenges for instrument design and implementation, including the time to administer additional questions. However, we recommend such questions wherever possible. Limitations We reviewed studies measuring the traditional forms of child maltreatment, and excluded studies of adverse childhood experiences conceptualised more broadly, such as peer bullying and community violence. Some researchers recommend that studies include both maltreat- ment and these other adversities [37] on the basis that chronic exposure to multiple adversities influences developmental trajectories through the lifespan. However, we applied rigorous eligi- bility criteria of four or five of the recognized maltreatment categories, all clearly associated with adverse sequelae, and which most closely reflect specific SDG targets of caregiver abuse and any sexual violence. Recent outcomes of the ACE study itself have only focused on these five types and three classes of household dysfunction [18]. Additionally, our data extraction method for the quality assessment was not formally validated, but we adopted an approach similar to that used elsewhere [32,35,45] considering key variables in detail. Similarly, while there were no previously validated risk of bias measures for this specific type of prevalence study, we used a method with high interrater agreement that has been used elsewhere [45], including in prevalence studies of abuse and interpersonal violence [105–106]. Our approach to risk of bias adopted a conservative approach, and reasonably concluded that studies scoring 9.5 or 10 had minimal bias. Conclusions This systematic review and analysis has shown nationwide studies of the prevalence of child maltreatment have been conducted, using methods of administration suited to the setting, and involving child participants, adult participants, or both. However, there are few such nation- wide studies of all five or even four maltreatment types, leaving substantial gaps in knowledge about the experience of childhood maltreatment in nearly all countries. Overall, our review and analysis indicates many of the completed studies are generally sound, but some take a more comprehensive and conceptually robust approach to provide nuanced, useful data for researchers and policymakers. To enable measurement of progress against the United Nations Agenda for Sustainable Development Goal 16 of reduction of child abuse, many countries need to invest in robust national prevalence studies. Such studies should measure exposure to domestic violence in addition to physical abuse, sexual abuse, emotional abuse, and neglect. Studies should use an instrument with demonstrated validity and reliability, and must ensure maltreatment types are operationalised appropriately in the questions asked. If participants are children or adolescents under age 18, studies should capture past year incidence, as well as childhood prevalence. Information should be captured about the specific nature, severity and frequency of the maltreatment, and the relationship of the child to the person who inflicted the acts. Such data can best inform the development and monitoring of nationwide prevention efforts. Acknowledgments We acknowledge Juliet Davis, Elizabeth Dallaston, and Andrea Boskovic for providing research assistance. We also thank the journal reviewers for their helpful comments. We acknowledge Juliet Davis, Elizabeth Dallaston, and Andrea Boskovic for providing research assistance. We also thank the journal reviewers for their helpful comments. Supporting information S1 Fig. PRISMA checklist. (DOCX) S1 File. Search strategy. (DOCX) S1 File. Search strategy. (DOCX) 16 / 22 PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 Improving measurement of child abuse and neglect: A systematic review S2 File. Data extraction template. (DOCX) S3 File. Extracted survey items. (DOCX) S4 File. Quality assessment tool. (DOCX) S5 File. Prevalence rates of included st (DOCX) S5 File. Prevalence rates of included studies. (DOCX) Author Contributions Conceptualization: Ben Mathews, Rosana Pacella, Michael P. Dunne, Cicely Marston. Conceptualization: Ben Mathews, Rosana Pacella, Michael P. Dunne, Cicely Marston Data curation: Marko Simunovic. Data curation: Marko Simunovic. Formal analysis: Ben Mathews, Rosana Pacella, Michael P. Dunne, Marko Simunovic, Cicely Marston. Formal analysis: Ben Mathews, Rosana Pacella, Michael P. Dunne, Marko Simunovic, Cicely Marston. Methodology: Ben Mathews, Rosana Pacella, Michael P. Dunne, Cicely Marston. Project administration: Ben Mathews, Marko Simunovic. Supervision: Ben Mathews. Supervision: Ben Mathews. Supervision: Ben Mathews. Writing – original draft: Ben Mathews. Writing – original draft: Ben Mathews. Writing – original draft: Ben Mathews. Writing – review & editing: Ben Mathews, Rosana Pacella, Michael P. Dunne, Marko Simu- novic, Cicely Marston. Writing – review & editing: Ben Mathews, Rosana Pacella, Michael P. Dunne, Marko Simu- novic, Cicely Marston. 7. Paolucci EO, Genuis ML, Violato C. A meta-analysis of the published research on the effects of child sexual abuse. J Psychol. 2001; 135: 17–36. https://doi.org/10.1080/00223980109603677 PMID: 11235837 PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 References Child Abuse Negl. 2010; 34: 734–41. https://doi.org/10.1016/j.chiabu.2010.03.001 PMID: 20850182 13. Kitzmann KM, Gaylord NK, Holt AR, Kenny ED. Child witnesses to domestic violence: a meta-analytic review. Consult Clin Psychol. 2003; 71: 339–52. https://doi.org/10.1037/0022-006X.71.2.339. 14. Finkelhor D, Ormrod R, Turner HA. Poly-victimization: A neglected component in child victimization. Child Abuse Negl. 2007; 31: 7–26. https://doi.org/10.1016/j.chiabu.2006.06.008 PMID: 17224181 15. Wolfe DA, Crooks CV, Lee V, McIntyre-Smith A, Jaffe PG. The effects of children’s exposure to domestic violence: a meta-analysis and critique. Clin Child Fam Psychol Rev. 2003; 6: 171–87. https:// doi.org/10.1023/a:1024910416164 PMID: 14620578 16. Hughes K, Bellis MA, Hardcastle KA, Sethi D, Butchart A, Mikton C, et al. The effect of multiple adverse childhood experiences on health: a systematic review and meta-analysis. Lancet Public Health. 2017; 2: e356–66. https://doi.org/10.1016/S2468-2667(17)30118-4 PMID: 29253477 17. Felitti VJ. Adverse childhood experiences and adult health. Acad Pediatr. 2009; 9: 131–32. https://doi. org/10.1016/j.acap.2009.03.001 PMID: 19450768 18. Flaherty EG, Thompson R, Dubowitz H et al. Adverse Childhood Experiences and Child Health in Early Adolescence. JAMA Pediatr. 2013; 167: 622–29. https://doi.org/10.1001/jamapediatrics.2013. 22 PMID: 23645114 19. Dong M, Giles WH, Felitti VJ, Dube SR, Williams JE, Chapman DP, et al. Insights into causal path- ways for ischemic heart disease. Circulation. 2004; 110: 1761–66. https://doi.org/10.1161/01.CIR. 0000143074.54995.7F PMID: 15381652 20. Danese A, McEwen BS. Adverse childhood experiences, allostasis, allostatic load, and age-related disease. Physiol Behav. 2012; 106: 29–39. https://doi.org/10.1016/j.physbeh.2011.08.019 PMID: 21888923 21. Danese A, Pariante CM, Caspi A, Taylor A, Poulton R. Childhood maltreatment predicts adult inflam- mation in a life-course study. Proc Natl Acad Sci. 2007; 104: 1319–24. https://doi.org/10.1073/pnas. 0610362104 PMID: 17229839 22. Danese A, Moffitt TE, Harrington H, Milne BJ, Polanczyk Z, Pariante CM, et al. Adverse childhood experiences and adult risk factors for age-related disease: Depression, inflammation, and clustering of metabolic risk markers. Arch Ped Adol Med. 2009; 163: 1135–43. https://doi.org/10.1001/ archpediatrics.2009.214. 23. Teicher MH, Samson JA. Annual research review: enduring neurobiological effects of childhood abuse and neglect. J Child Psychol Psychiatry. 2016; 57: 241–66. https://doi.org/10.1111/jcpp.12507 PMID: 26831814 24. Shalev I, Moffitt TE, Sugden K, Williams B, Houts RM, Danese A, et al. Exposure to violence during childhood is associated with telomere erosion from 5 to 10 years of age: a longitudinal study. Mol Psy- chiatry. 2013; 18: 576–81. https://doi.org/10.1038/mp.2012.32 PMID: 22525489 25. Danese A, Moffitt TE, Arseneault L, Bleiberg BA, Dinardo PB, Gandelman SB, et al. References 1. Gilbert R, Widom CS, Browne K, Fergusson D, Webb E, Janson S. Burden and consequences of child maltreatment in high-income countries. Lancet. 2009; 373(9657): 68–81. https://doi.org/10.1016/ S0140-6736(08)61706-7 PMID: 19056114 2. Hillis S, Mercy J, Amobi A, Kress H. Global prevalence of past-year violence against children: a sys- tematic review and minimum estimates. Pediatr. 2016; 137: e20154079. https://doi.org/10.1542/peds. 2015-4079. 3. Gershoff ET, Grogan-Kaylor A. Spanking and Child Outcomes: Old Controversies and New Meta- Analyses. J Fam Psychol. 2016; 30: 453–69. https://doi.org/10.1037/fam0000191 PMID: 27055181 4. Lansford JE, Dodge KA, Pettit GS, Bates JE, Crozier J, Kaplow J. A 12-year prospective study of the long-term effects of early child physical maltreatment on psychological, behavioral, and academic problems in adolescence. Arch Pediatr Adolesc Med. 2002; 156: 824–30. https://doi.org/10.1001/ archpedi.156.8.824 PMID: 12144375 5. Norman RE, Byambaa M, De R, Butchart A, Scott J, Vos T. The long-term health consequences of child physical abuse, emotional abuse, and neglect: a systematic review and meta-analysis. PLOS Med. 2012; 9: e1001349. https://doi.org/10.1371/journal.pmed.1001349 PMID: 23209385 6. Chen LP, Murad MH, Paras ML, Colbenson KM, Sattler AL, Goranson EN, et al. Sexual abuse and life- time diagnosis of psychiatric disorders: Systematic review and meta-analysis. Mayo Clin Proc. 2010; 85: 618–29. https://doi.org/10.4065/mcp.2009.0583 PMID: 20458101 7. Paolucci EO, Genuis ML, Violato C. A meta-analysis of the published research on the effects of child sexual abuse. J Psychol. 2001; 135: 17–36. https://doi.org/10.1080/00223980109603677 PMID: 11235837 17 / 22 PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 Improving measurement of child abuse and neglect: A systematic review 8. Egeland B. Taking stock: Childhood emotional maltreatment and developmental psychopathology. Child Abuse Negl. 2009; 33: 22–26. https://doi.org/10.1016/j.chiabu.2008.12.004 PMID: 19167068 9. Maguire S, Williams B, Naughton A, Cowley LE, Tempest V, Mann MK, et al. A systematic review of the emotional, behavioural and cognitive features exhibited by school-aged children experiencing neglect or emotional abuse. Child Care Health Dev. 2015; 41: 641–53. https://doi.org/10.1111/cch. 12227 PMID: 25733080 10. Taillieu TL, Brownridge DA, Sareen J, Afifi TO. Childhood emotional maltreatment and mental disor- ders: results from a nationally representative adult sample from the United States. Child Abuse Negl. 2016; 59: 1–12. https://doi.org/10.1016/j.chiabu.2016.07.005 PMID: 27490515 11. Hildyard KL, Wolfe DA. Child neglect: developmental issues and outcomes. Child Abuse Negl. 2002; 26: 679–95. https://doi.org/10.1016/s0145-2134(02)00341-1 PMID: 12201162 12. Hamby SL, Finkelhor D, Turner HA, Ormrod R. The overlap of witnessing partner violence with child maltreatment and other victimizations in a nationally representative survey of youth. References The origins of cog- nitive deficits in victimized children: implications for neuroscientists and clinicians. Am J Psychiatry. 2016; 174: 349–61. https://doi.org/10.1176/appi.ajp.2016.16030333 PMID: 27794691 26. Moffitt TE, the Klaus-Grawe Think Tank. Childhood exposure to violence and lifelong health: clinical intervention science and stress-biology research join forces. Dev Psychopathol. 2013; 25: 1619–34. https://doi.org/10.1017/S0954579413000801 PMID: 24342859 27. Fang X, Fry DA, Brown DS, Mercy JA, Dunne MP, Butchart AR, et al. The burden of child maltreatment in the East Asia and Pacific region. Child Abuse Negl. 2015; 42:146–62. https://doi.org/10.1016/j. chiabu.2015.02.012 PMID: 25757367 28. Sustainable Development Goals. United Nations General Assembly; 2015. Available from: https:// sustainabledevelopment.un.org/ 18 / 22 PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 Improving measurement of child abuse and neglect: A systematic review 29. Hammond WR, Whitaker DJ, Lutzker JR, Mercy J, Chin PM. Setting a violence prevention agenda at the centers for disease control and prevention. Aggress V Behav. 2006; 11: 112–19. https://doi.org/10. 1016/j.avb.2005.07.003 30. Sumner SA, Mercy A, Saul J, Motsa-Nzuza N, Kwesigabo G, Buluma R, et al. Prevalence of sexual violence against children and use of social services-seven countries, 2007–2013. Morb Mortal Wkly Rep. 2015; 64: 565–69. 31. Stoltenborgh M, Bakermans-Kranenburg M, Lenneke RA Alink, van Ijzendoorn M. The Prevalence of Child Maltreatment across the Globe: Review of a Series of Meta-Analyses. Child Abuse Rev. 2015; 24: 37–50. https://doi.org/10.1002/car.2353. 32. Moore SE, Scott JG, Ferrari AJ, Mills R, Dunne MP, Erskine HE, et al. Burden attributable to child mal- treatment in Australia. Child Abuse Negl. 2015; 48: 208–20. https://doi.org/10.1016/j.chiabu.2015.05. 006 PMID: 26056058 33. Manly JT. Advances in research definitions of child maltreatment. Child Abuse Negl. 2005; 29: 425– 39. https://doi.org/10.1016/j.chiabu.2005.04.001 PMID: 15970318 34. Hamby SL. Intimate partner and sexual violence research: Scientific progress, scientific challenges, and gender. Trauma Violence Abuse. 2014; 15: 149–58. https://doi.org/10.1177/1524838014520723 PMID: 24464246 35. Hovdestadt W, Campeau A, Potter D, Tonmyr L. A Systematic Review of Childhood Maltreatment Assessments in Population-Representative Surveys Since 1990. PLoS ONE. 2015; 10: e0123366. https://doi.org/10.1371/journal.pone.0123366 PMID: 25993481 36. Fisher B. The effects of survey question wording on rape estimates: Evidence from a quasi-experi- mental design. Violence Against Women. 2009; 15: 133–47. https://doi.org/10.1177/ 1077801208329391 PMID: 19126832 37. Anda RF, Butchart A, Felitti VJ, Brown DW. Building a Framework for Global Surveillance of the Public Health Implications of Adverse Childhood Experiences. Am J Prev Med. 2010; 39: 93–98. https://doi. org/10.1016/j.amepre.2010.03.015 PMID: 20547282 38. Laurin J, Wallace C, Draca J, Tonmyr L. PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 References Child Abuse Negl. 2014; 38: 1421–35. https:// doi.org/10.1016/j.chiabu.2014.05.005 PMID: 24953383 50. Finkelhor D, Turner HA, Shattuck A, Hamby SL. Prevalence of Childhood Exposure to Violence, Crime, and Abuse: Results From the National Survey of Children’s Exposure to Violence. JAMA Pedi- atrics. 2015; 169: 746–54. https://doi.org/10.1001/jamapediatrics.2015.0676 PMID: 26121291 51. Denholm R, Power C, Li L, Thomas C. Child Maltreatment and Household Dysfunction in a British Birth Cohort. Child Abuse Rev. 2013; 22: 340–53. https://doi.org/10.1093/ije/dyu071. 52. May-Chahal C, Cawson P. Measuring child maltreatment in the United Kingdom: a study of the preva- lence of child abuse and neglect. Child Abuse Negl. 2005; 29: 969–84. https://doi.org/10.1016/j. chiabu.2004.05.009 PMID: 16165212 53. Radford L, Corral S, Bradley C, Fisher HL. The prevalence and impact of child maltreatment and other types of victimization in the UK: findings from a population survey of caregivers, children and young people and young adults. Child Abuse Negl. 2013; 37: 801–13. https://doi.org/10.1016/j.chiabu.2013. 02.004 PMID: 23522961 54. Chan KL. Children exposed to child maltreatment and intimate partner violence: a study of co-occur- rence among Hong Kong Chinese families. Child Abuse Negl. 2011; 35: 532–42. https://doi.org/10. 1016/j.chiabu.2011.03.006 PMID: 21816472 55. Chan KL, Brownridge DA, Yan E, Fong DYT, Tiwari A. Child maltreatment polyvictimization: Rates and short-term effects on adjustment in a representative Hong Kong sample. Psychol Viol. 2011; 1: 4– 15. https://doi.org/10.1037/a0020284. 56. Feng JY, Chang YT, Chang HY, Fetzer S, Wang JD. Prevalence of different forms of child maltreat- ment among Taiwanese adolescents: a population-based study. Child Abuse Negl. 2015; 42: 10–19. https://doi.org/10.1016/j.chiabu.2014.11.010 PMID: 25477233 57. Shen AC-T, Feng JY, Feng J-Y, Wei H-S, Hsieh Y-P, Huang SC-Y, et al. Who gets protection? A national study of multiple victimization and child protection among Taiwanese children. J Interpers Vio- lence. 2016;1–25. https://doi.org/10.1177/0886260516670885. 58. Hauser W, Schmutzer G, Brahler E, Glaesmer H. Maltreatment in childhood and adolescence: results from a survey of a representative sample of the German population. Dtsch Arztebl Int. 2011; 108:287– 94. https://doi.org/10.3238/arztebl.2011.0287 PMID: 21629512 59. Witt A, Brown RC, Plener PL, Brahler E, Fegert JM. Child maltreatment in Germany: prevalence rates in the general population. Child Adolesc Psychiatry Ment Health. 2017; 11:47. https://doi.org/10.1186/ s13034-017-0185-0 PMID: 28974983 60. Christoffersen M, Armour C, Lasgaard M, Andersen T, Elklit A. The prevalence of four types of child- hood maltreatment in Denmark. Clin Pract Epidemiol Ment Health. 2013; 9: 149–56. https://doi.org/10. 2174/1745017901309010149 PMID: 24155769 61. References Youth self-report of child maltreatment in representative sur- veys: a systematic review. Health Promotion and Chronic Disease Prevention in Canada: Research, Policy and Practice. 2018; 38(2): 37–54. 39. Moher D, Liberati A, Tetzlaff J, Altman DG, The Prisma Group. Preferred reporting items for system- atic reviews and meta-analyses: the PRISMA statement. PLoS Med. 2009; 6: e1000097. https://doi. org/10.1371/journal.pmed.1000097 PMID: 19621072 40. Mathews B, Norman R, Dunne MP, Marston C. Improving measurement of child abuse and neglect: a systematic review and analysis of national prevalence studies. PROSPERO International prospective register of systematic reviews. #CRD42017068120, 8 June 2017 https://www.crd.york.ac.uk/ PROSPERO 41. World Health Organization and International Society for Prevention of Child Abuse and Neglect. Pre- venting child maltreatment: A guide to taking action and generating evidence. Geneva: World Health Organization; 2006. http://www.who.int/iris/handle/10665/43499. 42. Mathews B, Collin-Ve´zina D. Child Sexual Abuse: Towards a conceptual model and definition. Trauma, Violence Abuse 2019; 20: 131–148. https://doi.org/10.1177/1524838017738726 PMID: 29333990 43. Glaser D. How to deal with emotional abuse and neglect—Further development of a conceptual frame- work (FRAMEA). Child Abuse Negl. 2011; 35: 866–75. https://doi.org/10.1016/j.chiabu.2011.05.013 PMID: 22014553 44. Dubowitz H, Newton R, Litrownik A, Lewis T, Briggs EC, Thompson R, et al. Examination of a concep- tual model of child neglect. Child Maltreat. 2005; 10: 173–85. https://doi.org/10.1177/ 1077559505275014 PMID: 15798011 45. Hoy D, Brooks P, Woolf A, Blyth F, March L, Bain C, et al. Assessing risk of bias in prevalence studies: Modification of an existing tool and evidence of interrater agreement. J Clin Epidemiol. 2012; 65: 934– 39. https://doi.org/10.1016/j.jclinepi.2011.11.014 PMID: 22742910 46. Nikolaidis G, Petroulaki K, Zarokosta F, Tsirigoti A, Hazizaj A, Cenko E, et al. Lifetime and past-year prevalence of children’s exposure to violence in 9 Balkan countries: the BECAN study. Child Adolesc Psychiatry Ment Health. 2018; 12:1. https://doi.org/10.1186/s13034-017-0208-x PMID: 29308086 47. Finkelhor D, Ormrod RK, Turner HA, Hamby SL. The victimization of children and youth: a comprehen- sive, national survey. Child Maltreat. 2005; 10: 5–25. https://doi.org/10.1177/1077559504271287 PMID: 15611323 19 / 22 PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 Improving measurement of child abuse and neglect: A systematic review 48. Finkelhor D, Turner HA, Ormrod R, Hamby SL. Violence, abuse, and crime exposure in a national sample of children and youth. Pediatrics. 2009; 124: 1411–23. https://doi.org/10.1542/peds.2009- 0467 PMID: 19805459 49. Finkelhor D, Vanderminden J, Turner H, Hamby S, Shattuck A. Child maltreatment rates assessed in a national household survey of caregivers and youth. PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 References Euser S, Alink LR, Pannebakker F, Vogels T, Bakermans-Kranenburg MJ, Van IMH. The prevalence of child maltreatment in the Netherlands across a 5-year period. Child Abuse Negl 2013; 37(10): 841– 51. https://doi.org/10.1016/j.chiabu.2013.07.004 PMID: 23938018 62. Schick M, Schonbucher V, Landolt MA, Schnyder U, Xu W, Maier T, et al. Child Maltreatment and Migration: A Population-Based Study Among Immigrant and Native Adolescents in Switzerland. Child Maltreat. 2016; 21: 3–15. https://doi.org/10.1177/1077559515617019 PMID: 26590238 63. Tsuboi S, Yoshida H, Ae R, Kojo T, Nakamura Y, Kitamura K. Prevalence and demographic distribu- tion of adult survivors of child abuse in Japan. Asia Pac J Public Health. 2015; 27: 2578–86. https:// doi.org/10.1177/1010539513488626. 64. van der Kooij IW, Nieuwendam J, Bipat S, Boer F, Lindauer RJL, Graafsma TLG. A national study on the prevalence of child abuse and neglect in Suriname. Child Abuse Negl. 2015; 47: 153–61. https:// doi.org/10.1016/j.chiabu.2015.03.019 PMID: 25937450 65. Al Muneef M, El Choueiry N, Saleheen H, Al-Eissa M. The impact of Adverse Childhood Experiences on social determinants among Saudi adults. Jnl Public Health. 2017; 40(3): e219–e227. https://doi. org/10.1093/pubmed/fdx177. 66. Lev-Wiesel R, Eisikovits Z, First M, Gottfried R, Mehlhausen D. Prevalence of Child Maltreatment in Israel: A national epidemiological study. Journal Child Adolescent Trauma. 2018; 11:141–150. https:// doi.org/10.1007/s40653-016-0118-8. 20 / 22 PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 Improving measurement of child abuse and neglect: A systematic review 67. Ward C, Artz L, Leoschut L, Kassanjnee R, Burton P. Sexual violence against children in South Africa: a nationally representative cross-sectional study of prevalence and correlates. Lancet Glob Health. 2018; 6: e460–68. https://doi.org/10.1016/S2214-109X(18)30060-3 PMID: 29530424 68. Nagy AU, Szabo IK, Hann E, Kosa K. Measuring the prevalence of adverse childhood experiences by survey research methods. Int J Environ Res Public Health. 2019; 16: 1048. https://doi.org/10.3390/ ijerph16061048 69. Straus MA, Hamby SL, Finkelhor D, Moore DW, Runyan D. Identification of Child Maltreatment with the Parent-Child Conflict Tactics Scales: Development and Psychometric Data for a National Sample of American Parents. Child Abuse Negl. 1998; 22: 249–70. https://doi.org/10.1016/s0145-2134(97) 00174-9 PMID: 9589178 70. Straus MA, Hamby SL, Boney-McCoy S, Sugarman DB. The Revised Conflict Tactics Scales (CTS2). J Fam Issues. 1996; 17: 283–316. https://doi.org/10.1177/019251396017003001 71. Zolotor A, Runyan D, Dunne MP et al. ISPCAN Child Abuse Screening Tools Children’s Version (ICAST-C): Instrument development and multi-national pilot testing. Child Abuse Negl. 2009; 33: 833– 41. https://doi.org/10.1016/j.chiabu.2009.09.004 PMID: 19857897 72. Finkelhor D, Hamby SL, Ormrod R, Turner H. References The Juvenile Victimization Questionnaire: Reliability, validity, and national norms. Child Abuse Negl. 2005; 29: 383–412. https://doi.org/10.1016/j.chiabu. 2004.11.001 PMID: 15917079 73. Bernstein DP, Stein JA, Newcomb MD et al. Development and validation of a brief screening version of the Childhood Trauma Questionnaire. Child Abuse Negl. 2003; 27: 169–90. https://doi.org/10.1016/ s0145-2134(02)00541-0 PMID: 12615092 74. Douglas EM, Straus MA. Assault and injury of dating partners by university students in 19 countries and its relation to corporal punishment experienced as a child. Eur J Criminol. 2006; 3: 293–318. https://doi.org/10.1177/1477370806065584. 75. Straus MA, Douglas EM. A Short form of the Revised Conflict Tactics Scales, and Typologies for Severity and Mutuality. Violence Vict. 2004; 19: 507–21. https://doi.org/10.1891/vivi.19.5.507.63686 PMID: 15844722 76. Meinck F, Cluver LD, Boyes ME. Physical, emotional and sexual adolescent abuse victimisation in South Africa: prevalence, incidence, perpetrators and locations. J Epidemiol Community Health. 2016; 70: 910–916. https://doi.org/10.1136/jech-2015-205860 PMID: 26962202 77. Chiang L, Kress H, Sumner SA, Gleckel J, Kawemama P, Gordon RN. Violence Against Children Sur- veys (VACS): towards a global surveillance system. Inj Prev. 2016; 22(Suppl 1): i17–i22. http://dx.doi. org/10.1136/injuryprev-2015-041820 78. Nguyen KH, Kress H, Villaveces A and Massetti GM. Sampling design and methodology of the Vio- lence Against Children and Youth Surveys. Inj Prev. 2019; 25(4): 321–327. https://doi.org/10.1136/ injuryprev-2018-042916 PMID: 30472679 79. Huang L, Mossige S. Resilience and Poly-Victimization among Two Cohorts of Norwegian Youth. Int J Environ Res Public Health. 2018; 15: 2852. https://doi.org/10.3390/ijerph15122852 80. Mossige S, Huang L. Poly-victimization in a Norwegian adolescent population: Prevalence, social and psychological profile, and detrimental effects. PLoS One. 2017; 12(12): e0189637. https://doi.org/10. 1371/journal.pone.0189637 PMID: 29240805 81. Albert D, Steinberg L. Judgment and decision making in adolescence. J Research Adolescence. 2011; 21: 211–224. https://doi.org/10.1111/j.1532-7795.2010.00724.x 82. Casey BJ, Jones RM, Hare TA. The adolescent brain. Ann N Y Acad Sci. 2008; 1124: 111–126. https://doi.org/10.1196/annals.1440.010 PMID: 18400927 83. Steinberg L. Cognitive and affective development in adolescence. Trends Cogn Neurosci. 2005; 9(2): 69–74. https://doi.org/10.1016/j.tics.2004.12.005 84. Steinberg L, Icenogle G, Shulman E, Breiner K, Chein J, Bacchini D, et al. Around the world, adoles- cence is a time of heightened sensation seeking and immature self-regulation. Dev Sci. 2018; 21(2): e12532. https://doi.org/10.1111/desc.12532 85. Borgers N, De Leeuw E, Hox J. Children as respondents in survey research: cognitive development and response quality. Bull Methodol Sociol. 2000; 66: 60–75. https://doi.org/10.1177/ 075910630006600106 86. Riley AW. Evidence that school-age children can self-report on their health. Ambul Pediatr. 2004; 4(4): 371–376. https://doi.org/10.1367/A03-178R.1 87. References Finkelhor D, Hamby SL, Turner HA, Walsh W. Ethical Issues in Surveys about Children’s Exposure to Violence and Sexual Abuse. In: Cuevas CA, Rennison CM, editors. The Wiley Handbook on the Psy- chology of Violence. Chichester: Wiley; 2016. pp. 24–48. 21 / 22 PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 Improving measurement of child abuse and neglect: A systematic review 88. Allen B. Are researchers ethically obligated to report suspected child maltreatment? A critical analysis of opposing perspectives. Ethics & Behavior. 2009; 19: 15–24. https://doi.org/10.1080/ 10508420802623641 89. Cashmore J. Ethical issues concerning consent in obtaining children’s reports on their experience of violence. Child Abuse Negl. 2006; 30: 969–977. https://doi.org/10.1016/j.chiabu.2006.05.004 PMID: 17011032 90. Meinck F, Steinert J, Sethi D, Gilbert R, Bellis M, Mikton C, et al. Measuring and monitoring national prevalence of child maltreatment: a practical handbook. Geneva: World Health Organization; 2016. 91. Finkelhor D, Vanderminden J, Turner H, Hamby S, Shattuck A. Upset among youth in response to questions about exposure to violence, sexual assault and family maltreatment. Child Abuse Negl. 2014; 38: 217–223. https://doi.org/10.1016/j.chiabu.2013.07.021 PMID: 24004683 92. Widom CS, Raphael KG, DuMont KA. The case for prospective longitudinal studies in child maltreat- ment research: commentary on Dube, Williamson, Thompson, Felitti, and Anda (2004). Child Abuse & Negl. 2004; 28: 715–722. https://doi.org/10.1016/j.chiabu.2004.03.009 93. Widom CS. Are Retrospective Self-reports Accurate Representations or Existential Recollections? JAMA Psychiatry. 2019; 76(6): 567–568. https://doi.org/10.1001/jamapsychiatry.2018.4599 PMID: 30892563 94. Baldwin J, Reuben A, Newbury J, Danese A. Agreement Between Prospective and Retrospective Measures of Childhood Maltreatment: A Systematic Review and Meta-analysis. JAMA Psychiatry. 2019; 76(6): 584–593. https://doi.org/10.1001/jamapsychiatry.2019.0097 PMID: 30892562 95. Hardt J, Rutter M. Validity of adult retrospective reports of adverse childhood experiences: Review of the evidence. Scand J Child Adolesc Psychiatr Psychol. 2004; 45(2): 260–273. 96. Widom CS, Morris S. Accuracy of adult recollections of childhood victimization Part II: Childhood sex- ual abuse. Psychol Assess. 1997; 9(1): 34–46. 97. Williams LM. Recovered memories of abuse in women with documented child sexual victimization his- tories. J Trauma Stress Disord Treat. 1995; 8: 649–673. 98. Dube SR, Williamson D, Thompson T, Felitti VJ, Anda RF. Assessing the reliability of retrospective reports. Child Abuse Negl. 2004; 28: 729–737. https://doi.org/10.1016/j.chiabu.2003.08.009 PMID: 15261468 99. Herrenkohl T, Herrenkohl R. Assessing a Child’s Experience of Multiple Maltreatment Types: Some Unfinished Business. J Fam Viol. 2009; 24: 485–496. 100. Nussbaum M. Creating Capabilities. Cambridge, MA: Harvard University Press; 2011. 101. PLOS ONE \| https://doi.org/10.1371/journal.pone.0227884 January 28, 2020 References Jain S, Kurz K. New Insights on Preventing Child Marriage: A Global Analysis of Factors and Pro- grams. Washington, DC: ICRW; 2007. 102. Edwards V, Freyd J, Dube S, Anda R, Felitti V. Health outcomes by closeness of sexual abuse perpe- trator: A test of betrayal trauma theory. J Aggression, Maltreatment, Trauma. 2012; 21: 133–148. https://doi.org/10.1080/10926771.2012.648100 103. Trickett P, Noll J, Putnam F. The impact of sexual abuse on female development: Lessons from a multigenerational, longitudinal research study. Dev Psychopathol. 2011; 23: 453–476. https://doi.org/ 10.1017/S0954579411000174 PMID: 23786689 104. Stark L, Seff I, Hoover A, Gordon R, Ligiero D, Massetti G. Sex and age effects in past-year experi- ences of violence amongst adolescents in five countries. PLoS One. 2019; 14(7): e0219073. https:// doi.org/10.1371/journal.pone.0219073 PMID: 31283760 105. Elghossain T, Bott S, Akik C, Obermeyer CM. Prevalence of intimate partner violence against women in the Arab world: a systematic review. BMC Int Health Hum Rights. 2019; 19: 29. https://doi.org/10. 1186/s12914-019-0215-5 PMID: 31640716 106. Yon Y, Mikton C, Gassoumis ZD, Wilber KH. The Prevalence of Self-Reported Elder Abuse Among Older Women in Community Settings: A Systematic Review and Meta-Analysis. Trauma, Violence Abuse. 2019; 20(2): 245–259. https://doi.org/10.1177/1524838017697308 PMID: 29333977 22 / 22
https://openalex.org/W2752558536	https://www.nature.com/articles/s41598-017-10041-6.pdf	English	null	Zinc depletion promotes apoptosis-like death in drug-sensitive and antimony-resistance Leishmania donovani	Scientific reports	2,017	cc-by	10,388	Zinc depletion promotes apoptosis- like death in drug-sensitive and antimony-resistance Leishmania donovani Received: 9 March 2017 Accepted: 25 July 2017 Published: xx xx xxxx Shalini Saini1, Kavita Bharati1, Chandrima Shaha2 & Chinmay K. Mukhopadhyay1 Shalini Saini1, Kavita Bharati1, Chandrima Shaha2 & Chinmay K. Mukhopadhyay1 Micronutrients are essential for survival and growth for all the organisms including pathogens. In this manuscript, we report that zinc (Zn) chelator N,N,N’,N’-tetrakis(2-pyridinylmethyl)-1,2- ethylenediamine (TPEN) affects growth and viability of intracellular pathogen Leishmania donovani (LD) by a concentration and time dependent manner. Simultaneous addition of zinc salt reverses the effect of TPEN. Further experiments provide evidence of apoptosis-like death of the parasite due to Zn-depletion. TPEN treatment enhances caspase-like activity suggesting increase in apoptosis-like events in LD. Specific inhibitors of cathepsin B and Endoclease G block TPEN-induced leishmanial death. Evidences show involvement of reactive oxygen species (ROS) potentially of extra-mitochondrial origin in TPEN-induced LD death. Pentavalent antimonials remained the prime source of treatment against leishmaniasis for several decades; however, antimony-resistant Leishmania is now common source of the disease. We also reveal that Zn-depletion can promote apoptosis-like death in antimony-resistant parasites. In summary, we present a new finding about the role of zinc in the survival of drug sensitive and antimony-resistant LD. Zinc as a cofactor of several hundred enzymes is integral for a wide variety of cellular processes that control numerous biological functions1, 2. Its bioavailability is important for cell proliferation, differentiation, regulation of apoptosis and other various cellular homeostasis mechanisms3. Zinc is an essential trace element in almost all kingdoms of life including humans2, animals4, plants5 and microorganisms6. It plays a structural role in zinc fingers, twists and clusters to regulate gene expression7. Zinc can be stored in metallothionein both in micro- organisms and in the liver of animals6, 8, while inadequate zinc intake is also harmful to the organisms6, 9. The importance of zinc in biology is further underscored because it is the only metal that appears in all classes of enzymes5. As a redox-inactive metal it contributes to maintain cellular redox balance by various mechanisms10. Zinc is also well documented to be involved in apoptosis in mammalian cancer cells3 suggesting this micronutri- ent has various roles in maintaining crucial cellular mechanisms.l g Leishmaniasis is a vector-borne disease that is transmitted by phlebotomine sand flies and caused by more than 20 species of obligate intracellular protozoa of the genus Leishmania11. www.nature.com/scientificreports www.nature.com/scientificreports www.nature.com/scientificreports 1Special Centre for Molecular Medicine, Jawaharlal Nehru University, New Delhi, 110067, India. 2National Institute of Immunology, Aruna Asaf Ali Marg, New Delhi, 110067, India. Correspondence and requests for materials should be addressed to C.K.M. (email: ckm2300@mail.jnu.ac.in) Results Zi h FACS analysis showed marked increase in the intensity of annexin v staining but not any signifi- cant staining of propidium iodide in TPEN-treated (16 h) parasites in comparison to control parasites (Fig. 2A). Mitochondrial membrane depolarisation is another important characteristic of apoptotic cell death. So, we fur- ther examined whether Zn-depletion could influence mitochondrial membrane depolarisation in LD by using JC-1 dye. Mitochondrial membrane depolarization was usually accompanied by a decrease in the fluorescence intensity ratio (590 to 530 nm). Results from TPEN-treated LD for 24 h showed decreased ratio of 590/530 in a concentration dependent manner (Fig. 2B) suggesting zinc depletion could affect mitochondrial membrane depolarization. These results suggest that zinc depletion can promote apoptosis-like events in LD within 24 h of TPEN treatment. Activation of endonucleases and subsequent degradation of genomic DNA are hallmark and ultimate deter- minant of apoptosis that can be determined by TUNEL assay based on the incorporation of modified dUTPs by terminal deoxynucleotidyl transferase (TdT) enzyme at the 3′-OH ends of fragmented DNA19. Therefore, we per- formed TUNEL assay to confirm DNA fragmentation in TPEN-treated LD. We detected that TPEN even at 2 µM was able to cause DNA fragmentation in many cells and at 10 µM in almost all cells (Fig. 2C) further suggesting apoptosis-like death in Zn-depleted LD. Zn-depletion promotes caspase-like activity. Involvement of caspase-dependent and caspase-independent mechanisms of apoptosis has been well established in organisms20–22 and also advocated in Leishmania23–25. Interestingly, the presence of caspases has not yet been confirmed in LD but the presence of caspase-like activity is reported earlier26. So, we examined the role of caspase-like activity in death of TPEN treated LD by prior incubation of parasites with general caspase inhibitor Z-VAD-FMK (10 µM). Results detected by TUNEL assay showed complete blocking of DNA fragmentation by Z-VAD-FMK (Fig. 3A) suggesting involve- ment of caspase-like activity in Zn-depletion induced cell death in LD. The presence of two metacaspases (1 and 2) is reported in LD but their precise roles are not yet well defined27. An earlier report revealed the role of meta- caspase in disuccinyl betulin induced LD death28. So we investigated the role of metacaspase by pretreating LD with specific inhibitor antipain (50 nM). Antipain, which was reported to block disuccinyl betulin induced LD death28, was found ineffective in blocking TPEN induced DNA fragmentation suggesting metacaspases played little role in this process (Fig. 3A). Results Zi h Zinc chelation affects viability of LD promastigotes. LD promastigotes were treated with increasing concentrations of zinc chelator TPEN (0–10 µM) for up to 3-days and viability of the parasites were verified by MTT assay. Result showed that time and concentration dependent increase in cytotoxicity of LD by TPEN treatment (Fig. 1A). We detected about 37%, 23% and 15% cells were viable after 72 h in response to 2 µM, 5 µM and 10 µM TPEN treatment respectively (Fig. 1A). The IC50 was detected as 4.78 µM for 48 h TPEN treatment. We also examined LD growth in a similar condition and about 85% decrease was detected after 72 h by 10 µM TPEN treatment compared to untreated parasites; while 2 µM and 5 µM TPEN reduced LD growth about 40% and 75% respectively. Since, iron is crucial for survival of Leishmania17, 18 and TPEN is also known to chelate ferrous iron but with less affinity than zinc, so we further verified LD viability in presence of specific ferrous iron chelator BPS (4,7-Diphenyl-1,10-phenanthroline- disulfonic acid disodium salt trihydrate). Results showed that LD viability was affected by much higher concentration of BPS (Fig. 1B) compared to TPEN. Similarly, LD growth was affected only about 30% after 72 h treatment with BPS (1 mM). We used specific tetra anionic dye Fluozin-3AM to determine whether TPEN treatment actually resulted in depletion of zinc. We detected decreased Fluozin-3AM sensitivity with increasing concentration of TPEN treatment (Fig. 1C) suggesting TPEN actually could deplete available pool of intracellular zinc in LD. To further establish the role of zinc in LD survival, zinc sulphate (10 µM) was added along with TPEN (10 µM) and Fluozin-3AM sensitivity as well as LD viability were examined. We detected almost complete reversal of Fluozin-3AM sensitivity (Fig. 1C) and LD viability (Fig. 1D) by addition of zinc sulphate suggesting TPEN affected LD viability by chelation of available zinc. The result of MTT assay showed 82.2% of viability reversal from 13.4% by addition of zinc sulphate along with TPEN while only zinc sulphate treatment showed 93.8% viable cells (Fig. 1D). Zinc chelation promotes apoptosis-like death of LD promastigotes. Phosphatidylserine (PS) externalisation is one of the most remarkable features for apoptosis in almost all the cells. So we verified whether Zn-chelation induced LD death also featured PS externalization by using Alexaflour-488 conjugated-annexin v labelling. Zinc depletion promotes apoptosis- like death in drug-sensitive and antimony-resistance Leishmania donovani Leishmania donovani (LD) infection causes visceral leishmaniasis or kala-azar, which is associated with high mortality with an estimated 50,000 deaths each year12. The digenetic protozoan LD exists in two different forms in its life-cycle. One form is infective, flag- ellated promastigote, living in the gut of the sandfly and the other form is non-flagellated amastigotes residing in the mammalian host. The promastigote form is transmitted through sandfly bite and is taken up by macrophages in mammalian hosts. Inside the macrophage the promastigote is transformed into an aflagellated replicative amastigote. VL is widely prevalent in many tropical and subtropical regions of the world including the eastern part of India13. It is fatal if not treated properly. Moreover, drug resistance is a major problem in treating VL14. Current front line antileishmanial therapies are limited by their high costs, limited availability and/or toxicity and the widespread resistance in endemic areas12. Therefore, it is of utmost importance to look for effective new drugs for the treatment of leishmaniasis. Depriving essential micronutrients to parasites may be an effective way 1Special Centre for Molecular Medicine, Jawaharlal Nehru University, New Delhi, 110067, India. 2National Institute of Immunology, Aruna Asaf Ali Marg, New Delhi, 110067, India. Correspondence and requests for materials should be addressed to C.K.M. (email: ckm2300@mail.jnu.ac.in) Scientific REPorTS \| 7: 10488 \| DOI:10.1038/s41598-017-10041-6 1 www.nature.com/scientificreports/ to control VL. Although zinc is a well established micronutrient for most of the organisms; however, the effect of its depletion on LD or any other Leismania species has not been reported so far.h y Interestingly, zinc plays an integral part of virulence of Leishmania parasite. The promastigote form contains high gp63 protease that plays significant role in LD virulence. Zinc is an intergral part of gp63 as a cofactor15. Recently, the presence of Zn-transporter in Leishmania infantum has been reported underscoring the importance of zinc in crucial cellular functions of this parasite16. However, its role in growth and survival of any Leishmania species has not been reported so far. Here we report that Zn-depletion by specific chelator N,N,N′,N′-tetrakis (2-pyridinylmethyl)-1,2-ethylenediamine (TPEN) affects LD survival and growth by promoting cell death resem- bling apoptosis by a reactive oxygen species (ROS) dependent mechanism. We also reveal that antimony- resistant LD parasites are similarly affected by TPEN treatment. Our findings thus suggest an important role of zinc in survival of both drug-sensitive and antimony-resistant LD. Results Zi h Only Z-VAD-FMK or antipain did not show any influence on DNA frag- mentation in LD (data not shown). Further we investigated the effect of TPEN on caspase-like activity in LD by performing caspase 3/7 activity assay. Results showed about 2.3, 10.7 and 23.7-fold increase in caspase3/7 activity after treatment with 2 µM, 5 µM and 10 µM TPEN (48 h) respectively in comparison to untreated parasite (Fig. 3B). Scientific REPorTS \| 7: 10488 \| DOI:10.1038/s41598-017-10041-6 2 ntificreports/ Figure 1. TPEN treatment affects viability of LD promastigotes: (A) Cell viability was verified by MTT assay after TPEN treatment (0–10 µM) at various time point (24–72 h). Results were expressed as percentage of untreated control of each time point performed independently from three experiments. (B) MTT assay was performed after 72 h of BPS treatment (0–2 mM). (C) Parasites were incubated with different concentrations of TPEN (0–10 µM) for 48 h. In one set, 10 µM zinc sulphate was added 1 h prior to TPEN (10 µM) treatment. After the treatment, parasites were stained with FluoZin-3AM (5 µM) for 1 h. Left panel shows fluorescence-labelled Leishmania while right panel represents DIC image of the same parasite (Magnification × 100). (D) MTT assay was performed after 72 h of following treatments; Control, TPEN (10 µM), TPEN (10 µM) + Zinc Sulfate (10 µM) and Zinc Sulfate (10 µM). Zinc sulphate was added 1 h prior to TPEN treatment. Results are expressed as mean ± S.D. from three independent experiments performed in triplicate for A, B and D. www.nature.com/scientificreports/ Figure 1. TPEN treatment affects viability of LD promastigotes: (A) Cell viability was verified by MTT assay after TPEN treatment (0–10 µM) at various time point (24–72 h). Results were expressed as percentage of untreated control of each time point performed independently from three experiments. (B) MTT assay was performed after 72 h of BPS treatment (0–2 mM). (C) Parasites were incubated with different concentrations of TPEN (0–10 µM) for 48 h. In one set, 10 µM zinc sulphate was added 1 h prior to TPEN (10 µM) treatment. After the treatment, parasites were stained with FluoZin-3AM (5 µM) for 1 h. Left panel shows fluorescence-labelled Leishmania while right panel represents DIC image of the same parasite (Magnification × 100). Results Zi h (D) MTT assay was performed after 72 h of following treatments; Control, TPEN (10 µM), TPEN (10 µM) + Zinc Sulfate (10 µM) and Zinc Sulfate (10 µM). Zinc sulphate was added 1 h prior to TPEN treatment. Results are expressed as mean ± S.D. from three independent experiments performed in triplicate for A, B and D. Role of other proteases in TPEN-induced death in LD. It is now well-established that a number of proteases play crucial role in mediating apoptosis29, 30. Cysteine proteases like cathepsins also play important role in apoptotic cell death in mammalian cells31, 32. Therefore, we further explored whether any cysteine protease was involved in Zn-depletion induced death of LD. We performed TUNEL assay and found general cysteine protease inhibitor (E-64d) could rescue TPEN-induced DNA fragmentation in LD (Fig. 3C). To further detect involvement of any specific cysteine protease we used specific inhibitors of cathepsin B (CA074) and cathepsin L (SCP) and found that only cathepsin B inhibitor could rescue TPEN-induced DNA fragmentation but cathepsin L inhibitor showed no effect (Fig. 3C). f g Mitochondrial nuclease Endonuclease G (Endo G) is reported to be involved in mitochondrial replication33 and DNA fragmentation in LD24. So, we further examined the role of Endo G in TPEN-induced LD death by using specific inhibitor aurintricarboxylic acid (ATA). ATA functions by inhibiting binding of the nuclease to Scientific REPorTS \| 7: 10488 \| DOI:10.1038/s41598-017-10041-6 3 www.nature.com/scientificreports/ Figure 2. TPEN treatment promotes apoptosis-like death in LD. (A) Flow cytometric analysis for phosphatidylserine externalisation was detected by annexin V labelling in TPEN (0–10 µM) treated LD after 16 h. The left lower quadrant shows unstained parasite, right lower quadrant denotes annexin V positive cells, upper left quadrant shows only PI positive cells while upper right quadrant indicates both annexin V and PI positive cells. Dot plots are representative of one of three similar results. (B) Fluorometric analysis of mitochondrial membrane potential after TPEN treatment (0–10 µM, 24 h). Results represent as +/− SD from three independent experiments. (C) TUNEL staining was done in LD after TPEN treatment (0–10 µM, 72 h) for assessing DNA fragmentation. Microscopic images show red colour for TUNEL-positive LD (TdT staining) and blue colour for staining nucleus (Hoechst 33342) (magnification × 40). Figure 2. TPEN treatment promotes apoptosis-like death in LD. Results Zi h (A) TUNEL assay was performed in LD after 48 h treatment with 10 µM TPEN, 10 µM TPEN + Z-VAD-FMK (10 µM, added 30 min prior to TPEN) and Antipain (50 nM, added 30 min prior to TPEN). Hoechst 33342 was used for staining nucleus (magnification × 60). Results represent one of the five independent experiments. (B) LD promastigotes were treated with TPEN (0–10 µM) for 48 h. Caspase 3/7 activity assay was performed in cell lysates as per company’s protocol. Results represent data from three independent experiments; error bars +/− SD. (C) TPEN induced DNA fragmentation was detected by TUNEL assay after 48 h. Parasites were treated with TPEN (0 or 10 µM). E-64d (10 µM), CA074 (10 µM), ATA (50 µM) and SCP (10 µM) were added 30 min prior to TPEN treatment. Hoechst 33342 was used for staining nucleus (magnification × 60). Results represent one of the four independent experiments. www.nature.com/scientificreports/ Figure 3. Zn-depletion promotes caspase-like activity and other proteases for LD death. (A) TUNEL assay was performed in LD after 48 h treatment with 10 µM TPEN, 10 µM TPEN + Z-VAD-FMK (10 µM, added 30 min prior to TPEN) and Antipain (50 nM, added 30 min prior to TPEN). Hoechst 33342 was used for staining nucleus (magnification × 60). Results represent one of the five independent experiments. (B) LD promastigotes were treated with TPEN (0–10 µM) for 48 h. Caspase 3/7 activity assay was performed in cell lysates as per company’s protocol. Results represent data from three independent experiments; error bars +/− SD. (C) TPEN induced DNA fragmentation was detected by TUNEL assay after 48 h. Parasites were treated with TPEN (0 or 10 µM). E-64d (10 µM), CA074 (10 µM), ATA (50 µM) and SCP (10 µM) were added 30 min prior to TPEN treatment. Hoechst 33342 was used for staining nucleus (magnification × 60). Results represent one of the four independent experiments. Figure 3. Zn-depletion promotes caspase-like activity and other proteases for LD death. (A) TUNEL assay was performed in LD after 48 h treatment with 10 µM TPEN, 10 µM TPEN + Z-VAD-FMK (10 µM, added 30 min prior to TPEN) and Antipain (50 nM, added 30 min prior to TPEN). Hoechst 33342 was used for staining nucleus (magnification × 60). Results represent one of the five independent experiments. (B) LD promastigotes were treated with TPEN (0–10 µM) for 48 h. Results Zi h (A) Flow cytometric analysis for phosphatidylserine externalisation was detected by annexin V labelling in TPEN (0–10 µM) treated LD after 16 h. The left lower quadrant shows unstained parasite, right lower quadrant denotes annexin V positive cells, upper left quadrant shows only PI positive cells while upper right quadrant indicates both annexin V and PI positive cells. Dot plots are representative of one of three similar results. (B) Fluorometric analysis of mitochondrial membrane potential after TPEN treatment (0 10µM 24h) Results represent as+/ SD from Figure 2. TPEN treatment promotes apoptosis-like death in LD. (A) Flow cytometric analysis for phosphatidylserine externalisation was detected by annexin V labelling in TPEN (0–10 µM) treated LD after 16 h. The left lower quadrant shows unstained parasite, right lower quadrant denotes annexin V positive cells, upper left quadrant shows only PI positive cells while upper right quadrant indicates both annexin V and PI positive cells. Dot plots are representative of one of three similar results. (B) Fluorometric analysis of mitochondrial membrane potential after TPEN treatment (0–10 µM, 24 h). Results represent as +/− SD from three independent experiments. (C) TUNEL staining was done in LD after TPEN treatment (0–10 µM, 72 h) for assessing DNA fragmentation. Microscopic images show red colour for TUNEL-positive LD (TdT staining) and blue colour for staining nucleus (Hoechst 33342) (magnification × 40). DNA to block fragmentation24. TUNEL assay revealed that ATA could completely inhibit TPEN-induced DNA fragmentation in LD (Fig. 3C). To further determine the impact of cathepsin B and Endo G we verified growth rate of TPEN-treated LD in presence of their specific inhibitors (CA074 and ATA). Results showed affected LD growth by TPEN treatment was significantly reversed in presence of inhibitors of cathepsin B and Endo G (Supplemental Fig. 1). These experiments strongly suggest the role of cysteine protease cathepsin B and mito- chondrial endonuclease (Endo G) in death process of LD due to Zn-depletion. Zn-depletion induces ROS generation to promote LD death. ROS generation is suggested to be involved in promoting apoptosis in Leishmania parasites34, 35. Due to well documented role of zinc on cellular antioxidant capacity we considered that its depletion might increase ROS generation to promote apoptosis like Scientific REPorTS \| 7: 10488 \| DOI:10.1038/s41598-017-10041-6 4 entificreports/ Figure 3. Zn-depletion promotes caspase-like activity and other proteases for LD death. Results Zi h Caspase 3/7 activity assay was performed in cell lysates as per company’s protocol. Results represent data from three independent experiments; error bars +/− SD. (C) TPEN induced DNA fragmentation was detected by TUNEL assay after 48 h. Parasites were treated with TPEN (0 or 10 µM). E-64d (10 µM), CA074 (10 µM), ATA (50 µM) and SCP (10 µM) were added 30 min prior to TPEN treatment. Hoechst 33342 was used for staining nucleus (magnification × 60). Results represent one of the four independent experiments. Scientific REPorTS \| 7: 10488 \| DOI:10.1038/s41598-017-10041-6 5 www.nature.com/scientificreports/ Figure 4. Zn-depletion induces ROS generation in LD. (A) DCF- fluorescence was measured in TPEN (0 and10 µM, 36 h) treated LD. NAC (20 mM) was added 30 min prior to addition of TPEN. Data represent mean ± S.D. from three independent experiments. (B) Similarly, ROS generation was detected by fluorescence microscopy (magnification × 100) at 24 h in presence of TPEN (0 and 10 µM) and NAC (20 mM). (C) TUNEL assay was performed in a similar condition described in (A) (magnification × 60). Data represent one of the three independent experiments. Figure 4. Zn-depletion induces ROS generation in LD. (A) DCF- fluorescence was measured in TPEN (0 and10 µM, 36 h) treated LD. NAC (20 mM) was added 30 min prior to addition of TPEN. Data represent mean ± S.D. from three independent experiments. (B) Similarly, ROS generation was detected by fluorescence microscopy (magnification × 100) at 24 h in presence of TPEN (0 and 10 µM) and NAC (20 mM). (C) TUNEL assay was performed in a similar condition described in (A) (magnification × 60). Data represent one of the three independent experiments. events in LD. Therefore, we initially examined intracellular ROS level in TPEN treated LD. H2DCFDA was used to detect ROS level in TPEN treated LD promastigotes. A significant increase (~12 fold) in DCF-fluorescence was detected that was blocked by prior treatment of antioxidant NAC (N-acetyl cysteine) (Fig. 4A) suggesting increased ROS generation due to depletion of chelatable zinc pool in LD. We further verified ROS generation by fluorescence microscopy and found a similar result (Fig. 4B). To determine the role of ROS in Zn-depletion induced LD death process we performed TUNEL assay in presence of antioxidant NAC. Results revealed that NAC could protect TPEN-induced DNA fragmentation in LD (Fig. 4C). Zn-depletion does not induce mitochondrial ROS generation. Results Zi h Mitochondrial ROS generation has been suggested as the primary source for apoptotic cell death in general and in Leishmania parasite36, 37. Since increase in DCF-fluorescence usually represents cytosolic ROS generation, we further estimated mitochondrial ROS using MitoSox as a probe. Results showed no significant change in mitochondrial ROS generation (Fig. 5A). H2O2 (4 mM, 2 h) was used as a positive control. A similar data was also obtained by using fluorescence micros- copy (Fig. 5B). These results indicate that mitochondrial ROS generation has little role for promoting death in zinc depleted LD. Scientific REPorTS \| 7: 10488 \| DOI:10.1038/s41598-017-10041-6 6 www.nature.com/scientificreports/ www.nature.com/scientificreports/ Figure 5. TPEN does not influence mitochondrial ROS generation. (A) LD parasites were treated with TP (0–10 µM) for 48 h. MitoSox (5 µM) was added 15 min prior to the end of incubation period. Fluorescence w measured at 590 nm to estimate ROS generation. Data represent +/− SD from three independent experime (B) Similarly, fluorescence microscopy (magnification × 100) was used to detect mitochondrial ROS genera i Mit S H O (4 M) t t t f 2h d iti t l D t t f th th Role of zinc chelation on viability of antimony resistant LD. Our earlier experiments revealed that Zn-chelation could affect LD viability in wild type LD. So we further examined the effect of Zn-chelation b l f l k [ d l ( ) ] d Figure 5. TPEN does not influence mitochondrial ROS generation. (A) LD parasites were treated with TPEN (0–10 µM) for 48 h. MitoSox (5 µM) was added 15 min prior to the end of incubation period. Fluorescence was measured at 590 nm to estimate ROS generation. Data represent +/− SD from three independent experiments. (B) Similarly, fluorescence microscopy (magnification × 100) was used to detect mitochondrial ROS generation using MitoSox. H2O2 (4 mM) treatment for 2 h was used as a positive control. Data represent one of the three independent experiments. Figure 5. TPEN does not influence mitochondrial ROS generation. (A) LD parasites were treated with TPEN (0–10 µM) for 48 h. MitoSox (5 µM) was added 15 min prior to the end of incubation period. Fluorescence was measured at 590 nm to estimate ROS generation. Data represent +/− SD from three independent experiments. (B) Similarly, fluorescence microscopy (magnification × 100) was used to detect mitochondrial ROS generation using MitoSox. Results Zi h H2O2 (4 mM) treatment for 2 h was used as a positive control. Data represent one of the three independent experiments. Figure 5. TPEN does not influence mitochondrial ROS generation. (A) LD parasites were treated with TPEN (0–10 µM) for 48 h. MitoSox (5 µM) was added 15 min prior to the end of incubation period. Fluorescence was measured at 590 nm to estimate ROS generation. Data represent +/− SD from three independent experiments. (B) Similarly, fluorescence microscopy (magnification × 100) was used to detect mitochondrial ROS generation using MitoSox. H2O2 (4 mM) treatment for 2 h was used as a positive control. Data represent one of the three independent experiments. Figure 5. TPEN does not influence mitochondrial ROS generation. (A) LD p (0–10 µM) for 48 h. MitoSox (5 µM) was added 15 min prior to the end of inc measured at 590 nm to estimate ROS generation. Data represent +/− SD from (B) Similarly, fluorescence microscopy (magnification × 100) was used to det using MitoSox. H2O2 (4 mM) treatment for 2 h was used as a positive control independent experiments. Figure 5. TPEN does not influence mitochondrial ROS generation. (A) LD parasites were treated with TPEN (0–10 µM) for 48 h. MitoSox (5 µM) was added 15 min prior to the end of incubation period. Fluorescence was measured at 590 nm to estimate ROS generation. Data represent +/− SD from three independent experiments. (B) Similarly, fluorescence microscopy (magnification × 100) was used to detect mitochondrial ROS generation using MitoSox. H2O2 (4 mM) treatment for 2 h was used as a positive control. Data represent one of the three independent experiments. Role of zinc chelation on viability of antimony resistant LD. Our earlier experiments revealed that Zn-chelation could affect LD viability in wild type LD. So we further examined the effect of Zn-chelation on viability of antimony resistant parasites like GE1 [sodium antimony gluconate (SAG) resistant] and K39 Role of zinc chelation on viability of antimony resistant LD. Our earlier experiments revealed that Zn-chelation could affect LD viability in wild type LD. So we further examined the effect of Zn-chelation on viability of antimony resistant parasites like GE1 [sodium antimony gluconate (SAG) resistant] and K39 Role of zinc chelation on viability of antimony resistant LD. Our earlier experiments revealed that Zn-chelation could affect LD viability in wild type LD. Results Zi h So we further examined the effect of Zn-chelation on viability of antimony resistant parasites like GE1 [sodium antimony gluconate (SAG) resistant] and K39 Scientific REPorTS \| 7: 10488 \| DOI:10.1038/s41598-017-10041-6 7 ientificreports/ (Potassium-Antimony-Tartarate resistant). MTT assay revealed the ability and K39 (Fig 6A) substantially IC for GE1 and K39 strain for 48h TPEN Figure 6. Effect of TPEN on drug resistance LD. (A) Cell viability of drug- after 48 h incubation with TPEN (0, 5 and 10 µM) by MTT assay. Data repre independent experiments performed in triplicate. TUNEL assay was perfor GE1 (B) and K39 (C) after TPEN treatment for 48 h (0–10 µM) (magnificat three independent experiments. www.nature.com/scientificreports/ Figure 6. Effect of TPEN on drug resistance LD. (A) Cell viability of drug-resistant GE1 and after 48 h incubation with TPEN (0, 5 and 10 µM) by MTT assay. Data represent mean ± S.D. independent experiments performed in triplicate. TUNEL assay was performed in drug-resis GE1 (B) and K39 (C) after TPEN treatment for 48 h (0–10 µM) (magnification × 60). Data re th i d p d t p i t Figure 6. Effect of TPEN on drug resistance LD. (A) Cell viability of drug-resistant GE1 and K39 was detected after 48 h incubation with TPEN (0, 5 and 10 µM) by MTT assay. Data represent mean ± S.D. from three independent experiments performed in triplicate. TUNEL assay was performed in drug-resistant LD strains GE1 (B) and K39 (C) after TPEN treatment for 48 h (0–10 µM) (magnification × 60). Data represent one of the three independent experiments. Figure 6. Effect of TPEN on drug resistance LD. (A) Cell viability of drug-resistant GE1 and K39 was detected after 48 h incubation with TPEN (0, 5 and 10 µM) by MTT assay. Data represent mean ± S.D. from three independent experiments performed in triplicate. TUNEL assay was performed in drug-resistant LD strains GE1 (B) and K39 (C) after TPEN treatment for 48 h (0–10 µM) (magnification × 60). Data represent one of the three independent experiments. (Potassium-Antimony-Tartarate resistant). MTT assay revealed the ability of TPEN on affecting viability of GE1 and K39 (Fig. 6A) substantially. IC50 for GE1 and K39 strain for 48 h TPEN treatment was detected as 7.18 µM and 3.82 µM respectively. Discussion By using specific inhibitor we detected involvement of caspase or caspase-like activity in TPEN-induced LD cell death (Fig. 3A). We also detected a substantial increase in caspase3/7 activity by zinc depletion (Fig. 3B). Interestingly, the pres- ence of caspase in Leishmania parasites has not been confirmed; however, evidence of caspase-like activity in this parasite is reported45. Thus, our observation is probably attributed to the presence of caspase-like activity in LD, the precise molecular identification of which is still not substantiated.f i Apoptotic cell death is the result of increased activity of several other proteases in different organelles acting simultaneously and/or tandem46. Involvement of Endo G in leishmanial cell death is reported earlier24. There is also evidence of release of cathepsin B-like proteases from the lysosome that contribute to death of Leishmania spp47. We found evidence of involvement of lysosomal protease cathepsin B and mitochondrial protease Endo G during zinc depletion induced death of LD. However, the precise mechanism by which zinc depletion induces these different proteases in different organelles is not known so far and needs further study. ff ROS generation is a well established cause of apoptosis in general48 and also suggested in the Leishmania parasites26, 49. Given the suggested role of zinc as an antioxidant we assumed that zinc depletion might increase ROS generation in promoting death of LD. A strong increase in DCF-sensitive fluorescence was obtained due to TPEN treatment that was reversed by antioxidant NAC confirming that Zn-depletion could induce ROS gen- eration in drug-sensitive and antimony-resistant LD (Fig. 4A,B; Fig. 7A). Further, we detected reversal of LD death by NAC (Fig. 4C; Fig. 7B,C) suggesting that depletion of zinc might affect antioxidant capacity to initiate death of LD. We limited our study with NAC up to 36 h because we detected toxicity after this period with only NAC treatment; the reason of which is not clear so far. However, there are previous reports on NAC toxicity in different cell types50 supporting our observation. Mitochondrial ROS generation is suggested as key event in promoting apoptosis-like death in LD by a number of agents including H2O2 26, 36. We employed MitoSox, a sensor of mitochondrial ROS generation, and detected no significant alteration of MitoSox sensitivity suggesting ROS generation was from other cellular source than mitochondria due to zinc depletion (Fig. 5). Discussion All microbial pathogens require transition metals for their growth and survival as these metals participate in many structural, catalytic and signalling functions. Zinc is one of these essential elements required for growth and virulence of pathogens6. The role of zinc in regard to growth and survival of Leishmania parasites has not been addressed despite being a significant component of the virulence armoury of these parasites particularly in its promastigote stage15. The current work is thus the first to report the crucial role of zinc in survival of LD. We demonstrate a novel finding that depletion of zinc could promote death of LD. Interestingly; we also found that two antimony-resistant LD strains were also susceptible to depletion of zinc like drug-sensitive parasites. TPEN is widely used as cell permeable metal ion chelator mainly to limit intra- and extracellular concentra- tions of several transition metals; however, it is more effective in chelating zinc than ferrous iron38, 39. This study provides several evidences to show that TPEN actually chelated zinc to promote apoptosis-like death in LD. First, we found TPEN treatment affected LD viability and growth even in the presence of 10% serum that contained high amount of transferrin-bound iron, haemoglobin and other non-transferrin bound iron. Second, we detected specific ferrous iron chelator BPS could affect LD viability (Fig. 1B) but effectively in milliMolar concentration while TPEN was effective even at low (10 µM or less) concentration (Fig. 1A). Third, we found simultaneous addition of zinc salt along with TPEN could reverse the LD survival (Fig. 1D). Finally, we used an intracellular zinc sensor (Fluozin-3AM) to detect depletion of this transition metal in TPEN-treated LD (Fig. 1C). These exper- iments provided strong evidence that TPEN-induced death in LD was actually due to depletion of chelatable pool of zinc. Apoptosis is one of the major pathways of cell death in organisms and also proposed to be part of the leish- manial death mechanism37, 40. However, existence of classical programmed cell death mechanism in protozoa including Leishmania is controversial as presence of many of the dedicated molecular pathways for apoptosis have not been identified yet41. We provided several evidences like phosphatidylserine externalisation, mitochon- drial membrane depolarization, DNA fragmentation and involvement of Endo G in zinc depletion-induced death of LD (Figs 2 and 3) those recapitulated apoptosis-like cell death. Involvement of caspase-dependent and caspase-independent mechanisms for apoptosis has been well proposed in Leishmania spp42–44. Results Zi h The growth of these antimony resistant parasites was also decreased by about 90% after 3 days of TPEN (10 µM) treatment compared to untreated parasites. To verify whether Zn-depletion was Scientific REPorTS \| 7: 10488 \| DOI:10.1038/s41598-017-10041-6 8 www.nature.com/scientificreports/ also effective in promoting DNA fragmentation in these antimony resistant strains we performed TUNEL assay. Results showed TPEN treatment could promote DNA fragmentation in these drug resistant parasites (Fig. 6B and Fig. 6C). These results strongly suggest the effectiveness of Zn-chelation in promoting apoptosis-like death in antimony resistant strains of LD. TPEN treatment induces ROS generation to promote cell death in antimony resistant LD. Further we examined the effect of Zn-chelation on ROS generation in antimony-resistant GE1 and K39 LD strains. We detected substantial increase in DCF fluorescence in both the antimony-resistant strains of LD after TPEN treatment (Fig. 7A). We further observed that pre-treatment of antioxidant NAC blocked TPEN-induced DNA fragmentation by TUNEL assay (Fig. 7B and C). These results suggest Zn-depletion induced ROS genera- tion play critical role in promoting death in antimony-resistant LD strains. Discussion Usually, increase in ROS-sensitive DCF-fluorescence is a measure of cytosolic ROS because the entry of DCF is mainly limited to cytosol51, 52. It needs further study to understand the precise mechanism by which Zn-depletion promotes ROS generation in LD.ii One of the significant findings in this study is the ability of TPEN to induce ROS generation and apoptosis-like death in antimony-resistant LD. Most of the treatments of leishmaniasis so far have been centred on pentavalent antimonials53. Pentavalent antimonials like sodium antimony gluconate (SAG) are the standard first-line choice of drug against the disease54. We used two different strains of drug resistant LD; GE1 and K39 in the current study. GE1 strain is laboratory generated SAG-resistant strain54 while K39 is a clinical isolate of antimony resistant Scientific REPorTS \| 7: 10488 \| DOI:10.1038/s41598-017-10041-6 9 www.nature.com/scientificreports/ tificreports/ parasite53, 54. Earlier evidences suggested that most of the antileishmanial d totic death of the parasite, while drug resistant parasites became refractory to Figure 7. TPEN treatment induces ROS generation to promote apoptosis-l LD. (A) ROS generation was estimated using H2DCFDA in GE1 and K39 LD treatment (0, 5 and 10 µM). Data represent mean ± S.D. from three indepen assay was performed after TPEN treatment for 24 h (0 and 10 µM) in presen in GE1 (B) and K39 (C) strains. Results are representative of one of the thre (magnification × 60). Figure 7. TPEN treatment induces ROS generation to promote apoptosis-like death in antimony-resistant LD. (A) ROS generation was estimated using H2DCFDA in GE1 and K39 LD parasites after 48 h of TPEN treatment (0, 5 and 10 µM). Data represent mean ± S.D. from three independent experiments. TUNEL assay was performed after TPEN treatment for 24 h (0 and 10 µM) in presence and absence of NAC (20 mM) in GE1 (B) and K39 (C) strains. Results are representative of one of the three independent experiments (magnification × 60). Figure 7. TPEN treatment induces ROS generation to promote apoptosis-like death in antimony-resistant LD. (A) ROS generation was estimated using H2DCFDA in GE1 and K39 LD parasites after 48 h of TPEN treatment (0, 5 and 10 µM). Data represent mean ± S.D. from three independent experiments. TUNEL assay was performed after TPEN treatment for 24 h (0 and 10 µM) in presence and absence of NAC (20 mM) in GE1 (B) and K39 (C) strains. Materials and Methods Briefly, parasites were washed with 1x PBS after treatment with TPEN and incubated with 50 µM probe for 30 min in dark at 22 °C. In the presence of endogenous superoxide, non-fluorescent membrane permeable H2DCFDA was converted into impermeable fluorogenic 2′,7′-dichlorofluorescein, which was detected fluorimetrically as described earlier57. Results indicate primarily the cytosolic ROS level induced by TPEN treatment. Similarly, ROS production was also verified by fluorescence microscopy using the probe H2DCFDA58. Mitochondrial ROS generation. For detection of mitochondrial ROS generation MitoSox TM Red mito- chondrial superoxide indicator (Molecular Probe) was used as reported earlier45. It is live-cell-permeable, rapidly and selectively targeted to mitochondria. Once in the mitochondria, MitoSoxTM Red reagent is oxidized by super- oxide and show red fluorescence. LD (1 × 106 cells/mL) were treated with TPEN for 48 h at 22 °C. After treatment, cells were pelleted and washed with 1x PBS and further incubated with MitoSox Red reagent (5 µM) for 10 min. After incubation, cells were again washed with 1x PBS and transferred into 96 well plates. The fluorescence was measured in a fluorometer at ex/em of 543/590 nm. For microscopic detection, live leishmania cells were directly taken on the slide under cover-slip after washing with 1x PBS and examined under fluorescence microscope (Carl Zeiss-Axio-Vision-4.8). Caspase 3/7 activity assay. Caspase-like activity in LD was assayed with the help of a kit Apo-One® Homogenous Caspase-3/7Assay (Promega). The assay was performed according to manufacturer’s guidelines. Briefly, parasites after treatment were pelleted down and washed with 1x PBS. The pellet was resuspended in 1x PBS (50 µL) and mixed with caspase buffer in 1:1 ratio. The sample was then incubated for at least 12 h in dark at room temperature to get optimal fluorescence at an excitation wavelength range of 485 ± 20 nm and an emission wavelength range of 530 ± 25 nm. Estimation of mitochondrial membrane potential depolarisation. Mitochondrial membrane potential depolarisation was determined using mitochondria-specific probe JC-1 that could specifically accu- mulate within mitochondria according to membrane potential and provide fluorescence as per membrane potential59. To determine the mitochondrial membrane potential, parasites were treated with TPEN (0–10 µM), followed by washing with 1x PBS re-suspended in M199. Parasites were then incubated with JC-1 dye (10 µg/mL) for 20 min in dark. Then fluorescence was subsequently monitored in a fluorimeter (FLUOROSKAN ASCENT FL, Thermoscientific) at dual wavelengths as described earlier59. Materials and Methods Reagents were obtained from Sigma Chemical Company unless stated otherwise. Supplies related to tissue culture experiments were obtained from Corning, NY, USA. The MitoSox TM Red mitochondrial superoxide indicator, JC-1 mitochondrial potential sensor and Click-iT® Tunel Alexa Fluor® Imaging Assay from Molecular probes while Apo-One® Homogenous Caspase-3/7Assay kit was procured from Promega. Parasite culture. Leishmania donovani (MHOM\IN\1983\AG83) promastigotes were maintained in M199 medium supplemented with 10% FBS, 100 units/ml penicillin, 100 µg/ml streptomycin at 22 °C in BOD incubator as described earlier17, 55. Subculturing was done on every fourth or fifth day when the promastigotes reached the stationary phase of growth. Drug resistant parasites like K-39 (a clinical isolate resistant to PAT (SbIII)) and GE1 (a laboratory generated strain resistant to sodium antimony gluconate (SAG), the pentavalent antimony) were cultured as described earlier54. Study on LD growth. For growth study of LD, 1 × 106 cells/ml were seeded into the 25 mL flask with fresh M199 with 10% FBS and 1% PS. Parasites were treated with different concentrations of TPEN and counted at every 24 h using neubauer chamber under a light microscope at 40X magnification. LD was counted from all four 16-big squares and an average was considered for further calculations. The formula used was; No. of cells = Average counting by neubauer chamber × dilution factor × 104 cells/mL. Cytotoxicity assay. Viability of parasites was estimated using MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diph enyltetrazolium bromide) as described earlier56. Briefly, exponentially growing promastigotes (1 × 105 cells/mL) in M199 media (without phenol red) with 10% FBS were treated with TPEN for different time points. After treat- ments cells were washed with ice-cold 1x PBS and incubated in fresh M199 media with 10% FBS and 400 µg/ml MTT. After 3 h, 100 µL DMSO was added to solubilise the formazan crystals. The absorbance was measured on a microplate reader (Sunrise, Tecan) at 492 nm. The percentage of viability was calculated from O.D. The blank O.D. was subtracted from all the samples. The viability of the cells was calculated using the following formula - viabil- ity of cells = (Absorbance of treated cells − Absorbance of Blank)/(Absorbance of control cells − Absorbance of Blank) × 100. Reactive oxygen species (ROS) generation. To investigate the level of endogenous ROS production, we used the peroxide-sensitive fluorescent probe H2DCFDA (Sigma-Aldrich). Discussion Results are representative of one of the three independent experiments (magnification × 60). parasite53, 54. Earlier evidences suggested that most of the antileishmanial drugs functioned by promoting apop- totic death of the parasite, while drug resistant parasites became refractory to apoptosis35. In contrast, our findings suggest a very important fact that Zn-depletion could promote apoptosis-like death in antimony-resistant LD parasite53, 54. Earlier evidences suggested that most of the antileishmanial drugs functioned by promoting apop- totic death of the parasite, while drug resistant parasites became refractory to apoptosis35. In contrast, our findings suggest a very important fact that Zn-depletion could promote apoptosis-like death in antimony-resistant LD parasite53, 54. Earlier evidences suggested that most of the antileishmanial drugs functioned by promoting apop- totic death of the parasite, while drug resistant parasites became refractory to apoptosis35. In contrast, our findings suggest a very important fact that Zn-depletion could promote apoptosis-like death in antimony-resistant LD Scientific REPorTS \| 7: 10488 \| DOI:10.1038/s41598-017-10041-6 10 www.nature.com/scientificreports/ strains generated either in laboratory (GE1) or isolated clinically (K39). These results thus may open up possibil- ity of zinc depletion as a strategy to develop antileishmanials towards drug-sensitive and antimony-resistant LD. M t i l d M th d strains generated either in laboratory (GE1) or isolated clinically (K39). These results thus may open up possibil- ity of zinc depletion as a strategy to develop antileishmanials towards drug-sensitive and antimony-resistant LD. Materials and Methods All data are expressed as mean ± standard deviation (S.D.) and are represented as data of at least three different sets of experiments. Mean and standard deviation values were calculated with the help of Microsoft Excel. Materials and Methods The result is expressed as the ratio of reading at 590 nm to the reading at 530 nm. TUNEL assay. We used Click-iT® Tunel Alexa Fluor® Imaging Assay kit (cat no. C10246) to detect DNA fragmentation. Experiments were performed as per company’s protocol. Briefly, parasites after treatment with TPEN were fixed with 3.7% formaldehyde for 15 min and then permeabilized with 0.25% Triton X-100 for 20 min. Equilibrium was created by adding 100 µL of TdT reaction buffer (Component A) for 10 min at room temperature. Scientific REPorTS \| 7: 10488 \| DOI:10.1038/s41598-017-10041-6 11 www.nature.com/scientificreports/ After removing TdT reaction buffer, 100 µL of TdT reaction cocktail was added for 1 h at 37 °C. Coverslips were washed for 3% BSA for 2 min. After that Click-iT reaction cocktail was added for 15 min. All samples were coun- terstained for nuclei with Hoechst 33342 and analyzed under a fluorescence microscope. Annexin-V binding Assay. Phosphatidylserine externalization was assessed using a kit from Molecular Probes (Invitrogen) as per company’s protocol. Briefly, parasites were harvested after treatment and washed with 1x PBS and resuspended in the 1x annexin V binding buffer according to cell density (1 × 106 cells/mL). Alexa Fluor® 488 annexin v (component A) (5 µL) and 1 µL Propidium Iodide (100 µg/mL) were added to the sample. Samples were further incubated for 15 min at room temperature. After the incubation, 400 µL of 1x annexin V binding buffer was added to the cells and kept on ice until analyzed. Samples were analyzed by flow cytometry (within 1 h). Detection of intracellular zinc. Intracellular zinc level was detected with the help of sensor Fluozin™-3 (Molecular Probes) according to the protocol of the company. Fluozin™-3 is suitable for detection of Zn2+ con- centrations. The cell-permeable AM-ester is useful for detecting low intracellular Zn2+ levels and small concen- tration changes60. Briefly, 1 × 106 cells/mL were treated with zinc chelator TPEN for 48 h. Then cells were washed with1x PBS and resuspended in serum-free M199 medium (without phenol red). FluoZin™-3 (5 µM) was added to cells for 30 min at 22 °C in dark. Then cells were washed with only M199 (without phenol red) and further incubated for 45 min in dark at 22 °C to allow complete de-esterification of intracellular AM ester. After that cells were taken on glass slides and analysed under fluorescence microscope. Statistical analysis. References 1. Alam, S. & Kelleher, S. L. Cellular mechanisms of zinc dysregulation: A perspective on zinc homeostasis as an etiological factor in the development and progression of breast cancer. Nutrients 4, 875–903 (2012).f g 2. Beyersmann, D. & Haase, H. Functions of zinc in signaling, proliferation and differentiation of mammalian cells. BioMetals 14, 331–341 (2001). 3. Makhov, P., Golovine, K., Uzoo, R. G. et al. Zinc chelation induces rapid depletion of the X-linked inhibitor of apoptosis (XIAP) and sensitizes prostate cancer cells to TRAIL-mediated apoptosis. Cell Death and Differentiation 15, 1745–1751 (2008).f p p . Zinc in Human Health: Effect of Zinc on Immune Cells. Mol. Med f 4. Prasad, A. S. Zinc in Human Health: Effect of Zinc on Immune Cells. Mol. Med. 14, 353–357 (2008). f 5. Broadley, M. R., White, P. J., Hammond, J. P., Zelko, I. & Lux, A. Zinc in plants. New Phytologist. 173, 677–702 (20 f 5. Broadley, M. R., White, P. J., Hammond, J. P., Zelko, I. & Lux, A. Zinc in plants. New Phytologist. 173, 677–702 (2007). 6 S B Zi d i f ti R i f I f ti Di 5 137 47 (1983) 5. Broadley, M. R., White, P. J., Hammond, J. P., Zelko, I. & Lux, A. Zinc in plants. New Phytologist. 173, 677–702 (2007). 6. Sugarman, B. Zinc and infection. Review of Infectious Diseases. 5, 137–47 (1983).hi 6. Sugarman, B. Zinc and infection. Review of Infectious Diseases. 5, 137–47 (1983).hi n, B. Zinc and infection. Review of Infectious Diseases. 5, 137–47 (19 g f f 7. Klug, A. The discovery of zinc fingers and their applications in gene regulation and genome manipulation. Annu Rev Bi 213–231 (2010). 8. Robinson, N. J., Whitehall, S. K. & Cavet, J. S. Microbial metallothioneins. Adv Microb Physiol. 44, 183–213 (2001). 8. Robinson, N. J., Whitehall, S. K. & Cavet, J. S. Microbial metallo 9. Wessells, K. R., Singh, G. M. & Brown, K. H. Estimating the Global Prevalence of Inadequate Zinc Intake from Natio Balance Sheets: Effects of Methodological Assumptions. PLoS ONE 7, e50565 (2012). f 10. Oteiza, P. I. Zinc and the modulation of redox homeostasis. Free Radic. Biol. Med. 53, 1748–1759 (2012). 11. Ready, P. D. Epidemiology of visceral leishmaniasis. Clin Epidemiol. 6, 147–154 (2014). 2. Chappuis, F., Sundar, S., Hailu, A., Ghalib, H. & Rijal, S. Visceral leishmaniasis: what are the needs for diagnosis, treatment and control? Scientific REPorTS \| 7: 10488 \| DOI:10.1038/s41598-017-10041-6 References Nat Rev Microbiol 5, 873–882 (2007). 13. Das, B. B. et al. Topoisomerase research of kinetoplastid parasite Leishmania, with special reference to developme therapeutics. Indian J Med Res, 123, 221–232 (2006). p ( ) 14. Maltezou, H. C. Drug resistance in visceral leishmaniasis. J. Biomed. Biotechnol. 2010, 617521 (2010). 14. Maltezou, H. C. Drug resistance in visceral leishmani 15. Yang, D., Rogers, M. V., Brett, S. J. & Liew, F. Y. Immunological analysis of the zinc-binding peptides of surface metalloprote (gp63) of Leishmania major. Immunology. 78, 582–585 (1993). 16. Carvalho, S. et al. LiZIP3 is a cellular zinc transporter that mediates the tightly regulated import of zinc in Leishmania infantum parasites. Mol. Microbiol. 96, 581–595 (2015). p 7. Das, N. K., Biswas, S., Solanki, S. & Mukhopadhyay, C. K. Leishmania donovani depletes labile iron pool to exploit iron uptake capacity of macrophage for its intracellular growth. Cell Microbiol. 11, 83–94 (2009). p 17. Das, N. K., Biswas, S., Solanki, S. & Mukhopadhyay, C. K. Leishmania donovani depletes lab capacity of macrophage for its intracellular growth. Cell Microbiol. 11, 83–94 (2009). capacity of macrophage for its intracellular growth. Cell Microbiol. 11, 83–94 (2009). 18. Flannery, A. R., Renberg, R. L. & Andrews, N. W. Pathways of iron acquisition and utilization in. Leishmania. Curr. Opin. Microbiol. 16 716 721 (2013) 8. Flannery, A. R., Renberg, R. L. & Andrews, N. W. Pathways of iron acquisition and utilization in. Leishmania. Curr. Opin. Microbiol 16, 716–721 (2013).f , ( ) 19. Zangger, H., Mottram, J. C. & Fasel, N. Cell death in Leishmania induced by stress and differentiation: programmed cell death or necrosis? Cell Death Differ. 9, 1126–1139 (2002). ff 0. Galluzzi, L., López-Soto, A., Kumar, S. & Kroemer, G. Caspases Connect Cell-Death Signaling to Organismal Homeostasis Immunity 44, 221–231 (2016). y 21. Düssmann, H., Rehm, M., Kögel, D. & Prehn, J. H. M. Outer mitochondrial membrane permeabilization during apoptosis triggers caspase-independent mitochondrial and caspase-dependent plasma membrane potential depolarization: a single-cell analysis. J. Cell Sci. 116, 525–536 (2003). 22. Pradelli, L. A., Bénéteau, M. & Ricci, J.-E. Mitochondrial control of caspase-dependent and -independent cell death. Cell. Mol. Life Sci. 67, 1589–1597 (2010). 23. Arnoult, D. et al. On the evolution of programmed cell death: apoptosis of the unicellular eukaryote Leishmania major involves cysteine proteinase activation and mitochondrion permeabilization. Cell Death Differ. 9, 65–81 (2002).h y p p ff 24. BoseDasgupta, S. et al. Acknowledgementsh g This work was supported by the ICMR-CAR grant to Special Centre for Molecular Medicine, Jawaharlal Nehru University (JNU); by University with potential for excellence (UPE-II) to JNU and DST-PURSE programme to JNU. Authors also appreciate the infrastructural and all other supports of JNU. SS and KB acknowledge the support in terms of fellowship from UGC and CSIR respectively. References & Rego, A. C. Cytosolic and mitochondrial ROS in staurosporine-induced retinal cell apop Free Radic. Biol. Med. 35, 1500–1514 (2003). 53. Chakravarty, J. & Sundar, S. Drug Resistance in Leishmaniasis. J Glob Infect Dis. 2, 167–176 (2010). 53. Chakravarty, J. & Sundar, S. Drug Resistance in Leishmaniasis. J Glob Infect Dis. 2, 167–176 (2010). h k h h l d l h 54. Mathur, R. & Das, R. P. atrick, Ranjan, A. & Shaha, C. Elevated ergosterol protects Leishmania parasites against antimony-gene stress. FASEB J. 29, 4201–4213 (2015). 55. Singh, A. K., Mukhopadhyay, C., Biswas, S., Singh, V. K. & Mukhopadhyay, C. K. Intracellular Pathogen Leishmania donovani Activates Hypoxia Inducible Factor-1 by Dual Mechanism for Survival Advantage within Macrophage. PLoS One. 7(6), e38489 (2012).i 56. Dutta, A., Bandyopadhyay, S., Mandal, C. & Chatterjee, M. Development of a modified MTT assay for screening antimonial resistant field isolates of Indian visceral leishmaniasis. Parasitol. Int. 54, 119–122 (2005). i 57. Biswas, S., Gupta, M. K., Chattopadhyay, D. & Mukhopadhyay, C. K. Insulin induced activation of hypoxia inducible factor-1 requires generation of reactive oxygen species by NADPH oxidase. Am. J. Physiol. Heart and Circulatory Physiology 292, H758–766 (2007). ( ) 8. Tapryal, N. et al. Glutathione synthesis inhibitor butathione sulfoximine regulates ceruloplasmin by dual but opposite mechanism Implication in hepatic iron overload. Free Radic Biol Med. 48, 1492–1500 (2010). 58. Tapryal, N. et al. Glutathione synthesis inhibitor butathione sulfoximine regulates ceruloplasm Implication in hepatic iron overload. Free Radic Biol Med. 48, 1492–1500 (2010). et al. Glutathione synthesis inhibitor butathione sulfoximine regul 58. Tapryal, N. et al. Glutathione synthesis inhibitor butathione su Implication in hepatic iron overload. Free Radic Biol Med. 48, 1492–1500 (2010). 59. Mehta, A. & Shaha, C. Mechanism of metalloid-induced death in Leishmania spp.: Role of iron, reactive oxygen species, Ca2+, and glutathione. Free Radic. Biol. Med. 40, 1857–1868 (2006). p p 9. Mehta, A. & Shaha, C. Mechanism of metalloid-induced death in Leishmania spp.: Role of iron, reactive oxygen species, Ca2+, and glutathione. Free Radic. Biol. Med. 40, 1857–1868 (2006). g 0. Roh, H. C., Collier, S., Guthrie, J., Robertson, J. D. & Kornfeld, K. Lysosome-related organelles in intestinal cells are a zinc storage site in C. elegans. Cell Metab. 15, 88–99 (2012). References Cell death in parasitic protozoa: regulated or incidental? Nat Rev Microbiol. 11, 58–66 (2013). 2. Van Zandbergen, G., Lüder, C. G. K., Heussler, V. & Duszenko, M. Programmed cell death in unicellular parasites: A prerequisite fo sustained infection? Trends Parasitol. 26, 477–483 (2010). 43. Sen, N. et al. Camptothecin-induced imbalance in intracellular cation homeostasis regulates programmed cell death in unicellular hemoflagellate Leishmania donovani. J. Biol. Chem. 279, 52366–52375 (2004). hemoflagellate Leishmania donovani. J. Biol. Chem. 279, 52366– l g 44. Kulkarni, M. M., McMaster, W. R., Kamysz, W. & McGwire, B. S. Antimicrobial peptide-induced apoptotic death of Leishmania results from calcium-dependent, caspase-independent mitochondrial toxicity. J. Biol. Chem. 284, 15496–15504 (2009). y 45. Dolai, S., Pal, S., Yadav, R. K. & Adak, S. Endoplasmic reticulum stress-induced apoptosis in leishmania through Ca2 + −depe 45. Dolai, S., Pal, S., Yadav, R. K. & Adak, S. Endoplasmic reticulum stress-induced apoptosis in leishmania through C and caspase-independent mechanism. J. Biol. Chem. 286, 13638–13646 (2011).i 45. Dolai, S., Pal, S., Yadav, R. K. & Adak, S. Endoplasmic reticulum stress-induced ap and caspase-independent mechanism. J. Biol. Chem. 286, 13638–13646 (2011).i p p p and caspase-independent mechanism. J. Biol. Chem. 286, 13638–13646 (2011).i p p 46. Ferri, K. F. & Kroemer, G. Organelle-specific initiation of cell death pathways. Nat. Cell Biol. 3, E255–E263 (2001). erri, K. F. & Kroemer, G. Organelle specific initiation of cell death pathways. Nat. Cell Biol. 3, E255 E263 (2001). -Fadili, A. K. et al. Cathepsin B-like and cell death in the unicellular human pathogen Leishmania. Cell Death Dis 1, e71 (2010). gi y 47. El-Fadili, A. K. et al. Cathepsin B-like and cell death in the unicellular human pathogen Leishmania. Cell Death Dis 1, e71 (2 8. Redza-Dutordoir, M. & Averill-Bates, D. A. Activation of apoptosis signalling pathways by reactive oxygen species. Biochim. Biophys Acta - Mol. Cell Res. 1863, 2977–2992 (2016). 9. Das, R., Roy, A., Dutta, N. & Majumder, H. K. Reactive oxygen species and imbalance of calcium homeostasis contributes to curcumin induced programmed cell death in Leishmania donovani. Apoptosis 13, 867–882 (2008). 0. Zafarullaha, M., Lia, W. Q., Sylvestera, J. & Ahmadb, M. Molecular mechanisms of N-acetylcysteine actions. CMLS, Cell. Mol. Life Sci. 60, 6–20 (2003). 1. Coughlan, M. T. et al. RAGE-Induced Cytosolic ROS Promote Mitochondrial Superoxide Generation in Diabetes. J. Am. Soc Nephrol. 20, 742–752 (2009). p 52. Gil, J., Almeida, S., Oliveira, C. R. References 8, 324–326 (2001). 0. Bröker, L. E., Kruyt, F. A. E. & Giaccone, G. Cell death Independent of Caspases:A Review. Clin. Cancer Res. 11, 3155 62 (2005). 1. Leist, M. & Jäättelä, M. Triggering of apoptosis by cathepsins. Cell Death Differ. 8, 324–326 (2001). y p p 31. Leist, M. & Jäättelä, M. Triggering of apoptosis by cathepsins. Cell Death Differ. 8, 324–326 (2001). f 32. Chwieralski, C. E., Welte, T. & Bühling, F. Cathepsin-regulated apoptosis. Apoptosis 11, 143–149 (2006). 3. van Loo, G. et al. Endonuclease G: a mitochondrial protein released in apoptosis and involved in caspase-independent DNA degradation. Cell Death Differ. 8, 1136–1142 (2001). g ff 4. Mukherjee, S. B., Das, M., Sudhandiran, G. & Shaha, C. Increase in cytosolic Ca2+ levels through the activation of non-selective cation channels induced by oxidative stress causes mitochondrial depolarization leading to apoptosis-like death in Leishmania donovani promastigotes. J Biol Chem. 277, 24717–24727 (2002). g 35. Moreira, W., Leprohon, P. & Ouellette, M. Tolerance to drug-induced cell death favours the acquisition of multidrug resistance in Leishmania. Cell Death Dis. 2, e201 (2011). 6. Kathuria, M., Bhattacharjee, A., Sashidhara, K. V., Singh, S. P. & Mitra, K. Induction of mitochondrial dysfunction and oxidative stress in leishmania donovani by orally active clerodane diterpene. Antimicrob. Agents Chemother. 58, 5916–5928 (2014). K , , j , , , K , g , , K y stress in leishmania donovani by orally active clerodane diterpene. Antimicrob. Agents Chemother. 58, 5916–5928 (2014). tosis: a review of programmed cell death. Toxicol. Pathol. 35, 495–5 . Elmore, S. Apoptosis: a review of programmed cell death. Toxicol , p p p g , ( ) 38. Shumaker, D. K., Vann, L. R., Goldberg, M. W., Allen, T. D. & Wilson, K. L. TPEN, a Zn2+/Fe2+ chelator with low affinity for Ca2+, inhibits lamin assembly, destabilizes nuclear architecture and may independently protect nuclei from apoptosis in vitro. Cell Calcium. 23, 151–164 (1998). 9. Marvin, R. G. et al. Fluxes in “free” and total zinc are essential for progression of intraerythrocytic stages of Plasmodium falciparum Chem Biol. 19, 731–741 (2012). ( ) 40. Shaha, C. Apoptosis in Leishmania species & its relevance to disease pathogenesis. Indian J Med Res. 123, 233–244 (2006). b ll d h l d d l b l p p p p g 41. Proto, W. R., Coombs, G. H. & Mottram, J. C. Author Contributions S.S. designed, performed and analysed experiments; K.B. designed, performed and analysed experiments related to BPS-treated LD growth. C.S. designed, coordinated and analyzed experiments related to apoptosis and antimony-resistant parasites. C.K.M. conceived, designed and coordinated the study and wrote the paper. All authors reviewed the results and approved the final version of the manuscript. References The caspase-independent algorithm of programmed cell death in Leishmania indu of LdEndoG LdFEN 1 and LdTatD as a DNA ‘degradesome’ Cell Death Differ 15 1629 1640 (2008) y p p ff 24. BoseDasgupta, S. et al. The caspase-independent algorithm of programmed cell death in Leishmania induced by baicalein: th f 24. BoseDasgupta, S. et al. The caspase-independent algorithm of programmed cell death in Leishmania induce of LdEndoG, LdFEN-1 and LdTatD as a DNA ‘degradesome’. Cell Death Differ. 15, 1629–1640 (2008). oseDasgupta, S. et al. The caspase independent algorithm of programmed cell death in Leishmania induced by baicalein: the role LdEndoG, LdFEN-1 and LdTatD as a DNA ‘degradesome’. Cell Death Differ. 15, 1629–1640 (2008). g p ,h p p g p g of LdEndoG, LdFEN-1 and LdTatD as a DNA ‘degradesome’. Cell Death Differ. 15, 1629–1640 (2008). g ff 25. Sen, R., Saha, P., Sarkar, A., Ganguly, S. & Chatterjee, M. Iron enhances generation of free radicals by Artemisinin causing a caspase- independent, apoptotic death in Leishmania donovani promastigotes. Free Radic. Res. 44, 1289–1295 (2010). 26. Das, M., Mukherjee, S. B. & Shaha, C. Hydrogen peroxide induces apoptosis-like death in Leishmania donovani promastigotes. J. Cell Sci. 114, 2461–2469 (2001). 27. Lee, N., Gannavaram, S., Selvapandiyan, A. & Debrabant, A. Characterization of metacaspases with trypsin-like activity and putative role in programmed cell death in the protozoan parasite. Leishmania. Eukaryot. Cell 6, 1745–1757 (2007). h dh l l b l d d d l d d ll d h l p p g p p y 28. Chowdhury, S. et al. Disuccinyl betulin triggers metacaspase-dependent endonuclease G-mediated cell death in unicellular protozoan parasite Leishmania donovani. Antimicrob. Agents Chemother. 58, 2186–2201 (2014). 28. Chowdhury, S. et al. Disuccinyl betulin triggers metacaspase-dependent endonuclease G-mediated cell death in unicellular protozoan parasite Leishmania donovani. Antimicrob. Agents Chemother. 58, 2186–2201 (2014). 29. Turk, B. & Stoka, V. Protease signalling in cell death: caspases versus cysteine cathepsins. FEBS Lett. 581, 2761–2767 (2007). protozoan parasite Leishmania donovani. Antimicrob. Agents Chemother. 58, 2186 2201 (2014). 29. Turk, B. & Stoka, V. Protease signalling in cell death: caspases versus cysteine cathepsins. FEBS Lett. 581, 2761–2767 (2007). Scientific REPorTS \| 7: 10488 \| DOI:10.1038/s41598-017-10041-6 12 www.nature.com/scientificreports/ röker, L. E., Kruyt, F. A. E. & Giaccone, G. Cell death Independent of Caspases:A Review. Clin. Cancer Res. 11, 3155–62 (2005). eist, M. & Jäättelä, M. Triggering of apoptosis by cathepsins. Cell Death Differ. Scientific REPorTS \| 7: 10488 \| DOI:10.1038/s41598-017-10041-6 Additional Information Supplementary information accompanies this paper at doi:10.1038/s41598-017-10041-6 Competing Interests: The authors declare that they have no competing interests. Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps an institutional affiliations. Scientific REPorTS \| 7: 10488 \| DOI:10.1038/s41598-017-10041-6 13 nature.com/scientificreports/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. © Th A th ( ) 2017 www.nature.com/scientificreports/ www.nature.com/scientificreports/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2017 © The Author(s) 2017 Scientific REPorTS \| 7: 10488 \| DOI:10.1038/s41598-017-10041-6 14
https://openalex.org/W2807163776	https://sat.bntu.by/jour/article/download/1680/1579	Russian	null	ELECTROLYTE-PLASMA POLISHING OF TITANIUM AND NIOBIUM ALLOYS	Nauka i tehnika	2,018	cc-by	5,916	Mechanical Engineering Mechanical Engineering https://doi.org/10.21122/2227-1031-2018-17-3-211-219 © Белорусский национальный технический университет, 2018 Belarusian National Technical University, 2018 © Белорусский национальный технический университет, 2018 Belarusian National Technical University, 2018 Реферат. Титановые и ниобиевые сплавы широко применяются в настоящее время в самолетостроении, атомной энергетике, СВЧ-технике, космической и ультразвуковой технике, а также при производстве изделий медицинского назначения. В большинстве случаев технология изготовления таких изделий предусматривает выполнение качест- венного полирования поверхности. Традиционно для полирования изделий из титановых и ниобиевых сплавов ис- пользуются механические и электрохимические методы. Недостатки механических методов – малая производитель- ность, подверженность внедрению инородных частиц, затруднения при обработке сложных геометрических форм. Для электрохимических технологий указанные материалы являются труднообрабатываемыми, а процессы их полиро- вания требуют применения токсичных электролитов. Традиционно электрохимическое полирование титановых и ниобиевых сплавов осуществляют в кислотных электролитах, состоящих из токсичной плавиковой (20–25 %), сер- ной азотной и хлорной кислот. Недостатки таких растворов – их высокая агрессивность и вред, наносимый производ- ственному персоналу и окружающей среде. Предлагается использовать принципиально новые, разработанные авто- рами статьи режимы электролитно-плазменной обработки с целью электролитно-плазменного полирования и очистки изделий из титановых и ниобиевых сплавов с применением электролитов простого состава на основе водного раство- ра фторида аммония, обеспечивающие существенное повышение качества поверхности с высокой отражательной способностью. За счет применения водного электролита технология обладает высокой экологической безопасностью по сравнению с традиционным электрохимическим полированием. Приводятся результаты исследования влияния характеристик процесса электролитно-плазменного полирования титана и ниобия с применением разработанного режима на производительность, эффективность обработки, качество поверхности, а также на структуру и свойст- ва обрабатываемой поверхности. На основании полученных результатов отработаны процессы электролитно- плазменного полирования ряда изделий из титановых сплавов ВТ6 (Grade 5), применяемых в медицине и авиа- строении. Ключевые слова: электролитно-плазменная обработка, полирование, очистка, титан, ниобий, электролит, шерохова- тость, производительность, плотность тока, напряжение Для цитирования: Электролитно-плазменное полирование титановых и ниобиевых сплавов / Ю. Г. Алексеев [и др.] // Наука и техника. 2018. Т. 17, № 3. С. 211–219. https://doi.org/10.21122/2227-1031-2018-17-3-211-219 1)Белорусский национальный технический университет (Минск, Республика Беларусь) 1)Белорусский национальный технический университет (Минск, Республика Беларусь) Yu. G. Aliakseyeu1), A. Yu. Korolyov1), V. S. Niss1), A. E. Parshuto1), A. S. Budnitskiy1) Yu. G. Aliakseyeu1), A. Yu. Korolyov1), V. S. Niss1), A. E. Parshuto1), A. S. Budnitskiy1) Yu. G. Aliakseyeu1), A. Yu. Korolyov1), V. S. Niss1), A. E. Parshuto1), A. S. Budnitskiy1 1)Belarusian National Technical University (Minsk, Republic of Belarus) Abstract. Titanium and niobium alloys are widely used at present in aircraft, nuclear energy, microwave technology, space and ultrasonic technology, as well as in manufacture of medical products. In most cases production technology of such pro- ducts involves an implementation of a quality polishing surface. Mechanical and electrochemical methods are conventionally Address for correspondence Korolyov Aleksandr Yu. Belarusian National Technical University 24 Ya. Kolasa str., 220013, Minsk, Republic of Belarus Tel.: +375 17 292-25-98 korolyov@park.bntu.by Электролитно-плазменное полирование титановых и ниобиевых сплавов Канд. техн. наук, доц. Ю. Г. Алексеев1), канд. техн. наук А. Ю. Королёв1), канд. техн. наук, доц. В. С. Нисс1), инж. А. Э. Паршуто1), асп. А. С. Будницкий1) Канд. техн. наук, доц. Ю. Г. Алексеев1), канд. техн. наук А. Ю. Королёв1), канд. техн. наук, доц. В. С. Нисс1), инж. А. Э. Паршуто1), асп. А. С. Будницкий1) Адрес для переписки 211 Машиностроение used for polishing products made of titanium and niobium alloys. Disadvantages of mechanical methods are low productivi- ty, susceptibility to introduction of foreign particles, difficulties in processing complex geometric shapes. These materials are hard-to-machine for electrochemical technologies and processes of their polishing require the use of toxic electrolytes. Traditionally, electrochemical polishing of titanium and niobium alloys is carried out in acid electrolytes consisting of toxic hydrofluoric (20–25 %), sulfuric nitric and perchloric acids. The disadvantage of such solutions is their high aggressiveness and harmful effects for production personnel and environment. This paper proposes to use fundamentally new developed modes of electrolytic-plasma treatment for electrolyte-plasma polishing and cleaning products of titanium and niobium alloys while using simple electrolyte composition based on an aqueous ammonium fluoride solution providing a significant increase in surface quality that ensures high reflectivity. Due to the use of aqueous electrolyte the technology has a high ecological safety in comparison with traditional electrochemical polishing. The paper presents results of the study pertaining to the effect of titanium and niobium electrolytic-plasma polishing characteristics using the developed mode for productivity, processing efficiency, surface quality, and structure and properties of the surface to be treated. Based on the obtained results, processes of electrolytic-plasma polishing of a number of products made of titanium alloys BT6 (Grade 5), used in medicine and aircraft construction, have been worked out in the paper. Keywords: electrolyte-plasma treatment, polishing, cleaning, titanium, niobium, electrolyte, roughness, productivit density, voltage For citation: Aliakseyeu Yu. G., Korolyov A. Yu., Niss V. S., Parshuto A. E., Budnitskiy A. S. (2018) Electrolyte-Plasma Polishing of Titanium and Niobium Alloys. Science аnd Technique. 17 (3), 211–219. https://doi.org/10.21122/2227-1031- 2018-17-3-211-219 (in Russian) электролитов. Электрохимическое полирование титановых и ниобиевых сплавов осуществляют в кислотных электролитах, состоящих из токсич- ной плавиковой (20–25 %), серной азотной и хлорной кислот [7]. Недостатки таких раство- ров – их высокая агрессивность и вред, наноси- мый производственному персоналу и окружаю- щей среде. В последние годы были разработаны электролиты на основе растворов солей фтора в органических растворителях, таких как метанол или диметилформамид [8], которые также небез- опасны и токсичны. Введение Коммутационный режим Электролитно- плазменная обработка Режим электролиза Ток Напряжение, В А В С Цель работы – исследование влияния харак- теристик ЭПП титана и ниобия с применением разработанного режима на производительность, эффективность обработки, качество поверхно- сти, а также структуру и свойства обрабатыва- емой поверхности. Рис. 1. Вольтамперная характеристика анодного процесса в электролите (участок ВС соответствует режиму электролитно-плазменной обработки) Fig. 1. Voltage-current characteristic of anode process in electrolyte (segment BC corresponds to electrolyte-plasma treatment mode) Наука и техника. Т. 17, № 3 (2018) Science and Technique. V. 17, No 3 (2018) Введение Благодаря особым свойствам титановые и ниобиевые сплавы получили в настоящее время широкое распространение при производстве ряда ответственных изделий [1, 2]. Так, титано- вые и ниобиевые сплавы применяются в само- летостроении, атомной энергетике, СВЧ-тех- нике, космической и ультразвуковой технике, а также при производстве изделий медицинско- го назначения [3–5]. В большинстве случаев технология изготовления изделий предусмат- ривает выполнение качественного полирования поверхности. К таким изделиям относятся, например, зубные и костные имплантаты, им- плантаты для травматологии, черепные пласти- ны, фиксаторы позвоночника; турбинные ло- патки авиационных двигателей из титановых сплавов; листы, фольга и проволока, использу- емые для скрепления тканей, нервов, наложе- ния швов, изготовления протезов; детали теп- ловыделяющих и теплообменных элементов ядерно-энергетических систем; детали ускоря- ющих структур коллайдеров. Для снижения экологической нагрузки в ка- честве альтернативы существующим методам электрохимического полирования возможно использование электролитно-плазменной обра- ботки, которая применяется для полирования (электролитно-плазменное полирование – ЭПП), удаления заусенцев и очистки металлических изделий, а также с целью повышения физико- механических и химических свойств поверхно- сти [9]. Кроме того, электролитно-плазменная обработка может применяться для электролит- ного нагрева и электрохимикотермического упрочнения поверхности [10]. Электролитно- плазменная обработка выполняется при напря- жении более 200 В. Режим электролитно-плаз- менной обработки соответствует участку ВС на вольтамперной характеристике анодного процесса в электролите (рис. 1). На практике рабочее напряжение составляет 280–300 В при плотности тока 0,1–0,4 А/см2. В качестве элек- тролитов обычно используются растворы солей концентрацией 3–6 %. Традиционно для полирования изделий из титановых и ниобиевых сплавов используют- ся механические и электрохимические методы. Недостатки механических методов – малая про- изводительность, подверженность внедрению инородных частиц, затруднения при обработке сложных геометрических форм [6]. Для электро- химических технологий указанные материалы являются труднообрабатываемыми, а процессы их полирования требуют применения токсичных Наука техника. Т. 17, № 3 (2018) и ence and Technique V 17 No 3 (2018) Наука техника. Т. 17, № 3 (2018) и ence and Technique V 17 No 3 (2018) 212 Mechanical Engineering Рис. 1. Вольтамперная характеристика анодного процесса в электролите (участок ВС соответствует режиму электролитно-плазменной обработки) Fig. 1. Voltage-current characteristic of anode process in electrolyte (segment BC corresponds to electrolyte-plasma treatment mode) Коммутационный режим Электролитно- плазменная обработка Режим электролиза Ток Напряжение, В А В С обеспечивающие существенное повышение ка- чества поверхности с высокой отражательной способностью. За счет применения водного электролита технология обладает высокой эко- логической безопасностью по сравнению с тра- диционным электрохимическим полированием. Разработанный электролит легко корректирует- ся, что позволяет применять процесс для обра- ботки изделий из титановых и ниобиевых спла- вов в промышленных масштабах. Материалы, оборудование и методы исследований Экспериментальное оборудование для выполнения исследований: а – схема рабочей ванны: 1 – образец (анод); 2 – нагреватель; 3 – система перемешивания электролита; 4 – теплообменник; 5 – датчик температуры; 6 – ванна (катод); b – фотография оборудования Fig. 2. Experimental equipment for research: a – working bath scheme: 1 – sample (anode); 2 – heater; 3 – electrolyte mixing system; 4 – heat exchanger; 5 – temperature sensor; 6 – bath (cathode); b – equipment photo 6 1 2 3 5 4 (+) (–) b a 6 1 2 3 5 4 (+) (–) 4 5 (+) Рис. 2. Экспериментальное оборудование для выполнения исследований: а – схема рабочей ванны: ец (анод); 2 – нагреватель; 3 – система перемешивания электролита; 4 – теплообменник; 5 – датчик тем 6 – ванна (катод); b – фотография оборудования Рис. 2. Экспериментальное оборудование для выполнения исследований: а – схема рабочей ванны: 1 – образец (анод); 2 – нагреватель; 3 – система перемешивания электролита; 4 – теплообменник; 5 – датчик температуры; 6 – ванна (катод); b – фотография оборудования Fig. 2. Experimental equipment for research: a – working bath scheme: 1 – sample (anode); 2 – heater; 3 – electrolyte mixing system; 4 – heat exchanger; 5 – temperature sensor; 6 – bath (cathode); b – equipment photo 6 – ванна (катод); b – фотография оборудования Fig. 2. Experimental equipment for research: a – working bath scheme: 1 – sample (anode); 2 – heater; 3 – electrolyte mixing system; 4 – heat exchanger; 5 – temperature sensor; 6 – bath (cathode); b – equipment photo 6 ванна (катод); b фотография оборудования Fig. 2. Experimental equipment for research: a – working bath scheme: 1 – sample (anode); 2 – heater; 3 – electrolyte mixing system; 4 – heat exchanger; 5 – temperature sensor; 6 – bath (cathode); b – equipment photo Fig. 2. Experimental equipment for research: a – working bath scheme: 1 – sample (anode); 2 – heater; 3 – electroly 4 – heat exchanger; 5 – temperature sensor; 6 – bath (cathode); b – equipment photo tal equipment for research: a – working bath scheme: 1 – sample (anode); 2 – heater; 3 – electrolyte mixing system; 4 – heat exchanger; 5 – temperature sensor; 6 – bath (cathode); b – equipment photo Оценка производительности выполнялась по изменению массы образцов в результате об- работки. Материалы, оборудование и методы исследований ЭПП по сравнению с механическим и элек- трохимическим полированием обладает рядом существенных преимуществ: высокая экологиче- ская безопасность по сравнению с классическим электрохимическим полированием за счет при- менения электролитов на основе водных раство- ров солей; возможность обработки деталей и из- делий любой конфигурации; возможность полу- чения зеркальной поверхности с высотой микронеровностей вплоть до Ra = 0,01 мкм; устранение в процессе обработки некондицион- ного поверхностного слоя и остаточных на- пряжений, что улучшает физико-механические и химические свойства поверхности; достаточно короткая продолжительность процесса полирова- ния; существенное снижение ручного труда; воз- можность обработки высокотвердых и вязких материалов [11–14]. Однако массовое использо- вание технологии ЭПП ограничивается тем, что в промышленных масштабах к настоящему време- ни освоены процессы ЭПП только небольшого перечня материалов: низкоуглеродистые и корро- зионностойкие стали, алюминиевые сплавы, бронзы и латуни. Технологии обработки таких материалов, как, например, титан и ниобий, отра- ботаны лишь в лабораторных условиях [15] и не получили распространения в промышленности. Исследования проводили на плоских образ- цах технически чистого титана ВТ1-0 (Grade 2) размерами 30×15×1,5 мм и технически чисто- го ниобия ВН (Nb-1) размерами 20×30×2 мм. Образцы из титана предварительно были об- работаны шлифовальной бумагой SiC зернис- тостью Р600, образцы из ниобия – шлифоваль- ной бумагой размерностью Р300. Среднее зна- чение шероховатости поверхности Ra исход- ных образцов из титана и ниобия составило 0,365 и 0,706 мкм соответственно. Для выполнения ЭПП использовалась специ- ально разработанная экспериментальная установ- ка, включающая рабочую ванну (катод), нагрева- тель, теплообменник, датчик температуры, си- стему перемешивания электролита (рис. 2). ЭПП образцов выполняли в водном раство- ре фторида аммония (NH4F) концентрацией 4 %. Значение рабочего напряжения изменялось в диа- пазоне от 260 до 300 В с шагом 10 В. Продол- жительность обработки каждого образца 2 мин. При исследовании влияния напряжения на эф- фективность обработки, качество поверхности, производительность обработки температура электролита составляла 90 оС, продолжитель- ность обработки 1, 3 и 5 мин. При исследова- нии влияния плотности тока на эффективность обработки, качество поверхности, производи- тельность обработки регулирование плотности тока осуществлялось путем изменения темпе- ратуры электролита в диапазоне от 75 до 95 оС (величина напряжения составляла 300 В, про- должительность обработки 3 мин). Для решения указанных проблем предлага- ется использовать принципиально новые, раз- работанные авторами статьи режимы электро- литно-плазменного полирования и очистки изделий из титановых и ниобиевых сплавов с применением электролитов простого состава на основе водного раствора фторида аммония, 213 Машиностроение a b Рис. 2. Материалы, оборудование и методы исследований Массу образцов до и после обработки определяли с помощью аналитических весов Ohaus Pioneer PA214. Силу тока находили с помощью токовых клещей UNIT-203. Плот- ность тока устанавливалась как отношение си- лы тока к площади обрабатываемой поверхно- сти. Эффективность обработки при различных режимах определялась как отношение измене- ния шероховатости к удельной массе удаленно- го материала в процессе обработки (∆Ra/∆mуд). видно, что с увеличением напряжения плот- ность тока незначительно уменьшается. Это в целом характерно для процессов электролитно- плазменной обработки и связано с тем, что с повышением рабочего напряжения возрастает энергия, выделяющаяся в парогазовой оболоч- ке. Это приводит к росту температуры анода и увеличению толщины парогазовой оболочки. Плотность тока для ниобия соответствует ана- логичным значениям плотности тока при обра- ботке коррозионностойких сталей в 4%-м рас- творе сульфата аммония согласно данным, по- лученным в [13]. Сравнение зависимостей на рис. 4а, b показывает, что для образцов из тита- на плотность тока больше, чем для образцов из ниобия. Так, из экспериментальных данных следует, что, например, при температуре элек- тролита 90 оС плотность тока при обработке образцов из титана в среднем больше на 20,3 %, чем при обработке ниобия. Кроме того, вольт- амперные характеристики для титана являются более пологими, чем для ниобия. Соответ- ственно влияние напряжения на плотность тока для титана менее выраженное. Микрофотографии поверхности образцов до и после ЭПП получены с помощью сканирую- щего электронного микроскопа VEGA II LMU с микроанализатором INCA350. т. е. соблюдается корреляция между плотностью тока и съемом металла в соответствии с законом Фарадея. Экспериментальные зависимости удельного съема ∆mуд от рабочего напряжения, полученные для образцов из титана и ниобия при обработке с различной продолжительностью, представле- ны на рис. 5. С увеличением напряжения умень- шается производительность обработки. Зависи- мости удельного съема от напряжения для ти- тана (рис. 5а), как и в случае с зависимостями для плотности тока от напряжения (рис. 4а), яв- ляются более пологими по сравнению с анало- гичными зависимостями для ниобия (рис. 4b, 5b), т. е. соблюдается корреляция между плотностью тока и съемом металла в соответствии с законом Фарадея. Снижение производительности ЭПП с уве- личением рабочего напряжения не обусловли- вает ухудшения качества формируемой поверх- ности, в частности параметра шероховатости Ra. Экспериментальные зависимости, демонстри- рующие динамику изменения шероховатости поверхности Ra при обработке образцов из ти- тана и ниобия, представлены на рис. 6. b 250 260 270 280 290 300 310 Напряжение U, В 0,012 0,010 0,008 0,006 0,004 0,002 0 Удельный съем ∆mуд, г/см2 5 мин 3 мин 1 мин а b 250 260 270 280 290 300 310 Напряжение U, В 250 260 270 280 290 300 310 Напряжение U, В Рис. 5. Влияние напряжения электролитно-плазменного полирования на удельный съем: a – титан; b – ниобий Fig. 5. Effect of electrolyte-plasma polishing voltage on specific removal: a – titanium; b – niobium 0,010 0,008 0,006 0,004 0,002 0 Удельный съем ∆mуд, г/см2 5 мин 3 мин 1 мин 0,012 0,010 0,008 0,006 0,004 0,002 0 Удельный съем ∆mуд, г/см2 5 мин 3 мин 1 мин а 250 260 270 280 290 300 310 Напряжение U, В 0,010 0,008 0,006 0,004 0,002 0 Удельный съем ∆mуд, г/см2 5 мин 3 мин 1 мин Напряжение U, В Напряжение U, В Рис. 5. Влияние напряжения электролитно-плазменного полирования на удельный съем: a – титан; b – ниобий Fig. 5. Effect of electrolyte-plasma polishing voltage on specific removal: a – titanium; b – niobium Рис. 5. Влияние напряжения электролитно-плазменного полирования на удельный съем: a – титан; b – ниобий Fig. 5. Результаты и обсуждение Фотографии образцов титана и ниобия до и после обработки представлены на рис. 3. Зависимости плотности тока от рабочего напряжения (вольтамперные характеристики) в процессе ЭПП титана и ниобия, полученные при различных значениях температуры элек- тролита, представлены на рис. 4. Из графиков а b Рис. 3. Внешний вид образцов титана (a) и ниобия (b) до и после электролитно-плазменного полирования Fig. 3. Appearance of titanium (a) and niobium (b) samples before and after electrolyte-plasma polishing Рис. 3. Внешний вид образцов титана (a) и ниобия (b) до и после электролитно-плазменного полирования Fig. 3. Appearance of titanium (a) and niobium (b) samples before and after electrolyte-plasma polishing 214 Mechanical Engineering а b 220 240 260 280 300 320 Напряжение U, В 220 240 260 280 300 320 Напряжение U, В Рис. 4. Вольтамперные характеристики электролитно-плазменного полирования титана (а) и ниобия (b) при различных значениях температуры электролита Fig. 4. Voltage-current characteristics of titanium (а) and niobium (b) electrolyte-plasma polishing at different values of electrolyte temperature 0,7 0,6 0,5 0,4 0,3 0,2 0,1 0 Плотность тока j, А/см2 70 °C 80 °C 90 °C 70 °C 80 °C 90 °C 0,7 0,6 0,5 0,4 0,3 0,2 0,1 0 Плотность тока j, А/см2 b 220 240 260 280 300 320 Напряжение U, В 70 °C 80 °C 90 °C 0,7 0,6 0,5 0,4 0,3 0,2 0,1 0 Плотность тока j, А/см2 Рис. 4. Вольтамперные характеристики электролитно-плазменного полирования титана (а) и ниобия (b) при различных значениях температуры электролита Fig. 4. Voltage-current characteristics of titanium (а) and niobium (b) electrolyte-plasma polishing at different values of electrolyte temperature Fig. 4. Voltage-current characteristics of titanium (а) and niobium (b) electrolyte-plasma polishing at different values of electrolyte temperature т. е. соблюдается корреляция между плотностью тока и съемом металла в соответствии с законом Фарадея. Effect of electrolyte-plasma polishing voltage on specific removal: a – titanium; b – niobium b 0 1 2 3 4 5 6 Продолжительность обработки, мин 2,2 1,8 1,4 1,0 0,6 0,2 Шероховатость Ra, мкм 260 В 280 В 300 В а 0 1 2 3 4 5 6 Продолжительность обработки, мин 0,40 0,35 0,30 0,25 0,20 0,15 0,10 0,05 Шероховатость Ra, мкм 260 В 280 В 300 В а b 0 1 2 3 4 5 6 Продолжительность обработки, мин 0 1 2 3 4 5 6 Продолжительность обработки, мин Рис. 6. Влияние продолжительности электролитно-плазменного полирования на шероховатость поверхности образцов при различных значения напряжения: а – титан; b – ниобий Fig. 6. Effect of electrolyte-plasma polishing duration on surface roughness of samples at different voltage values: a – titanium; b – niobium 0,40 0,35 0,30 0,25 0,20 0,15 0,10 0,05 2,2 1,8 1,4 1,0 0,6 0,2 Шероховатость Ra, мкм Шероховатость Ra, мкм 260 В 280 В 300 В 260 В 280 В 300 В b а Рис. 6. Влияние продолжительности электролитно-плазменного полирования на шероховатость поверхности образцов при различных значения напряжения: а – титан; b – ниобий Fig. 6. Effect of electrolyte-plasma polishing duration on surface roughness of samples at different voltage values: a – titanium; b – niobium Наука и техника. Т. 17, № 3 (2018) 215 Машиностроение 0,18–0,45 А/см2, для ниобия – 0,19–0,48 А/см2) экспериментально установленные значения ве- личины изменения шероховатости поверхно- сти ∆Ra имеют существенный разброс как для образцов из титана, так и для образцов из нио- бия. При этом наблюдается тенденция к незначи- тельному росту величины изменения шерохова- тости поверхности с увеличением плотности то- ка. Значения ∆Ra для ниобия существенно выше аналогичных значений для титана. Из графиков следует, что с увеличением ра- бочего напряжения в исследуемом диапазо- не (от 260 до 300 В) обеспечивается снижение достигаемых значений параметра шерохова- тости поверхности Ra. При этом в результа- те обработки ниобия при значении напряже- ния 260 В вместо полирования происходит растравливание поверхности с увеличением ше- роховатости, а значение параметра шерохова- тости Ra интенсивно увеличивается с повыше- нием продолжительности обработки (рис. 6b). 0,1 0,2 0,3 0,4 0,5 0,6 Плотность тока j, А/см2 0,40 0,35 0,30 0,25 0,20 0,15 0,10 0,05 0 Изменение шероховатости ∆Ra, мкм 0,1 0,2 0,3 0,4 0,5 0,6 Плотность тока j, А/см2 Рис. 8. Влияние плотности тока на изменение шероховатости поверхности ∆Ra образцов титана и ниобия Fig. 8. т. е. соблюдается корреляция между плотностью тока и съемом металла в соответствии с законом Фарадея. Effect of current density on surface roughness change ∆Ra of titanium and niobium samples 0,40 0,35 0,30 0,25 0,20 0,15 0,10 0,05 0 Изменение шероховатости ∆Ra, мкм В процессах электрохимической обработ- ки (в том числе электролитно-плазменной) съем металла с поверхности выполняется по закону Фарадея, согласно которому объем (или масса) металла, удаленного с заготовки, прямо пропорционален электрическому заряду, про- шедшему через электролит. Таким образом, чем больше плотность тока, тем выше произво- дительность обработки. Экспериментальные зависимости удельного съема от плотности то- ка для титана и ниобия представлены на рис. 7. Полученные зависимости имеют линейный ха- рактер. Съем металла увеличивается с повы- шением плотности тока. При этом удель- ный съем для образцов из ниобия почти в два раза превышает удельный съем для титана, что связано с более высокими значениями элект- рохимического эквивалента (Ti – 0,162 мг/K, Nb – 0,192 мг/К) [16] и, вероятно, коэффициен- та выхода по току. Рис. 8. Влияние плотности тока на изменение шероховатости поверхности ∆Ra образцов титана и ниобия Fig. 8. Effect of current density on surface roughness change ∆Ra of titanium and niobium samples Диаграммы, характеризующие влияние ра- бочего напряжения и плотности тока на эффек- тивность обработки титана и ниобия по изме- нению шероховатости, представлены на рис. 9. Показатели эффективности обработки поверх- ности ∆Ra/∆mуд как для титана, так и для нио- бия имеют примерно равные значения, за ис- ключением значения, полученного для ниобия при напряжении 260 В, когда эффективность является отрицательной. Анализ представлен- ных диаграмм показывает, что для достижения высоких значений эффективности одновре- менно необходимо выполнять обработку при следующих режимах: для титана – напряже- ние 300 В, плотность тока 0,18–0,30 А/см2, для ниобия – напряжение 280–300 В, плотность то- ка 0,18–0,20 А/см2. 0,1 0,2 0,3 0,4 0,5 0,6 Плотность тока j, А/см2 Рис. 7. Влияние плотности тока на удельный съем Fig. 7. Effect of current density on specific removal 0,012 0,010 0,008 0,006 0,004 0,002 0 Удельный съем ∆mуд, г/см2 Nb Ti 0,1 0,2 0,3 0,4 0,5 0,6 Плотность тока j, А/см2 0,012 0,010 0,008 0,006 0,004 0,002 0 Удельный съем ∆mуд, г/см2 Nb Ti Рис. 7. Влияние плотности тока на удельный съем Микрофотографии поверхности образцов титана и ниобия до и после обработки пред- ставлены на рис. 10. Поверхность титана до ЭПП (рис. 10a) характеризуется наличием про- дольных полос, образованных в результате предварительного шлифования образцов. После Fig. 7. Effect of current density on specific removal Fig. 7. т. е. соблюдается корреляция между плотностью тока и съемом металла в соответствии с законом Фарадея. Effect of current density on specific removal Зависимости изменения шероховатости по- верхности образцов титана и ниобия от плотно- сти тока представлены на рис. 8. В исследуемом диапазоне значений плотности тока (для титана – Наука техника. Т. 17, № 3 (2018) и ence and Technique V 17 No 3 (2018) Наука техника. Т. 17, № 3 (2018) и ence and Technique V 17 No 3 (2018) 216 Mechanical Engineering ность с присутствием незначительного количе- ства питтингов размерами до 5 мкм (рис. 10d). ЭПП поверхность сглаживается, присутствуют только следы от наиболее глубоких цара- пин (рис. 10b). Поверхность исходных образцов из ниобия, кроме продольных полос, получен- ных при шлифовании, характеризуется также наличием достаточно крупных задиров с раз- мерами в плане до 12 мкм (рис. 10c). В резуль- тате ЭПП ниобия формируется гладкая поверх- На основании полученных результатов от- работаны процессы ЭПП ряда изделий из тита- новых сплавов ВТ6 (Grade 5), применяемых в медицине и авиастроении. Примеры обработки деталей с помощью разработанной технологии представлены на рис. 11. b 0,18–0,19 0,20–0,23 0,30–0,33 0,45–0,48 Плотность тока j А/см2 – Nb – Ti 12 10 8 6 4 2 0 Эффективность ∆Ra/∆mуд, мкм⋅см2/г а b 260 280 300 Напряжение U, B 0,18–0,19 0,20–0,23 0,30–0,33 0,45–0,48 Плотность тока j, А/см2 Рис. 9. Влияние электрических параметров на эффективность электролитно-плазменного полирования титана и ниобия: а – напряжения; b – плотности тока Fig 9 Influence of electrical parameters on electrolyte-plasma polishing efficiency of titanium and niobium: 8 6 4 2 0 –2 –4 Эффективность ∆Ra/∆mуд, мкм⋅см2/г – Nb – Ti – Nb – Ti 12 10 8 6 4 2 0 Эффективность ∆Ra/∆mуд, мкм⋅см2/г b а 260 280 300 Напряжение U, B 8 6 4 2 0 –2 –4 Эффективность ∆Ra/∆mуд, мкм⋅см2/г – Nb – Ti Влияние электрических параметров на эффективность электролитно-плазменного полирования титана и ниоб а – напряжения; b – плотности тока Рис. 9. Влияние электрических параметров на эффективность электролитно-пла а – напряжения; b – плотности тока Fig. 9. Influence of electrical parameters on electrolyte-plasma polishing efficiency of titanium and niobium: a – voltage; b – current density b d b d а b c d Рис. 10. Микрофотографии поверхности образцов титана и ниобия до и после электролитно-плазменного полирования (ЭПП): a, b – титан до и после ЭПП; c, d – ниобий до и после ЭПП Fig. 10. Наука и техника. Т. 17, № 3 (2018) т. е. соблюдается корреляция между плотностью тока и съемом металла в соответствии с законом Фарадея. Surface microphotographs of titanium and niobium samples before and after electrolyte-plasma polishing (EPP): a, b – titanium before and after EPP; c, d – niobium before and after EPP а c а d d Рис. 10. Микрофотографии поверхности образцов титана и ниобия до и после электролитно-плазменного полирования (ЭПП): a, b – титан до и после ЭПП; c, d – ниобий до и после ЭПП Fig. 10. Surface microphotographs of titanium and niobium samples before and after electrolyte-plasma polishing (EPP): a, b – titanium before and after EPP; c, d – niobium before and after EPP Наука и техника. Т. 17, № 3 (2018) S i d T h i V 17 N 3 (2 217 Машиностроение Рис. 11. Примеры электролитно-плазменного полирования изделий из титанового сплава ВТ6 (Grade 5) пазоне наблюдается постепенный рост величи- ны изменения шероховатости поверхности ∆Ra. Наибольшее изменение шероховатости ∆Ra при обработке ниобия достигается в диапазоне зна- чений напряжения от 280 до 300 В. пазоне наблюдается постепенный рост величи- ны изменения шероховатости поверхности ∆Ra. Наибольшее изменение шероховатости ∆Ra при обработке ниобия достигается в диапазоне зна- чений напряжения от 280 до 300 В. 3. По результатам исследования влияния плотности тока на качество электролитно-плаз- менного полирования титана и ниобия установ- лено, что с увеличением плотности тока в ис- следуемом диапазоне значений (для титана – 0,18–0,45 А/см2, для ниобия – 0,19–0,48 А/см2) наблюдается тенденция к незначительному ро- сту величины изменения шероховатости по- верхности ∆Ra как для титана, так и для нио- бия. При этом значения ∆Ra для ниобия су- щественно больше аналогичных значений для титана. Рис. 11. Примеры электролитно-плазменного полирования изделий из титанового сплава ВТ6 (Grade 5) Рис. 11. Примеры электролитно-плазменного полирования изделий из титанового сплава ВТ6 (Grade 5) 4. Наибольшие значения эффективности элек- тролитно-плазменного полирования достигаются при следующих режимах: для титана – напряже- ние 300 В, плотность тока 0,18–0,30 А/см2, для ниобия – напряжение 280–300 В, плотность то- ка 0,18–0,20 А/см2. Fig. 11. Examples of electrolyte-plasma polishing (EPP) products made of titanium alloy ВТ6 (Grade 5) ВЫВОДЫ 1. Разработан новый метод электролитно- плазменного полирования титановых и ниобие- вых сплавов, обладающий высокой экологиче- ской безопасностью по сравнению с класси- ческим электрохимическим полированием за счет применения безвредных электролитов на основе водных растворов солей общей концен- трацией не более 6 %. Для сравнения, традици- онное полирование, например, титановых спла- вов выполняется в электролитах с температу- рой не ниже 80 оС, содержащих помимо серной, азотной и хлорной кислот токсичную плавико- вую кислоту концентрацией 20–25 %. 1. Разработан новый метод электролитно- плазменного полирования титановых и ниобие- вых сплавов, обладающий высокой экологиче- ской безопасностью по сравнению с класси- ческим электрохимическим полированием за счет применения безвредных электролитов на основе водных растворов солей общей концен- трацией не более 6 %. Для сравнения, традици- онное полирование, например, титановых спла- вов выполняется в электролитах с температу- рой не ниже 80 оС, содержащих помимо серной, азотной и хлорной кислот токсичную плавико- вую кислоту концентрацией 20–25 %. ЛИТЕРАТУРА (2009) Workpie- ce Surface Integrity of Ti-6-4 Heat-Resistant Alloy when Employing Different Polishing Methods. Journal of Materials Processing Technology, 209 (4), 1843–1852. https://doi.org/10.1016/j.jmatprotec.2008.04.046. 9. Surface Finish Machining of Medical Parts using Plasma Electrolytic Polishing / H. Zeidler [et al.] // Procedia CIRP. 2016. Vol. 49. Р. 83–87. https://doi.org/10.1016/ j.procir.2015.07.038. 10. Особенности электролитно-плазменного нагрева при электрохимико-термической обработке стали / Ю. Г. Алексеев [и др.] // Наука и техника. 2013. № 6. С. 20–24. 7. Tian H., Corcoran S. G., Reece C. E., Kelley M. J. (2008) The Mechanism of Electropolishing of Niobium in Hydro- fluoric-Sulfuric Acid Electrolyte. Journal of the Electro- chemical Society, 155 (9), D563–D568. https://doi.org/10. 1149/1.2945913. 11. Aliakseyeu, Yu. Electrolyte-Plasma Treatment of Metal Materials Surfaces / Yu. Aliakseyeu, A. Korolyov, A. Be- zyazychnaya // 14 International Scientific Conference “CO-MAT-TECH-2006”, Slovakia, Trnava, 19–20 oct. 2006. Р. 6. 8. Tajima K., Hironaka M., Chen K.-K., Nagamatsu Y., Kakigawa H., Kozono Y. (2008) Electropolishing of CP Titanium and its Alloys in an Alcoholic Solution-Based Electrolyte. Dental Materials Journal, 27 (2), 258–265. https://doi.org/10.4012/dmj.27.258. 12. Влияние электролитно-плазменной обработки на структуру и свойства поверхностного слоя стали 12Х18Н10Т / И. В. Фомихина [и др.] // Весцi Нац. акад. навук Беларусі. Сер. фiз.-тэхн. навук. 2008. № 3. С. 24–29. 9. Zeidler H., Boettger-Hiller F., Edelmann J., Schubert A. (2016) Surface Finish Machining of Medical Parts using Plasma Electrolytic Polishing. Procedia CIRP, 49, 83–87. https://doi.org/10.1016/j.procir.2015.07.038. 13. Модель размерного съема материала при электролит- но-плазменной обработке цилиндрических поверхно- стей / Ю. Г. Алексеев [и др.] // Наука и техника. 2012. № 3. С. 3–6. 10. Alekseev Yu. G., Niss V. S., Korolev A. Yu., Parshuto A. E. (2013) Peculiar Features of Electrolytic-Plasma Heating at Electrochemical-Heat Treatment of Steel. Nauka i Tekhni- ka = Science & Technique, (6), 20–24. 14. Электролитно-плазменная обработка при нестационар- ных режимах в условиях высокоградиентного электри- ческого поля / Ю. Г. Алексеев [и др.] // Наука и техни- ка. 2017. Т. 16, № 5. С. 391–399. https://doi.org/10. 21122/2227-1031-2017-16-5-391-399. 11. Aliakseyeu Yu., Korolyov A., Bezyazychnaya A. (2006) Electrolyte-Plasma Treatment of Metal Materials Surfaces. CO-MAT-TECH-2006: Proceeding of the Abstracts of 14th In- ternational Scientific Conference, Slovak University of Technology, 19–20 October 2006. Slovakia, Trnava, 6. 15. Многоэтапная электролитно-плазменная обработка изделий из титана и титановых сплавов / А. М. Смыс- лов [и др.] // Вестник Уфимского государственного авиационного технического университета. 2009. Т. 13, №. 1. C. 141–145. 12. Fomikhina I. V., Litovskaya Yu. O., Alekseev Yu. G., Korolev A. ЛИТЕРАТУРА 1. Application and Features of Titanium for the Aerospace Industry / I. Inagaki [et al.] // Nippon Steel & Sumitomo Metal Technical Report. 2014. Vol. 106. Р. 22–27. 2. Nikishina, E. E. Niobium and Tantalum: State of the World Market, Fields of Application, and Raw Sources. Part I / Е. Е. Nikishina, D. V. Drobot, E. N. Lebedeva // Russian Journal of Non-Ferrous Metals. 2013. Vol. 54, No 6. Р. 446–452. 3. Froes, F. H. Cost Affordable Developments in Titanium Technology and Applications / F. H. Froes, A. M. Imam // Key Engineering Materials. 2010. Vol. 436. Р. 1–11. https://doi.org/10.4028/www.scientific.net/kem.436.1 2. Установлено, что с увеличением рабочего напряжения в исследуемом диапазоне (от 260 до 300 В) в процессе электролитно-плазмен- ного полирования обеспечивается снижение до- стигаемых значений параметра шероховатости поверхности Ra. При этом в результате обра- ботки ниобия при значении напряжения 260 В вместо полирования происходит растравлива- ние поверхности с увеличением шероховато- сти, а значение параметра шероховатости Ra интенсивно возрастает с повышением продол- жительности обработки. При электролитно- плазменном полировании титана с увеличением рабочего напряжения во всем исследуемом диа- 4. Zardiackas, L. D. Titanium, Niobium, Zirconium, and Tan- talum for Medical and Surgical Applications / L. D. Zar- diackas, M. J. Kraay, H. L. Freese. Astm, 2006. 5. Veiga, C. Properties and Applications of Titanium Alloys: a Brief Review / С. Veiga, J. P. Davim, A. J. R. Loureiro // Rev. Adv. Mater. Sci. 2012. Vol. 32, No 2. Р. 133–148. 6. Axinte, D. A. Workpiece Surface Integrity of Ti-6-4 Heat- Resistant Alloy when Employing Different Polishing Methods / D. A. Axinte, J. Kwong, M. C. Kong // Jour- nal of Materials Processing Technology. 2009. Vol. 209, No 4. Р. 1843–1852. https://doi.org/10.1016/j.jmatprotec. 2008.04.046. 7. The Mechanism of Electropolishing of Niobium in Hydro- fluoric-Sulfuric Acid Electrolyte / H. Tian [et al.] // Jour- nal of the Electrochemical Society. 2008. Vol. 155, No 9. Р. D563–D568. https://doi.org/10.1149/1.2945913. Наука техника. Т. 17, № 3 (2018) и ence and Technique V 17 No 3 (2018) 218 Mechanical Engineering 8. Electropolishing of CP Titanium and its Alloys in an Al- coholic Solution-Based Electrolyte / K. Tajima [et al.] // Dental Materials Journal. 2008. Vol. 27, No 2. Р. 258–265. https://doi.org/10.4012/dmj.27.258. 5. Veiga C., Davim J. P., Loureiro A. J. R. (2012) Properties and Applications of Titanium Alloys: a Brief Review. Re- views on Advanced Materials Science, 32 (2), 133–148. 6. Axinte D. A., Kwong J., Kong M. C. ЛИТЕРАТУРА Yu., Niss V. S. (2008). Influence of Electroly- tic-Plasma Treatment on Structure and Properties in Sur- face Layer of Authentic Stainless 12Х18Н10Т – Steel. Vestsi Natsionalnoy Akademii Navuk Belarusi. Ser. Fiz.- Tekhn. Navuk = Bulletin of National Academy of Science of Belarus. Series of Physical and Technical Sciences, (3), 24–29 (in Russian). 16. Справочник по электрохимическим и электрофизиче- ским методам обработки / Г. Л. Амитан [и др.]; под. общ. ред. В. А. Волосатова. Л.: Машиностроение. Ле- нингр. отд-ние, 1988. 719 с. 13. Alekseev Yu., Korolev A., Parshuta A., Niss V. (2012). Model for Metal Removal in Electrolyte-Plazma Treat- ment of Cylindrical Surfaces. Nauka i Tekhnika = Science & Technique, (3), 3–6 (in Russian). Поступила 23.11.2017 Подписана в печать 19.01.2018 Опубликована онлайн 29.05.2018 Опубликована онлайн 29.05.2018 14. Aliakseyeu Y. G., Korolyov A. Y., Parshuto A. E., Niss V. S. (2017) Electrolyte-Plasma Treatment under Non-Stationa- ry Mode in a High-Gradient Electric Field. Nauka i Tekh- nika = Science & Technique, 16 (5), 391–399 (in Rus- sian). https://doi.org/10.21122/2227-1031-2017-16-5-391- 399. 8. Electropolishing of CP Titanium and its Alloys in an Al- coholic Solution-Based Electrolyte / K. Tajima [et al.] // Dental Materials Journal. 2008. Vol. 27, No 2. Р. 258–265. https://doi.org/10.4012/dmj.27.258. REFERENCES 1. Inagaki I., Takechi T., Shirai Y., Ariyasu N. (2014) Appli- cation and Features of Titanium for the Aerospace Indus- try. Nippon Steel & Sumitomo Metal Technical Report, 106, 22–27. 15. Smyslov A. M., Smyslova M. K., Mingazhev A. D., Se- livanov K. S. (2009) Multistage Electrolytic-Plasma Treatment of Products from Titan and Titanic Alloys. Vestnik Ufimskogo Gosudarstvennogo Aviatsionnogo Tekh- nicheskogo Universiteta [Bulletin of Ufa State Aviation Technical University], 13 (1), 141–145 (in Russian). 2. Nikishina E. E., Drobot D. V., Lebedeva E. N. (2013) Ni- obium and Tantalum: State of the World Market, Fields of Application, and Raw Sources. Part I. Russian Journal of Non-Ferrous Metals, 54 (6), 446–452. https://doi.org/10. 3103/s1067821213060187. 3. Froes F. H., Imam A. M. (2010) Cost Affordable Deve- lopments in Titanium Technology and Applications. Key Engineering Materials, 436, 1–11. https://doi.org/10.4028/ www.scientific.net/kem.436.1. 16. Amitan G. L., Baisupov I. A., Baron Yu. M., Volosa- tov V. A. (ed.) (1988) Reference Book on Electrochemical and Electro-Physical Processing Methods. Leningrad, Mashi- nostroenie Publ., Leningrad Branch. 719 (in Russian). 4. Zardiackas L. D., Kraay M. J., Freese H. L. (2006) Tita- nium, Niobium, Zirconium, and Tantalum for Medical and Surgical Applications. ASTM International. 265. https://doi.org/10.1520/stp1471-eb. Received: 23.11.2017 Accepted: 19.01.2018 Published online: 29.05.2018 Received: 23.11.2017 Accepted: 19.01.2018 Published online: 29.05.2018 Наука и техника. Т. 17, № 3 (2018) Science and Technique V 17 No 3 (2 Наука и техника. Т. 17, № 3 (2018) Science and Technique V 17 No 3 (2 219
https://openalex.org/W4378223486	https://journal.uny.ac.id/index.php/jpms/article/download/36135/pdf	English	null	Science Teacher Education Considering the Covid-19 Pandemic: The African Account	Jurnal pendidikan matematika dan sains/Jurnal Pendidikan Matematika dan Sains	2,022	cc-by-sa	4,418	Olalekan Taofeek Badmus1,, Afees Akanni Amuda2, Abdulrasaq Oladimeji Akanbi3, Esther Ore Omosewo4 1,3,4Department of Science Education, University of Ilorin, Ilorin, Nigeria 2Department of Science and Vocational Education, Usmanu Danfodiyo University, Sokoto, Nigeria Korespondensi Penulis. E-mail: badmus.ot@unilorin.edu.ng 1,3,4Department of Science Education, University of Ilorin, Ilorin, Nigeria 2Department of Science and Vocational Education, Usmanu Danfodiyo University, Sokoto, Nigeria Korespondensi Penulis. E-mail: badmus.ot@unilorin.edu.ng Jurnal Pendidikan Matematika dan Sains, 10 (2), 2022, 93-99 Jurnal Pendidikan Matematika dan Sains, 10 (2), 2022, 93-99 Abstract The outbreak of Covid-19 the world over is devastating. The responses to the disruptions experienced had left many nations in doubt of a possible recovery in years to come. Various sectors of human needs were distorted, and the socioeconomic impact cannot be overestimated. The education sector is one of the first sectors to be locked down immediately after the emergence of the pandemic and about the last to be recalled after the first wave of Covid-19. Education institutions are experiencing massive reforms globally amid the Covid-19 outbreak and the effect of the lockdown is enormous. The economic impact of the pandemic is more pronounced than the academic, especially in science and science teacher education. This paper investigated science teacher education considering the Covid-19 pandemic. Specifically, this research explored the African account of Covid-19, the effect of Covid-19 on science teacher education, the rationale for a rethink in science teacher education amidst Covid-19, the limitations to science teacher education post-Covid-19 and the way forward for stakeholders amid/post Covid-19. This article concluded among others that more is required in terms of investment in the education sector to actualize the modern approaches required to cope with the pandemic situations in Africa. Keywords: science, teacher, education, Covid-19, pandemic Keywords: science, teacher, education, Covid-19, pandemic How to Cite: Badmus, O. T., Amuda, A. A., Akanbi, A. O., & Omosewo, E. O. (2022). Science teacher education considering the Covid-19 pandemic: The African account. Jurnal Pendidikan Matematika dan Sains, 10(2), 93-99. doi:http://dx.doi.org/10.21831/jpms.v10i2.36135 How to Cite: Badmus, O. T., Amuda, A. A., Akanbi, A. O., & Omosewo, E. O. (2022). Science teacher education considering the Covid-19 pandemic: The African account. Jurnal Pendidikan Matematika dan Sains, 10(2), 93-99. doi:http://dx.doi.org/10.21831/jpms.v10i2.36135 Science Teacher Education Considering the Covid-19 Pandemic: The African Account Olalekan Taofeek Badmus1,, Afees Akanni Amuda2, Abdulrasaq Oladimeji Akanbi3, Esther Ore Omosewo4 1,3,4Department of Science Education, University of Ilorin, Ilorin, Nigeria 2Department of Science and Vocational Education, Usmanu Danfodiyo University, Sokoto, Nigeria *Korespondensi Penulis. E-mail: badmus.ot@unilorin.edu.ng Available online at: http://journal.uny.ac.id/index.php/jpms Available online at: http://journal.uny.ac.id/index.php/jpms The African Account of Covid-19 The first case of coronavirus was discovered towards the end of February 2020 in Africa. A nationwide lockdown was declared in March, 23rd 2020 in Nigeria. Across Africa, the lockdown was about the only meditated strategy to combat the scourge. During this period, only workers and artisans delivering essential services were permitted to carry on their daily activities, other citizens were required to obey the new normal through law enforcement strategies (Stocklin-Weinberg et al., 2019). While this position seems commendable at the time, many African countries were only willing to survive through the period, as other media for learning and education were not considered especially in the first four months of the pandemic. Surviving the pandemic emerged as the way forward after China’s systemic reopening and dolling out coping strategies to sustain economic dominance (Aknin et al., 2022). After about 7 months of the global crisis, Nigerian and many African countries have started realizing that Covid-19 is here to stay and its attendant consequence will have to be confronted head-on (Mbombo, 2022). This crisis can be looked upon as an opportunity to restructure our longstanding education systems for better and update practices in an academic institution to suit the present generation of learners. As Science teachers/educators, we must prepare ourselves for the changing world when the Covid-19 pandemic is over (Fishbane & Tomer, 2020). The reality resulting from the Covid-19 crisis raises questions about the nature of teaching and ways of supporting the learning of student teachers, but it also challenges teachers to rethink ways of re-educating themselves for scenarios that are unpredictable but raise questions related to equity and social justice in the face of Covid-19 pandemic. Until June 2020, most businesses and education institutions were on total lockdown without any attempt to explore blended learning or online classes to either remediate or palliate academic paralysis. Educational divide already exists in African societies owing to school location, parents' socio-economic ability, and even access to the internet (Strauss, 2020). The pandemic may further widen the gap for the already disadvantaged students. Moving forward, universities and other institutions may have no choice but to continue to offer online instruction (Dhawan, 2020). This will mean that educators around Africa will have to move content, classes, materials, and opportunities for experiential instruction to online classrooms. Jurnal Pendidikan Matematika dan Sains, 10 (2), 2022, 94 Olalekan Taofeek Badmus, Afees Akanni Amuda, Abdulrasaq Oladimeji Akanbi, Esther Ore Omosewo On 10th May 2020, the Covid-19 pandemic was reported to have gripped 215 countries across the globe, and many of these countries were on lockdown (Dein et al., 2020). Education was among the first few sectors to face a rapid shutdown of all its activities. Thousands of schools and millions of students are affected by the lockdown due to the Covid- 19 pandemic. The first response from the education sector was to completely halt its operations (Rajhans et al., 2020). This directive was advised by the WHO to various heads of government all over the world. Many responded swiftly to this directive while others delayed the action. The consequence of such delays was first experienced across Europe and later America with limited casualties from Asia and Africa (UNESCO, 2020). The coronavirus pandemic triggered significant changes in the management of various sectors of nations. Chief of these sectors is the education community globally. They are expected to quickly adjust to homeschooling, online learning, and/or other versions of new learning modalities. Similarly, the basic, post-basis, and tertiary schools have had to result in remote/online learning in the middle of this pandemic. This transition to online learning has come with its share of challenges. The first assumption to successfully participate in these new online learning spaces is that families must have access to multiple internet-enabled devices. However, many families, especially those in rural areas lack access to mobile devices as well as reliable internet services (Akerlof & Kranton, 2002; Carrillo & Flores, 2020). INTRODUCTION Consequently, teaching and learning activities were instantly suspended during the early period of the lockdown. Beyond the early stage, the need to rapidly adapt to new contexts of teaching and learning to cope with the pandemic exposed teachers, educators, and learners alike to unusual teaching mediums, especially for those in rural communities (Butler-Henderson et al., 2020). The experiences and challenges present rare opportunities to explore new ways to cope with such unexpected circumstances. In this period, it is important to look at how teachers have adapted to the restriction in interaction and moved to new ways of teaching and learning in preparation for future learners in a world marked with uncertainty (Bozkurt et al., 2020; Johns Hopkins University & Medicine, 2020). The expert awareness of Covid-19 as a pandemic in many African countries was through the Director-General of the World Health Organization (WHO). The emergence of Covid-19 at the start of the year 2020 met the world unprepared. Scientists alike were left dazed at the rate of spread with little or no answer to how best to prevent, manage and treat the new pandemic. Starting from Wuhan (a city in China), the virus traveled all over the world and changed the way humans live, interact, work, teach and learn (Mellish et al., 2020). While staying at home have calculated economic implications in term of projections, profit and loss. The implication of the pandemic on education is yet to be accurately described, but it will surely be more challenging for educators and learners in more fragile and unstable contexts. Most countries of the world experienced total or partial lockdowns which led to the immediate closure of universities and other schools (Flores & Swennen, 2020). Copyright © 2022, JPMS, p-ISSN: 1410-1866, e-ISSN: 2549-1458 Jurnal Pendidikan Matematika dan Sains, 10 (2), 2022, 94 Olalekan Taofeek Badmus, Afees Akanni Amuda, Abdulrasaq Oladimeji Akanbi, Esther Ore Omosewo Jurnal Pendidikan Matematika dan Sains, 10 (2), 2022, 94 ek Badmus, Afees Akanni Amuda, Abdulrasaq Oladimeji Akanbi, Esther Ore Omosewo Effects of Covid-19 Pandemic on Science Teacher Education The covid-19 pandemic has affected educational systems worldwide, leading to the near-total closures of schools, universities, and colleges. Most governments around the world have temporarily closed educational institutions in an attempt to reduce the spread of Covid-19. As of 30 September 2020, approximately 1.077 billion learners were currently affected due to school closures in response to the pandemic. According to the UNICEF monitoring team, 53 countries implemented nationwide closures and 27 implemented local closures, impacting about 61.6 percent of the world's student population. School closures impact not only students, teachers, and families, but have far-reaching economic and societal consequences. School closures in response to the pandemic have shed light on various social and economic issues, including student debt, digital learning, food insecurity, and homelessness as well as access to childcare, health care, housing, internet, and disability services. A sizable number of studies concluded that online classes can be as effective as physical classrooms with the studies mostly done in tertiary institutions (Lin & Gao, 2020). Other scholars have speculated on the alternative. In situations like this, online classes are preferred for continuous education (Woolley et al., 2020). With the relaxation of various aspects of life at the tail end of the year 2020. Academic activities in Nigeria and other African schools were rapidly halted from March up until August 2020 when exit classes were allowed to sit for general examinations. With the current trends of events, a science teacher in Nigeria and Africa must strive to provide quality education, and ensure uniformity, equity, and universal accessibility to all (Ikoni & Ogundele, 2020). The impact was more severe for disadvantaged children and their families, causing interrupted learning, compromised nutrition, childcare problems, and consequent economic cost to families who could not work (UNESCO, 2020). In response to school closures, UNESCO recommended the use of distance learning programs and open educational applications and platforms that schools and teachers can use to reach learners remotely and limit the disruption of academic activities. Efforts to slow the spread of COVID- 19 through non-pharmaceutical interventions and preventive measures, such as social distancing and self-isolation have prompted widespread closure of primary, secondary, and tertiary institutions schooling across Africa (UNESCO, 2020). The African Account of Covid-19 While we struggle to make sense of our lives during the global pandemic, questions on whether the health and well-being of citizens should be prioritized over reviving the economy have ensued. Caught in the middle are businesses, industries, and educational institutions whose business models depend on and thrive on interactions among people, places, and experiences in various spaces (Verma et al., 2020). Prominent among dependent spaces are our children, teenagers, and students who have experienced an unrivaled disruption in their academic experiences (Arendt et al., 2019). Copyright © 2022, JPMS, p-ISSN: 1410-1866, e-ISSN: 2549-1458 Jurnal Pendidikan Matematika dan Sains, 10 (2), 2022, 95 Olalekan Taofeek Badmus, Afees Akanni Amuda, Abdulrasaq Oladimeji Akanbi, Esther Ore Omosewo Jurnal Pendidikan Matematika dan Sains, 10 (2), 2022, 95 Olalekan Taofeek Badmus, Afees Akanni Amuda, Abdulrasaq Oladimeji Akanbi, Esther Ore Omosewo Jurnal Pendidikan Matematika dan Sains, 10 (2), 2022, 95 T f k B d Af Ak i A d Abd l Ol di ji Ak bi E h O O Jurnal Pendidikan Matematika dan Sains, 10 (2), 2022, 95 Olalekan Taofeek Badmus, Afees Akanni Amuda, Abdulrasaq Oladimeji Akanbi, Esther Ore Omosewo The absence of a reliable treatment and/or vaccine for Covid-19 around the globe is a new normal to the way Africans live, work, study, travel, gather, and how access day-to-day services (Jandrić et al., 2022). For science educators, the need to acquire the requisite skills to traverse this unfortunate occurrence is most desired. Although, the economic status of science educators in developing countries remains poor (Quezada et al., 2020). However, more is desired from science educators and teachers alike despite their limited resources. Science educators during this period are expected to introduce learning approaches that are alien to their usual practice (Onishchuk et al., 2020). Among these are distance learning, blended learning. Several empirical studies have posited differently on these approaches. need for imminent change in our practice of teaching. An appraisal of how and why the quick adaptation of new strategies is required, along with the challenges facing science teacher education during this Covid-19 transition period (Santika et al., 2022). Effects of Covid-19 Pandemic on Science Teacher Education Since face-to-face interaction between teachers and students is minimized owing to the rapid transmission of Covid-19, innovative use and promotion of technology in ushering educational reforms to create a vibrant knowledge society are desired (Muralidharan et al., 2019). There is an urgent need for the transformation of science education, from a traditional/conventional system to an e-learning environment, imparting updated competencies to our science teachers. Some universities across Africa have responded quickly to this crisis, considering guidelines issued by various Governments. Holding online conferences, classes, and webinars and the usage of other online media have alleviated the situation to a reasonable extent. Although, these opportunities are not equally available at various locations owing to remote/under-developed communities across Africa (Gupta et al., 2021). There is a The rationale for Rethinking Science Teacher Education amidst COVID-19 Pandemic Copyright © 2022, JPMS, p-ISSN: 1410-1866, e-ISSN: 2549-1458 Jurnal Pendidikan Matematika dan Sains, 10 (2), 2022, 96 ekan Taofeek Badmus, Afees Akanni Amuda, Abdulrasaq Oladimeji Akanbi, Esther Ore Omosewo The Covid-19 pandemic may well change our world and our global outlook; it can also teach us about how education needs to change to be able to better prepare our young learners for what the future holds. So, as science educators grapple with new ways of communicating with our students away from our classrooms and lecture theatres, it is a good time to reflect on how this disruptive crisis can help us define what learning should look like for our students. The following ways have been identified as rethinking strategies that could be employed amidst the Covid-19 pandemic (Tsagakis & Papatriantafyllou, 2020). role of the science teacher in the classroom and lecture theatre. This may mean that the role of science teachers will need to move towards facilitating young people’s development as contributing members of society (Larimore, 2020). Teaching life skills needed for the future: In this ever-changing global world, young students require resilience and adaptability – skills that are proving to be essential to navigating effectively through this pandemic. Looking into the future, some of the most important skills that employers will be looking for will be creativity, communication, and collaboration, alongside empathy and emotional intelligence; and being able to work across demographic lines of differences to harness the power of collectiveness through effective teamwork (Edmondson et al., 2020). Usage of social media platforms (WhatsApp, Telegram, Twitter, video conferencing, dedicated educational portals, virtual classroom, or blended learning): These are promising applications or avenues to foster a self-learning attitude in science education students. An increasing number of institutions are encouraging the use of technology in day-to- day teaching as students find it enjoyable. Covid-19 has resulted in educational institutions across the world being compelled to suddenly harness and utilize the suite of available technological tools to create content for remote learning for students. Science educators across the world are experiencing new possibilities to do things differently and with greater flexibility resulting in potential benefits (Ferri et al., 2020). to With students being able to gain access to knowledge and even learn a technical skill, through a few clicks on their phones, tablets, and computers, there is a need to redefine the The Way Forward Review of our Curriculum to Accommodate Changes Resulting from the The Way Forward Review of our Curriculum to Accommodate Changes Resulting from the Copyright © 2022, JPMS, p-ISSN: 1410-1866, e-ISSN: 2549-1458 Jurnal Pendidikan Matematika dan Sains, 10 (2), 2022, 97 Olalekan Taofeek Badmus, Afees Akanni Amuda, Abdulrasaq Oladimeji Akanbi, Esther Ore Omosewo other hand, failure to meet the needs of this period may leave us with yet another regret on the list of things we should have done as a people but refused. Covid-19 pandemic: The trend in pedagogy for theoretical content should be shifted from monotonous didactic lectures to interactive online lectures using video conferencing tools (e.g. Google meet, Microsoft teams, Zoom, etc.); reading and writing assignments should be done using various teaching-learning apps such as Google Classroom, Microsoft Team, etc. Re- orientation of our Students’ Information Seeking Behavior: Students should be encouraged to become active learners by contributing to the inputs in the e-learning environment. REFERENCES Akerlof, G.A. & Kranton, R.E. (2002). Identity and schooling: Some lessons for the economics of education. Journal of Economic Literature, 40(4), 1167-1201. Aknin, L. B., De Neve, J. E., Dunn, E. W., Fancourt, D. E., Goldberg, E., Helliwell, J. F., & Ben Amor, Y. (2022). Mental health during the first year of the COVID- 19 pandemic: A review and recommendations for moving forward. Perspectives on Psychological Science, 17(4), 915-936. Training and Retraining of Science Teachers Toward the use of Technological Tools: Science teachers should be formally trained for efficient usage of various tools and online pedagogy why through training or teachers’ development programs. Provision of highly subsidize computers and other gadgets to the students to give way to easy delivery of classes online or offline. Development of application packages that can be used by students offline to facilitate their learning. Science educators should use a wide variety of online tools and apps to keep going with their teaching. Arendt, F., Scherr, S., & Romer, D. (2019). Effects of exposure to self-harm on social media: Evidence from a two-wave panel study among young adults. New Media & Society, 21(11-12), 2422-2442. Bozkurt, A., Jung, I., Xiao, J., Vladimirschi, V., Schuwer, R., Egorov, G., & Paskevicius, M. (2020). A global outlook to the interruption of education due to the Covid-19 pandemic: Navigating in a time of uncertainty and crisis. Asian Journal of Distance Education, 15(1), 1-126. to g Resources Availability: The majority of Nigerian schools (both Public and Private) are under-resourced. They are ill-equipped to respond to the teaching and learning challenges of the 21st century – let alone the latest demands of the Covid-19 pandemic. Availability of Network: The current lockdown has suddenly compelled teachers to adopt predominantly online, blended learning teaching practices. But most households in Nigeria are still without access to the internet. Very few schools had adapted to blended learning before the lockdown and few schools would be able to adopt it during the lockdown. School Location: Schools located in rural areas and villages may not be able to adopt e-learning as most of them do not have access to either network. The use of media houses such as Television and radio stations amongst others: Many countries introduced educational packages that were presented through media houses before the lockdown was lifted. Such media houses should continue because students can benefit from such presentations (Sahlberg, 2021). Educating Citizens in an Interconnected World: Covid-19 is a pandemic that illustrates how globally interconnected we are – there is no longer such a thing as isolated issues and actions. Successful people in the coming decades need to be able to understand this interrelatedness and navigate across boundaries to leverage their differences and work in a globally collaborative way. Redefining the Role of Science Instructors: The notion of a teacher as the knowledge-holder who imparts wisdom to their pupils is no longer fit for 21st-century instruction. High Cost of Mobile Devices: There are challenges in having appropriate high-tech gadgets, computers, or laptops for every student. More often these electronic gadgets with students are shared between siblings or among two or more students. This does not favor e- learning among students. Technological Know- How: Most science teachers do not have the requisite knowledge to deliver this online teaching or lack knowledge of how to organize the virtual classroom. This will certainly affect how they discharge their duty during this crisis period. Educational Technology Systems, 49(1), 5-22. courses. Asian Journal of Distance Education, 15(1), 169-179. courses. Asian Journal of Distance Education, 15(1), 169-179. Edmondson, J., Formica, P., & Mitra, J. (2020). Empathy, sensibility, and graduate employment–Can the humanities help? Industry and Higher Education, 34(4), 223-229. Mbombo, J. M. K. (2022). Peace in the face of the Covid‐19 pandemic: Making sense of the paralysis at the UN Security Council. Peace & Change, 47(1), 11-21. Mellish, T. I., Luzmore, N. J., & Shahbaz, A. A. (2020). Why were the UK and USA unprepared for the COVID-19 pandemic? The systemic weaknesses of neoliberalism: A comparison between the UK, USA, Germany, and South Korea. Journal of Global Faultlines, 7(1), 9-45. Ferri, F., Grifoni, P., & Guzzo, T. (2020). Online learning and emergency remote teaching: Opportunities and challenges in emergencies. Societies, 10(4), 86-95. Fishbane, L., & Tomer, A. (2020). As classes move online during COVID-19, what are disconnected students to do? Brookings. Retrieved from https://www.brookings.edu/blog/the- avenue /2020/03/20/as-classes-move- online-during-covid-19-what-are- disconnected-students-to-do/ Muralidharan, K., Singh, A., & Ganimian, A. J. (2019). Disrupting education? Experimental evidence on technology- aided instruction in India. American Economic Review, 109(4), 1426-60. Flores, M. A., & Swennen, A. (2020). The COVID-19 pandemic and its effects on teacher education. European Journal of Teacher Education, 43(4), 453-456. Onishchuk, I., Ikonnikova, M., Antonenko, T., Kharchenko, I., Shestakova, S., Kuzmenko, N., & Maksymchuk, B. (2020). Characteristics of foreign language education in foreign countries and ways of applying foreign experience in pedagogical universities of Ukraine. Revista Romaneasca Pentru Educatie Multidimensionala, 12(3), 44- 65. Gupta, S., Gupta, N., Yadav, P., & Patil, D. (2021). Ebola virus outbreak preparedness plan for developing Nations: Lessons learned from affected countries. Journal of Infection and Public Health, 14(3), 293-305. Quezada, R. L., Talbot, C., & Quezada-Parker, K. B. (2020). From bricks and mortar to remote teaching: A teacher education program‘s response to COVID- 19. Journal of Education for Teaching, 46(4), 472-483. Ikoni, O., & Ogundele, M. O. (2020). The disruption of the novel coronavirus (COVID-19) pandemic and the quality of higher education in Nigeria. KIU Journal of Humanities, 5(2), 25-36. Jandrić, P., Martinez, A. F., Reitz, C., Jackson, L., Grauslund, D., Hayes, D., & Hayes, S. (2022). Teaching in the age of Covid- 19—The new normal. Postdigital Science and Education, 4(3), 877-1015. Rajhans, V., Memon, U., Patil, V., & Goyal, A. (2020). Impact of Covid-19 on academic activities and way forward in Indian Optometry. Journal of Optometry, 13(4), 216-226. CONCLUSION The Covid-19 pandemic situation is a reality check for the African government and stakeholders in education. The need to look within and bring about a meaningful, impactful, and purposeful direction to alleviate the needs of teachers permanently should be paramount. This will go a long way in solving the challenges associated with teachers' standard of living, teaching standards, teaching environment, and teaching resources. These are extraordinary times, with special needs to cope with the present-day rigor. The pandemic is proving to be a creative disruption with tough challenges for all educational systems allowing restructuring of the present conventional classroom-based educational system. Butler-Henderson, K., Crawford, J., Rudolph, J., Lalani, K., & Sabu, K. M. (2020). Covid- 19 in higher education literature database (CHELD V1): An open access systematic literature review database with coding rules. Journal of Applied Learning & Teaching, 3(2), 1-6. Carrillo, C., & Flores, M. A. (2020). Covid-19 and teacher education: a literature review of online teaching and learning practices. European Journal of Teacher Education, 43(4), 466-487. At the same time, there is a strong opportunity for us to adopt newer techniques that are more suitable for the present generation of learners. The rapid transition to online education will not only benefits science educators and students but will also create a momentum of continued education for practicing science educator in the continent. We will do well, as scholars if this situation is properly managed and come out better, on the Dein, S., Loewenthal, K., Lewis, C. A., & Pargament, K. I. (2020). Covid-19, mental health, and religion: An agenda for future research. Mental Health, Religion & Culture, 23(1), 1-9. Dhawan, S. (2020). Online learning: A panacea in the time of Covid-19 crisis. Journal of Copyright © 2022, JPMS, p-ISSN: 1410-1866, e-ISSN: 2549-1458 Jurnal Pendidikan Matematika dan Sains, 10 (2), 2022, 98 ek Badmus, Afees Akanni Amuda, Abdulrasaq Oladimeji Akanbi, Esther Ore Omosewo Educational Technology Systems, 49(1), 5-22. Johns Hopkins University & Medicine. (2020). COVID-19 case tracker: Follow global cases and trends. Retrieved from https://coronavirus.jhu.edu Sahlberg, P. (2021). Does the pandemic help us make education more equitable? Educational Research for Policy and Practice, 20(1), 11-18. Larimore, R. A. (2020). Preschool science education: A vision for the future. Early Childhood Education Journal, 48(6), 703- 714. Santika, V., Indriayu, M., & Sangka, K. B. (2022). Policy in Covid-19 pandemic to increase ICT integration: What is the role of economics teacher' TPACK? Journal of Positive School Psychology, 6(8), 8503-8517. Lin, X., & Gao, L. (2020). Students’ sense of community and perspectives of taking synchronous and asynchronous online Copyright © 2022, JPMS, p-ISSN: 1410-1866, e-ISSN: 2549-1458 Jurnal Pendidikan Matematika dan Sains, 10 (2), 2022, 99 Olalekan Taofeek Badmus, Afees Akanni Amuda, Abdulrasaq Oladimeji Akanbi, Esther Ore Omosewo Jurnal Pendidikan Matematika dan Sains, 10 (2), 2022, 99 ek Badmus, Afees Akanni Amuda, Abdulrasaq Oladimeji Akanbi, Esther Ore Omosewo Stocklin-Weinberg, R., Veiga, M. M., & Marshall, B. G. (2019). Training artisanal miners: A proposed framework with performance evaluation indicators. Science of The Total Environment, 660(1), 1533-1541. UNESCO. (2020). Covid-19 educational disruption and response. Retrieved from http://en.unesco.org/covid19/educationres ponse on 2020-10-28 Verma, G., Campbell, T., Melville, W., & Park, B. Y. (2020). Science teacher education in the times of the Covid-19 pandemic. Journal of Science Teacher Education, 31(5), 483-490. Strauss, V. (2020). Why covid-19 will ‘explode’ existing academic achievement gaps. The Washington Post. Retrieved from https://www.washingtonpost.com/educati on/2020/04/17/whycovid-19-will- explode-existing-academic-achievement- gaps/ Woolley, S., Sattiraju, N., & Moritz, S. (2020). U.S. schools trying to teach online highlight a digital divide: The children who lack internet and computers now also lack access to education. Bloomberg. Retrieved from https://www.bloomberg.com/news/articles /2020-03-26/covid-19-school-closures- reveal-disparity-in-access-to-internet Tsagakis, I., & Papatriantafyllou, M. (2020). Safeguarding cancer research funding by European charities amidst the COVID‐19 pandemic. Molecular Oncology, 14(12), 2987-2993. Copyright © 2022, JPMS, p-ISSN: 1410-1866, e-ISSN: 2549-1458
https://openalex.org/W2580364606	https://dash.harvard.edu/bitstream/1/41483215/1/91460%201548.full.pdf	English	null	Nucleotide-dependent switch in proteasome assembly mediated by the Nas6 chaperone	Proceedings of the National Academy of Sciences of the United States of America	2,017	cc-by	9,613	Terms of Use This article was downloaded from Harvard University’s DASH repository, and is made available under the terms and conditions applicable to Other Posted Material, as set forth at http:// nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of-use#LAA Permanent link http://nrs.harvard.edu/urn-3:HUL.InstRepos:41483215 Citation Li, Frances, Geng Tian, Deanna Langager, Vladyslava Sokolova, Daniel Finley, and Soyeon Park. 2017. “Nucleotide-Dependent Switch in Proteasome Assembly Mediated by the Nas6 Chaperone.” Proceedings of the National Academy of Sciences 114 (7): 1548–53. https:// doi.org/10.1073/pnas.1612922114. Share Your Story The Harvard community has made this article openly available. Please share how this access benefits you. Submit a story . Accessibility Nucleotide-dependent switch in proteasome assembly mediated by the Nas6 chaperone Frances Lia, Geng Tianb, Deanna Langagera, Vladyslava Sokolovaa, Daniel Finleyb, and Soyeon Parka,1 aDepartment of Molecular, Cellular, and Developmental Biology, University of Colorado Boulder, Boulder, CO 80309; and bDepartment of Cell Biology, Harvard Medical School, Boston, MA 02115 er, Max Planck Institute of Chemistry, Martinsried, Germany, and approved December 15, 2016 (received for review Augu Edited by Wolfgang Baumeister, Max Planck Institute of Chemistry, Martinsried, Germany, and approved December 15 4, 2016) is unhydrolyzed. A hybrid state sharing features of both ATPh and ATPγS proteasomes has also been resolved (22). ATPγS protea- somes presumably represent a physiological state because they closely resemble proteasomes trapped with substrate and ATP (20, 21). The C-domain chaperones obstruct base–CP association when base is in the ATPh state, but not the ATPγS state (17). Through this mechanism, the C-domain chaperones can prevent premature base–CP complex formation. The proteasome is assembled via the nine-subunit lid, nine-subunit base, and 28-subunit core particle (CP). Previous work has shown that the chaperones Rpn14, Nas6, Hsm3, and Nas2 each bind a specific ATPase subunit of the base and antagonize base–CP in- teraction. Here, we show that the Nas6 chaperone also obstructs base–lid association. Nas6 alternates between these two inhibi- tory modes according to the nucleotide state of the base. When ATP cannot be hydrolyzed, Nas6 interferes with base–lid, but not base–CP, association. In contrast, under conditions of ATP hydro- lysis, Nas6 obstructs base–CP, but not base–lid, association. Mod- eling of Nas6 into cryoelectron microscopy structures of the proteasome suggests that Nas6 controls both base–lid affinity and base–CP affinity through steric hindrance; Nas6 clashes with the lid in the ATP-hydrolysis–blocked proteasome, but clashes in- stead with the CP in the ATP-hydrolysis–competent proteasome. Thus, Nas6 provides a dual mechanism to control assembly at both major interfaces of the proteasome. Contrary to the extensive knowledge of base–CP complex formation (7, 12, 13, 17, 23–27), mechanistic understanding of base–lid complex formation is lacking. The nine-subunit lid is thought to self-assemble without dedicated chaperone proteins (28–31), and then joins with the base to form the RP. Based on biochemical and proteomic identification of assembly interme- diates, the base and lid appear to be assembled via distinct precursor complexes, which do not overlap in their subunit composition (9–14, 16, 28, 29). These data raise a question as to whether premature interactions between base and lid complexes are prevented until they are mature. Importantly, the C-domain chaperones have been considered to be critical for the RP as- sembly process (9–14), although the role of these chaperones in base–lid joining has not been examined. proteasome \| chaperone \| AAA+ ATPase \| assembly \| Nas6 T he proteasome, a central component of the ubiquitin-dependent protein degradation system, is the most complex protease known (1–4). In the proteasome, the regulatory particle (RP) rec- ognizes ubiquitinated proteins and translocates them in an ATP- dependent process into the core particle (CP) to be degraded. The barrel-shaped CP is a stack of four heteroheptameric rings com- prising seven α-type and β-type subunits (α1–7β1–7β1–7α1–7). The in- ner β-rings contain proteolytic sites facing the interior space of the CP. The outer α-rings form a gated entry pore into the CP, thus controlling substrate access to the proteolytic sites (5). The α-ring also provides radially arrayed docking sites for the RP through seven “α-pockets” formed between subunits. The six ATPases of the RP (Rpt1–Rpt6) insert their C-terminal tails into individual α-pockets, thus joining RP and CP as well as opening the axial gate of the CP for substrate entry (6–8). T Nucleotide-dependent switch in proteasome assembly mediated by the Nas6 chaperone Frances Lia, Geng Tianb, Deanna Langagera, Vladyslava Sokolovaa, Daniel Finleyb, and Soyeon Parka,1 aDepartment of Molecular, Cellular, and Developmental Biology, University of Colorado Boulder, Boulder, CO 80309; and bDepartment of Cell Biology, Harvard Medical School, Boston, MA 02115 In the present study, we investigated whether the C-domain chaperones may regulate base–lid complex formation. Our data demonstrate that the Nas6 chaperone antagonizes both lid–base joining and base–CP join- ing. Importantly, the specificity of Nas6 activity is determined by the nucleotide state of the base. When ATP hydrolysis is blocked, Nas6 interferes with base–lid association and does not obstruct base–CP association. Conversely, under the conditions proteasome \| chaperone \| AAA+ ATPase \| assembly \| Nas6 1548–1553 \| PNAS \| February 14, 2017 \| vol. 114 \| no. 7 Author contributions: S.P. designed research; F.L., D.L., V.S., and S.P. performed research; G.T. contributed new reagents/analytic tools; D.F. provided helpful comments on the paper; and S.P. wrote the paper. Significance The proteasome is a molecular machine essential for protein degradation. The proteasome holoenzyme assembles from the association of three subcomplexes: lid, base, and core particle (CP). The base binding chaperones, Nas6, Rpn14, Hsm3, and Nas2, antagonize base–CP interaction. Here, we demonstrate that the Nas6 chaperone also obstructs base–lid association. The specificity of Nas6 activity is determined by the nucleotide state of the base. When ATP hydrolysis is blocked, Nas6 inter- feres with base–lid association. In contrast, under conditions of ATP hydrolysis, Nas6 interferes with base–CP association. Nas6 uses the mechanism of steric hindrance at both major inter- faces of the proteasome: base–lid and base–CP. We demon- strate that Nas6 provides a dual mechanism to control proteasome assembly. ( ) Five chaperones have been found to promote formation of the Rpt ring during RP assembly (9–15). The chaperones bind spe- cific Rpt proteins in a pairwise manner: Rpn14-Rpt6, Adc17- Rpt6, Nas6-Rpt3, Hsm3-Rpt1, and Nas2-Rpt5 (10, 11, 13–15). Distinct chaperone modules have been identified, with each containing an Rpt dimer: Rpn14-Rpt6-Rpt3-Nas6, Hsm3-Rpt1- Rpt2-Rpn1, and Nas2-Rpt5-Rpt4 (9, 11, 14, 16). These modules join to complete the heterohexameric Rpt ring of the base, and the chaperones are expelled as the proteasome holoenzyme (RP- CP) matures (12, 13, 17). Although phylogenetically distinct, the chaperones, with the exception of Adc17, bind the C domain of cognate Rpt proteins and occlude the proximal C-terminal tails from docking into the CP (12, 13, 17–19); we will refer to them as C-domain chaperones henceforth. The C-domain chaperones can thus prevent free base and free RP from associating with CP. Recent cryoelectron microscopy (cryo-EM) studies have de- fined two major states of the proteasome: ATP-hydrolyzing (ATPh) and ATP-bound (ATPγS) (20–22). Proteasomes in the basal ATPh state are active in ATP hydrolysis, whereas in ATPγS proteasomes, representing a high-energy state, all bound nucleotide Author contributions: S.P. designed research; F.L., D.L., V.S., and S.P. performed research; G.T. contributed new reagents/analytic tools; D.F. provided helpful comments on the paper; and S.P. wrote the paper. Recent cryoelectron microscopy (cryo-EM) studies have de- fined two major states of the proteasome: ATP-hydrolyzing (ATPh) and ATP-bound (ATPγS) (20–22). Proteasomes in the basal ATPh state are active in ATP hydrolysis, whereas in ATPγS proteasomes, representing a high-energy state, all bound nucleotide The authors declare no conflict of interest. This article is a PNAS Direct Submission. 1To whom correspondence should be addressed. Email: soyeon.park-1@colorado.edu. Author contributions: S.P. designed research; F.L., D.L., V.S., and S.P. performed research; G.T. contributed new reagents/analytic tools; D.F. provided helpful comments on the paper; and S.P. wrote the paper. The authors declare no conflict of interest. This article is a PNAS Direct Submission. 1To whom correspondence should be addressed. Email: soyeon.park-1@colorado.edu. This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. 1073/pnas.1612922114/-/DCSupplemental. The authors declare no conflict of interest. Results The Target of Nas6 Steric Hindrance Is Governed by Nucleotide. To understand the molecular basis of the apparent obstruction of the lid by Nas6, and its nucleotide dependence, we superimposed the previously reported crystal structure of the known Rpt3– Nas6 co-complex (33) onto the cryo-EM structure of the pro- teasome (22). Both ATPh and ATPγS states of the proteasome were examined. In the ATPh model, no clash was seen between Rpt3–Nas6 co-complex and lid subunit Rpn6 (Fig. 2A). In con- trast, Nas6 clashed dramatically with Rpn6 in the ATPγS model (Fig. 2B). The overlap includes the N-terminal α-solenoid of Rpn6 (Fig. 2B). Interestingly, Rpn6 has been postulated to serve as a molecular clamp at the base–CP interface (1, 3, 32), and is also thought to be a pivotal regulator of proteasome levels (34). Our modeling indicates that the positioning of both Nas6 and Rpn6 undergoes dramatic rearrangement in the transition be- tween the ATPh and ATPγS states of the proteasome because their mutual interaction partner, Rpt3, is at a critical location for conformational changes in the Rpt ring (20–22) (Fig. S3). In addition, the lid exhibits a global conformational change (3, 31) upon incorporation into the proteasome, including the reposi- tioning of Rpn6. Consistent with our previous work (13, 17), the clash between Nas6 and the CP α2 subunit was prominent in the ATPh state, but absent in the ATPγS state (Fig. 2 C and D). In summary, modeling the known Nas6 structure into proteasome structures is in close agreement with our experimental data that Nas6 acts as a negative regulator during proteasome assembly (12, 13, 17) and suggests, more specifically, that a clash between Nas6 and the lid subunit Rpn6 is responsible for inhibition of lid joining to the base in the ATPγS state (Figs. 1 and 2B). Nas6 Interferes with Base–Lid Complex Formation When ATP Hydrolysis Is Blocked. In the proteasome holoenzyme, the lid binds laterally to the base and projects toward the CP, making close contacts with Rpt3, Rpt6, α1, and α2 (1, 3, 32). Specifically, we noted that the lid contact sites of Rpt3 within the base appear to overlap with the binding sites of its cognate chaperone, Nas6 (33). Therefore, we tested whether Nas6 sterically interferes with incoming lid to the base, using the lid–base association assay in vitro. Results When purified lid was added to base under physiological ATP concentrations, it readily associated with base (Fig. 1A, lane 2). The Nas6 chaperone did not interfere with this process (Fig. 1A, lane 3). Whether in the ATPh or ATPγS state, the base was competent to serve as a lid acceptor (Fig. 1A, lanes 2 and 4). With the addition of Nas6, however, the lid–base interaction was specifically attenuated when the base was in the ATPγS state (Fig. 1A, lane 5). Nas6 associated with the base to a comparable extent in the two nucleotide conditions (Fig. 1A, lanes 3 and 5), consistent with previous findings (17). With wild-type base, copurifying endogenous Nas6 was sufficient to antagonize lid binding upon incubation with ATPγS (Fig. S1). We then exam- ined whether Nas6 also induces dissociation of purified RP into lid and base in the ATPγS state. We immobilized the RP via a lid subunit and added Nas6 in the presence of ATP and ATPγS (Fig. 1B). Nas6 resulted in a loss of base subunits in the ATPγS state (Fig. 1B, lanes 5 and 6), whereas lid subunit levels remained comparable in both nucleotide conditions, serving as a control that a comparable amount of resin-bound material was used. These data demonstrate that Nas6 also induces dissociation of j g γ ( g ) Given the Nas6-mediated steric occlusion in both the ATPh and ATPγS proteasome states (Fig. 2 B and C), we investigated how Nas6 can promote proteasome formation rather than constitu- tively inhibit it. We hypothesized that these inhibitory interactions might be self-limiting if Nas6-mediated interference could be re- lieved by the eviction of Nas6 from the base as the lid and CP each bind. Therefore, we loaded the Nas6 chaperone on immobilized base and monitored Nas6 release as the base associates with lid or CP (Fig. 2E). With ongoing ATP hydrolysis, Nas6 was released as the base–CP complex forms (Fig. 2E, compare lanes 2 and 4). The lid further joined to form the holoenzyme (Fig. 2E, lane 5). When ATP hydrolysis was inhibited, Nas6 was partially displaced from the base upon lid addition (Fig. 2F, lane 3). However, stable base– lid complexes failed to form, indicating robust Nas6 interference with the formation of free base–lid complexes, consistent with the results of Fig. 1A (lane 5). Interestingly, Nas6 interference with base–lid association is neutralized as the CP joins to form the proteasome (Fig. Results 2F, lane 5, lid–base–CP complex), even though Nas6 remains present in the complex. Thus, when both base–lid and base–CP interfaces form, the clash between Nas6 and Rpn6 at the base–lid interface (Fig. 2B) can apparently be relieved by the base–CP interface specifically in the ATPγS state (Fig. 2D). Nas6 interference in base–lid association is therefore highly selective for assembly intermediates as opposed to the holoenzyme. d h l f i i i i Fig. 1. Nas6 antagonizes base–lid association when nucleotide hydrolysis is blocked. (A) Base–lid association assay showing Nas6 interference against the lid in the ATPγS state. Base lacking Nas6 was affinity-purified from nas6Δ strain via a chromosomally integrated protein A-tobacco etch virus (TEV) af- finity tag appended to the N terminus of RPT1, and retained on IgG resin (17). Immobilized nas6Δ base (∼0.8 pmol) was incubated with lid (2 pmol) with or without recombinant Nas6 (18 pmol) in the presence of ATP or ATPγS (1 mM). The base-bound materials were washed with buffer containing 150 mM NaCl, and eluted with TEV protease. The eluates were analyzed by 12% (vol/vol) SDS/PAGE and immunoblotting for lid (Rpn8 and Rpn12) and base (Rpt3 and Rpt6) subunits. (B) Nas6-mediated dissociation of the base from the RP in the ATPγS state. Immobilized nas6Δ RP (∼1 pmol) was incubated in the presence of indicated nucleotides (1 mM) with or without Nas6 (++, 36 pmol; +, 18 pmol). Following washing and elution as in A, the eluates were analyzed for base (Rpt1 and Rpt5) and lid (Rpn8 and Rpn12) subunits. Note that the RP is immobilized via the lid subunit Rpn11. Nas6-mediated inhibition of the interaction between the base and lid results in loss of the base. Fig. 1. Nas6 antagonizes base–lid association when nucleotide hydrolysis is blocked. (A) Base–lid association assay showing Nas6 interference against the lid in the ATPγS state. Base lacking Nas6 was affinity-purified from nas6Δ strain via a chromosomally integrated protein A-tobacco etch virus (TEV) af- finity tag appended to the N terminus of RPT1, and retained on IgG resin (17). Immobilized nas6Δ base (∼0.8 pmol) was incubated with lid (2 pmol) with or without recombinant Nas6 (18 pmol) in the presence of ATP or ATPγS (1 mM). The base-bound materials were washed with buffer containing 150 mM NaCl, and eluted with TEV protease. Significance This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. 1073/pnas.1612922114/-/DCSupplemental. This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. 1073/pnas.1612922114/-/DCSupplemental. 1548–1553 \| PNAS \| February 14, 2017 \| vol. 114 \| no. 7 www.pnas.org/cgi/doi/10.1073/pnas.1612922114 of ATP hydrolysis, Nas6 interferes with base–CP association but does not obstruct base–lid association. Our data demonstrate that Nas6 uses the same mechanism of steric hindrance at both major interfaces of the proteasome: base–lid and base–CP in- terfaces. We propose that Nas6 provides a finely regulated mechanism to order the events of proteasome assembly in time. of ATP hydrolysis, Nas6 interferes with base–CP association but does not obstruct base–lid association. Our data demonstrate that Nas6 uses the same mechanism of steric hindrance at both major interfaces of the proteasome: base–lid and base–CP in- terfaces. We propose that Nas6 provides a finely regulated mechanism to order the events of proteasome assembly in time. the base from purified RP upon incubation with ATPγS. Nas6 in the eluates decreases in the ATPγS state because Nas6 dissoci- ates together with the base, and they are removed from the resin- bound lid during wash. Taken together, these results show that Nas6 interference with base–lid association is determined by the nucleotide state of the base. Other C-domain chaperones, such as Rpn14 and Hsm3, lack this activity (Fig. S2). PNAS \| February 14, 2017 \| vol. 114 \| no. 7 \| 1549 Results 2F, lane 5), Nas6 does not appear to antagonize the lid when the lid is in direct contact with the base and CP within the proteasome holoenzyme, because base-CP can accommodate Nas6 without steric clash in ATPγS (Fig. 2D). could reflect the existence of a class of proteasome that is in neither the ATPh nor ATPγS state. A hybrid form of the pro- teasome was recently described by cryo-EM in which the lid and base are found in the ATPγS and ATPh states, respectively (22). A proteasome in which, conversely, the base is in the ATPγS state and the lid is in the ATPh state would be predicted to be com- petent for Nas6 binding. Nas6 Clashes Against the Lid Subunit Rpn6. To understand the basis for the effects of Nas6 on the lid–base interface, we examined whether the steric hindrance of Nas6 against the lid subunit Rpn6 is due to their positioning depending on nucleotide- dependent contacts between Rpn6 and the ATPases in the base (Fig. 2B). We disabled Rpn6 contacts with ATPase subunits of the base specifically to relieve the Nas6 clash at the lid–base interface. Based on the crystal structure of Rpn6 (32), three substitutions were introduced into Rpn6 at its contact sites with Rpt3 to form the Rpn6-SYE allele (S237, Y238, E241), whereas two additional substitutions at the nearby Rpt6 contact sites were added as well to create the Rpn6-TTSYE allele (Fig. 3A; T199, T203, S237, Y238, E241). Rpn6 SYE residues prominently contact Rpt3 in the ATPγS-state proteasome, but not in the ATPh-state proteasome. In contrast, Rpn6 TT (T199, T203) residues contact Rpt6 in both the ATPγS and ATPh states (Fig. S4). Abrogating Rpn6 contacts with both Rpt3 and Rpt6 is predicted to relieve Nas6 interference in the lid–base interface and to result in a failure to evict Nas6 from the proteasome holoenzyme. Indeed, endogenous Nas6 was retained on the proteasome holoenzymes isolated from the rpn6-TTSYE cells in both the ATPh and ATPγS states (Fig. 3B, i, lanes 2 and 6); we Taken together, these results suggest that the two interfaces of the proteasome are linked in their formation. In particular, for- mation of the base–CP interface in the ATPγS state may promote base–lid formation by relieving Nas6-mediated inhibition against the lid, thus facilitating ordered addition of subcomplexes to form the nascent proteasome. Results The eluates were analyzed by 12% (vol/vol) SDS/PAGE and immunoblotting for lid (Rpn8 and Rpn12) and base (Rpt3 and Rpt6) subunits. (B) Nas6-mediated dissociation of the base from the RP in the ATPγS state. Immobilized nas6Δ RP (∼1 pmol) was incubated in the presence of indicated nucleotides (1 mM) with or without Nas6 (++, 36 pmol; +, 18 pmol). Following washing and elution as in A, the eluates were analyzed for base (Rpt1 and Rpt5) and lid (Rpn8 and Rpn12) subunits. Note that the RP is immobilized via the lid subunit Rpn11. Nas6-mediated inhibition of the interaction between the base and lid results in loss of the base. Based on the results from proteasome association assays in vitro (Fig. 2F), we examined the possibility that Nas6 can be present in the proteasome holoenzyme without destabilizing it in vivo. Note that in wild-type cells, Nas6 is substoichiometric to proteasome subunits (35) and associates preferentially with assembly inter- mediates versus the holoenzyme (9, 11, 13, 14). We overexpressed Nas6 and isolated endogenous proteasomes (Fig. 2 G and H), using PNAS \| February 14, 2017 \| vol. 114 \| no. 7 \| 1549 Li et al. Fig. 2. Nas6 clashes at the base–lid interface of ATPγS proteasomes and at the base–CP interface of ATPh proteasomes. Docking of Nas6 (cyan) at the base–lid interface of the proteasome holoenzyme structures in the ATPh state [A, Protein Data Bank (PDB) ID code 4CR2 (22)] and ATPγS state [B, PDB ID code 4CR4 (22)]. The Nas6-Rpt3 crystal structure [PDB ID code 2DZN (33)] was used to dock Nas6 onto the proteasome. PyMOL (39) was used to superimpose the Rpt3 C domain (green) from the Nas6–Rpt3 complex and its counterpart (red) from the proteasome. The Rpt ring (orange) of the base and Rpn6 (magenta) of the lid are shown. (C and D) Docking of Nas6 (cyan) at the base–CP interface of the proteasome holoenzyme structures as in A and B. The CP α2 subunit (orange), α-ring (yellow), and Rpt3 C domain, as in A and B, are shown. In the ATPγS state, the Nas6 C terminus is sterically occluded by Rpn6 aa 209–215 (helix 8–9), aa 244–252 (helix 10–11), and aa 280–291 (helix 12–14) (B), but is distant from α2 (D). In the ATPh state, Nas6 does not clash significantly with Rpn6 (A) but rather with α2 (C). (E and F) Proteasome assembly assays. Results Base-associated material was analyzed via 12% (vol/vol) SDS/PAGE and immunoblotting for lid (Rpn8, Rpn12) and base (Rpn1, Rpt3) subunits. (G) Proteasome holoenzymes were purified via Pre1-TEV-ProA from Nas6- overexpressing cells (Nas6 o.e.) in ATP and ATPγS (1 mM). Nas6 level was assessed via 12% (vol/vol) SDS/PAGE and immunoblotting. (H, Top) Purified proteasomes from G were subjected to 3.5% (vol/vol) native PAGE and activity assay using the fluorogenic peptide substrate LLVY-AMC. (H, Middle) SDS (0.02%) activates latent CP activity (5). (H, Bottom) Proteasome levels were assessed by immunoblotting. Rpn8 is a lid subunit, and RP2-CP and RP-CP are proteasome holoenzymes. Fig. 2. Nas6 clashes at the base–lid interface of ATPγS proteasomes and at the base–CP interface of ATPh proteasomes. Docking of Nas6 (cyan) at the base–lid interface of the proteasome holoenzyme structures in the ATPh state [A, Protein Data Bank (PDB) ID code 4CR2 (22)] and ATPγS state [B, PDB ID code 4CR4 (22)]. The Nas6-Rpt3 crystal structure [PDB ID code 2DZN (33)] was used to dock Nas6 onto the proteasome. PyMOL (39) was used to superimpose the Rpt3 C domain (green) from the Nas6–Rpt3 complex and its counterpart (red) from the proteasome. The Rpt ring (orange) of the base and Rpn6 (magenta) of the lid are shown. (C and D) Docking of Nas6 (cyan) at the base–CP interface of the proteasome holoenzyme structures as in A and B. The CP α2 subunit (orange), α-ring (yellow), and Rpt3 C domain, as in A and B, are shown. In the ATPγS state, the Nas6 C terminus is sterically occluded by Rpn6 aa 209–215 (helix 8–9), aa 244–252 (helix 10–11), and aa 280–291 (helix 12–14) (B), but is distant from α2 (D). In the ATPh state, Nas6 does not clash significantly with Rpn6 (A) but rather with α2 (C). (E and F) Proteasome assembly assays. Nas6 releases from ATPh proteasomes (E), but remains on ATPγS proteasomes (F). Immobilized nas6Δ base (∼0.8 pmol) was loaded with Nas6 (∼21 pmol) with ATP (E) or ATPγS (F), and incubated with CP (2 pmol), lid (3.3 pmol), or both. Base-associated material was analyzed via 12% (vol/vol) SDS/PAGE and immunoblotting for lid (Rpn8, Rpn12) and base (Rpn1, Rpt3) subunits. (G) Proteasome holoenzymes were purified via Pre1-TEV-ProA from Nas6- overexpressing cells (Nas6 o.e.) in ATP and ATPγS (1 mM). Nas6 level was assessed via 12% (vol/vol) SDS/PAGE and immunoblotting. Results (H, Top) Purified proteasomes from G were subjected to 3.5% (vol/vol) native PAGE and activity assay using the fluorogenic peptide substrate LLVY-AMC. (H, Middle) SDS (0.02%) activates latent CP activity (5). (H, Bottom) Proteasome levels were assessed by immunoblotting. Rpn8 is a lid subunit, and RP2-CP and RP-CP are proteasome holoenzymes. an affinity tag appended to a CP subunit, β4, to exclude purification of free RP that is in complex with chaperones. Nas6 was readily detectable on proteasomes purified from Nas6-overexpressing cells in both the ATPh and ATPγS states (Fig. 2G), indicating that Nas6 was retained on the proteasome holoenzyme. We then analyzed the overall assembly status of the proteasome holoenzyme via native PAGE and in-gel peptidase assays using the fluorogenic substrate LLVY-AMC (Leu-Leu-Val-Tyr-7-amino-4-methylcoumarin). Nas6 overexpression decreased the level of proteasome holoenzyme as isolated in the ATPh state (Fig. 2H; RP2-CP, ATP) with a con- current increase in free CP, indicating that Nas6 interferes with the base–CP interface as predicted (12, 13) (Fig. 2C). However, Nas6 had no effect on the stability of proteasomes in the ATPγS state (Fig. 2H; ATPγS). Nas6-containing proteasomes did not exhibit an accumulation of free lid or base-CP. In agreement with our in vitro data (Fig. 2F, lane 5), Nas6 does not appear to antagonize the lid when the lid is in direct contact with the base and CP within the proteasome holoenzyme, because base-CP can accommodate Nas6 without steric clash in ATPγS (Fig. 2D). an affinity tag appended to a CP subunit, β4, to exclude purification of free RP that is in complex with chaperones. Nas6 was readily detectable on proteasomes purified from Nas6-overexpressing cells in both the ATPh and ATPγS states (Fig. 2G), indicating that Nas6 was retained on the proteasome holoenzyme. We then analyzed the overall assembly status of the proteasome holoenzyme via native PAGE and in-gel peptidase assays using the fluorogenic substrate LLVY-AMC (Leu-Leu-Val-Tyr-7-amino-4-methylcoumarin). Nas6 overexpression decreased the level of proteasome holoenzyme as isolated in the ATPh state (Fig. 2H; RP2-CP, ATP) with a con- current increase in free CP, indicating that Nas6 interferes with the base–CP interface as predicted (12, 13) (Fig. 2C). However, Nas6 had no effect on the stability of proteasomes in the ATPγS state (Fig. 2H; ATPγS). Nas6-containing proteasomes did not exhibit an accumulation of free lid or base-CP. In agreement with our in vitro data (Fig. Results Nas6 releases from ATPh proteasomes (E), but remains on ATPγS proteasomes (F). Immobilized nas6Δ base (∼0.8 pmol) was loaded with Nas6 (∼21 pmol) with ATP (E) or ATPγS (F), and incubated with CP (2 pmol), lid (3.3 pmol), or both. Base-associated material was analyzed via 12% (vol/vol) SDS/PAGE and immunoblotting for lid (Rpn8, Rpn12) and base (Rpn1, Rpt3) subunits. (G) Proteasome holoenzymes were purified via Pre1-TEV-ProA from Nas6- overexpressing cells (Nas6 o.e.) in ATP and ATPγS (1 mM). Nas6 level was assessed via 12% (vol/vol) SDS/PAGE and immunoblotting. (H, Top) Purified proteasomes from G were subjected to 3.5% (vol/vol) native PAGE and activity assay using the fluorogenic peptide substrate LLVY-AMC. (H, Middle) SDS (0.02%) activates latent CP activity (5). (H, Bottom) Proteasome levels were assessed by immunoblotting. Rpn8 is a lid subunit, and RP2-CP and RP-CP are proteasome holoenzymes. Fig. 2. Nas6 clashes at the base–lid interface of ATPγS proteasomes and at the base–CP interface of ATPh proteasomes. Docking of Nas6 (cyan) at the base–lid interface of the proteasome holoenzyme structures in the ATPh state [A, Protein Data Bank (PDB) ID code 4CR2 (22)] and ATPγS state [B, PDB ID code 4CR4 (22)]. The Nas6-Rpt3 crystal structure [PDB ID code 2DZN (33)] was used to dock Nas6 onto the proteasome. PyMOL (39) was used to superimpose the Rpt3 C domain (green) from the Nas6–Rpt3 complex and its counterpart (red) from the proteasome. The Rpt ring (orange) of the base and Rpn6 (magenta) of the lid are shown. (C and D) Docking of Nas6 (cyan) at the base–CP interface of the proteasome holoenzyme structures as in A and B. The CP α2 subunit (orange), α-ring (yellow), and Rpt3 C domain, as in A and B, are shown. In the ATPγS state, the Nas6 C terminus is sterically occluded by Rpn6 aa 209–215 (helix 8–9), aa 244–252 (helix 10–11), and aa 280–291 (helix 12–14) (B), but is distant from α2 (D). In the ATPh state, Nas6 does not clash significantly with Rpn6 (A) but rather with α2 (C). (E and F) Proteasome assembly assays. Nas6 releases from ATPh proteasomes (E), but remains on ATPγS proteasomes (F). Immobilized nas6Δ base (∼0.8 pmol) was loaded with Nas6 (∼21 pmol) with ATP (E) or ATPγS (F), and incubated with CP (2 pmol), lid (3.3 pmol), or both. 1550 \| www.pnas.org/cgi/doi/10.1073/pnas.1612922114 Results 3 B–D), the nas6Δ rpn6-mut double mutants did not show additional growth defects relative to the rpn6-mut single mutants in the presence of canavanine. Fig. 3. Nas6–lid clash is driven by Rpn6. (A) Using a segmented density map of the 26S proteasome (3), the Rpn6 crystal structure [PDB ID code 3TXN (32)] was fit into the segments corresponding to the Rpn6 in chimera (40). Rpn6 (gray), Rpts (blue), and CP (green) are shown. (Bottom) Rpn6 amino acid residues that contact Rpt3 (orange; S237, Y238, E241) and Rpt6 (red; T199, T203) are highlighted in yellow. (B) Proteasome holoenzymes from rpn6- TTSYE mutants were subjected to SDS/PAGE and immunoblotting for Nas6 and Rpt6 [i], and were analyzed via native PAGE and LLVY-AMC activity assays [ii, iii]. The rpn6-mut indicates the rpn6-TTSYE allele harboring amino acid substitutions: T199K, T203K, S237A, Y238A, and E241A. wt, wild type. (C and D) Base–lid association assays showing the loss of Nas6 interference against the rpn6-mut lid. Immobilized nas6Δ base (∼0.8 pmol) was incubated with or without Nas6 (21 pmol) and mixed with the rpn6-mut lid (∼4.4 pmol). Base-bound materials were analyzed via 12% (vol/vol) SDS/PAGE and im- munoblotting. Rpn5 and Rpn8 are lid subunits, and Rpn1 is a base subunit. Nas6 Regulates Endogenous Base–Lid Association and Proteasome Function. Given robust Nas6 interference against the lid binding to base before holoenzyme formation, we examined whether endogenous Nas6 at its physiological level also obstructs lid–base association in vivo. We supplemented ATP or ATPγS to the yeast cell lysates and affinity-purified Nas6-associated complexes through a FLAG-affinity tag integrated into its chromosomal locus (Fig. 4A). Nas6 associated with RP more than base in the ATPh state (Fig. 4A; ATP). Importantly, with ATPγS present in cell lysates, Nas6-bound RP diminished, whereas Nas6-associ- ated base, serving as an internal control, remained largely the same (Fig. 4A; ATPγS). These results are as predicted by the hypothesis that the lid and Nas6 compete for binding to the base in a nucleotide-dependent manner. We confirmed that ATPγS decreased the yield of lid subunits by immunoblotting (Fig. 4B; will refer to the rpn6-TTSYE allele as rpn6-mut henceforth. These results show that the Nas6 steric clash against the lid is due to the positioning of Rpn6 via its contacts with Rpt3 and Rpt6. On native gels, proteasome holoenzymes from rpn6-mut were comparable to wild type in the ATPγS state (Fig. Results In the ATPγS state, the CP can attenuate Nas6 interference without releasing it, whereas in the ATPh state, the CP attenuates Nas6 interference with lid–base joining by promoting eviction of Nas6, thereby completing the formation of the proteasome holoenzyme. Also, Nas6-containing proteasomes Li et al. Li et al. Fig. 3. Nas6–lid clash is driven by Rpn6. (A) Using a segmented density map of the 26S proteasome (3), the Rpn6 crystal structure [PDB ID code 3TXN (32)] was fit into the segments corresponding to the Rpn6 in chimera (40). Rpn6 (gray), Rpts (blue), and CP (green) are shown. (Bottom) Rpn6 amino acid residues that contact Rpt3 (orange; S237, Y238, E241) and Rpt6 (red; T199, T203) are highlighted in yellow. (B) Proteasome holoenzymes from rpn6- TTSYE mutants were subjected to SDS/PAGE and immunoblotting for Nas6 and Rpt6 [i], and were analyzed via native PAGE and LLVY-AMC activity assays [ii, iii]. The rpn6-mut indicates the rpn6-TTSYE allele harboring amino acid substitutions: T199K, T203K, S237A, Y238A, and E241A. wt, wild type. (C and D) Base–lid association assays showing the loss of Nas6 interference against the rpn6-mut lid. Immobilized nas6Δ base (∼0.8 pmol) was incubated with or without Nas6 (21 pmol) and mixed with the rpn6-mut lid (∼4.4 pmol). Base-bound materials were analyzed via 12% (vol/vol) SDS/PAGE and im- munoblotting. Rpn5 and Rpn8 are lid subunits, and Rpn1 is a base subunit. complex when Nas6 is present. The rpn6-mut lid interacted with the base comparably in both the ATPh and ATPγS states (Fig. 3C, lanes 2 and 4), indicating that the compound substitution did not affect binding of the rpn6-mut lid to base. Importantly, when Nas6 was added, the rpn6-mut lid maintained a stable complex with the base (Fig. 3C, lanes 3 and 5). In addition, when Nas6 was loaded on the base and incubated with the rpn6-mut lid, Nas6 did not release from the base in either the ATPh or ATPγS state (Fig. 3D, lanes 3 and 6). The rpn6-mut cells exhibited an increased sensitivity to cana- vanine, an arginine analog that generates protein misfolding and, consequently, proteasome stress (Fig. S5B). The growth pheno- types of nas6Δ mutants alone are not readily detectable due to compensatory mechanisms, including Rpn4-mediated protea- some induction (13) and the cooperative function of other C-domain chaperones (9, 13, 14). Consistent with loss of Nas6 interference in rpn6-mut (Fig. Results 3B, ii, compare lanes 5 and 6; RP2-CP, RP1-CP), indicating that altered posi- tioning of Rpn6 does not compromise proteasome stability. In agreement with the loss of steric hindrance of Nas6 against the lid subunit Rpn6, proteasome holoenzyme levels appeared similar between rpn6-mut and nas6Δ rpn6-mut (Fig. 3B, ii, compare lanes 6 and 8; RP2-CP, RP1-CP). Free CP in rpn6-mut became un- detectable in nas6Δ rpn6-mut in the ATPγS state (Fig. 3B, iii, compare lanes 6 and 8), indicating that RP-CP association stabi- lizes the proteasome upon the removal of Nas6, which is otherwise retained in the rpn6-mut proteasome (Fig. 3B, i, lane 6). Fig. 4. Nas6 regulates endogenous RP abundance via nucleotide-mediated in- hibitory control. (A) Endogenous Nas6 interferes with base–lid association in the ATPγS state. Endogenous chaperone complexes were affinity-purified via 3×FLAG tags appended to the C termini of the individual chaperones, and were subjected to 3.5% (vol/vol) native PAGE and Coomassie staining. (B) Loss of lid subunits from endogenous Nas6–RP complex in the ATPγS state. Affinity-purified chaperone complexes as in A were analyzed by 12% (vol/vol) SDS/PAGE and immunoblotting (IB) with antibodies for lid (Rpn8, Rpn12) and base (Rpt5, Rpn1) subunits. (C) Whole-cell lysates (80 μg) were subjected to 3.5% native PAGE and immunoblotting for base (Rpt5) and lid (Rpn5) subunits. (D) Yeast cell growth assay. Individual chaperone knockout strains were transformed with a URA3- marked Sec61-2L–expressing plasmid and grown in threefold serial dilutions on media lacking uracil. Deficient proteolysis, leading to the accumulation of Leu2 protein, was monitored on synthetic media lacking both uracil and leucine. BIOCHEMISTRY p p ( g ) Both rpn6-mut and nas6Δ rpn6-mut in the ATPh state exhibited a decrease in doubly capped holoenzyme species (RP2-CP), with an accompanying increase in free CP (Fig. 3B, iii, lanes 2 and 4, and Fig. S5A). The level of free CP slightly decreased in nas6Δ rpn6-mut relative to rpn6-mut in the ATPh state (Fig. 3B, iii, lanes 2 and 4), but to a lesser extent than in their counterparts in the ATPγS state (Fig. 3B, iii, lanes 6 and 8). Discussion During proteasome biogenesis, the base, lid, and CP assemblies join. The centrally positioned base is the linchpin of these asso- ciations (1–4). These late steps in assembly are intricately regu- lated by a cohort of base binding molecular chaperones, including Rpn14, Nas6, Hsm3, and Nas2. Their common function is to negatively regulate the interaction between the base and CP (12, 13, 17–19, 26). Here, we report a complementary mechanism that is specific to the Nas6 chaperone and that governs the base–lid interaction (Fig. 5 and Fig. S10). Nas6 potently antagonizes base– lid association through steric hindrance, the same type of mech- anism used by the C-domain chaperones to regulate base–CP association. However, the negative regulation of the base–lid in- teraction by Nas6 only occurs in one of the two major states as- sumed by the base, the ATPγS state. In contrast, for Nas6, as well as its relatives Hsm3 and Rpn14, the suppression of base–CP joining is relieved by ATPγS (17). Thus, the dynamic nucleotide state of the proteasome governs its assembly through comple- mentary mechanisms under the direction of Nas6. We examined whether loss of Nas6 results in premature lid– base association in the chaperone mutants, thereby disrupting base assembly. Because the lid binding site is present in the Rpt3-Rpt6 module (Fig. 3A), we conducted affinity purification with 3×FLAG-tagged Rpn14, the cognate chaperone for Rpt6, using the nas6 double mutants nas6Δhsm3Δ and nas6Δnas2Δ, as well as their single mutants (Fig. S7). Importantly, the lid species are readily detected in Rpn14-associated complexes in both double mutants lacking NAS6, suggesting premature joining of the lid to the base species (Fig. S7, lanes 5 and 6). Multiple diffuse lid–base bands indicate that deregulated lid joining dis- rupts ongoing base assembly. The ability of Nas6 to regulate the base–lid interface is linked to the assembly state of the proteasome not only via its de- pendence on nucleotide hydrolysis but also through the base–CP Fig. 5. Model of Nas6-mediated proteasome assembly. (Left) In early-assembly intermediates of the heterohexameric Rpt ring (orange), ATP hydrolysis is blocked (24, 37) [s3 (ATPγS) state (22)]. Numbers indicate the positions of indi- vidual Rpt subunits (36). In this state, Nas6 (red) obstructs the lid (light blue) binding to Rpt3 and Rpt6 to promote the heterohexameric Rpt ring formation. This diagram depicts a representative early intermediate (36). Early-assembly intermediates may also contact the CP (Fig. Results 4D, −Ura/−Leu). We confirmed that proteasomes in nas6Δ cells are functionally defective and cannot efficiently degrade Sec61- 2L using a cycloheximide chase assay (Fig. S8). In addition to Sec61-2L, the nas6Δ mutants accumulate polyubiquitinated proteins upon heat stress at 37 °C, indicating general defects in cellular protein degradation (Fig. S9). These results suggest that the distinct role of Nas6 in the regulation of proteasome as- sembly is also reflected in perturbed proteasome function in vivo. ATPγS). Unlike Nas6, the Rpn14 and Hsm3 chaperones asso- ciated mainly with the base and did not copurify with lid subunits (Fig. 4 A and B). Nas2 did not associate with the base due to its being released before completion of the Rpt ring formation (36) (Fig. 4 A and B). In summary, Nas6 interference can control the abundance of the endogenous RP species depending on the nucleotide state of the complex, in agreement with the data from the RP association experiments in vitro (Figs. 1 and 2). p ( g ) Given that base assembly intermediates bind ATP but lack the hydrolytic activity (24, 37), Nas6 interference with the lid in the ATPγS state (Figs. 1–3) suggests that Nas6 obstructs lid binding during base assembly to complete the heterohexameric Rpt ring before base–lid association. To test this hypothesis, we deleted NAS6 in a panel of chaperone knockout strains that lack each chaperone singly and in combination, and examined the base assembly inter- mediates, the Rpt5-Rpt4 and Rpt1-Rpt2 modules, together with the base, lid, and proteasome holoenzyme via native PAGE and immu- noblotting. First, we examined the Rpt5-Rpt4 module. The nas6Δ single mutants accumulate the Rpt5-Rpt4 module relative to wild type (Fig. 4C, Top, compare lanes 1 and 2). Importantly, the Rpt5- Rpt4 module noticeably accumulates in all double mutants lacking NAS6, nas6Δrpn14Δ, nas6Δhsm3Δ, and nas6Δnas2Δ relative to the single mutants (Fig. 4C, Top, compare lanes 4, 6, and 8 with lanes 3, 5, and 7, respectively). The base is nearly undetectable in these mu- tants (Fig. 4C, Top, lanes 4, 6, and 8). As long as Nas6 is present, the Rpt5-Rpt4 module does not substantially accumulate in other chaperone mutants: rpn14Δhsm3Δ, hsm3Δnas2Δ, rpn14Δnas2Δ, and rpn14Δhsm3Δnas2Δ (Fig. 4C, Top, lanes 9, 11, 13, and 15). When NAS6 is deleted in each of these mutants, the Rpt5-Rpt4 module accumulates and the RP2-CP species diminish (Fig. 4C, Top, lanes 10, 12, 14, and 16). Results Thus, the interaction between RP and CP becomes unstable in rpn6-mut in the ATPh state, consistent with a unique capacity of Rpn6 to serve as a molecular clamp for the proteasome (32): Rpn6 establishes contacts with the CP through the same N-terminal tetratricopeptide repeat (TPR) domain that pro- jects over Rpt3 and Rpt6 in the ATPh state (Fig. S3B). In summary, our results suggest that successful contacts of Rpn6 with Rpt3 and Rpt6 of the base are critical for Rpn6 to interact productively with the CP, and thus stabilize the ATPh state of the proteasome. Fig. 4. Nas6 regulates endogenous RP abundance via nucleotide-mediated in- hibitory control. (A) Endogenous Nas6 interferes with base–lid association in the ATPγS state. Endogenous chaperone complexes were affinity-purified via 3×FLAG tags appended to the C termini of the individual chaperones, and were subjected to 3.5% (vol/vol) native PAGE and Coomassie staining. (B) Loss of lid subunits from endogenous Nas6–RP complex in the ATPγS state. Affinity-purified chaperone complexes as in A were analyzed by 12% (vol/vol) SDS/PAGE and immunoblotting (IB) with antibodies for lid (Rpn8, Rpn12) and base (Rpt5, Rpn1) subunits. (C) Whole-cell lysates (80 μg) were subjected to 3.5% native PAGE and immunoblotting for base (Rpt5) and lid (Rpn5) subunits. (D) Yeast cell growth assay. Individual chaperone knockout strains were transformed with a URA3- marked Sec61-2L–expressing plasmid and grown in threefold serial dilutions on media lacking uracil. Deficient proteolysis, leading to the accumulation of Leu2 protein, was monitored on synthetic media lacking both uracil and leucine. p To validate that the contacts of Rpn6 with Rpt3 and Rpt6 form the basis for a steric clash with Nas6, we examined whether the rpn6-TTSYE substitution abrogates Nas6 interference by using base–lid association assays. If the rpn6-mut lid still clashes with Nas6, the rpn6-mut lid should not form a stable lid–base PNAS \| February 14, 2017 \| vol. 114 \| no. 7 \| 1551 Li et al. proteasome-mediated degradation of Sec61, a protein of the endoplasmic reticulum, in nas6 mutants (Fig. 4D). A sensitive Sec61-derived reporter, Sec61-2L, in which the Leu2 protein is fused to Sec61, provided for a plate assay in which defective protein degradation allows for cell growth in the absence of leucine (38). General growth properties were comparable among individual chaperone knockout cells (Fig. 4D, −Ura). However, nas6Δ cells uniquely exhibited restoration of growth on media lacking leucine, indicating a defect in Sec61-2L degradation (Fig. Discussion S10), which can serve as a template for Rpt ring assembly (17). Non-ATPase subunits of the base (Rpn1, Rpn2, Rpn13) are omitted for clarity. (Center) Fully formed Rpt ring initiates ATP hydrolysis. Nas6 and the other chaperones interfere with Rpt ring–CP interaction (17), but they now permit base–lid assembly due to a conformational change in the base from the s3 (ATPγS) to s1 (ATPh) state (22). In addition, chaperone–base–CP complexes may form (Fig. S10) because the base also exists in the s3 (ATPγS) state (22, 41), which can accommodate the chaperones at the base–CP interface (17) (Fig. 2D). (Right) Lid–base–CP complex forms, resulting in eviction of the chap- erones and completion of the proteasome holoenzyme. p g g y The lid has been shown to self-assemble (29). Deficient base assembly leads to accumulation of free lid in whole-cell extracts in all double mutants lacking Nas6, nas6Δrpn14Δ, nas6Δhsm3Δ, and nas6Δnas2Δ, supporting that Nas6 action is crucial for proper base– lid association (Fig. 4C, Bottom, lanes 4, 6, and 8). The free lid in hsm3Δ does not reflect direct regulation of the lid contact sites by Hsm3 (Fig. 4C, Bottom, lane 5), but indicates deficient formation of the cognate Rpt1-Rpt2 module (10, 18) (Fig. S6B, lane 5). Fur- thermore, free lid accumulates specifically when NAS6 is deleted in rpn14Δhsm3Δ, hsm3Δnas2Δ, rpn14Δnas2Δ, and rpn14Δhsm3Δnas2Δ mutants (Fig. 4C, Bottom, compare lanes 9, 11, 13, and 15 with lanes 10, 12, 14, and 16, respectively). Therefore, lid accumulation mainly results from loss of Nas6 function rather than general defects in base assembly. Taken together, these results support that Nas6 in- terference with the lid (Figs. 2B, 2F, and 3) provides a mechanism to order the events of Rpt ring assembly and lid joining for the formation of the proteasome holoenzyme. Fig. 5. Model of Nas6-mediated proteasome assembly. (Left) In early-assembly intermediates of the heterohexameric Rpt ring (orange), ATP hydrolysis is blocked (24, 37) [s3 (ATPγS) state (22)]. Numbers indicate the positions of indi- vidual Rpt subunits (36). In this state, Nas6 (red) obstructs the lid (light blue) binding to Rpt3 and Rpt6 to promote the heterohexameric Rpt ring formation. This diagram depicts a representative early intermediate (36). Early-assembly intermediates may also contact the CP (Fig. S10), which can serve as a template for Rpt ring assembly (17). Non-ATPase subunits of the base (Rpn1, Rpn2, Rpn13) are omitted for clarity. Results The levels of the proteasome holoenzymes are consistent with their activities (Fig. S6A). Similarly, immunoblotting analysis of the Rpt1-Rpt2 module also shows that it accumulates whenever NAS6 is deleted (Fig. S6B). These results demonstrate that Nas6 is distinctly required for all three Rpt modules to assemble productively into the Rpt ring, supporting the unique role of Nas6 in obstructing the lid during base assembly. 1552 \| www.pnas.org/cgi/doi/10.1073/pnas.1612922114 Materials and Methods Groll M, et al. (2000) A gated channel into the proteasome core particle. Nat Struct Biol 7(11):1062–1067. p g y p y j g ( ) 30. Tomko RJ, Jr, Hochstrasser M (2014) The intrinsically disordered Sem1 protein func- 30. Tomko RJ, Jr, Hochstrasser M (2014) The intrinsically disordered Sem1 protein func- tions as a molecular tether during proteasome lid biogenesis. Mol Cell 53(3):433–443. 6. Förster A, Masters EI, Whitby FG, Robinson H, Hill CP (2005) The 1.9 A structure of a proteasome-11S activator complex and implications for proteasome-PAN/PA700 in- teractions. Mol Cell 18(5):589–599. , , ( ) y p tions as a molecular tether during proteasome lid biogenesis. Mol Cell 53(3):433–443. g p g 31. Tomko RJ, Jr, et al. (2015) A single α helix drives extensive remodeling of the pro- 31. Tomko RJ, Jr, et al. (2015) A single α helix drives extensive remodeling of the pro- teasome lid and completion of regulatory particle assembly. Cell 163(2):432–444. 31. Tomko RJ, Jr, et al. (2015) A single α helix drives extensive remodeling of the pro teasome lid and completion of regulatory particle assembly. Cell 163(2):432–444. 7. Smith DM, et al. (2007) Docking of the proteasomal ATPases’ carboxyl termini in the 20S proteasome’s alpha ring opens the gate for substrate entry. Mol Cell 27(5):731–744. 32. Pathare GR, et al. (2012) The proteasomal subunit Rpn6 is a molecular clamp holding the core and regulatory subcomplexes together. Proc Natl Acad Sci USA 109(1):149–154. 8. Rabl J, et al. (2008) Mechanism of gate opening in the 20S proteasome by the pro- teasomal ATPases. Mol Cell 30(3):360–368. 8. Rabl J, et al. (2008) Mechanism of gate op teasomal ATPases. Mol Cell 30(3):360–368. teasomal ATPases. Mol Cell 30(3):360–368. 33. Nakamura Y, et al. (2007) Structural basis for the recognition between the regulatory particles Nas6 and Rpt3 of the yeast 26S proteasome. Biochem Biophys Res Commun 359(3):503–509. 9. Funakoshi M, Tomko RJ, Jr, Kobayashi H, Hochstrasser M (2009) Multiple assembly chap- erones govern biogenesis of the proteasome regulatory particle base. Cell 137(5):887–899. 34. Vilchez D, et al. (2012) Increased proteasome activity in human embryonic stem cells is regulated by PSMD11. Nature 489(7415):304–308. 10. Le Tallec B, Barrault MB, Guérois R, Carré T, Peyroche A (2009) Hsm3/S5b participates in the assembly pathway of the 19S regulatory particle of the proteasome. Mol Cell 33(3):389–399. 35. Ghaemmaghami S, et al. (2003) Global analysis of protein expression in yeast. Materials and Methods Nature 425(6959):737–741. 11. Kaneko T, et al. (2009) Assembly pathway of the Mammalian proteasome base sub- complex is mediated by multiple specific chaperones. Cell 137(5):914–925. 36. Tomko RJ, Jr, Funakoshi M, Schneider K, Wang J, Hochstrasser M (2010) Hetero- hexameric ring arrangement of the eukaryotic proteasomal ATPases: Implications for proteasome structure and assembly. Mol Cell 38(3):393–403. 12. Park S, et al. (2009) Hexameric assembly of the proteasomal ATPases is templated through their C termini. Nature 459(7248):866–870. 37. Lee SH, Moon JH, Yoon SK, Yoon JB (2012) Stable incorporation of ATPase subunits into 19 S regulatory particle of human proteasome requires nucleotide binding and C-terminal tails. J Biol Chem 287(12):9269–9279. 13. Roelofs J, et al. (2009) Chaperone-mediated pathway of proteasome regulatory particle assembly. Nature 459(7248):861–865. p y 14. Saeki Y, Toh-E A, Kudo T, Kawamura H, Tanaka K (2009) Multiple proteasome-interacting proteins assist the assembly of the yeast 19S regulatory particle. Cell 137(5):900–913. 38. Kohlmann S, Schäfer A, Wolf DH (2008) Ubiquitin ligase Hul5 is required for frag- ment-specific substrate degradation in endoplasmic reticulum-associated degrada- tion. J Biol Chem 283(24):16374–16383. 15. Hanssum A, et al. (2014) An inducible chaperone adapts proteasome assembly to stress. Mol Cell 55(4):566–577. 16. Sakata E, et al. (2011) The catalytic activity of Ubp6 enhances maturation of the proteasomal regulatory particle. Mol Cell 42(5):637–649. 39. Schrödinger, LLC (2010) The PyMOL Molecular Graphics System (Schrödinger, LLC, New York), Version 1.3r1. 40. Pettersen EF, et al. (2004) UCSF Chimera–a visualization system for exploratory re- search and analysis. J Comput Chem 25(13):1605–1612. 17. Park S, et al. (2013) Reconfiguration of the proteasome during chaperone-mediated assembly. Nature 497(7450):512–516. 18. Barrault MB, et al. (2012) Dual functions of the Hsm3 protein in chaperoning and scaffolding regulatory particle subunits during the proteasome assembly. Proc Natl Acad Sci USA 109(17):E1001–E1010. 41. Luan B, et al. (2016) Structure of an endogenous yeast 26S proteasome reveals two major conformational states. Proc Natl Acad Sci USA 113(10):2642–2647. 42. Özalp VC, Pedersen TR, Nielsen LJ, Olsen LF (2010) Time-resolved measurements of intracellular ATP in the yeast Saccharomyces cerevisiae using a new type of nano- biosensor. J Biol Chem 285(48):37579–37588. 19. Satoh T, et al. (2014) Structural basis for proteasome formation controlled by an as- sembly chaperone nas2. Structure 22(5):731–743. 20. Matyskiela ME, Lander GC, Martin A (2013) Conformational switching of the 26S proteasome enables substrate degradation. Nat Struct Mol Biol 20(7):781–788. 43. Discussion (Center) Fully formed Rpt ring initiates ATP hydrolysis. Nas6 and the other chaperones interfere with Rpt ring–CP interaction (17), but they now permit base–lid assembly due to a conformational change in the base from the s3 (ATPγS) to s1 (ATPh) state (22). In addition, chaperone–base–CP complexes may form (Fig. S10) because the base also exists in the s3 (ATPγS) state (22, 41), which can accommodate the chaperones at the base–CP interface (17) (Fig. 2D). (Right) Lid–base–CP complex forms, resulting in eviction of the chap- erones and completion of the proteasome holoenzyme. To determine whether the distinct role of Nas6 in proteasome assembly is reflected in proteasome function in vivo, we tracked Li et al. interaction. Thus, Nas6 readily splits the RP into base and lid, but does so poorly, if at all, when challenged with mature proteasomes, despite the ability of Nas6 to bind these proteasomes (Fig. 2 F–H). The specificity of Nas6 for regulating nascent forms of the protea- some was found to be linked to Rpn6, as indicated by the ability of targeted mutations in Rpn6 to abrogate base–lid interference (Fig. 3). Thus, the steric hindrance between these two proteins is desta- bilizing to the RP but is accommodated in the holoenzyme, perhaps because holoenzyme-specific Rpn6-CP contacts suffice to stabilize this complex (3, 32). to mature proteasomes by chaperone eviction, which is principally accomplished by the ability of CP to outcompete the chaperones in the ATPh state (13, 17). Because both the base and lid assemble through a multitude of intermediates (9–14, 16, 29), Nas6 may serve to suppress not only the interaction of base and lid but also inter- actions between partially assembled precursors of these complexes, and thus help to ensure that only a fully and properly assembled base and lid can join to form the RP. 1. da Fonseca PC, He J, Morris EP (2012) Molecular model of the human 26S proteasome. Mol Cell 46(1):54–66. 2. Kish-Trier E, Hill CP (2013) Structural biology of the proteasome. Annu Rev Biophys 42:29–49. 3. Lander GC, et al. (2012) Complete subunit architecture of the proteasome regulatory particle. Nature 482(7384):186–191. 4. Lasker K, et al. (2012) Molecular architecture of the 26S proteasome holocomplex determined by an integrative approach. Proc Natl Acad Sci USA 109(5):1380–1387. 5. Groll M, et al. (2000) A gated channel into the proteasome core particle. Nat Struct Biol 7(11):1062–1067. 6. Förster A, Masters EI, Whitby FG, Robinson H, Hill CP (2005) The 1.9 A structure of a proteasome-11S activator complex and implications for proteasome-PAN/PA700 in- teractions. Mol Cell 18(5):589–599. 7. Smith DM, et al. (2007) Docking of the proteasomal ATPases’ carboxyl termini in the 20S proteasome’s alpha ring opens the gate for substrate entry. Mol Cell 27(5):731–744. 8. Rabl J, et al. (2008) Mechanism of gate opening in the 20S proteasome by the pro- teasomal ATPases. Mol Cell 30(3):360–368. 9. Funakoshi M, Tomko RJ, Jr, Kobayashi H, Hochstrasser M (2009) Multiple assembly chap- erones govern biogenesis of the proteasome regulatory particle base. Cell 137(5):887–899. Materials and Methods The intricate nucleotide dependence on the action of the C-domain chaperones is expected to link the mode of chaperone activity to the assembly state of the nascent proteasome, because intermediates in assembly of the Rpt ring are thought to bind, but not hydrolyze, nucleotides (24, 37). Therefore, in our model of assembly, the base is formed in the ATPγS state and only transitions to the ATPh state after Rpt ring closure (Fig. 5 and Fig. S10). In early intermediates, Nas6 will hinder base–lid in- teractions but allow base-CP contacts, whereas in late interme- diates, in which the ATPh state predominates over the ATPγS state, Nas6 and the other C-domain chaperones hinder base-CP contacts but allow base-lid contacts. Late intermediates give rise A complete list of yeast strains is provided in Table S1. All biochemical and genetic experiments were performed at least twice. Detailed procedures for base–lid association assays and affinity purification of proteasomes and chaperone complexes are described in Supporting Information. ACKNOWLEDGMENTS. We thank J. D. Orth, S. Elsasser, and J. Hanna for comments on the manuscript. We thank J. Roelofs for sharing Nas2 antibody and Escherichia coli expression plasmids for Nas6, Rpn14, and Hsm3, as well as a 2μ high-copy expression plasmid for Nas6; D. Wolf for providing Sec61-2L plasmid; and M. Hochstrasser for sharing chaperone strains. We also thank G. Polovin for his assistance with biochemical experiments. This work was supported by Uni- versity of Colorado startup funds, by a Boettcher Webb–Waring Biomedical Re- search Award to S.P., and by NIH Grant R37GM43601 (to D.F.). 26. Lee SY, De la Mota-Peynado A, Roelofs J (2011) Loss of Rpt5 protein interactions with the core particle and Nas2 protein causes the formation of faulty proteasomes that are inhibited by Ecm29 protein. J Biol Chem 286(42):36641–36651. 27. Tian G, et al. (2011) An asymmetric interface between the regulatory and core par- ticles of the proteasome. Nat Struct Mol Biol 18(11):1259–1267. 28. Estrin E, Lopez-Blanco JR, Chacón P, Martin A (2013) Formation o 28. Estrin E, Lopez-Blanco JR, Chacón P, Martin A (2013) Formation of an intricate helical bundle dictates the assembly of the 26S proteasome lid. Structure 21(9):1624–1635. y 29. Tomko RJ, Jr, Hochstrasser M (2011) Incorporation of the Rpn12 subunit couples completion 29. Tomko RJ, Jr, Hochstrasser M (2011) Incorporation of the Rpn12 subunit couples completion of proteasome regulatory particle lid assembly to lid-base joining. Mol Cell 44(6):907–917. 5. 1. da Fonseca PC, He J, Morris EP (2012) Molecular model of the human 26S proteasome. Mol Cell 46(1):54–66. 1. da Fonseca PC, He J, Morris EP (2012) Molecular model of the human 26S proteasome. Mol Cell 46(1):54–66. 2. Kish-Trier E, Hill CP (2013) Structural biology of the proteasome. Annu Rev Biophys 42:29–49. 3. Lander GC, et al. (2012) Complete subunit architecture of the proteasome regulatory particle. Nature 482(7384):186–191. 4. Lasker K, et al. (2012) Molecular architecture of the 26S proteasome holocomplex determined by an integrative approach. Proc Natl Acad Sci USA 109(5):1380–1387. 5. Groll M, et al. (2000) A gated channel into the proteasome core particle. Nat Struct Biol 7(11):1062–1067. Materials and Methods Beckwith R, Estrin E, Worden EJ, Martin A (2013) Reconstitution of the 26S proteasome reveals functional asymmetries in its AAA+ unfoldase. Nat Struct Mol Biol 20(10):1164–1172. 44. Russell SJ, Steger KA, Johnston SA (1999) Subcellular localization, stoichiometry, and protein levels of 26 S proteasome subunits in yeast. J Biol Chem 274(31):21943–21952. 21. Sledz P, et al. (2013) Structure of the 26S proteasome with ATP-γS bound provides insights into the mechanism of nucleotide-dependent substrate translocation. Proc Natl Acad Sci USA 110(18):7264–7269. 45. Sokolova V, Li F, Polovin G, Park S (2015) Proteasome activation is mediated via a functional switch of the Rpt6 C-terminal tail following chaperone-dependent as- sembly. Sci Rep 5:14909. ( ) 22. Unverdorben P, et al. (2014) Deep classification of a large cryo-EM dataset defines the con- formational landscape of the 26S proteasome. Proc Natl Acad Sci USA 111(15):5544–5549. 46. Park S, Kim W, Tian G, Gygi SP, Finley D (2011) Structural defects in the regulatory particle-core particle interface of the proteasome induce a novel proteasome stress response. J Biol Chem 286(42):36652–36666. 23. Kim YC, DeMartino GN (2011) C termini of proteasomal ATPases play nonequivalent roles in cellular assembly of mammalian 26 S proteasome. J Biol Chem 286(30):26652–26666. 24. Thompson D, Hakala K, DeMartino GN (2009) Subcomplexes of PA700, the 19 S reg- ulator of the 26 S proteasome, reveal relative roles of AAA subunits in 26 S proteasome assembly and activation and ATPase activity. J Biol Chem 284(37):24891–24903. 47. Finley D, Ozkaynak E, Varshavsky A (1987) The yeast polyubiquitin gene is essential for resistance to high temperatures, starvation, and other stresses. Cell 48(6):1035–1046. 25. Kumar B, Kim YC, DeMartino GN (2010) The C terminus of Rpt3, an ATPase subunit of PA700 (19 S) regulatory complex, is essential for 26 S proteasome assembly but not for activation. J Biol Chem 285(50):39523–39535. 48. Leggett DS, Glickman MH, Finley D (2005) Purification of proteasomes, proteasome subcomplexes, and proteasome-associated proteins from budding yeast. Methods Mol Biol 301:57–70. PNAS \| February 14, 2017 \| vol. 114 \| no. 7 \| 1553 Li et al.
https://openalex.org/W2378627613	https://europepmc.org/articles/pmc5015556?pdf=render	English	null	How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People?	International journal of integrated care	2,016	cc-by	11,804	Hanna Maria van Dijk, Jane Murray Cramm† and Anna Petra Nieboer‡ Background: Although the need for integrated neighborhood approaches (INAs) is widely recognized, we lack insight into strategies like INA. We describe diverse Dutch INA partners’ experiences to provide integrated person- and population-centered support to community-dwelling older people using an adapted version of Valentijn and colleagues’ integrated care model. Our main objective was to explore the experi ences with INA participation. We sought to increase our understanding of the challenges facing these partners and identify factors facilitating and inhibiting integration within and among multiple levels. Methods: Twenty-one interviews with INA partners (including local health and social care organizations, older people, municipal officers, and a health insurer) were conducted and subjected to latent content analysis. Background: Although the need for integrated neighborhood approaches (INAs) is widely recognized, we lack insight into strategies like INA. We describe diverse Dutch INA partners’ experiences to provide integrated person- and population-centered support to community-dwelling older people using an adapted version of Valentijn and colleagues’ integrated care model. Our main objective was to explore the experi ences with INA participation. We sought to increase our understanding of the challenges facing these partners and identify factors facilitating and inhibiting integration within and among multiple levels. j g g j p p ences with INA participation. We sought to increase our understanding of the challenges facing these partners and identify factors facilitating and inhibiting integration within and among multiple levels. Methods: Twenty-one interviews with INA partners (including local health and social care organizations, older people, municipal officers, and a health insurer) were conducted and subjected to latent content analysis. Methods: Twenty-one interviews with INA partners (including local health and social care organizations, older people, municipal officers, and a health insurer) were conducted and subjected to latent content analysis. y Results: This study showed that integrated care and support provision through an INA is a complex, dynamic process requiring multilevel alignment of activities. The INA achieved integration at the personal, service, and professional levels only occasionally. Micro-level bottom-up initiatives were not aligned with top-down incentives, forcing community workers to establish integration despite rather than because of meso- and macro-level contexts. Conclusions: Top-down incentives should be better aligned with bottom-up initiatives. This study further demonstrated the importance of community-level engagement in integrated care and support provision. RESEARCH AND THEORY How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? RESEARCH AND THEORY Hanna Maria van Dijk, Jane Murray Cramm† and Anna Petra Nieboer‡ Keywords: integrated care and support; integrated neighborhood approach; community level; community- dwelling older people; informal support; the Netherlands of older people continue to live at home, an integrated neighborhood approach (INA) is needed [7]. Integrated care approaches need to incorporate the recognition that communities are co-producers of health and well-being [8–10]. INAs, consisting of collaboration among munici palities, health and social care, informal care providers, voluntary/third sector, churches, schools and the private sector are therefore increasingly advocated as means to overcome current service fragmentation and co-ordinate care and support according to people’s (complex) needs [11–13]. INAs aim is to use available neighborhood resources effectively and increase responsiveness to citi zens’ specific needs, ensuring the provision of person- and population-centered support [13, 14]. Person-centered care and co-ordination among both formal and informal care are now widely recognized as a critical component of privately and publicly funded health care despite significantly different systems of care [15]. * Institute of Health Policy and Management, Erasmus University Rotterdam, the Netherlands hanna.vandijk@bmg.eur.nl † Associate Professor, Institute of Health Policy and Management, Erasmus University Rotterdam, the Netherlands ‡ Professor, Institute of Health Policy and Management, Erasmus University Rotterdam, the Netherlands Corresponding author: Hanna van Dijk How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? Hanna Maria van Dijk, Jane Murray Cramm† and Anna Petra Nieboer‡ Background g Many Western countries face the challenge of meeting the needs of increasing numbers of care-dependent older people using limited health and social care budgets. The development of sustainable long-term care systems that adequately address these needs strains countries’ innova tive capacities forcing nations to restructure the division of responsibilities among the state, market, and commu nity [1, 2]. Instead of the state serving as the main pro vider of (social) care, such burdens have been allocated to communities in many Western countries [3–6]. In this framework, public protection is provided only when the community cannot provide care for objective rea sons, such as the absence of informal caregivers and/or insufficient economic means [2]. As increasing numbers Institute of Health Policy and Management, Erasmus University Rotterdam, the Netherlands hanna.vandijk@bmg.eur.nl † Associate Professor, Institute of Health Policy and Management, Erasmus University Rotterdam, the Netherlands ‡ Professor, Institute of Health Policy and Management, Erasmus University Rotterdam, the Netherlands Corresponding author: Hanna van Dijk van Dijk, H M et al 2016 How To Build an Integrated Neighborhood Approach to Support Community- Dwelling Older People? International Journal of Integrated Care, 16(2): 4, pp. 1–15, DOI: http://dx.doi. org/10.5334/ijic.1596 van Dijk, H M et al 2016 How To Build an Integrated Neighborhood Approach to Support Community- Dwelling Older People? International Journal of Integrated Care, 16(2): 4, pp. 1–15, DOI: http://dx.doi. org/10.5334/ijic.1596 An INA in Rotterdam An INA in Rotterdam INA was supported by a grant provided by the Nether lands Organization for Health Research and Development (ZonMw) as part of the National Care for the Elderly Program, which was designed to improve care for elderly people with complex care needs throughout the Art. 4, page 2 of 15 Art. 4, page 2 of 15 van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? Netherlands. The National Care for the Elderly program started in April 2008 and will run until 2016. Funds for INA were also received from the Dutch Healthcare Author ity (NZa), Geriatric Network Rotterdam and the municipality of Rotterdam. In 2011, the Rotterdam municipality, local health and social care organizations, Erasmus University Rotterdam, the University of Applied Sciences, and Geriatric Network Rotterdam initiated the INA for community-dwelling older people in Rotterdam called ‘Let’s Talk’ (Even Buurten) in which the municipal ity took the lead [16]. Its overarching aim was to create a supportive environment allowing community-dwelling older people to live independently. Although health and social care services are widely available in Rotterdam, they are often fragmented and lack outreach activities that foster early identification of frail older people. The need to invest in (preventive) strategies facilitating older peo ple’s ability to continue living at home has increased with municipal legal responsibilities related to social services (e.g. home care and support of older people and informal support-givers). To achieve this goal the INA needs to over come barriers associated with the provision of care and support for older people in the Netherlands, reinforce networks among health and social care providers and informal support-givers in the community, based on recognition of their mutual dependence in efforts to optimize current services [16, 17]. report manifestations of frailty among older people to INA community workers (Fig. 1). Early detection and case finding is crucial to support older people to age in place. Residents often notice changes and deteriorations in older people’s lives at an earlier stage than professionals. Key figures who are the ‘eyes and ears’ of the neighborhoods are represented by active residents as well as professionals working in these areas (e.g. general practitioners, social workers, police officers). An INA in Rotterdam If key figures notice an older person might be in need of support they are expected to reach out to the community worker of the INA via a signal. These community workers have health and social care backgrounds and have been temporarily reassigned to INA teams, which often include at least one social worker and one community nurse familiar with the neighborhood. Community workers visit older people at home and map their wishes and needs via phased interviews. In consultation with older people, community workers seek appropriate solutions within (preferably informal) networks. The project’s study protocol [17] contains more information on its scope and aims. Theoretical framework As INAs depend on stakeholders’ continuous involvement and interdependence across multiple levels, gaining insight into factors that hinder or facilitate community- based integrated care and support is important. INA’s success depends on integration within and among the micro-level (primary delivery of care and support), meso- level (community, professional, organizational contexts), macro-level (broader policy context of care and support systems), functional integration, and normative integration (Fig. 2) [13, 18, 19]. The INA’s success in providing person- and population- centered care and support requires collaboration among formal and informal community partners on aspects of care ranging from the signaling of problems to preven tion and support. Within the INA context, professionals and residents were asked to watch over neighbors and Figure 1: Working method INA. Figure 1: Working method INA. Figure 1: Working method INA. Art. 4, page 3 of 15 van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? Figure 2: Integrated care model van Dijk, Cramm and Nieboer (adapted from Valentijn et al., 2013). Figure 2: Integrated care model van Dijk, Cramm and Nieboer (adapted from Valentijn et al., 2013). At the micro-level, personal integration involves a holistic and coordinated approach to an older person’s health and well-being, requiring professionals’ active engagement and support of his/her self-management abilities [18]. Service integration ensures the provision of tailored and coherent services across time, space, and disciplines [20]. To support a person-centered, rather than disease-oriented, approach, assessment tools and instruments should account for overall well-being [19]. Micro-level integration thus requires collective actions of partners across the entire care and support continuum. financial incentives that stimulate integrated care and support realization on the meso- and micro-levels [19]. Integration should also focus on the multilevel alignment of activities [18, 19]. Functional integration focuses on the coordination of support functions, such as information management, skilled leadership, and qual ity improvement. Normative integration is a less tangible, yet essential, dimension involving the creation of an inte grated mind-set and common set of values [21]. Aims Although the need for INAs to achieve a better balance between supporting increasing numbers of care- dependent older people and reducing public spending is widely recognized, we lack insight into strategies like INA. In this case study of an INA in Rotterdam, the Netherlands, we describe collaboration among diverse community partners to provide person- and population-centered integrated care and support to community-dwelling older people using an adapted version of Valentijn and colleagues’ [19] integrated care model. Our main objec tive was to explore the experiences of municipal officers, health insurers, health and social care organizations, and older people with INA participation. We sought to increase our understanding of the challenges facing these partners and identify factors facilitating and inhibiting integration within and among multiple levels. The meso-level encompasses structures that exceed community, professional, and organizational boundaries [13]. Integrated care and support models often neglect the community level, which is crucial in increasing responsiveness to older people’s needs and bundling resources available among formal and informal support-givers [13]. Therefore, we added community integration to the meso-level in the adapted model. On a professional level, partnerships within and among health and social care organizations are needed. These partnerships ideally cover a range of specialist and generalist skills to enable a holistic approach to older peoples’ needs. Organizational integration aims to overcome organizational boundaries that may hamper collaboration among health and social care professionals. It provides structural activities that promote collaboration among organizations [18, 19]. Methods Design and setting On the macro-level, the system should account for the complexity of issues that arise locally with respect to person- and population-centered support provision. It should thus provide regulatory, accountability, and This qualitative, descriptive study was based on face-to- face interviews with INA participants conducted in several districts of Rotterdam over a 4-month period in 2013. van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? Art. 4, page 4 of 15 van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? health and social care backgrounds, four managers/ directors of health and social care organizations, seven municipal officers, one health insurer, and one former politician who remained actively engaged in the field of long-term care (see Table 1 for more details). Profes sionals were selected on the basis of their participation within INA, ensuring a variety of participants in terms of (professional) background (e.g. health or social care background) and responsibilities (e.g. on an execu tive, managerial or policy level). In addition to our own selection of participants that we deemed essential to interview, we used snowball sampling to identify new participants that could contribute to our study. Last, we relied on INA’s community workers to identify The first author also made field notes and audio-recordings at several INA-related meetings, ranging from those of community-based teams and civic steering committees to educational meetings for community workers. The INA was initiated in two districts of Rotterdam in April 2011 and extended to two additional districts 1 year later. The ethics committee of Erasmus University Medical Centre of Rotterdam approved the project in June 2011 (MEC-2011-197). and invite older people who received INA support to participate in the study. and invite older people who received INA support to participate in the study. education team (HET)]. Elements contributing to func tional and normative integration among levels were also examined. Micro-level: personal integration Gaining trust Obtaining older people’s trust was identified as a key pre requisite for the provision of person-centered support: “older people are very suspicious. And from that distrust they need trust, someone they can trust”. (HCD) Continuity is a precondition for gaining trust. Rapid fluctuations in projects often have resulted in discon tinuities in care and support co-ordination, rendering older people distrustful of new projects and faces. Their awareness of their frail condition exacerbates this distrust. INA community workers thus had to invest much time in becoming familiar faces in neighborhoods. The use of business cards and posters with their photographs contributed to their familiarity, and older people reported that they kept these business cards at hand. Older people who built relationships with community workers felt reas sured that they could confide in them and rely on them when in need. Acknowledging and strengthening older people’s capabilities The INA uses individualized support plans based on assessments of older people’s physical and social needs and capabilities (e.g. housing, mobility issues, social activities). One community worker argued that filling in a support plan was itself an intervention, as it encouraged older people to articulate needs and reflect on their capa bilities. Community workers felt that older people needed guidance in using and strengthening capabilities, taking responsibility for their own health and well-being (e.g. applying for a walker, learning to manage finances). Older people often felt entitled to health and social care ser vices, which they ‘had been working for all their life’ (OP). Community workers thus played important roles in generating awareness of and strengthening older peo ple’s capabilities before turning to (in)formal support, which required careful consideration of when (not) to intervene. Results The first author conducted all interviews (60–90 minutes) at participants’ offices or homes; one interview involved three municipal officers simultaneously. Interviews were audio-taped with participants’ permission and transcribed. The interviews aimed to elicit participants’ reflections on their experiences with the INA from their (professional) perspectives. Because relevant research is sparse, performing interviews with a limited number of preconceived categories was most appropriate [22]. To gain new insight, we allowed themes to arise from the data [22, 23]. Participants were encouraged to describe and reflect on their experiences with (collaborative or competitive) interaction among community partners and the perceived roles and responsibilities with respect to integrated care and support provision to older people. More specifically, older people were asked to describe their experiences with INA; i.e. how they got involved with INA, their initial expectations of INA, their perspec tives on INA’s core principles (e.g. with respect to informal support-giving and encouragement of self-management abilities) and their overall experiences with INA support-giving. Next, the interviews with INA community workers and project manager aimed to elicit their roles and responsibilities within INA, their experiences with successful/difficult parts of their work, their perspectives on INA’s core principles and their views on micro-, meso- and macro-level factors that impeded or facilitated these principles. Last, the goal of the interviews with the other participants (i.e. managers/directors of health and social care organizations, municipal officers, the health insurer and former politician) was to gain an overview of the different micro-, meso- and macro-level facilitators and barriers they faced when dealing with integrated care and support-giving to community-dwelling older people in general as well as with INA in particular. Results Table 2 presents the main findings of our study. Micro-level: personal integration Gaining trust Sample h The sample consisted of 21 INA participants, including the INA project manager, three older people who received INA support, four INA community workers with Participant Gender Background Community workers Participant 1 woman Community nurse INA with a social care background (specialized in coordinating voluntary work) Participant 2 woman Community nurse INA with a health care background (specialized as a nurse practitioner) Participant 3 man Community nurse INA with a social care background (specialized in community work) Participant 4 woman Community nurse INA with a social care background (specialized in community work) Managers/directors Participant 5 man Manager health care organization Participant 6 man Manager social care organization Participant 7 woman Manager health care organization Participant 8 woman Director health care organization Municipal officers Participant 9 woman Alderman (with a portfolio responsibility on participation and integration) Participant 10 woman Alderman sub-municipality (with a portfolio responsibility on health and social care) Participant 11 man Senior policy officer Social Support Act Participant 12 man Program manager assisted living Participant 13 woman Policy officer sub-municipality health and social care Participant 14 woman Policy officer sub-municipality health and social care Participant 15 man Policy officer health and social care Older people Participant 16 woman Older person who received INA support and resided in Oude Westen Participant 17 woman Older person who received INA support and resided in Lombardijen Participant 18 woman Older person who received INA support and resided in Kralingen Other Participant 19 man Project manager of INA Participant 20 man Director procurement and policy of a health insurance company Participant 21 woman Former politician who remained actively engaged in the field of long-term care (e.g. through her participation as a program member of the National Care for the Elderly Program) Table 1: Study participants. Art. 4, page 5 of 15 Art. 4, page 5 of 15 van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? Analysis L Personal integration Overcoming resistance to informal support Community workers repor difficulty relying on inform to ask for help and strongl Service integration Engaging community resources Community workers tried services, which was not alw Service integration Community workers must set up and track responses to interventions To ensure service integrati be integrated throughout supporting older people. M support provision requires simultaneously at multiple Meso-level Community integration Building community awareness and trust Community workers noted took time and that commu to alert them to frail older someone’s life. Community integration Familiarity with the neighborhood INA community workers m sometimes prejudices, of s support into account. Community integration Adaption to new roles The need for community in to reinvent their roles and Community integration Sustaining relationships To overcome barriers to co workers perceived that sus in gaining access to frail ol assessing potential suppor Professional integration Individual skills Recruitment of ‘entreprene generalist and specialist sk crucial for professional int Professional integration Team skills Discontinuity and a lack of hamper professional integ Organizational integration Conflicting organizational interests Although health and socia the need to collaborate, pr containments are forcing t interests over the common Organizational integration Lack of organizational commitment Organizational integration organizational interests an conditions, i.e. through a f managers and through hig previous collaborations. St creation of opportunities f insight in each other’s add integration. Macro-level Key observations Obtaining older people’s trust was identified as a key prerequisite for the provision of person-centered support. Continuity, in turn, is a precondition for gaining trust. The INA uses individualized support plans based on assessments of older people’s physical and social needs and capabilities. Community workers reported that older people had difficulty relying on informal networks; they were reluctant to ask for help and strongly desired independence. Community workers tried to mobilize volunteers to set up services, which was not always successful. To ensure service integration, community resources must be integrated throughout the process of signaling and supporting older people. Moreover, integrated care and support provision requires community workers to operate simultaneously at multiple levels. Analysis L Latent content analysis of narrative text was performed, which yields a rich understanding of a phenomenon [22, 24]. To avoid loss of nuance, participants’ narratives were translated into English only in the report-writing stage. To obtain a holistic perspective, entire transcribed texts were first read open-mindedly several times. Transcriptions from each group were then read separately to comprehend overall meaning. Then, we read texts word by word, extracting ‘units of meaning’ that were coded and categorized using atlas.ti. Finally, the under lying meanings (i.e. latent content) of categories were formulated into themes [24]. Overcoming resistance to informal support Within the INA, informal support is sought before profes sional support for older people who cannot meet their own needs. Community workers, however, reported that older people had difficulty relying on informal networks; they were reluctant to ask for help and strongly desired independence: “People first must drop dead so to speak, before they will turn to help, in other cases they feel you just shouldn’t whine”. (CW) Barriers to and facilitators of integration were identi fied within our theoretical framework of the adapted integrated care model [19]. Results are reported by inte gration level, with quotations identified by participants’ backgrounds [community worker (CW), older person (OP), project manager (PM), health or social care director/ manager (HCD/SCD/HCM), sub-alderman (SUBALD), municipal officer (MO), health insurer (HI), and head Older people especially struggle to ask for social sup port; despite recognizing its importance, many avoid social contact: “When they have defined it, it becomes real and that’s so confronting. They got used to being alone and isolated; it became part of their own structure, making them extremely afraid of any change”. (CW) Art. 4, page 6 of 15 van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? Integration level Challenge Key observations Micro-level Personal integration Gaining trust Obtaining older people’s t prerequisite for the provis Continuity, in turn, is a pre Personal integration Acknowledging and strengthening older people’s capabilities The INA uses individualize assessments of older peop capabilities. Table 2: Overview of our study findings. Community workers play a crucial role in breaking through this structure and supporting them in seeking contact (Box 1). required services are unavailable, community workers are expected to mobilize volunteers to set up services. In practice, such interventions are not always successful. Two community workers, for example, explained that an informal grocery delivery service they set up at older peo ple’s request remained unused because older people felt it would ‘threaten their sense of independence’ (CW) and were anxious about having ‘an unknown volunteer in their house’ (OP). After building relationships and familiarity, community workers are thus crucial in raising older people’s awareness of their (social) needs and capabilities, encouraging self- management, and facilitating informal support-giving. Normative integration throughout all levels Normative integration throughout all levels For older people, tender practices and policy changes often implied the rationing of publicly funded health and social care services and discontinuity in service delivery. Municipalities were similarly affected by a high degree of insecurity. Box 1: real-life case of an INA participant in Rotterdam. Challenge Community workers are told that the provision of high- quality support requires innovation and collaboration among community partners while being required to bureaucratically account for all actions and meet targets. Functional integration throughout all levels The risk of excessive professional autonomy Lack of support tools Functional integration throughout all levels The risk of excessive professional autonomy Professional autonomy provided by project management was at odds with guidance in adopting a new professional role that matched the INA’s core principles. The INA’s innovative character increased community workers’ need for guidance and supportive tools. The lack of material (i.e. decision-support tools or guidelines) and immaterial (i.e. acknowledgement) resources hampered the creation of shared values and aligned professional standards. In exchanging information, community workers often applied a ‘high touch, low tech’ approach. Rather than using the web-based portal developed for the INA, community workers preferred to consult each other by telephone or in person. Meso-level Community workers noted that conveying the INA’s message took time and that community members often hesitated to alert them to frail older persons, reluctant to interfere in someone’s life. INA community workers must take the preferences, and sometimes prejudices, of support-givers and those in need of support into account. The need for community integration requires professionals to reinvent their roles and serve as community workers. To overcome barriers to community integration, community workers perceived that sustaining relationships was crucial in gaining access to frail older people and adequately assessing potential support-givers. Recruitment of ‘entrepreneurial’ professionals with generalist and specialist skills to form diverse teams was crucial for professional integration. Discontinuity and a lack of mutual goals were found to hamper professional integration. Although health and social care organizations recognize the need to collaborate, professionals feel that cost containments are forcing the prioritization of organizations’ interests over the common good. Organizational integration was impeded by conflicting organizational interests and achieved only under favorable conditions, i.e. through a few willing professionals or managers and through high levels of trust built during previous collaborations. Structural incentives, such as the creation of opportunities for professionals to meet and gain insight in each other’s added value, facilitate organizational integration. Participants identified divergent flows of funds as the main cause for the lack of adequate financial incentives, affecting health and social care organizations and municipalities. Health and social care organizations urged the municipality to reconsider its accountability incentives, annoyed by the focus on how they do things. Contd. Art. 4, page 7 of 15 van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? Micro-level: service integration Engaging community resources 4, page 8 of 15 van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? distinct preferences, standards, and values, which must be considered carefully when providing support to older people: ‘Although Vreewijk is a very cohesive neighborhood, along the way we learned that they uphold the principle of “not washing your dirty laundry in public”, feeling most comfortable in leaving their concerns private’ (PM). Such norms may lead an older person to prefer a support-giver from a different apartment block or street due to fear of gossip. Furthermore, (cultural) differences between neighbors giving and receiving support (e.g. dif ferent expectations about support-giving intensity and tasks) may cause problems. One older woman in need of support explained that she knew the match would fail as soon as she saw her potential support-giver walking down the street. INA community workers must take the preferences, and sometimes prejudices, of support-givers and those in need of support into account. Once they find good matches, they notice improvements: ‘People who previously spent their time in their homes now come alive in the neighborhood. There was this isolated man, who now comes to our coffee morning every week’ (CW). older people), professional (seeking a multidisciplinary approach to support), and community (establishing a well- functioning network and engaging informal support- givers) levels. The INA project manager emphasized the divergence of these tasks: “Mobilizing the community is completely different from assessing what older people are capable of, which again is different from seeking informal support-givers, without having to throw in a gift card so that they feel valued for what they do. So we expect quite a lot of them”. (PM) Participants perceived generalist skills as indispensable, but a health care manager argued that older people may be more inclined to approach community workers based on their specialist, rather than generalist, backgrounds: ‘I’m not sure whether the ease and trust with which you reach people increases when you position yourself as “being everything” [. . .] I notice that people talk more eas ily to a caretaker on safety in the neighborhood than their care issues [. . .] The reverse is true as well; it’s easier to talk with a nurse about your prostate disturbances than with a caretaker’ (HCM). To ensure service integration, community resources must be integrated throughout the process of signaling and supporting older people. Meso-level: community integration Building community awareness and trust u d g co u ty a a e ess a d t ust Within the INA, community integration relies on community workers’ ability to generate community members’ awareness and trust. Community workers often faced skepticism related to ‘being yet another community project’ and about the INA’s main goals, as utilization of older people’s capabilities and informal networks was often perceived as a way to cut public spending. Moreover, participants wondered whether community members were ready to lose some personal autonomy ‘in favor of doing something for or with others’ (HCD). Community workers noted that conveying the INA’s message took time and that community members often hesitated to alert them to frail older persons, reluctant to interfere in someone’s life. Community members, for example, shared only ‘justified’ concerns about very frail older people in great need with INA community workers, instead of signaling related to the INA’s target population of those at risk of becoming (more) frail. Micro-level: service integration Engaging community resources Moreover, integrated care and support provision requires community workers to operate simultaneously at multiple levels. Community integration: what it takes from professionals The need for community integration requires profes sionals to reinvent their roles and serve as community workers. A health care organization manager identi fied this challenge as her greatest concern, wondering whether professionals would successfully attract informal support-givers and perceive collaboration with the com munity as a self-evident part of their working methods. INA community workers admitted that they struggled to shift from providing to facilitating support: ‘I find it very hard and contradictory to gain trust among older people on the one hand, while I should withdraw and facilitate support in the informal network on the other hand’ (CW). Community workers further argued that redirecting older people to professional networks or well-known volunteers was often less time consuming and more reliable than seeking informal support. Although they agreed that neighbors were willing to provide informal support, they emphasized the difficulty of appealing to this sense of willingness. Along the way, they have learned that people’s willingness to support one another is best addressed by articulating concrete, clear requests (e.g. asking whether someone is willing to bring groceries or provide assistance in the garden, rather than whether he/she is willing to do ‘something’ for someone else) and by preventing the excessive formalization of informal sup port, which would undermine its spontaneous and volun tary character. Micro-level: service integration Engaging community resources Rather than professional resources, INA community workers utilize locally available community resources and older people’s social networks as much as possible. They engage the community in supporting older people and alerting them to potentially frail individuals. When Service integration: what it takes from professionals The previous example illustrates that community workers must set up and track responses to interventions to support frail older people’s needs. They must play liaison roles at the personal (supporting and monitoring Mrs. Jansen, a 75-year-old, moved from a big house in a village-like neighborhood to a senior apartment block in an adjacent neighborhood at her children’s encouragement after her husband’s death 6 years previously. Although she initially enjoyed this new home, with nearby shops and well-organized activities, she had lost her sense of belonging and struggled to relate to newcomers: ‘new people are moving in who don’t even bother to say good morning or good evening[. . .] They pay so much attention to how you walk or how you dress your hair[. . .] I’m not like that. I’m just an ordinary woman’. After negative experiences (ridicule at coffee socials, avoidance of invitations to visit), Mrs. Jansen was reluctant to seek social contact, which she missed. She occasionally cried about her husband’s death and longed for someone to talk to. Mrs. Jansen was thus positively surprised when a community worker approached her in the apartment lobby; they sat together and Mrs. Jansen was able to share her story with a neutral person. At Mrs. Jansen’s agreement (and within a week), the community worker arranged for a neighbor to have coffee with her on Monday mornings, which pleases them both: ‘She tells me how happy she is having me over and I also feel very comfortable around her’. Mrs. Jansen explained that the community worker was essential in setting up this contact. She also appreciated the community worker’s updates about neighborhood activities (e.g. social activities and informational meetings, for example on current reforms in domestic help) and the ability to call someone she trusted whenever she needed support. She was more comfortable opening up to a professional than sharing with one of her hardworking children or neighbors, ‘who have worries of their own’. van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? Art. Familiarity with the neighborhood y g Neighborhood-specific familiarity with the preferences of support-givers and those in need of support is crucial for the successful engagement of community members in providing support. One community worker described difficulties in finding a neighbor willing to deliver bread weekly to an older man estranged from society: ‘the whole flat ignored him completely. Although there are quite a few people in that neighborhood supporting others, they seem unwilling to support a person living on the edge of society’ (CW). Neighborhoods may have Sustaining relationships as a prerequisite for community integration Sustaining relationships as a prerequisite for community integration To overcome these barriers to community integration, community workers perceived that sustaining relation ships was crucial in gaining access to frail older people and adequately assessing potential support-givers: ‘it’s about sustaining relationships, that’s why I go to the community center every week, to connect with people, Art. 4, page 9 of 15 Art. 4, page 9 of 15 van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? only then do they open up and become willing to collaborate’ (CW). Community workers emphasized that relationships were often person-specific and not easily transferred to other community workers. They thus advocated minimal weekly working hours and project durations to allow professionals to invest in integration among community members and other professionals: ‘At minimum it requires a year to get a grip on the neighborhood, your own role within INA and the working method of INA. After that you’re able to further refine it’ (CW). peoples’ sexual impulses[. . .] and the other community worker had great entrepreneurial energy, which I found very stimulating” (CW). Membership in a diverse community team seemed to generate more than the sum of its parts: ‘soon I started to feel that we could conquer the world[. . .] you learn to recognize symptoms that up until then weren’t natural for me’ (CW). ( ) Community workers, however, emphasized that team synergy could be achieved only when team members were receptive to professionals from other disciplines and were able to address relational issues that may hamper the establishment of mutual goals. Continuity within the team was thus perceived as a prerequisite for professional integration. One community worker explained that changes in team composition harmed collaboration: ‘Every time we needed to start from scratch, how do we communicate, what are our intentions?’ (CW). Moreover, community workers perceived the imposition of output criteria and targets as the greatest threat to collaboration: ‘If one of us generates a lot of clients, and the others don’t, it sure causes friction’ (CW). Community workers expressed concern about meeting the target of identifying frail older people: ‘Although we planned to go to a senior apartment block together, one of the community workers went there before; it sure makes you doubt whether there are any elderly people left for you’ (CW). Sustaining relationships as a prerequisite for community integration The establishment of team, rather than individual, targets may overcome this barrier. Community integration was thus found to rely on community workers’ ability to gain community members’ trust and the extent to which they became familiar with the neighborhood. Community integration further requires community workers to facilitate, rather than provide, support. Meso-level: professional integration Individual skills Meso-level: organizational integration Conflicting organizational interests Although health and social care organizations recog nize the need to collaborate, professionals feel that cost containments are forcing the prioritization of organi zations’ interests over the common good: ‘in times of reforming a structure, in times of insecurity concerning the survival of organizations, you have to save your own skin, and that’s when the power of the institute becomes way too large in the procedure’ (HCD). Although INA directors and managers displayed a lack of confidence in achiev ing organizational integration on an institutional level, due to their need to meet organizational targets in order to ‘survive’, they did feel that collaboration succeeds on an operational level on the basis of mutual under standing and acknowledgement. Although they seemed confident, community workers constantly noted that competition among professionals hampered organiza tional integration. In addition to expressing the gen eral fear of failing to meet their targets, profession als identified the ‘blurring’ of professional identities, i.e. lack of clear roles, as an important impediment to organizational integration; one community worker com mented: ‘I still find it very strange that community nurses Meso-level: professional integration Individual skills Professional integration starts with selecting appro priate people for the job. Although INA community workers were initially selected for their health and social care backgrounds and familiarity with neighborhoods, the project team learned that entrepreneurial skills were most important. Given the INA’s innovative and complex character, creative community workers who constantly tried new ways to actively reach frail older people and supportive community members most suc cessfully established integration. Community workers’ employment by health and social care organizations in addition to INA work sometimes hampered profes sional integration. For example, some professionals had to combine INA community work with other functions that necessitated more commercial approaches, which may ‘lead to a schizophrenic situation in which community workers have to unite a neutral with a commercial attitude; only a few succeed in’ (HET). The question of whether INA community work could best be accomplished by allo cating specific tasks–functionalities– to existing pro fessions or by creating new, specific INA professions was a recurrent theme during meetings. Most partners agreed that community workers should combine a generalist scope with specialist backgrounds, enabling determina tion of when (another) specialty is required to support an older person and ensuring high-quality person-centered support. Recruitment of ‘entrepreneurial’ professionals with generalist and specialist skills to form diverse teams was thus found to be crucial for professional integration in the support of older people with varying and complex needs. Although teams may generate more than the sum of their parts, discontinuity and a lack of mutual goals were found to hamper professional integration. Macro-level: system integration Inadequate financial incentives d f d d Participants identified divergent flows of funds as the main cause for the lack of adequate financial incentives, affecting health and social care organizations and municipalities: ‘When the municipality performs its tasks regarding the Social Support Act [The Social Support Act took effect in 2007 and requires municipalities to meet increasing legal responsibilities regarding support of people (with disabilities)] well and arranges prevention properly, it won’t benefit the municipality; it will only lead to lower expenditures among health insurers. Well, do you think that’s of any interest to the average civil servant?’ (HI). Lack of organizational commitment Community workers were equally disappointed in their organizations’ and managers’ lack of engagement dur ing the project. They were seldom asked about their INA experiences, and meetings called by management merely involved elaboration on practical issues, such as sick leave or investment of time in the INA. One community worker felt appreciated only ‘for delivering clients through the project’, whereas she had hoped her INA experience would foster innovation in her organization. Similarly, the INA project manager stated that he had placed too much trust in health and social care organizations’ com mitment. Furthermore, he identified a lack of structural incentives that would generate organizational integra tion; during an advisory group meeting within the INA context, ‘they kept going on about who was responsible for which domain and about the sense of competition or col laboration among health and social care [. . .] And suddenly it struck me that there was no other meeting where they encountered each other’ (PM). Thus, successful partner ships often involved willing professionals or managers or depended on high levels of trust built through previ ous collaboration. This specifically accounted for the INA engagement only of general practitioners who felt affiliated with the need to support community-dwelling older people. Managers’ active interference was felt to promote organizational integration. For example, when INA community workers indicated that community nurses from a similar but more health care–oriented program perceived them as valuable only when no other way to support an older person had been found, both program managers held a meeting to integrate services provided by community workers and nurses. The managers’ expres sion of mutual commitment to collaboration, through organization of this meeting and on-site articulation of their engagement, made community workers perceive collaboration as an indispensable part of their job. Team skills To facilitate professional integration, community teams must incorporate various specialties, ‘combining their skills to ensure a generalist and holistic approach’ (PM). The availability of an appropriate range of skills and expertise on a team was perceived as a prerequisite for professional integration, and particularly relevant for the INA’s focus on improving overall well-being. One community worker, for example, commented: ‘I brought my knowledge about health care to the table and how to approach older people [. . .] And I taught the social worker how to cope with older van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? Art. 4, page 10 of 15 van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? people to INA workers enabled professionals to see their complementary values. [community workers with similar but more health care– related tasks] are getting all these extra tasks. They may say the former community nurse did the same, but then they’re talking about the 50s, when the milkman put the bottles on the curb; it’s a completely different and complicated world now’ (CW). Ill-defined roles not only led to confusion among older people and professionals, it also encour aged INA community workers to constantly explain and justify their roles, even within their own organizations. This sense of competition hampered INA community workers’ provision of support to older people; one community worker explained that she was opposed by an activity coordinator when she tried to organize activities in a flat that was crowded with isolated older people: ‘We had put an INA folder in the mailbox, which made the activity supervisor very irritated. She argued that they didn’t need it and that they had their own activities[. . .]but it’s a three-by-three apartment with no balcony, it’s like a prison’ (CW). Within the INA, organizational integration was thus impeded by conflicting organizational interests and achieved only under favorable conditions, i.e. through a few willing professionals or managers and through high levels of trust built during previous collabora tions. Structural incentives, such as the creation of opportunities for professionals to meet and gain insight in each other’s added value, facilitate organizational integration. Macro-level: system integration Inadequate financial incentives d f d d But when the job is to attract as many volunteers to empower social networks, then you should provide them the necessary space to do so’ (SUBALD). Instead of focusing on process, creating a bureaucratic accountability system, many participants would prefer the municipality to promote results and focus on result-oriented indicators. Lack of support tools pp The lack of clear interventions or decision support tools paralyzed community workers, forcing them to rely on usual working methods. They were expected to develop support plans, but the process had not been fine-tuned for everyday practice. Given the lack of tools and guide lines to support community workers’ decision making, the plans became a formality instead of a supportive tool. Team meetings neither were a resource for aligning pro fessional standards or gaining confidence in the value of the work. Whereas most community workers needed to describe their struggles and discuss community issues, meetings were focused predominantly on practical issues (e.g. whether targets were being met): ‘I have a strong need to get inspired and informed. I wonder why we don’t discuss such things, I read stuff in the local newspaper which I think we should address and which is being addressed by residents in the community centers’ (CW). Discussion of local and broader (e.g. transitions in municipal and central government) issues would also contribute to workers’ understanding of the context in which they operated, fostering their sense of purpose. Encouraged by INA’s education team, the project manager decided to include discussions of successful cases/situations and those with which community workers struggled in team meetings. For example, one community worker was reassured that she was allowed to spend 2 hours on a bench in front of the supermarket if it facilitated her acquaintance with the neighborhood and its (older) inhabitants. Inadequate regulatory incentives Paradoxically, professionals experienced similar restric tions. Community workers are told that the provision of high-quality support requires innovation and collabora tion among community partners while being required to bureaucratically account for all actions and meet targets. A health care director and a health insurer expressed the wish that professionals would seek ways around these constraints, taking the system and its operational rules rather lightly. Macro-level: system integration Inadequate financial incentives d f d d Moreo ver, regular discussion of clients or provision of feedback to professionals who had identified potentially frail older ( ) The health insurer and municipal officers argued that incentives should ensure that incremental improve ments bring economic benefits for all stakeholders, facilitating work toward the same goal, i.e. integrated care and support provision to older people. Current financial systems lack stimuli for innovation; the health insurer commented: ‘I’d also prefer to build in a reward for innovative behavior. And that’s very complicated, what you basically see is that those who act last in this transition, or focus on its production, win this race finan cially’ (HI). During meetings held to discuss whether and how the INA could be sustained after project funding ended, partners looked to each other with the hope of (financial) commitment. Participants emphasized the need for broader (financial) commitment to sustain approaches such as the INA: ‘On the content, we agree with each other, but there are other sides to consider as well. The question is whether we can commit ourselves jointly [. . .] if you ask an individual organization if they’re willing to commit, while having to cut back intensely. . . it demands a broader approach in which all parties commit’ (HCD). Inadequate accountability incentives y Similarly, health and social care organizations urged the municipality to reconsider its accountability incentives, annoyed by the focus on how they do things: ‘They use accountable performance indicators such as the amount of hours spent. . . because that’s the most measurable aspect [. . .] When I’m talking with a municipal officer, 90% of the conversation turns to talking about a spreadsheet with the amount of hours of our employees’ (SCD). A sub-alderman argued that this focus compromised municipalities’ Art. 4, page 11 of 15 van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? interests, namely the need to innovate and empower citizens to participate: ‘we currently steer on “fifteen minutes of this and fifteen minutes of that [. . .] and it should also meet these and these conditions”; so we’re very much on the details of how to do it. Macro-level: system integration Inadequate financial incentives d f d d The health care director explained how he would like two community nurses from different organi zations to collaborate in the community: ‘I actually hope they’re being smart about it, like “you know what, I’ll give you a hand, or I’ll do it for you this time”. Without having to send an invoice on either side’ (HCD). However, operational rules seem to restrain professionals’ autonomy on such occasions. On a macro-level, the INA was affected by the system’s failure to provide adequate financial, regulatory, and accountability incentives. Current system incentives lack a clear division of tasks and fail to generate broad engage ment. To enable successful integrated care and support efforts, incentives should carefully anticipate the needs for innovation and collaboration. This approach requires financial incentives that account for aligned incremental improvements and accountability measures that provide professional autonomy. In exchanging information, community workers often applied a ‘high touch, low tech’ approach. Rather than using the web-based portal developed for the INA, community workers preferred to consult each other by telephone or in person. These ‘short lines of communication’ (CW) were considered to be most valuable for team collaboration. One community worker, however, expressed her preference for a handheld tablet to assist with field work: ‘I can’t bring all my paperwork on the street. Give me an iPad and I can access all the information: which volunteer is available, for example. I’m racking my brains out there’ (CW). Discussion This study showed that integrated care and support provision through an INA is a complex, dynamic pro cess requiring multilevel alignment of activities [18]. The INA achieved integration at the personal, service, and professional levels only occasionally. Micro-level bottom-up initiatives were not aligned with top-down incentives, forcing community workers to establish integration despite rather than because of meso- and macro-level contexts. Functional and normative integra tion were lacking, with excessive reliance on professionals to achieve integration. Incoherent macro-level policies have been identified as main barriers to the pursuit of integration. Current sys tem incentives are not aligned to achieve collaboration and innovation and do not account for the complexity and nature of issues arising locally. In line with previous findings [11], health and social care partners identified divergent flows of funds and the lack of joint budgets as significant obstacles to collaboration. Current perfor mance indicators prioritize accountability and control, rather than creating a learning environment that allows partners to try innovative approaches [25]. Thus, health and social care partners advocate that the government is ‘tight on ends and loose on means’. However, munici pal officers and health insurers expressed concern that allowing local variations in means may cause (frail older) people (to whom they are legally required to provide support) to ‘fall through the cracks’. Tender practices also generated mistrust among health and social care professionals. Many professionals com mented that they did not understand ‘why the municipality first imposed major cutbacks’ (CW), leading to community center closures and job losses among very experienced community workers, and then forced them to rebuild ser vices. These practices drew energy away from the support of older people through the INA. The project manager argued that this situation paralyzed community workers and prevented the INA from making a real transition: ‘It caused a standstill. The community workers were caught by insecurity and passivity for at least half a year. There was only room for bereavement’ (PM). The INA’s expectations concerning deprofessionalization further increased professionals’ mistrust, causing a conflict of loyalty toward the INA. Municipalities were similarly affected by a high degree of insecurity: ‘Until January 2015 we won’t know how much money we’ll get from the state[. . Functional integration throughout all levels The risk of excessive professional autonomy The INA’s innovative character, specifically with respect to active community engagement, created a paradoxical situation. The project leader gave community workers autonomy to create their own working methods, with no guideline or restriction on how they spent their hours. This autonomy was a main motivation to become an INA community worker, as professionals missed it in their regular jobs. However, joint training conducted 1.5 years after INA initiation revealed a discrepancy between community workers’ and project management’s perspectives on core tasks. The trainer concluded that community workers did not yet perceive community engagement and a facilitating role as self-evident parts of their job. She stated that community workers remained ‘bound by the conventional way of organizing things, i.e. from the perspective of helping/fixing problems’ (HET). Professional autonomy provided by project manage ment was at odds with guidance in adopting a new pro fessional role that matched the INA’s core principles. The INA’s innovative character increased community work ers’ need for guidance and supportive tools. The lack of material (i.e. decision-support tools or guidelines) and immaterial (i.e. acknowledgement) resources hampered the creation of shared values and aligned professional standards. Normative integration throughout all levels The dynamic environment in which the INA oper ated seemed to overshadow the urgency to facilitate an integrated mind-set. Rotterdam’s use of competitive tender practices to appoint (new) providers and award Art. 4, page 12 of 15 van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? legislation, which makes no sense’ (MO). Paradoxically, municipalities’ constant tendency to control and prevent risks so that frail people don’t ‘fall through the cracks’ causes mutual distrust, undermining collaboration and innovation; a municipal program manager commented: ‘We face very complex strategic decisions, and of course there is no mutual trust. It seems very simple, but trust in one another is a key driver in this sector; are we actually supporting people who are in need or are we just earning money on their backs?’ (MO). The widespread culture of accountability thus causes organizations to focus on their own interests instead of committing to an integrated mind-set that focuses on the best interests of (frail) citizens. contracts impacted INA organizations and commu nity workers. Insecurity and mistrust For older people, tender practices and policy changes often implied the rationing of publicly funded health and social care services and discontinuity in service delivery. Older people thus have become insecure and feel that they are burdening society: ‘in roughly one and a half years they restructured all home care services. . . And they may argue that volunteers will cover those things that remain to be done, but we must wait to see who’s coming [. . .] It feels like we don’t matter anymore’ (OP). The INA’s anticipation of these transformations by shifting respon sibilities back to the community frightened older people and confirmed the idea that the INA was ‘no more than a hidden economic measure’ (OP). Furthermore, based on previous experiences, older people associated the INA with a negative form of social control: ‘The problem is that those of the younger generation are not familiar with a cohesive community and I think that those from the older generation who still remember that world can´t relate to it anymore’ (HCD). Functional integration throughout all levels The risk of excessive professional autonomy Although these practices and other policy changes (mainly in home care) aimed to increase the efficiency of integrated care and support provision, they created marked insecurity, impeding the INA’s ability to generate multilevel integration. Conclusions Thi d This study enabled us to identify factors facilitating and inhibiting integration within and among levels defined by Valentijn and colleagues [19]. This integrated care model enabled us to acquire a rich understanding of the INA’s underlying processes, which most integrated care and support initiatives fail to do. Although this model, like most integrated care models, focuses predominantly on the improvement of health outcomes, instead of aim ing to improve overall well-being, it was useful for the detection of contextual factors and mechanisms that may hinder or facilitate an INA. However, our findings highlighted the need for further refinement of the model by adding the community level [33]. This level often is not ‘incorporated in our theorising on integrated care’, as Nies [8, p. 3] and Goodwin [9] recently remarked. Our study indicated that this community level is indispen sable in engaging community members and resources to meet older people’s needs. Given the general shift in the primary provision of (social) care from the state to the community, community engagement is increas ingly essential. Our research identified several barriers to the pursuit of community integration. To overcome these barriers, neighborhood-specific familiarity with the preferences of support-givers and those in need of support may be crucial for the successful engagement of the community. Our study also enhanced our understanding of the importance of normative inte gration in INA development. Relational and normative aspects may be best accounted for in what Goodwin [9, p. 2] describes as a culturally sensitive approach; an approach that aims to build community aware ness and trust among formal and informal partners. Through our multifaceted and thorough description of the experiences of diverse INA partners, we were able to test Valentijn and colleagues’ (2013) integrated care model and provide a richer account of its impli cations. These findings are especially important in a time of ageing populations and a general shift in the primary provision of (social) care from the state to the community. Our study also revealed a lack of functional integration. Material and immaterial support tools were insufficient for the creation of shared values and aligned professional standards. Although a protocol-driven approach would conflict with the need to provide tailored care to older people with complex needs, the INA’s innovative char acter increased the need to support change and direct professionals toward mutually agreed-upon objectives and practices [11]. Discussion .]But what’s even more fundamental, is that the Bill of the Social Support Act won’t be ready until mid-2014, and that should pro vide us with the instructions and the conditions under which we must operate. But by that time our procurement should have long been realized. So that’s a very strange situation’ (MO). In an interview conducted 2 days before his resignation, a municipal officer described the result ing risk-averse culture within municipalities, which prohibited ‘thinking out of the box and trying innovative approaches’ (MO). Although municipalities shift respon sibilities to (social) care organizations and communi ties, they concurrently try to retain top-down control; the same municipal officer commented: ‘we supposedly have marketed it, but on the other hand, we still held on to This tendency to control and prevent risks while being in need of innovation and collaboration affected the pro fessional and organizational levels. Although managers and directors were confident that professionals would seek ways around system constraints, our research dem onstrates that professional and organizational collabo ration requires appropriate structural incentives. The creation of opportunities for professionals and managers to meet and gain insight into their complementary roles is crucial. Without an aligned macro-level policy narra tive, bottom-up initiatives such as the INA will struggle to make impacts. Overcoming these macro-level barriers is necessary – but not sufficient – for integration [18, 21, 26, 27]. The lack of normative integration fundamentally prevented the INA’s integration of care and support. The rate and Art. 4, page 13 of 15 van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? complexity of current reforms were detrimental to estab lished community relationships and generated high levels of mutual distrust and insecurity throughout system levels. Professionals and organizations re-focused energy on individual interests [26, 28, 29], rather than working toward the common goal of improving care and support for older people. In line with previous findings, such dynamic environments hampered the development of an innovative culture [30]. time, continuity, and broad commitment throughout levels, with evolution toward aligned norms and practices. Moreover, we demonstrate that the community level should be included in integrated care and support models, as specific (social) community characteristics must be considered when improving community-based integrated care and support. Competing Interests Competing Interests The authors declare that they have no competing interests. Conclusions Thi d Support tools must be responsive to professionals’ struggles and the need for innovation while respecting professional autonomy and diversity. Although not addressed in many integrated care and support models, the community level was found to be critical in engaging community members and resources when meeting older people’s needs. Our study indicated the importance of community workers’ understanding of community standards and norms. Furthermore, pro fessionals struggled to perceive community members’ roles as integral to the support-giving process [33]; guidance of professionals in engaging informal support-givers is thus crucial in promoting community integration. Our study revealed clear barriers to informal support, suggesting that its provision and receipt require a paradigm shift toward more natural occurrence and self-evidence. Although this study provides knowledge about factors that promote or hinder integration at the micro-, meso- and macro-levels, the context-specific nature limits the generalizability of its findings. However, we feel that our detailed and multi-faceted description of diverse INA partners’ experiences provides useful insights for future research. The INA took place in a highly dynamic envi ronment with intense external forces, which impacted the success of integrated care and support provision. Further research should account for interactions between external factors and local integrated care and support delivery processes from all perspectives including the perspectives of older people. Successful integration within a complex program such as the INA requires Discussion Future research should also focus on the development of validated measurement tools to assess the ‘strength’ of integration throughout levels and its impact on (cost) effectiveness. To promote normative integration, trust may be more determinant than streamlined structures [31]. Trust was a recurrent theme at the personal, community (as a prereq uisite for older people’s and community members’ com mitment) and professional (as a pre-existing factor built through previous collaboration that enabled professional integration) levels. These findings emphasize the impor tance of continuous relationships that allow the develop ment of trust and social capital in pursuing integration [1, 31, 32]. Restructuring efforts may cause ‘cultural damage’ by undermining the importance of trust and relationships for normative integration [29]. References 1. Humphries, R and Curry, N. Integrating health and social care. Where next? London: The King’s Fund; 2011. 1. Humphries, R and Curry, N. Integrating health and social care. Where next? London: The King’s Fund; 2011. 14. Lowndes, V and Sullivan, H. How low can you go? Rationales and challenges for neighbourhood governance. Public Administration. 2008; 86(1): 53–74. DOI: http://dx.doi.org/10.1111/j.1467-9299. 2007.00696.x 2. Pavolini, E and Ranci, C. Restructuring welfare states: reforms in long-term care in Western European countries. Journal of European Social Policy. 2008; 18(3): 246–259. DOI: http://dx.doi.org/10.1177/ 0958928708091058 15. Goodwin, N, Sonola, L, Thiel, V and Kodner, DL. Co-ordinated care for people with complex chronic conditions. Key lessons and markers for success. London: The King’s Fund; 2013. 3. Daly, M and Lewis, J. The Concept of Social Care and the Analysis of Contemporary Welfare States. British Journal of Sociology. 2000; 51(2): 281–98. DOI: http://dx.doi.org/10.1080/00071310050030181 16. van Dijk, HM. Neighbourhoods for ageing in place. Thesis. Rotterdam: Erasmus University Rot terdam; 2015. 4. Tonkens, E. The embrace of responsibility: citi zenship and governance of social care in the Netherlands. In Newman, J and Tonkens, E (Eds.), Partic ipation, responsibility and choice: summoning the active citizen in Western European welfare states. Amsterdam, the Netherlands: Amsterdam University Press. 2011; pp. 45–65. 17. Cramm, JM, van Dijk, H, Lötters, F, van Exel, J and Nieboer, AP. Evaluation of an integrated neighbourhood approach to improve well-being of frail elderly in a Dutch community: a study protocol. BMC Research Notes. 2011; 4: 532. DOI: http:// dx.doi.org/10.1186/1756-0500-4-532 5. Verhoeven, I and Tonkens, E. Talking Active Citizenship: Framing Welfare State Reform in Eng land and the Netherlands. Social Policy and Society. 2013; 12(3): 415–426. DOI: http://dx.doi.org/ 10.1017/S1474746413000158 18. Goodwin, N, Dixon, A, Anderson, G and Wodchis, W. Providing integrated care for older people with complex needs. London: The King’s Fund; 2014. 19. Valentijn, P, Schepman, S, Opheij, W and Bruijnzeels, M. Understanding integrated care: a comprehensive conceptual framework based on the integrative functions of primary care. International Journal of Integrated Care. 2013; 13: 1–12. Available from URN:NBN:NL:UI:10-1-114415. 6. Grootegoed, E and van Dijk, D. The return of the family? Welfare state retrenchment and client autonomy in long-term care. Journal of Social Policy. 2012; 41(4): 677–694. DOI: http://dx.doi. org/10.1017/S0047279412000311 20. Kodner, DL. All together now: a conceptual exploration of integrated care. Healthcare Quarterly. 2009; 13: 6–15. DOI: http://dx.doi.org/10.12927/ hcq.2009.21091 7. Anderson, GF and Hussey, PS. Reviewers Judith Sixsmith, Professor of Public Health Improvement and Implementation, University of Northampton, UK. Judith Sixsmith, Professor of Public Health Improvement and Implementation, University of Northampton, UK. 13. Plochg, T and Klazinga, NS. Community-based integrated care: myth or must? International Journal for Quality in Health Care. 2002; 14(2): 91–101. DOI: http://dx.doi.org/10.1093/oxfordjournals. intqhc.a002606 One anonymous reviewer. Acknowledgements The research was supported by a grant provided by the Netherlands Organization for Health Research and Devel opment (ZonMw, project number 314030201). The views Art. 4, page 14 of 15 van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? expressed in the paper are those of the authors. The authors declare that they have no competing interests. expressed in the paper are those of the authors. The authors declare that they have no competing interests. 12. Morikawa, M. Towards community-based integrated care: trends and issues in Japan’s long-term care policy. International Journal of Integrated Care. 2014; 14: 1–10. Available from URN:NBN:NL:UI:10-1-114772. References Population aging: a comparison among industrialized countries. Health Affairs. 2011; 19: 191–203. DOI: http://dx.doi. org/10.1377/hlthaff.19.3.191 21. Petch, A. Delivering integrated care and support. Glasgow: IRISS; 2013. 8. Nies, H. Communities as co-producers in integrated care. International Journal of Integrated Care. 2014; 14: 1–4. Available from URN:NBN:NL:UI:10-1-114789. 22. Hsieh, H-F and Shannon, SE. Three approaches to qualitative content analysis. Journal of Qualitative Health Research. 2005; 15(9): 1277–1288. DOI: http://dx.doi.org/10.1177/1049732305276687 9. Goodwin, N. Thinking differently about integration: people-centred care and the role of local communities. International Journal of Integrated Care. 2014; 14(3). DOI: http://dx.doi.org/10.5334/ijic.1736 23. Kondracki, NL and Wellman, NS. Content analysis: review of methods and their applications in nutrition education. Journal of Nutrition Educa tion and Behavior. 2002; 34: 224–230. DOI: http:// dx.doi.org/10.1016/S1499-4046(06)60097-3 10. Nieboer, AP. Sustainable care in a time of crisis. Inaugural lecture. Rotterdam: Erasmus University Rotterdam; 2013 [cited 2016 9 march]. Available from http://www.bmg.eur.nl/fileadmin/ASSETS/ bmg/Onderzoek/Oraties/Nieboer/Oratie_Anna_ Nieboer.pdf. 24. Graneheim, UH and Lundman, B. Qualitative content analysis in nursing research, concepts, pro cedures and measures to achieve trustworthiness. Nursing Education Today. 2004; 24: 105–112. DOI: http://dx.doi.org/10.1016/j.nedt.2003.10.001 11. Leichsenring, K. Developing integrated health and social care services for older persons in Europe. International Journal of Integrated Care. 2004; 4(10): 1–15. DOI: http://dx.doi.org/10.5334/ijic.107 25. Ham, C and Walsh, N. Making integrated care happen at scale and pace. London: The King’s Fund; 2013. Art. 4, page 15 of 15 van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? van Dijk et al: How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? organizational characteristics. Health Care Manage ment Review. 2012; 37(2): 165–174. DOI: http:// dx.doi.org/10.1097/HMR.0b013e318222416b 26. Cumming, J. Integrated care in New Zealand. Inter national Journal of Integrated Care. 2011; 11: 1–13. DOI: http://dx.doi.org/10.5334/ijic.678 organizational characteristics. Health Care Manage ment Review. 2012; 37(2): 165–174. DOI: http:// dx.doi.org/10.1097/HMR.0b013e318222416b 27. Glendinning, C. Breaking down barriers: integrating health and care services for older people in England. Health Policy. 2003; 65: 139–151. DOI: http://dx.doi. org/10.1016/S0168-8510(02)00205-1 31. Williams, P and Sullivan, H. Faces of integration. International Journal of Integrated Care. 2009; 9(22): 1–13. DOI: http://dx.doi.org/10.5334/ijic.509 32. Nolan, M, Davies, S and Brown, J. Transitions in care homes: towards relationship-centred care using the ‘Senses Framework’. Quality in Ageing. 2006; 7(3): 5–15. DOI: http://dx.doi. org/10.1108/14717794200600015 28. Demers, L. Mergers and integrated care: the Quebec experience. References International Journal of Integrated Care 2013; 13: 1–4. Available from URN:NBN:NL: UI:10-1-114229. 29. Hudson, B. Ten years of jointly commissioning health and social care in England. International Jour nal of Integrated Care. 2011; 7: 1–9. Available from URN:NBN:NL:UI:10-1-101296. 33. van Dijk, HM, Cramm, JM and Nieboer, AP. The experiences of neighbour, volunteer and professional support-givers in supporting community dwelling older people. Health and Social Care in the Community. 2013; 21: 150–158. DOI: http://dx.doi. org/10.1111/hsc.12006 30. Nieboer, AP and Strating, MMH. Innovative cul ture in long-term care settings: the influence of How to cite this article: van Dijk, H M, Cramm, J M and Nieboer, A P 2016 How To Build an Integrated Neighborhood Approach to Support Community-Dwelling Older People? International Journal of Integrated Care, 16(2): 4, pp. 1–15, DOI: http://dx.doi. org/10.5334/ijic.1596 Submitted: 30 July 2014 Accepted: 25 April 2016 Published: 12 May 2016 Copyright: © 2016 The Author(s). This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International License (CC-BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. See http://creativecommons.org/licenses/by/4.0/. OPEN ACCESS International Journal of Integrated Care is a peer-reviewed open access journal published by Ubiquity Press. OPEN ACCESS
https://openalex.org/W4362486548	https://figshare.com/articles/journal_contribution/Figure_S4_from_AP-1_Signaling_by_Fra-1_Directly_Regulates_HMGA1_Oncogene_Transcription_in_Triple-Negative_Breast_Cancers/22513309/1/files/39975334.pdf	English	null	Figure S7 from AP-1 Signaling by Fra-1 Directly Regulates HMGA1 Oncogene Transcription in Triple-Negative Breast Cancers	null	2,023	cc-by	201	Tolza et al. - Supplementary Figure S4 Tolza et al. - Supplementary Figure S4 GM12878 Hi-C data from Rao et al., 2014 HMEC HUVEC K562 HMGA1 Fra-1 Fra-2 HMGA1 Supplementary Figure S4 : Contact matrices from Hi-C data from on the Chr6 33,724,700-34,948,488 region that contain the HMGA1 gene (black rectangle). The data are taken from Rao et al., 2014 (54). The positions of Fra-1 and Fra-2 binding sites located 350 kb upstream and 685 kb downstream of HMGA1 TSS in MDA-MB-231 cells are indicated by doted arrows. Fra-1 Fra-2 Fra-1 Fra-2 Fra-1 Fra-2 HMGA1 Fra-1 Fra-2 HMGA1 Fra-1 Fra-2 Fra-1 Fra-2 Fra-1 Fra-2 GM12878 HMEC HMGA1 Fra-1 Fra-2 HMGA1 Fra-1 Fra-2 Fra-1 Fra-2 Fra-1 Fra-2 Hi-C data from Rao et al., 2014 HUVEC K562 HMGA1 Fra-1 Fra-2 HMGA1 Fra-1 Fra-2 Fra-1 Fra-2 Fra-1 Fra-2 Hi-C data from Rao et al., 2014 Supplementary Figure S4 : Contact matrices from Hi-C data from on the Chr6 33,724,700-34,948,488 region that contain the HMGA1 gene (black rectangle). The data are taken from Rao et al., 2014 (54). The positions of Fra-1 and Fra-2 binding sites located 350 kb upstream and 685 kb downstream of HMGA1 TSS in MDA-MB-231 cells are indicated by doted arrows.
https://openalex.org/W2096771547	https://www.scielo.br/j/jbchs/a/t9CnLKS7jL3stLxfyGnpmYC/?lang=en&format=pdf	English	null	Green and roasted arabica coffees differentiated by ripeness, process and cup quality via electrospray ionization mass spectrometry fingerprinting	Journal of the Brazilian Chemical Society	2,009	cc-by	6,332	aInstituto de Química, Universidade Federal do Rio de Janeiro, 21945-970 Rio de Janeiro-RJ, Brazil bInstituto de Química, Universidade Estadual de Campinas, 13083-970 Campinas-SP, Brazil cEmbrapa Agroindústria de Alimentos, 23020-470 Rio de Janeiro-RJ, Brazil e Ciências Médicas, Universidade Estadual de Campinas, CP 6111, 13083-887 Campinas-SP, Braz A habilidade da técnica de espectrometria de massas com infusão direta e ionização por eletronebulização (IES-EM), nos modos de íons positivos e negativos, foi avaliada na diferenciação de cafés Arábica verdes e torrados e com diferentes estágios de amadurecimento (verde, maduro e passado), processo pós-colheita (seco, úmido e semi-úmido) e cafés classificados por prova de xícara. No modo negativo, a análise dos cafés verdes mostrou que os íons correspondentes aos ácidos graxos e ácidos clorogênicos desprotonados são os mais importantes para a discriminação da maturidade. No modo positivo, a maturidade é diferenciada através de íons correspondentes a cafeína, ácidos clorogênicos protonados e adutos de K+ de ácidos graxos. Na diferenciação da pós-colheita, em ambos os modos de ionização, são mais importantes os íons correspondentes aos ácidos graxos, ácidos clorogênicos, açúcares e ácidos carboxílicos formados da fermentação. Cafés Arábica torrados também são discriminados com eficiência. No modo negativo, são importantes os íons correspondentes aos ácidos clorogênicos e ácidos orgânicos de cadeia curta, derivados de açúcares. No modo positivo, a discriminação é realizada por íons de baixa m/z tais como piridina e alquil piridinas protonadas, formadas através da degradação da trigonelina. Ambos os IES(+)-EM e IES(-)-EM são capazes de discriminar diferentes cafés Arábica torrados classificados por prova de xícara e os íons que permitem esta diferenciação são os mesmos descritos para a maturidade e processos pós-colheita. Direct infusion electrospray ionization mass spectrometry in both the negative ESI(-)-MS and positive ESI(+)-MS ion modes are investigated to differentiate green and roasted Arabica coffees with different stages of ripeness (green, ripe and overripe), post-harvesting process (dry, wet and semi-wet) and coffees with different cup qualities. In the ESI(-)-MS of green coffees, ions from deprotonated fatty acids and chlorogenic acids are the most important for ripeness discrimination. In the ESI(+)-MS, maturity is differentiated by ions from protonated caffeine, chlorogenic acids and K+ adducts of fatty acids. To differentiate between post-harvesting process in both ionization modes, ions from fatty acids, chlorogenic acids, sugars and carboxylic acids generated in the fermentation process are the most representative. Roasted Arabica coffees are also well discriminated: in the ESI(-)-MS, ions from chlorogenic acids and short-chain organic acids derived from sugars are important. In the ESI(+)-MS, discrimination are mainly performed by low m/z ions such as protonated pyridine and alkylpiridines formed via trigonelline degradation. aInstituto de Química, Universidade Federal do Rio de Janeiro, 21945-970 Rio de Janeiro-RJ, Brazil bInstituto de Química, Universidade Estadual de Campinas, 13083-970 Campinas-SP, Brazil cEmbrapa Agroindústria de Alimentos, 23020-470 Rio de Janeiro-RJ, Brazil Both ESI(+)-MS and ESI(-)-MS are able to differentiate cup quality for Arabica roasted coffees and the ions used to perform discrimination are the same ones described in ripeness and post-harvesting processes. Keywords: Arabica coffee, ripeness, post-harvest, cup quality, ESI-MS fingerprinting e-mail: crezende@iq.ufrj.br J. Braz. Chem. Soc., Vol. 20, No. 2, 313-321, 2009. Printed in Brazil - ©2009 Sociedade Brasileira de Química 0103 - 5053 $6.00+0.00 J. Braz. Chem. Soc., Vol. 20, No. 2, 313-321, 2009. Printed in Brazil - ©2009 Sociedade Brasileira de Química 0103 - 5053 $6.00+0.00 Article Ana Carolina L. Amorim,a Ana Maria C. Hovell,a Angelo C. Pinto,a Marcos N. Eberlin,b Neusa P. Arruda,a Elenilda J. Pereira,a Humberto R. Bizzo,c Rodrigo R. Catharino,d Zenildo B. Morais Filhoa and Claudia M. Rezende,a aInstituto de Química, Universidade Federal do Rio de Janeiro, 21945-970 Rio de Janeiro-RJ, Brazil bInstituto de Química, Universidade Estadual de Campinas, 13083-970 Campinas-SP, Brazil cEmbrapa Agroindústria de Alimentos, 23020-470 Rio de Janeiro-RJ, Brazil Green and Roasted Arabica Coffees Differentiated by Ripeness, Process and Cup Quality via Electrospray Ionization Mass Spectrometry Fingerprinting Ana Carolina L. Amorim,a Ana Maria C. Hovell,a Angelo C. Pinto,a Marcos N. Eberlin,b Neusa P. Arruda,a Elenilda J. Pereira,a Humberto R. Bizzo,c Rodrigo R. Catharino,d Zenildo B. Morais Filhoa and Claudia M. Rezende*,a J. Braz. Chem. Soc., Vol. 20, No. 2, 313-321, 2009. Printed in Brazil - ©2009 Sociedade Brasileira de Química 0103 - 5053 $6.00+0.00 Introduction There are several species in the genus Coffea (Rubiaceae), but Coffea arabica and Coffea canephora are the two most commercialized around the world, which are commonly known as Arabica and Robusta coffees, respectively. Coffee is the most popular beverage in the world and a commodity of extreme importance to developing countries. A considerable amount of research has been undertaken on the chemical composition and flavors in green and Green and Roasted Arabica Coffees Differentiated by Ripeness, Process and Cup Quality 314 J. Braz. Chem. Soc. roasted coffees, which helps to understand botanical, agricultural and process influences on coffee quality and also to tentatively assure origin authentication and avoid adulteration.1 Most studies on differentiation deals with Arabica and Robusta green and roasted coffees in relation to their major chemical composition such as metal,2 lipids focused on sterolic and alcohol diterpenes,3 alkaloids,4 chlorogenic acids5 and sugar profiles.6 Chemometric discrimination using principal component analysis (PCA) has been successfully used as an important tool to differentiate roasting processes,7 geographical origin of green and roasted coffees8 and other aspects of coffee process. To search for faster methods for coffee discrimination based on fingerprinting analysis, NIR and NMR studies were also applied.9,10 and the harvest was made by strip-picking with the cherries being collected on a sheet. The harvested cherries were then sorted and cleaned. The green (immature) and overripe (overripe) cherries were transferred to dry on a sun terrace. The harvested ripe coffee cherries were split in three parts, one part being dry processed in the same way than green and overripe cherries (ripe or dry), and the other two parts were semi-wet and wet processed. In the semi-wet process, the pulp was machine removed and the resulting cherry was transferred to dry on a sun terrace (semi-wet). In the wet process the pulp was also machine removed from the cherries, after which the cherries were fermented in tanks with water for 16 h. After this, the coffee cherries were thoroughly washed and transferred to dry on a sun terrace (wet). All samples were dried for 5 days on the sun terrace, with temperatures ranging from 35 to 40 ºC and were finally dried in a hot air wood burner until reaching ± 11% moisture (hot air temperature of 50 ºC and coffee temperature of 40 ºC). A new and potential technique is direct infusion electrospray ionization mass spectrometry (ESI-MS). Introduction ESI-MS employs little sample preparation and gives immediate compositional information for ESI-ionizable compounds. It has been very efficient for fingerprinting complex mixtures such as that of soybean extracts,11 tea,12 spices,13 vegetable oil14 and more recently, it has been applied to investigate the differences between Arabica and Robusta defective green coffees.15i Coffee samples were roasted in a convection oven at 155 ºC for 15 min and stored in sealed plastic bags in a freezer at −8 ºC until used. Prior to each analysis, the coffee was brought to ambient temperature and ground in a Wiley mill (Tecnal Te-650, Brazil) for 60 s. We report herein the ESI-MS fingerprinting of the methanolic extracts of Arabica coffees in the positive and negative ion modes, ESI(+)-MS and ESI(-)-MS, respectively. The ESI-MS data were handled by principal component analysis (PCA) to differentiate both green and roasted Arabica coffees with different stages of ripeness (green or immature, cherry or ripe and overripe), post-harvesting process (dry, wet and semi-wet) and also coffees with different cup qualities. Cup qualities is a Brazilian classification method for Arabica coffees based on the brewing method of steeping, that generates a classification list, from the best to the worst, using the following denominations: strictly soft, soft, barely soft, hard, rioysh, rio and rio zona.16 For the analysis of coffees with different cup qualities, discriminant analysis was applied on PCA data, since PCA alone was unable to perform such discrimination. Extraction Ground coffee samples (1 g) were extracted by reflux with 4 mL of methanol for 2 h (condenser at 5 ± 2 °C). After that time, samples were filtered in a qualitative Whatman filter paper. Samples were kept in a freezer at −8 °C until used. For each sample, three different extractions were made, identified in the figures by the number following the name of the sample. Mass spectrometry Mass spectra were acquired by using a quadrupole/ time-of-flight (Q-ToF) mass spectrometer (Micromass, Manchester, U.K.). General conditions were as follows: source temperature of 100 °C, capillary voltage of 2.1 kV, and cone voltage of 15 to 30 V. Prior to the ESI-MS analysis, 1 μL of an aqueous solution of 0.1% ammonium hydroxide (v/v) or formic acid was added to 1.00 mL of each sample and the mixture vigorously stirred for 15 s. Sample introduction was performed by using a syringe pump (Harvard Apparatus, Pump 11) at a flow rate of Cup quality samples Five samples with medium roast, considered to be typical of different cup quality, identified as soft, hard, rioysh, rio and rio zona, were kindly supplied by ABIC (Associação Brasileira das Indústrias do Café). Ripeness and post-harvest process samples Five samples of Brazilian Arabica coffee from the 2006 crop were directly obtained from the Fazenda São José, São José do Rio Preto, RJ, Brazil. The coffee plants of the 2.5 ha area chosen for the experiment had 20 years of production Amorim et al. Vol. 20, No. 2, 2009 315 10.0 μL min-1 and pumped through an uncoated fused silica capillary. Each analysis required about 60 s. Mass spectra were acquired by scanning over the 50-1000 m/z range. ESI tandem mass spectra were obtained by selection of a specific ion by Q1, by using a unitary m/z window, which was then submitted to collision-induced dissociation (CID) with argon in the hexapole collision chamber at energies of 15-25 eV. The product ion MS analysis was accomplished with the orthogonal TOF (time-of-flight) analyzer. ones with the highest positive or negative values are the most important for the model, the variables related to them were selected for running the PCA with autoscaled data. Data to determine cup quality was first submitted to PCA and the first eight principal components were calculated and submitted to discriminant analysis using the software Statistica, version 7.0. ESI-MS fingerprints of green and roasted Arabica coffees Besides this, other ions observed are those of m/z 195 related to protonated caffeine [M + H]+ and a plethora of positive ions originated from K+ adducts, the most abundant metal ion in coffee seeds. Chlorogenic acids are observed as the ions of m/z 393 [CQA + K]+ and 407 [FQA + K]+, saccharose of m/z 381 [M + K]+, linoleic acid of m/z 319 [M + K]+ and caffeine of m/z 233 [M + K]+. In roasted coffee, innumerous low molecular mass ions below m/z 250 were present in all spectra, mainly originated from the degradation of carbohydrates (almost all degraded in roasting), amino acids, chlorogenic acids and lipids. In the ESI(-)-MS, deprotonated low molecular mass carboxylic acids such as glycolic (m/z 75), lactic and oxalic (m/z 89) and fumaric (m/z 115), generated from sugars, along with citric, quinic and ferulic acids are present. Short-chain carboxylic acids in roasted coffee have been quantified by different methodologies22,23 and, besides phenolics and chlorogenic acids, are responsible for the sourness of coffee brews, an important attribute of coffee beverage quality.24 The concentration of chlorogenic acid oligomers are known to decrease along ripeness, which reflects in the better quality of the beverage obtained with ripe cherry beans. The variability of lipids concentration was studied by Jham et al.,28 who showed a pronounced decrease of diacylglycerols in coffees beans from unripe to ripe, the same happening with fatty acids. These previous results justify the relevance of these ions to discriminate between coffee ripeness. From a matrix of 9 samples and 23 variables, the ions selected as being the most important for the separation in ESI (-)-MS are those of m/z 191, 192, 255, 279, 353, 511, 535, 557 and 695. In this plot, PC1 explains 55% and PC2 44% of the total data variation, summing up to 99% in total. Although the separation among the different ripeness samples has been achieved with the use of only PC1 or only PC2, a better separation is obtained with the two PCs. The immature samples are positively correlated with ions of m/z 511 and 535, the ripe samples are positively correlated with ions of m/z 191 and 192 but negatively correlated with that of m/z 353 and the overripe samples are positively correlated with ions of m/z 695 but negatively correlated with those of m/z 255 and 279. ESI-MS fingerprints of green and roasted Arabica coffees The color of cherry is a good marker of maturation and coffee cherries can be classified into three maturation classes: green or immature, mature or ripe and overripe. Maturation clearly favors the development of high quality flavors in coffee brew. The main volatile precursors identified in coffees are trigonelline, aminoacids, sugars, chlorogenic acids, lipids and carotenoids.25 be associated with cafeoylquinic acid isomers (3, 4 and 5-CQA). Other characteristic ions are those of m/z 529, 515, 367 due to cafeoylferuloyl (CFQA), dicaffeoyl (CDQA) and feruolylquinic (FQA) acids, respectively, besides their fragment ions of m/z 191, 179, 173 and 163 (for example) as extensively discussed by Clifford and co-workers.17-19 Chlorogenic acids are well-known constituents of coffee and proved to exert important antioxidative effects in vitro and in vivo biological systems.20,21 Maturation clearly favors the development of high quality flavors in coffee brew. The main volatile precursors identified in coffees are trigonelline, aminoacids, sugars, chlorogenic acids, lipids and carotenoids.25 Green Arabica coffees (immature, ripe and overripe) are differentiated via ESI(-)-MS by ions related to deprotonated fatty acids (m/z 255, 279, 283 from palmitic, linoleic acid and stearic acids) and to the monomers, dimers and trimers of chlorogenic acids (m/z 353, 557 and 695). Other common ions observed in the ESI(-)-MS of green Arabica coffees are those of m/z 341, associated with deprotonated saccharose [M–H]–, the major sugar in coffees and those of m/z 279 and 255, associated with palmitic and linoleic acids, the two major fatty acids present in green coffees. Another relevant group in ripeness discrimination is observed between m/z 500 and 600. Two of these ions, those of m/z 535 and 511, show in their ESI-MS/MS the fragment ions of m/z 353, 191, 179 and 173, suggesting the presence of caffeoylquinic acid. The difference of the molecular mass from 353 Da and the presence of ions of m/z 279 and 255 with their CO2 loss fragments observed in the ESI-MS/MS suggest the presence of 2-methylbut-2-enoic acid (tiglic acid, C5H8O2) and 2-butenoic acid (crotonic acid, C4H11O2) acids esterifying cafeoyl quinic moieties. Similar compounds have been described in studies of phenylpropane metabolism of tomato (Solanaceae) cotyledons and in chrysanthemum (Asteraceae) phytochemical screening.26,27 In ESI(+)-MS of green Arabica coffees, the major ion is of m/z 138 associated with protonated trigonelline (1-methylnicotinic acid). ESI-MS fingerprints of green and roasted Arabica coffees All mass spectra were accumulated over 60 s, centered, aligned, and handled using MassLynx 3.5 software (Waters, Manchester, U.K.). The abundance readings, for each mass spectrum, were summed into integral m/z ion readings and normalized to maximum abundance value using an in-house built program. To discard noise, only the ions with a relative abundance higher than 5% were included in the final data matrix. The remaining m/z ion values were aligned and compiled to generate a final matrix where each line was a sample and each column a variable (m/z ratios and relative intensities of detected ions). Multivariate analyses by PCA and Partial Least Squares (PLS) were performed using the software Unscrambler, version 9.1. PLS was performed, with autoscaled data, to select the most important variables able to discriminate between different groups. The regression coefficients were calculated and, since the Green and roasted coffees were ground and refluxed with methanol. The samples were not defatted, differently to other reported procedures, to investigate the real contribution of each class of compounds present in coffee. The contents of lipids have been shown to be an important group in coffee differentiation.3,16i Figure 1 shows the ESI-MS fingerprints both in the negative and positive ion modes of two typical samples of green and roasted Arabica coffees. Due to the complex chemical nature of coffee,2-6 both ionization modes contribute with important ions to characterize coffee composition. In general, ESI(-)-MS of green coffees showed [M–H]– ions of chlorogenic acids as the most abundant compounds. Frequently, the most intense ion is of m/z 353, which can Figure 1. ESI-MS fingerprints of methanol extracts of coffees: (A) crude in the negative ion mode, (B) crude in the positive ion mode, (C) roasted in the negative ion mode and (D) roasted in the positive ion mode. Figure 1. ESI-MS fingerprints of methanol extracts of coffees: (A) crude in the negative ion mode, (B) crude in the positive ion mode, (C) roasted in the negative ion mode and (D) roasted in the positive ion mode. Figure 1. ESI-MS fingerprints of methanol extracts of coffees: (A) crude in the negative ion mode, (B) crude in the positive ion mode, (C) roasted in the negative ion mode and (D) roasted in the positive ion mode. Green and Roasted Arabica Coffees Differentiated by Ripeness, Process and Cup Quality 316 J. Braz. Chem. Soc. J. Braz. Chem. Soc. ESI-MS fingerprints of green and roasted Arabica coffees In the ESI(+)-MS of roasted coffees, the ion of m/z 94 is abundant in almost all spectra and could be associated with the methyl pyridinium cation formed by trigonelline degradation under high temperature. Other ions present are protonated trigonelline (m/z 138), its potassium adduct (m/z 176), protonated methyl pyrazine (m/z 95), protonated caffeine (m/z 195) and its potassium adduct (m/z 233). Up to m/z 250, ions relative to K+ adducts of chlorogenic acid were also observed. Ripeness Coffee cherries turn red when ripe due to the replacement of chlorophyll in the pericarp by red flavonoid pigments. In the ESI(+)-MS of green coffees, the main ions responsible for maturity differentiation are relative to Vol. 20, No. 2, 2009 Amorim et al. 317 Figure 2. PCA scores () and loadings () bi-plot obtained with selected ESI-MS data of immature, ripe and overripe green coffees. (a) ESI(-)-MS; (b) ESI(+)-MS. protonated caffein (m/z 195) and chlorogenic acids. Caffein concentration decreases along ripeness but not so much as chlorogenic acids, as it has been pointed out by Clifford and Kazi.29 K+ adducts of fatty acids such as linoleic acid (m/z 319) are also representative. Another important ion, commonly observed in all spectra, is that of m/z 104. By ESI-MS/MS, it could be associated with malonic acid by the loss of CH3CO2H (m/z 60), as observed for the deprotonated species of m/z 103. Malonic acid was found in roasted coffees but in minor amounts and, one supposition is that it could be generated in the rearrangement of malonylglycosides. O- and C-glycosylflavonoids have been found to be acylated by malonic, succinic and malic acids, as pointed out by Harborne in his extensive study of anthocyanins and related phenolics.30 These compounds could give rise to malonic acid during fragmentation. From a matrix of 9 samples and 24 variables, the ions selected as being the most important for this separation are those of m/z 104, 132, 133, 195, 221, 355, 394, 407, 419 and 747. In this plot, PC1 explains 75% and PC2 20% of the total data variation, summing up to 95% in total. The separation among the immature, ripe and overripe has been achieved with the use of only PC1 and the variable with more influence in PC1 is that of m/z 104, which is positively correlated with the immature samples.i Figure 2. PCA scores () and loadings () bi-plot obtained with selected ESI-MS data of immature, ripe and overripe green coffees. (a) ESI(-)-MS; (b) ESI(+)-MS. Although carbohydrate contents show very significant alterations during coffee maturity, ions related to reduced and non-reduced sugars were not between the 10 first variables used to discriminate ripeness in this work. Only minor carboxylic acids such as oxaloacetic acid of m/z 133 are seen probably related to its presence as an intermediate in the citric acid cycle. Ripeness The separation between the overripe and the Green and Roasted Arabica Coffees Differentiated by Ripeness, Process and Cup Quality 318 J. Braz. Chem. Soc. other samples has been achieved in PC1, with the variables of m/z 144, 165 and 176 being positively correlated with the overripe samples. The separation between ripe and immature was obtained in PC2 and the variables positively correlated with the immature samples were ions at m/z 80, 187 and 352, while the variables positively correlated with the ripe samples are those of m/z 94, 108, 122, 154 and 173. fermentation occurs in water at controlled temperatures, giving rise to reduced levels of undesirable flavors. The wet method has been associated with better quality coffees. Greatest amounts of reduced sugars were found in cherry dry processed beans followed by pulped coffees and, at last, lower levels were found for demucilated wet beans.31 No significant differences were seen by the same authors for soluble solids. Figure 3 shows the PCA scores and loadings bi-plot obtained with selected ESI-MS, (a) negative ion mode, (b) positive ion mode, data of roasted coffees with different ripeness. The ESI-MS of green coffees in the positive and negative ion modes are found to be effective to discriminate between post-harvesting processes. In both ionization modes, ions from fatty acids such as palmitic, oleic and linoleic acids are characterized via ESI-MS/MS analysis and found to be representative on differentiation, along with chlorogenic acids. Figure 3. PCA scores () and loadings () bi-plot obtained with selected ESI–MS data of immature, ripe and overripe roasted coffees. (a) ESI(-)-MS; (b) ESI(+)-MS. In the ESI(-)-MS, from a matrix of 9 samples and 25 variables, the ions selected as being the most important for this separation are those of m/z 75, 192, 264, 279, 281, 341, 353, 354, 367 and 695. It can be seen that 99% of total data variation is explained in this plot, PC1 explaining 69% and PC2 30%.The separation among the differently processed samples has been achieved in PC1, with PC2 discriminating the wet samples from the others. The variables of m/z 192, 279, 281 and 341 are positively correlated with the dry samples, the variables of 353, 354 and 695 with the wet samples and variables of 75, 264 and 367 with the semi-wet samples. Most of these ions are also present in the previous ESI(-)-MS ripeness analyses already discussed. Ripeness In the ESI(+)-MS for green coffees, ions are also seen from sugars such as saccharose of m/z 383 [saccharose + Na]+, m/z 219 [glucose + K]+ and K+ adducts of carboxylic acids generated in the fermentation process along the wet treatment. From a matrix of 9 samples and 32 variables, the ions selected as being the most important to differentiate between dry and wet methods via ESI-(+)-MS are those of m/z 104, 133, 175, 176, 219, 220, 221, 231, 266 and 383. In this plot, PC1 explains 56% and PC2 42% of the total data variation, summing up to 98% in total. The separation among the differently processed samples has been achieved in PC1, with PC2 discriminating the dry samples from the others. The variables of m/z 176, 221, 231 and 383 are positively correlated with the dry samples, the variable of m/z 175 with the wet samples and variables of m/z 220 and 266 with the semi-wet samples. Figure 3. PCA scores () and loadings () bi-plot obtained with selected ESI–MS data of immature, ripe and overripe roasted coffees. (a) ESI(-)-MS; (b) ESI(+)-MS. Ripeness been achieved with the use of only PC1, a better separation is obtained with the two PCs. The immature samples are negatively correlated with all variables, whereas the ripe and overripe samples are positively correlated with all. PC2 discriminates between ripe and overripe samples, with ions of m/z 109, 128 and 337 are positively correlated with ripe samples and ions of m/z 133, 337, 353, 354 and 527 positively correlated with overripe samples. Figure 2 shows the PCA scores and loadings bi-plot obtained with selected ESI-MS, (a) negative ion mode, (b) positive ion mode, data of green coffees with different ripeness. In the ESI(+)-MS, discrimination of roasted coffees are mainly performed by low m/z ions such as protonated pyridine (m/z 80) and alkylpiridines (m/z 94 and 108 due to methylpyridinium and 1,4-dimethylpyridinium, respectively), formed via trigonelline degradation. Because N-alkylpyridinium ions are charged species, they are readily detected from solution. Carboxylic acids formed on the roasting process such as octanoic and decanoic acids (m/z 145 and m/z 173) and also the phenolics benzoic, coumaric and cinnamic acids [m/z 123, 165 and 187 in the [M + K]+ form, respectively] are also representative. The different ripeness stages of roasted Arabica coffee were well discriminated by ESI-MS. In the negative ion mode, contributions due to ions of higher m/z from chlorogenic acids are the most important for discrimination. Due to roasting, low m/z ions are observed in the spectra and short-chain organic acids ions such as glycolic (m/z 75), oxalic (m/z 89) and mesaconic acids (m/z 129 and 128) together with ferulic, cinnamic and cafeic acids and phenols as catechol (m/z 109) are also observed. From a matrix of 9 samples and 48 variables, the ions selected as being the most important for ESI(-)-MS separation are those of m/z 109, 128, 133, 337, 353, 354, 373 and 527. In this plot, PC1 explains 66% and PC2 25% of the total data variation, summing up to 91% in total. Although the separation among the different ripeness samples has In the ESI(+)-MS, from a matrix of 9 samples and 418 variables the ions selected as being the most important for this separation are those of m/z 80, 94, 108, 122, 144, 154, 165, 173, 176, 187 and 352. In this plot, PC1 explains 49% and PC2 28% of the total data variation, summing up to 77% in total. Post-Harvest process Dry and wet methods were employed to process green cherry coffees. In the dry method, coffee beans were dried as a whole in the cherry state, with pulp and mucilage. In the wet method, coffee beans were pulped (removal of fruit skin) or pulped and demucilated (removal of mucilaginous mesocarp under fermentation). Dry process is slow, leading to the translocation of chemical constituents from the pulp to the inner bean and also to chemical transformations that depend to ambient conditions whereas in the wet method, Important even mass ions for discrimination are detected via ESI (+)-MS of green coffees accompanied by their [M + H]+ counterpartners. This hypothesis was investigated in the generation of malonic acid and for other relevant ions such as those of m/z 220 and 221. ESI-MS/MS shows a fragment ion of m/z 148, typical of a pentose glycoside Vol. 20, No. 2, 2009 Amorim et al. 319 this separation are those of m/z 94, 139, 200, 210, 429, 430, 486, 758, 796 and 820. In this plot, PC1 explains 51% and PC2 34% of the total data variation, summing up to 95% in total. The separation among the dry, and the other samples has been achieved in PC1, with the variables of m/z 91, 200, 210, 486, 758, 796 and 820 being the most important in this separation. The separation among wet and the other samples was obtained in PC2 and the variables with the stronger influence are those of m/z 139, 200, 210, 429, 430, 486 and 758. Ions of m/z 200 and 210 could not be associated to any constituent but they seem to bear a flavonoidic moiety due to the loss of water and fragment ions from an aromatic nucleus. The low intensity of the ions in the m/z 400-500 range makes any structural proposal difficult. moiety accompanied by loss of CH3CO2H, suggesting the presence of a monoacetylated pentose. Butanoyl and butenoylsucroses have already been identified by Weckerle et al.32 in green Arabica coffees. Figure 4 shows the PCA scores and loadings bi-plot obtained with selected ESI-MS, (a) negative ion mode, (b) positive ion mode, data of green coffee obtained with different post-harvest process. Figure 4. PCA scores () and loadings () bi-plot obtained with selected ESI-MS data of green coffee obtained with dry, semi-wet and wet process. (a) ESI(-)-MS; (b) ESI(+)-MS. Post-Harvest process Figure 5 shows the PCA scores and loadings bi-plot obtained with selected ESI-MS, (a) negative ion mode, (b) positive ion mode, of roasted coffee obtained with different post harvest processes. Figure 5. PCA scores () and loadings () bi-plot obtained with selected ESI-MS data of roasted coffee obtained with dry, semi-wet and wet process. (a) ESI(-)-MS; (b) ESI(+)-MS. Figure 4. PCA scores () and loadings () bi-plot obtained with selected ESI-MS data of green coffee obtained with dry, semi-wet and wet process. (a) ESI(-)-MS; (b) ESI(+)-MS. In roasted coffees, from a matrix of 9 samples and 65 variables the ions selected as being the most important for the separation in ESI (-)-MS are those of m/z 117, 133, 191, 249, 350, 353, 365, 373 and 383. In this plot, PC1 explains 47% and PC2 42% of the total data variation, summing up to 89% in total. The separation among the differently processed samples has been achieved separately in both PCs. The variables of m/z 117 and 373 are positively correlated with the dry samples, the variables of m/z 133 and 383 with the wet samples and variables of m/z 350 and 353 with the semi-wet samples. The ion of m/z 353 is tentatively associated to the K+ adduct of kaveol from the ESI-MS/MS data showing the loss of 18 Da (H2O) from the ion of m/z 314. Figure 5. PCA scores () and loadings () bi-plot obtained with selected ESI-MS data of roasted coffee obtained with dry, semi-wet and wet process. (a) ESI(-)-MS; (b) ESI(+)-MS. Cup quality In Brazil, coffees are graded through a unique classification by using type and cup evaluations. Type is related to the appearance and number of defects of the beans, whereas in cup test sensory evaluations such as flavor In the ESI(+)-MS, from a matrix of 9 samples and 418 variables, the ions selected as being the most important for Green and Roasted Arabica Coffees Differentiated by Ripeness, Process and Cup Quality 320 J. Braz. Chem. Soc. J. Braz. Chem. Soc. and aroma are more relevant. In this test, trained “cuppers” describe their sensations so as to classify coffees as: strictly soft, soft, barely soft, hard, rioysh, rio and rio zona, going from the best to the worst coffee. The beverages for the Brazilian cup test are made by pouring boiling water onto the ground roasted coffees and describing the flavor and smell for some minutes. Following the official Brazilian coffee classification beverage,33 the description of these coffees are: strictly soft, that has a very smooth flavor, slightly sweet and low acidity; soft, also with a smooth flavor and slightly sweetness; barely soft, that is similar but with a slight astringency; hard, with an astringent flavor; rough taste that lacks sweetness; rioysh, with a slight taste of iodoform; rio, that has a strong unpleasant taste reminding iodoform and rio zona, with an intolerable taste and smell.34 These differences were tentatively associated to the presence of defects occasioned by excess of fermentation, skin oxidation, immature and broken beans, for example. Franca and co-workers16 suggested that these defects were mostly related to fermentation processes and could be monitored by sugar and chlorogenic acids contents. Again, both ESI-MS in the positive and negative ion modes are able to differentiate cup quality Arabica roasted coffees. In the ESI(-)-MS, the ions (most deprotonated molecules) used to perform discrimination are the same ones as previously discussed: m/z 75 (glycolic acid), 133 (malic acid), 176/175 (dimethoxycinnamic acid), 191 (quinic acid), 255 (palmitic acid), 257, 279 (linoleic acid), 341 (saccharose), 353 (cafeoylquinic acid), 367 (feruloylquinic acid) and 515 (dicafeoylquinic acid). Cup quality In the ESI-(+)-MS, the major ions detected as either protonated molecules or K+ adducts are, for the protonated molecules, those of m/z 83, 104 (malonic acid), 138 (trigonelline), 139 (p-hydroxybenzoic acid), 151 (pentose) and 195 (caffeine), and for the K+ adducts those of 175, 176 (trigonelline), 233 (caffeine), 317 (linolenic acid), 319 (linolenic acid), 381 (saccharose), 382 and 393 (caffeoylquinic acid). Attempts to correlate the Brazilian cup quality classification to chemical composition of coffees pointed to major differences on chlorogenic acids, trigonelline and caffeine contents.16 By ESI-MS discrimination, it is clear that lipids and sugars imparted a very important contribution to discriminate between Brazilian cup quality coffees. The matrices obtained for these analyses had 15 samples and 12 variables for the negative ion mode and 15 variables for the positive ion mode. PCA was initially used but provided no clustering. So, discriminant analysis was performed on PCA scores and Figure 6 shows the plots obtained for ESI(-)-MS and ESI(+)-MS data. It must be emphasized, however, that this is a preliminary work, which, in the future, should be increased, with samples from other crops, allowing for further development on the conclusions on the chemical compounds responsible for coffee ripeness and processing methods. Figure 6. Discriminant analysis scatterplot obtained for ESI-MS data of soft, hard, rioysh, rio and rio zona coffee samples. (a) ESI(-)-MS; (b) ESI(+)-MS. Acknowledgments The authors gratefully acknowledge the financial support of CNPq, CAPES, FAPERJ and EMBRAPA CAFÉ, Brazil. References 1. Clarke, R. J.; Macrae, R.; Coffee Chemistry, Elsevier Applied Science Publisher: London , 1985, vol. 1. 1. Clarke, R. J.; Macrae, R.; Coffee Chemistry, Elsevier Applied Science Publisher: London , 1985, vol. 1. 2. Martin, M. J.; Pablos, F.; González, A. G.; Food Chem. 1999, 66, 365; Amorim Filho, V. R.; Polito, W. L.; Gomes Neto, J. A.; J. Braz. Chem. Soc. 2007, 18, 47. 2. Martin, M. J.; Pablos, F.; González, A. G.; Food Chem. 1999, 66, 365; Amorim Filho, V. R.; Polito, W. L.; Gomes Neto, J. A.; J. Braz. Chem. Soc. 2007, 18, 47. 3. Kurzrock, T.; Speer, K.; Food Rev. Int. 2001, 17, 433 4. Koshiro, Y.; Zheng, X. Q.; Wang, M. L.; Nagai, C.; Ashihara, H.; Plant Sci. 2006, 171, 242. 5. Martin, M. J.; Pablos, F.; Gonzalez, A. G.; Talanta 1998, 46, 1259. 6. Rogers, W. J.; Michaux, S.; Bastin, M.; Bucheli, P.; Plant Sci. 1999, 149, 115. Figure 6. Discriminant analysis scatterplot obtained for ESI-MS data of soft, hard, rioysh, rio and rio zona coffee samples. (a) ESI(-)-MS; (b) ESI(+)-MS. 7. Zambonin, C. G.; Balest, L.; De Benedetto, G. E.; Palmisano, F.; Talanta 2005, 66, 261. Vol. 20, No. 2, 2009 321 Amorim et al. 8. Maeztu, L.; Andueza, S.; Ibanez, C.; de Pena M. P.; Bello, J.; Cid, C.; J. Agric. Food Chem. 2001, 49, 4743. 23. Jham, G. N.; Fernandes, S. A.; Garcia, C. F.; Phytochem. Anal. 2002, 13, 1399. 24. Galli, V.; Barbas, C.; J. Chromatogr., A 2004, 1032, 299. 9. Rubayiza, A. B.; Meurens, M.; J. Agric. Food Chem. 2005, 53, 4654. 25. De Maria, C. A. B.; Moreira, R. F. A; Trugo, L. C.; Quim. Nova 1999, 22, 209. 10. Charlton, A. J ; Farrington, W. H. H; Brereton, P.; J. Agric. Food Chem. 2002, 50, 3098. 26. Strack, D.; Gross, W.; Wray, V.; Grotjahn, L.; Plant Physiol. 1987, 83, 475. 11. Santos, L. S.; Catharino, R. R.; Aguiar, C. L.; Tsai, S. M.; Eberlin, M. N.; J. Radioanal. Nucl. Chem. 2006, 269, 505. 27. Clifford, M. N.; Wu, W. G.; Kirkpatrick, J.; Kuhnert, N.; J. Agric. Food Chem. 2007, 55, 929. 12. Bastos, D. H. M.; Saldanha, L. A.; Catharino, R. R.; Sawaya, A.; Cunha, I. B. S.; Carvalho, P. O.; Eberlin, M. N.; Molecules 2007, 12, 423. 28. Jham, G. N.; Velikova, R.; Muller, H. V.; Nikolova-Damyanova, 28. Jham, G. N.; Velikova, R.; Muller, H. References V.; Nikolova-Damyanova, B.; Cecon, P. R.; Food Res. Int. 2001, 34, 111. B.; Cecon, P. R.; Food Res. Int. 2001, 34, 111. 29. Clifford, M. N.; Kazi, T.; Food Chem. 1987, 26, 59. 13. Moller, J. K. S.; Catharino, R. R.; Eberlin, M. N.; Food Chem. 2007, 100, 1283. 30. Harborne, J. B.; Phytochemistry 1986, 25, 1887. 14. Wu, Z. G.; Rodgers, R. P.; Marshall, A. G.; J. Agric. Food Chem. 2004, 52, 5322. 31. Gonzalez-Rios, O.; Suarez-Quiroz, M. L.; Boulanger, R.; Barel, M.; Guyot, B.; Guiraud, J. P.; Schorr-Galindo, S.; J. Food Comp. Anal. 2007, 20, 289; ibid., 2007, 20, 297; De Moraes, H. M.; Luchese, R. H.; J. Agric. Food Chem. 1996, 51, 5824. 15. Mendonça, J. C. F.; Franca, A. S. ; Oliveira, L. S.; Nunes, M.; Food Chem. 2008, 111, 490. Luchese, R. H.; J. Agric. Food Chem. 1996, 51, 5824. 16. Farah, A.; Monteiro, M. C. ; Calado, V. ; Franca, A. ; Trugo, L. C.; Food Chem. 2006, 98, 373. 32. Weckerle, B.; Gati, T.; Toth, G.; Schreier, P.; Phytochemistry 2002, 60, 409. 17. Clifford, M. N.; Johnston, K. L.; Knight, S.; Kuhnert, N. A.; J. Agric. Food Chem. 2003, 51, 2900. 33. Bartholo, G. F.; Guimarães, P. T. G.; Informe Agropecuário 1997, 18, 33. 18. Clifford, M. N.; Knight, S.; Surucu, B.; Kuhnert, N.; J. Agric. Food Chem. 2006, 54, 1957. 34. Lingle, T. R.; The Basics of Cupping Coffee, Specialty Coffee Association of America: Long Beach, 1993; Spadone, J. C.; Takeoka, G.; Liardon, R.; J. Agric. Food Chem. 1990, 38 , 226. 19. Clifford, M. N.; Marks, S.; Knight, S.; J. Agric. Food Chem. 2006, 54, 4095. Takeoka, G.; Liardon, R.; J. Agric. Food Chem. 1990, 38 , 226. 20. Chen, J. H.; Ho, C. T.; J. Agric. Food Chem. 1997, 45, 2374. 21. Daglia, M.; Racchi, M.; Papetti, A.; J. Agric. Food Chem. 2004, 52, 1700. FAPESP helped in meeting the publication costs of this article. 22. Balzer, H. H. In Coffee: Recent Developments; Clarke, R. J.; Vitzthum, O. G., eds.; Blackwell Scientific Publications: UK, 2001.
https://openalex.org/W2800856661	https://europepmc.org/articles/pmc5896095?pdf=render	English	null	A genetically and functionally diverse group of non-diazotrophic Bradyrhizobium spp. colonizes the root endophytic compartment of Arabidopsis thaliana	BMC plant biology	2,018	cc-by	8,327	© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Schneijderberg et al. BMC Plant Biology (2018) 18:61 https://doi.org/10.1186/s12870-018-1272-y Schneijderberg et al. BMC Plant Biology (2018) 18:61 https://doi.org/10.1186/s12870-018-1272-y Abstract Background: Diazotrophic Bradyrhizobium spp. are well known for their ability to trigger nodule formation on a variety of legume species. In nodules, Bradyrhizobium utilizes plant-derived carbohydrates in exchange for fixed nitrogen. The genes essential for the nodulation and nitrogen-fixation trait are clustered in a genomic region, which is known as the ‘symbiotic island’. Recently, novel non-diazotrophic Bradyrhizobium spp. have been found to be highly abundant in soils, suggesting that these species can also have a ‘free-living’ life history. However, whether non-diazotrophic Bradyrhizobium spp. can live in association with plants remains elusive. Results: In this study, we show that Bradyrhizobium spp. are common root endophytes of non-legume plant species – including Arabidopsis thaliana (Arabidopsis) – grown in an ecological setting. From a single Arabidopsis root, four Bradyrhizobium sp. strains (designated MOS001 to MOS004) were isolated. Comparative genome analysis revealed that these strains were genetically and functionally highly diverse, but did not harbour the nodulation and the nitrogen fixation gene clusters. Comparative colonization experiments, with MOS strains and nitrogen-fixing symbiotic strains, revealed that all tested Bradyrhizobium spp. can colonize the root endophytic compartment of Arabidopsis. Conclusion: This study provides evidence that both diazotrophic and non-diazotrophic Bradyrhizobium spp. colonize the root endophytic compartment of a wide variety of plant species, including the model species Arabidopsis. This demonstrates that plant roots form a major ecological niche for Bradyrhizobium spp., which might be ancestral to the evolution of the nodulation and nitrogen-fixation trait in this genus. Keywords: Bradyrhizobium, Arabidopsis, Root colonization, Endophytic compartment * Correspondence: ton.bisseling@wur.nl Department of Plant Sciences, Laboratory of Molecular Biology, Wageningen University, Droevendaalsesteeg 1, 6708 PB Wageningen, The Netherlands A genetically and functionally diverse group of non-diazotrophic Bradyrhizobium spp. colonizes the root endophytic compartment of Arabidopsis thaliana Martinus Schneijderberg, Lucas Schmitz, Xu Cheng, Sharon Polman, Carolien Franken, Rene Geurts and Ton Bisseling* Background facilitate rhizobia to convert atmospheric nitrogen into ammonium, which subsequently can be utilized by the plant [2]. The bacterial genes underlying this nitrogen- fixing nodule symbiosis are the nodulation (nod and nol) and nitrogen fixation (nif and fix) genes. These genes are organized in one or more clusters on the genome or sym- biotic plasmid(s). The nod/nol and nif/fix genes – as iden- tified in the diverse range of rhizobial genera – are highly homologous and it is therefore widely accepted that these symbiotic genes were transmitted by horizontal gene transfer [3]. This suggests that the lifestyle of ancestral rhi- zobium species was unrelated to nodule symbiosis. Plants can develop relationships with soil bacteria, which vary from loose associations in the rhizosphere up to in- timate inter- and intracellular hosting in plant tissues. An eminent and well-studied example of an intimate inter- action is the relationship between legumes and a paraphy- letic group of nitrogen-fixing α- and β-proteobacteria, collectively known as rhizobia [1]. Legumes can form spe- cial organs on their roots – known as nodules – that * Correspondence: ton.bisseling@wur.nl Department of Plant Sciences, Laboratory of Molecular Biology, Wageningen University, Droevendaalsesteeg 1, 6708 PB Wageningen, The Netherlands Schneijderberg et al. BMC Plant Biology (2018) 18:61 Page 2 of 9 Page 2 of 9 One of the most profound clades of symbiotic rhizobia is the genus Bradyrhizobium (within the family of the Bradyrhizobiaceae). Defined as “slow-growing, non-acid- producing root nodule bacteria of leguminous plants” [4], this genus encompasses now 37 species [5]. The type strains of these species all nodulate and have a wide range of legumes (including Glycine max; soybean), and the non-legume Parasponia spp. as host [6, 7]. The genomes of Bradyrhizobium spp. are relatively large (7–10 Mb), and the nodulation (nod/nol) and nitrogen fixation (nif/ fix) gene clusters are located on a symbiotic island within the genome [8]. dataset comprises a field experiment at the Veluwe area (a region called the Mossel) in The Netherlands and includes nine plant species representing six taxonomic orders (Table 1). Eight of these species are non-legumes, among which is Arabidopsis (accession Mossel; Msl). The bacter- ial communities of the root endophytic compartments of these species have been determined by 16S rDNA ampli- con sequencing, after growing for 8 weeks in the summer of 2016 on an experimental plot in the Mossel (Schneij- derberg et al. in preparation). Background Analysis revealed four oper- ational taxonomic units (OTUs) affiliated with the genus Bradyrhizobium, of which one OTU (#7) was highly abun- dant in both the soil (0.5%) and the plant root endophytic compartment (Fig. 1). This OTU was not only abundant in the root endophytic compartment of the legume Lotus corniculatus, but also in root samples of all the eight non- legumes, ranging from 1.5% in Leucanthemum vulgare to more than 3% in Plantago lanceolata (Fig. 1). The wide- spread abundance of Bradyrhizobium spp. in the root endophytic compartment of all studied plant species indi- cates that root colonization by strains of this genus is a generic phenomenon in the Mossel area. g Recent studies revealed that Bradyrhizobium spp. are also highly abundant in soils in absence of legume plants. At first, evidence came from culture-independent studies on the bacterial communities in soils and rhizospheres of oak trees. These communities harboured large popula- tions of Bradyrhizobium spp. [9–11]. Additional evidence for the abundance of Bradyrhizobium spp. in absence of legumes was found in coniferous forest soils of North America [12]. Through quantitative population genomics and whole-genome sequencing, it was shown that the identified Bradyrhizobium spp. lack the nod/nol and nif/ fix gene clusters, and consequently are incapable to nodu- late legumes or to fix atmospheric nitrogen. Instead, these strains possess multiple gene clusters affiliated with com- plex carbon metabolism and aromatic compound degrad- ation. Likewise, two Bradyrhizobium spp. strains isolated from soils of a bare fallow field and a grassland did not possess the symbiotic island [13]. Together, these studies indicate that Bradyrhizobium spp. can not only live in symbiosis with legumes, but can also have a ‘free-living’ life history. However, it remains unknown whether such non-nitrogen fixing Bradyrhizobium spp. can have intim- ate relationships with plants. To determine whether Bradyrhizobium spp. are root endophytes in other ecosystems as well, we analysed publicly available 16S rDNA amplicon datasets. This showed that Bradyrhizobium OTUs are present in the endophytic compartment of a variety of plants, including Arabidopsis, rice (Oryza sativa) and maize (Zea mays) (Table 2). This strongly supports that Bradyrhizobium spp. are common endophytes in plant roots. However, whether the strains that make up the Bradyrhizobium OTUs in those studies possess the symbiotic island, re- mains elusive. Arabidopsis endophytic Bradyrhizobium sp. MOS strains lack the nitrogen fixation trait As nodulating and nitrogen-fixing Bradyrhizobium spp. can establish an intracellular lifestyle with their legume hosts, we hypothesised that Bradyrhizobium spp. lacking the symbiotic island can live in a close intimate association with plants. In this study, we investigated whether such Bradyrhizobium spp. occur on or inside the roots of non- legumes in a natural ecosystem. We made use of a dataset from an ecological test field in The Netherlands at the Veluwe area called the Mossel, where we previously stud- ied the root microbiome of nine species, among which Arabidopsis thaliana (Arabidopsis) (Schneijderberg et al., in preparation). An OTU based on the 16S rDNA V4 region can repre- sent multiple strains or even species. To determine whether there are multiple strains belonging to OTU 7 Table 1 Plant species native to the Veluwe Mossel area in the Netherlands that were used in the 16S rDNA meta amplicon Table 1 Plant species native to the Veluwe Mossel area in the Netherlands that were used in the 16S rDNA meta amplicon analysis Species Order Common name Arabidopsis thaliana Brassicales Thale cress / Arabidopsis Crepis capillaris Asterales Smooth hawksbeard Hypericum perforatum Malpighiales St John’s wort Leucathemum vulgare Asterales Ox-eye daisy Lotus corniculatus Fabales Bird’s-foot trefoil Myosotis arvensis Boraginales Field forget-me-not Plantago lanceolata Lamiales Narrowleaf plantain Tanacetum vulgare Asterales Tansy Bradyrhizobium spp. colonize plant roots Bradyrhizobium spp. colonize plant roots To determine whether Bradyrhizobium species can colonize non-legume plant roots, we analysed a 16S rDNA amplicon dataset from a separate but simultaneous experiment (Schneijderberg et al. in preparation). This Schneijderberg et al. BMC Plant Biology (2018) 18:61 Page 3 of 9 Fig. 1 16S rDNA amplicon sequencing reveals high abundance of Bradyrhizobium in all tested plant species. The plant species (listed in Table 1) were grown for 7 weeks on an experimental plot in the Veluwe Mossel area (The Netherlands). Shown is the relative abundance of Bradyrhizobium OTU 7 in the root endophytic compartment. Each dot represents one replicate (n = 8) strains Bradyrhizobium sp. MOS001 to Bradyrhizobium sp. MOS004, after the area from which they originated. A fifth strain designated MOS005, classified as Bradyrhizo- bium based on the 16S rDNA gene, was also sequenced. However, since the full genome sequence could not ensure that this was a Bradyrhizobium sp., we chose to exclude it from the analysis (Additional file 1). y To determine whether these Bradyrhizobium spp. MOS strains possess the nitrogen fixation trait, we as- sembled draft genome sequences based on a 150 base pair paired-end library sequenced on the Illumina HiSeq platform. This resulted in draft assemblies ranging in size from 7.6 (MOS003) to 9.1 Mb (MOS001), represent- ing 6879 (MOS003) to 8622 (MOS001) annotated gene models (Additional file 2). Such genome sizes, as well as the corresponding GC contents (64–66%), are compar- able to previously sequenced Bradyrhizobium spp. ge- nomes [12, 13, 15, 16]. To verify the completeness of the draft genome assemblies of Bradyrhizobium sp. MOS strains, Busco analysis was conducted [17]. This revealed full-length presence of 97.6% in MOS004 and 99.1% in the other MOS strains of the Rhizobiales test set of 686 genes, indicating close to complete genome coverages (Additional file 2). Fig. 1 16S rDNA amplicon sequencing reveals high abundance of Bradyrhizobium in all tested plant species. The plant species (listed in Table 1) were grown for 7 weeks on an experimental plot in the Veluwe Mossel area (The Netherlands). Shown is the relative abundance of Bradyrhizobium OTU 7 in the root endophytic compartment. Each dot represents one replicate (n = 8) To determine whether Bradyrhizobium sp. MOS strains possess nodulation and/or nitrogen fixation genes we used a reciprocal best blast hit algorithm with the nitrogen-fixing symbiont Bradyrhizobium diazoefficiens strain USDA110 as reference [16]. Bradyrhizobium spp. colonize plant roots This revealed that the four Bradyrhizobium sp. MOS strains do not have any hit in either the nod/nol gene cluster or the nif/fix gene cluster (Fig. 2). Therefore we conclude that Bradyrhizo- bium sp. MOS strains are incapable of nodulation and nitrogen fixation. and whether these strains possess the nodulation and ni- trogen fixation gene clusters, we set out on an isolation and characterization strategy for Bradyrhizobium spp. From a single Arabidopsis root grown in the Mossel ex- perimental plot, we isolated 102 strains that showed growth characteristics of Bradyrhizobium. These strains were classified in 12 groups based on DNA-fingerprinting using BOX-PCR (Additional file 1). Next, we sequenced the 16S rDNA locus of a representative strain of each group. This revealed four strains (representing 85% of the isolates) that matched with Bradyrhizobium in the RDP database [14] (Additional file 1). The V4 region of the 16S rDNA gene of these four strains showed more than 99% identity to the consensus sequence of OTU 7, indicating that these isolates represent this OTU. We named these and whether these strains possess the nodulation and ni- trogen fixation gene clusters, we set out on an isolation and characterization strategy for Bradyrhizobium spp. From a single Arabidopsis root grown in the Mossel ex- perimental plot, we isolated 102 strains that showed growth characteristics of Bradyrhizobium. These strains were classified in 12 groups based on DNA-fingerprinting using BOX-PCR (Additional file 1). Next, we sequenced the 16S rDNA locus of a representative strain of each group. This revealed four strains (representing 85% of the isolates) that matched with Bradyrhizobium in the RDP database [14] (Additional file 1). The V4 region of the 16S rDNA gene of these four strains showed more than 99% identity to the consensus sequence of OTU 7, indicating that these isolates represent this OTU. We named these Arabidopsis endophytic Bradyrhizobium sp. strains are highly diverse On the right, the strain name (color-coded as followed. Black: the type strains from Fig. 3; red: strains possessing only the nif/fix gene cluster; purple: the non-symbiotic LTSP strains [12]; green: the non-symbiotic strains G22 and BF49 [13]: orange, the MOS strains. Dark grey bar means a hit on the reference genome, with a coverage of ≥50% and a similarity of ≥70%. The Bradyrhizobium sp. MOS strains do not have a hit in the nod/nol or nif/fix gene clusters, suggesting absence of the symbiotic island. As expected, the LTSP strains as well as the G22 / BF49 strains show no hits in the symbiotic genes either Bradyrhizobium sp. MOS001 and MOS002 belong to dif- ferent lineages of a clade represented by the type strain B. japonicum USDA6 and cluster with the two non- symbiotic strains isolated from grassland (namely Bradyr- hizobium sp. G22 and BF49, respectively). Bradyrhizobium sp. MOS003 falls within a large clade, which includes the type strains B. arachidis LMG26795 and B. yuanmingense CCBAU10071, whereas Bradyrhizobium sp. MOS004 shows high sequence homology to Bradyrhizobium sp. DFCI-1, within the B. elkanii USDA76 clade. capacity to colonize a plant root endophytic compartment in an experimental system. Furthermore, we questioned to what extent this colonization capacity is similar to the colonization capacity of diazotrophic Bradyrhizobium spp. To investigate this, we set up a growth assay in where 7 days-old in vitro grown Arabidopsis seedlings (Acces- sion Msl) are transplanted into autoclaved river sand inoc- ulated with a Bradyrhizobium strain to a density comparable with the density of Bradyrhizobium in the na- tive Mossel soil, which is approximately 2.5 × 105 cells per gram of soil. After 2 weeks, plants were analysed. None of the four Bradyrhizobium sp. MOS strains, nor any of 3 tested nitrogen-fixing symbiotic strains affected the fresh weight of the Arabidopsis plants (Additional file 3). To quantify root colonization of the different strains, we used qPCR on the highly conserved bacterial rpoB gene and normalized that against tubulin binding co-factor C of Arabidopsis [23]. While detection of bacterial DNA in the roots of non-inoculated control plants was close to zero, DNA of all strains was detected in the root endophytic compartment samples in the quantity of approximately 2 bacterial genome copies for every plant genome copies (Fig. 5). Comparing the four Bradyrhizobium sp. Arabidopsis endophytic Bradyrhizobium sp. strains are highly diverse Based on the genetic fingerprinting the four isolates are different. In order to get insight in the taxonomic diver- sity represented by the four isolated Bradyrhizobium sp. MOS strains, the phylogenetic relation was determined. Since we obtained full genome sequences, we based our phylogenetic analysis on the nucleotide sequences of 31 AMPHORA genes [18], which comprise a set of highly conserved marker genes. We aligned these genes against all Bradyrhizobium spp. for which genomes have been published, including non-symbiotic strains identified in forest areas (designated LTSP [12]) and grassland (desig- nated G22 and BF49 [13]). Subsequently, the phylogen- etic relation was reconstructed (Fig. 3). The topology of the phylogenetic tree is in line with previously reported phylogenies based on single or multiple sequence align- ments [12, 19, 20]. The Bradyrhizobium sp. MOS strains represent four distinct lineages within the genus (Fig. 3). Table 2 Bradyrhizobium identified in root endophytic compartment Species Relative Abundance (%) Reference Arabidopsis thaliana 1 [44] Arabidopsis thaliana 3 [45] Oryza sativa 1 [46] Zea mays 0,3 [47] Table 2 Bradyrhizobium identified in root endophytic compartment Schneijderberg et al. BMC Plant Biology (2018) 18:61 Page 4 of 9 Fig. 2 Reciprocal best blast hit analysis shows that Bradyrhizobium sp. MOS strains lack the symbiotic island. On top a schematic representation of the symbiotic reference strain B. diazoefficiens USDA110, with in black the conserved regions, in red the nif/fix gene cluster and in blue the nod/ nol gene cluster. On the right, the strain name (color-coded as followed. Black: the type strains from Fig. 3; red: strains possessing only the nif/fix gene cluster; purple: the non-symbiotic LTSP strains [12]; green: the non-symbiotic strains G22 and BF49 [13]: orange, the MOS strains. Dark grey bar means a hit on the reference genome, with a coverage of ≥50% and a similarity of ≥70%. The Bradyrhizobium sp. MOS strains do not have a hit in the nod/nol or nif/fix gene clusters, suggesting absence of the symbiotic island. As expected, the LTSP strains as well as the G22 / BF49 strains show no hits in the symbiotic genes either Fig. 2 Reciprocal best blast hit analysis shows that Bradyrhizobium sp. MOS strains lack the symbiotic island. On top a schematic representation of the symbiotic reference strain B. diazoefficiens USDA110, with in black the conserved regions, in red the nif/fix gene cluster and in blue the nod/ nol gene cluster. Arabidopsis endophytic Bradyrhizobium sp. strains are highly diverse MOS strains with the three nitrogen-fixing symbionts did not reveal a significant difference in colonization capacity. This shows that symbiotic and non-symbiotic Bradyrhizo- bium strains colonize the Arabidopsis root equally well in our growth system (Fig. 5). Bradyrhizobium sp. MOS001 and MOS002 belong to dif- ferent lineages of a clade represented by the type strain B. japonicum USDA6 and cluster with the two non- symbiotic strains isolated from grassland (namely Bradyr- hizobium sp. G22 and BF49, respectively). Bradyrhizobium sp. MOS003 falls within a large clade, which includes the type strains B. arachidis LMG26795 and B. yuanmingense CCBAU10071, whereas Bradyrhizobium sp. MOS004 shows high sequence homology to Bradyrhizobium sp. DFCI-1, within the B. elkanii USDA76 clade. Since the four strains originate from a single Arabidopsis root, we questioned whether the isolates are functionally similar, despite their genetic divergence. We used a previ- ously published custom R pipeline [21] to predict the func- tional groups of all the open reading frames in each of the isolates using the KEGG Orthology (KO) database [22]. After determining presence and absence of the KO groups and creating a pair-wise dissimilarity matrix using a binary distance measure, we plotted each functional profile along the first two Principal Coordinates. This resulted in a PCoA in which the distances between the species resemble the phylogenetic relationship of the strains (Fig. 4). Along the first two principal coordinates, the Bradyrhizobium sp. MOS001, MOS002 and MOS003 strains cluster relatively close to B. diazoefficiens USDA110 and B. japonicum USDA6, while MOS004 are relatively close to the B. elka- nii clade. This indicates that the group of MOS isolates is not only genetically, but also functionally diverse. The MOS isolates recolonize the Arabidopsis root We aimed to determine whether Bradyrhizobium sp. MOS001, MOS002, MOS003 and/or MOS004 have the Discussion Here we showed that the endophytic compartment of a variety of non-legume plants – including Arabidopsis – is We aimed to determine whether Bradyrhizobium sp. MOS001, MOS002, MOS003 and/or MOS004 have the Schneijderberg et al. BMC Plant Biology (2018) 18:61 Page 5 of 9 Fig. 3 Phylogenetic analysis shows that the Bradyrhizobium sp. MOS strains are genetically diverse. The tree was constructed using a concatenated nucleotide alignment of 31 AMPHORA genes [18]. For each strain, the presence of the nod/nol gene cluster (filled squares) and the nif/fix gene cluster (filled circles) are indicated. A red triangle means that the particular strain is a type strain for that species [5]. The Bradyrhizobium sp. MOS strains fall into four different clades. Afipia broomae ATCC49717 was used as outgroup Fig. 3 Phylogenetic analysis shows that the Bradyrhizobium sp. MOS strains are genetically diverse. The tree was constructed using a concatenated nucleotide alignment of 31 AMPHORA genes [18]. For each strain, the presence of the nod/nol gene cluster (filled squares) and the nif/fix gene cluster (filled circles) are indicated. A red triangle means that the particular strain is a type strain for that species [5]. The Bradyrhizobium sp. MOS strains fall into four different clades. Afipia broomae ATCC49717 was used as outgroup Fig. 3 Phylogenetic analysis shows that the Bradyrhizobium sp. MOS strains are genetically diverse. The tree was constructed using a concatenated nucleotide alignment of 31 AMPHORA genes [18]. For each strain, the presence of the nod/nol gene cluster (filled squares) and the nif/fix gene cluster (filled circles) are indicated. A red triangle means that the particular strain is a type strain for that species [5]. The Bradyrhizobium sp. MOS strains fall into four different clades. Afipia broomae ATCC49717 was used as outgroup the endophytic compartment by Bradyrhizobium spp. is a common trait and that these Bradyrhizobium spp. most probably have a wide host range. colonized by a diverse range of non-nitrogen fixing Bra- dyrhizobium species. Earlier studies revealed that Bra- dyrhizobium species are enriched in rhizosphere soils of non-legumes. For example, in a recent study Bradyrhizo- bium, but also Rhizobium and Bulkholderia OTUs were found in the core rhizosphere microbiome of a tropical chronosequence comprising 31 plant species [24]. This is in line with a growing body of data that suggests rhizobium-plant interactions are evolutionary conserved and that a non-endosymbiotic interaction may be wide- spread [12, 13, 24]. Discussion Green: strains containing the symbiotic island; orange: strains only possessing the nitrogen fixation gene cluster; red: strains lacking the symbiotic island; black: Bradyrhizobium sp. MOS strains, also lacking this island Fig. 4 PCoA on dissimilarities matrix shows that the Bradyrhizobium sp. MOS strains are functionally diverse. Plotted are the first two principal coordinates, which explain 20% and 11%, respectively. Each circle in the plot represents one functional profile. Green: strains containing the symbiotic island; orange: strains only possessing the nitrogen fixation gene cluster; red: strains lacking the symbiotic island; black: Bradyrhizobium sp. MOS strains, also lacking this island Conclusions This study shows that Bradyrhizobium spp. colonize the root endophytic compartment of a wide variety of plant species, including the model species Arabidopsis. Four isolates from a single Arabidopsis root have high genetic and functional variation, but all lack the genes for nodule formation and nitrogen fixation. These non-diazotrophic Bradyrhizobium sp. MOS strains, as well as diazotrophic strains, re-colonize the root endophytic compartment of Arabidopsis. Taken together, this study demonstrates that plant roots form a major ecological niche for Bradyrhizo- bium spp., which might be ancestral to the evolution of the nodulation and nitrogen-fixation trait in this genus. Fig. 5 qPCR-based quantification shows Bradyrhizobium spp. colonization of Arabidopsis roots. On the y-axis an approximation of the number of bacterial genome copies per plant genome copy. This is calculated by exponentiating 2 by the difference between the Ct values of rpoB (targeting bacterial DNA) and Btub (targeting B tubulin in Arabidopsis). Two plant roots grown in autoclaved river sand supplemented with bacteria were pooled, a total of four repli- cates per treatment was measured. Arabidopsis: mock treated plants. USDA110: Bradyrhizobium diazoefficiens USDA110. RPG001: Bradyrhi- zobium sp. isolated from Chamaecrista mimosoides nodule. KLD001: Bradyrhizobium sp. isolated from Parasponia andersonii nodule. These strains possess the symbiotic island Discussion Plotted are the first two principal coordinates, which explain 20% and 11%, respectively. Each circle in the plot represents one functional profile. Green: strains containing the symbiotic island; orange: strains only possessing the nitrogen fixation gene cluster; red: strains lacking the symbiotic island; black: Bradyrhizobium sp. MOS strains, also lacking this island lifestyle, but that the absence of the symbiotic island is widespread and that these non-diazotrophic species have diverse life strategies. Re-inoculation experiments on Arabidopsis did not show any plant growth effects of any of our strains under the conditions that we tested. However, it is well possible that within the setting of a complex micro- biome, plants as well as the microbes benefit from the close association. We showed that the isolated Bradyrhi- zobium sp. MOS strains colonize roots with an efficiency comparable to nitrogen-fixing symbionts. In line with this, we argue that an endophytic lifestyle is likely a common trait of Bradyrhizobium sp. Previously, it has been shown that the nodulation and nitrogen fixation genes have been distributed by horizontal gene transfer within the various genera that can establish a nodule symbiosis with legumes [3, 26]. Therefore it seems prob- able that Bradyrhizobium strains with an endophytic life style acquired the nod/nol and nif/fix genes by horizon- tal gene transfer and in this way obtained the specialized nitrogen-fixing symbiotic lifestyle. Fig. 4 PCoA on dissimilarities matrix shows that the Bradyrhizobium sp. MOS strains are functionally diverse. Plotted are the first two principal coordinates, which explain 20% and 11%, respectively. Each circle in the plot represents one functional profile. Green: strains containing the symbiotic island; orange: strains only possessing the nitrogen fixation gene cluster; red: strains lacking the symbiotic island; black: Bradyrhizobium sp. MOS strains, also lacking this island Fig. 4 PCoA on dissimilarities matrix shows that the Bradyrhizobium sp. MOS strains are functionally diverse. Plotted are the first two principal coordinates, which explain 20% and 11%, respectively. Each circle in the plot represents one functional profile. Green: strains containing the symbiotic island; orange: strains only possessing the nitrogen fixation gene cluster; red: strains lacking the symbiotic island; black: Bradyrhizobium sp. MOS strains, also lacking this island Fig. 4 PCoA on dissimilarities matrix shows that the Bradyrhizobium sp. MOS strains are functionally diverse. Plotted are the first two principal coordinates, which explain 20% and 11%, respectively. Each circle in the plot represents one functional profile. Discussion Whether the rhizobia identified in those studies possess the symbiotic genes remains to be tested. The fact that the Bradyrhizobium sp. MOS strains that we identified in the root endophytic com- partment of Arabidopsis, and other non-legume species, do not have the symbiotic island, shows that the inter- action is independent of the known nodulation and fix- ation genes. Also, we conclude that the colonization of p y g The four Bradyrhizobium sp. MOS strains were iso- lated from a single Arabidopsis root. Surprisingly, we found that they are genetically highly diverse. Using 31 conserved amphora genes, we analysed the phylogeny of the MOS strains and all publicly available Bradyrhizo- bium isolates. Based on analysis by the Average Nucleo- tide Identity software [25], we conclude that MOS001 represents most probably a B. japonicum strain as it shows ~ 95% sequence identity with the type strain USDA6. In contrast, MOS002, MOS003 and MOS004 the sequence similarity with known type strains is less than 95%, and therefore these strains possibly represent new species. The fact that the isolates have highly di- verse functional profiles suggests that the absence of the symbiotic island is not linked to a specific non-symbiotic Schneijderberg et al. BMC Plant Biology (2018) 18:61 Page 6 of 9 Page 6 of 9 Fig. 4 PCoA on dissimilarities matrix shows that the Bradyrhizobium sp. MOS strains are functionally diverse. Plotted are the first two principal coordinates, which explain 20% and 11%, respectively. Each circle in the plot represents one functional profile. Green: strains containing the symbiotic island; orange: strains only possessing the nitrogen fixation gene cluster; red: strains lacking the symbiotic island; black: Bradyrhizobium sp. MOS strains, also lacking this island Fig. 5 qPCR-based quantification shows Bradyrhizobium spp. colonization of Arabidopsis roots. On the y-axis an approximation of the number of bacterial genome copies per plant genome copy. This is calculated by exponentiating 2 by the difference between the Ct values of rpoB (targeting bacterial DNA) and Btub (targeting B tubulin in Arabidopsis). Two plant roots grown in autoclaved river sand supplemented with bacteria were pooled, a total of four repli- cates per treatment was measured. Arabidopsis: mock treated plants. USDA110: Bradyrhizobium diazoefficiens USDA110. RPG001: Bradyrhi- zobium sp. isolated from Chamaecrista mimosoides nodule. KLD001: Bradyrhizobium sp isolated from Parasponia andersonii nodule These Fig. 4 PCoA on dissimilarities matrix shows that the Bradyrhizobium sp. MOS strains are functionally diverse. Soil collection and field experiment p Soil was collected in May 2016 at the Mossel area at the ‘Hoge Veluwe’ in the Netherlands (coordinates: N52°03′ 35.5″ E5°45′06.4″), from four different spots within a radius of 100 m. If there was any vegetation present, the top 5–10 cm soil was removed. The soil was homoge- nized and all detritus was removed. The soil was kept in a cold room at 4 °C until use. All plant species were ster- ilized in 4× diluted household bleach for 10 min, washed seven times with sterile MQ water and transferred to plates with a wet filter paper, placed at 4 °C for 48 h and then moved to a 21 °C incubator in the dark. Seeds of Arabidopsis Msl also got a short rinse with 70% ethanol and the seeds of L. corniculatus and were treated 2 min with H2SO4 before exposure to the household bleach so- lution. Seeds of T. officinale, T. vulgare, L. vulgare and M. arvensis were placed directly in the 21 °C incubator Fig. 5 qPCR-based quantification shows Bradyrhizobium spp. colonization of Arabidopsis roots. On the y-axis an approximation of the number of bacterial genome copies per plant genome copy. This is calculated by exponentiating 2 by the difference between the Ct values of rpoB (targeting bacterial DNA) and Btub (targeting B tubulin in Arabidopsis). Two plant roots grown in autoclaved river sand supplemented with bacteria were pooled, a total of four repli- cates per treatment was measured. Arabidopsis: mock treated plants. USDA110: Bradyrhizobium diazoefficiens USDA110. RPG001: Bradyrhi- zobium sp. isolated from Chamaecrista mimosoides nodule. KLD001: Bradyrhizobium sp. isolated from Parasponia andersonii nodule. These strains possess the symbiotic island Fig. 5 qPCR-based quantification shows Bradyrhizobium spp. colonization of Arabidopsis roots. On the y-axis an approximation of the number of bacterial genome copies per plant genome copy. This is calculated by exponentiating 2 by the difference between the Ct values of rpoB (targeting bacterial DNA) and Btub (targeting B tubulin in Arabidopsis). Two plant roots grown in autoclaved river sand supplemented with bacteria were pooled, a total of four repli- cates per treatment was measured. Arabidopsis: mock treated plants. USDA110: Bradyrhizobium diazoefficiens USDA110. RPG001: Bradyrhi- zobium sp. isolated from Chamaecrista mimosoides nodule. KLD001: Bradyrhizobium sp. isolated from Parasponia andersonii nodule. These strains possess the symbiotic island Schneijderberg et al. Soil collection and field experiment BMC Plant Biology (2018) 18:61 Page 7 of 9 Page 7 of 9 control was treated with ¼ Hoagland’s only. During the grow period, 5 individual pots were weighed to estimate water loss and if necessary plants were irrigated with sterile water. After 2 weeks, the plants were carefully dug out from the sand and the shoot was separated from the root. Shoot biomass was immediately measured. The roots underwent the harvesting protocol previously described [27] and were stored at −80 °C until DNA was extracted as follows: roots were grinded with metal beads in a Tissue Lyzer after which 500 μL of CTAB buffer was added. Samples were incubated for 30 min at 65 °C. Next, 500 μL of chloroform was added and the sample was spun down for 5 min at 14,000 g. Then, the upper phase was transferred to a new Eppendorf tube containing 400 μL of isopropanol. The sample was incubated at −80 °C for 1 h. After incubation, the mixture was spun down for 10 min at 14,000 g and the DNA pellet was washed with 500 μL of 70% ethanol. After another step of centrifugation, the ethanol was removed and the DNA pellet was dried for 15 min at room temperature. Finally, the pellet was resuspended in 50 μL of MilliQ water and stored at −2 °C until further use. without cold treatment. Mossel soil was placed in a tray with 3 × 3 cm pots and watered. To remove the en- dogenous seed population, the tray was placed in the greenhouse for 2 days. After weeding, the sterile seed- lings on the plates were transplanted to the tray with Mossel soil and after 7 days the plants including the soil were planted into to the Mossel field. Arabidopsis growth assay p g y Sterile Arabidopsis Msl Seeds were transferred to ½ MS- plates and incubated at 21 °C with a 16 h/8 h photo- period. After 7 days, roots were ca. 3 cm in length, and transplanted to pots which were prepared as follows: bacterial cultures were spun down, washed with 10 mM MgSO4 and added to 15 ml of ¼ Hoagland’s medium [31]. This medium was added to 4 × 4 cm pots filled with sterile river sand. The soil humidity was set at 70% of the water holding capacity with a bacterial cell density of 4 × 105 cells per gram of sterile sand. The axenic Plant harvesting and DNA isolation of the microbial community After 6 weeks of growth in the field, plants were excavated using a small shovel 3–4 cm around the base of the plant. Consequently, the holes were 6–8 cm wide and around 10 cm deep. After transporting the plants (including the clod) to the laboratory, we applied the harvesting protocol as described before [27]. Four individual plants were pooled into one sample. DNA was isolated from soil and endophytic compartment samples using the Mo Bio PowerSoil kit (Qiagen) and the Fast DNA Spin Kit for Soil (MP Biomedicals) respectively. Quality and quantity of the DNA was checked by nanodrop and gel electrophoresis. Around 400 ng was sent for 16S rDNA sequencing at Beijing Genomics Institute (BGI). 16S rDNA amplicon sequencing and data processing Primers targeting the rpoB gene in Bradyrhizobium spp. (rpoB_Fw1; 5’-CGCTGAAGAACCTCGACGAAGCC-3′ and rpoB_Rv1; 5’-CGGCGTGATCTTGCCGACCAG-3′) were designed using Geneious 8.1. The Arabidopsis B- tubulin gene primers (Forward: 5’-AGAAAACCGGA AACGAGAGC-3′ and Reverse 5’-ACAAGACACTTT CCGCTTGG-3′) were used to normalize against plant DNA [23]. Both primer sets were tested for efficiency prior to the qPCR. The qPCR was performed on four plant DNA samples per treatment, by following the manufacturer’s protocol (Power SYBR Green, Life Tech- nologies). To determine bacterial versus plant DNA, we used the formula: 16S rDNA amplicon sequencing and data processing Using primers 515F and 806R [28], the V4 region was sequenced at BGI on the HiSeq2500 sequencing plat- form (Illumina). Raw data from BGI was processed using a previously reported custom implementation [29] of QIIME [30] with minor modifications which are de- scribed in detail in Schneijderberg et al. (in preparation). In short, reads were quality filtered and filtered for chimeras using ChimeraSlayer. Using a 97% identity threshold, de novo OTUs were determined, which were taxonomically assigned using the RDP classifier with the GreenGenes database. OTUs related to mitochondrial and chloroplast sequences were removed, as were the OTUs that did not have 25 reads in at least 5 samples. To obtain relative abundance of Bradyrhizobium, reads from OTUs matching Bradyrhizobium were added up per sample, and then divided by the total number of reads of that sample after filtering rare taxa. Y ¼ 2 Ct rpoB ð Þ−Ct tub ð Þ ð Þ which is an approximation of the number of bacterial genome copies versus plant genome copy number. Strain isolation and culturing Sonicated and cleaned Arabidopsis root tissue was ground with mortar and pestle in 1 ml phosphate buffer (per litre: 6.33 g of NaH2PO4·H2O, 10.96 g of Na2HPO4·2H2O and 200 μL Silwet L-77). 100 μL of the resulting solution was plated in duplicate on 1/10 TSA and YEM medium plates. 500 μL of the remaining solution was stored in 40% glycerol for later use. In addition, a 100× and 1000× fold dilution were also plated in duplicate. Colonies appearing after Schneijderberg et al. BMC Plant Biology (2018) 18:61 Page 8 of 9 3 days and 7 days were picked (Additional file 2), and stored in YEM medium in 96 well format. with Gblocks [41]. Using the neighbour-joining method in Geneious 8.1, the phylogenetic tree was generated with 1000 bootstraps. The tree was visualized by using Interactive Tree of Life (iTol) platform [42]. Symbiotic island homology g This research was funded by ERC grant number 3100000843. All publicly available draft and complete assemblies be- longing to the Bradyrhizobium genus were pulled from NCBI in January 2017. Symbiotic genes were identified with a BLAST reciprocal best hit algorithm [39]. The genes on the symbiotic island of the B. diazoefficiens USDA110 assembly was used as the reference for ortho- logue searching. Presence or absence was defined as a putative orthologue with a minimum 70% identity and 50% coverage as compared to the reference gene. Sequencing of the bacterial strains Sequencing was performed at BGI, using the Illumina HiSeq 2500 PE 150 platform, with a 350 bp insert library size. Paired-end Illumina reads were quality assessed with FastQC [32], and trimmed accordingly with Trim- momatic v. 0.35 [33], and all samples were filtered for a PHRED33 threshold score above 20. Raw reads were as- sembled with SPAdes v. 3.9.0 [34] under different k-mer sizes with the best assembly judged as containing the smallest number of contigs. Contigs < 1000 bp (only present in MOS002) were removed from the genomes. Assembly statistics were evaluated with QUAST v. 4.5 [35]. The GC content of the contigs revealed contamin- ation in MOS001 and MOS004. The contamination was removed with MaxBin v. 2.2.4 [36]. The ORFs and proteome of the Mossel assemblies were predicted with Prodigal v. 2.6.1 [37]. Assemblies were annotated with the Prokka v1.11 tool [38]. A local BLAST database was made from RefSeq proteins belonging to the Bradyrhizo- bium genus with the BLAST+ package [39]. By specify- ing the genus flag, Prokka first annotated genes with respect to this database. Availability of data and materials The Bradyrhizobium sp. MOS draft genomes are available on the NCBI under accession numbers SAMN07975002 (MOS001) and SAMN08687197 - SAMN08687199 (MOS002 – MOS004). The 16S amplicon data is part of a manuscript in preparation. A draft version of that manuscript and the corresponding 16S amplicon data is available upon request. Additional files Additional file 1: Table S1. DNA fingerprinting of 102 isolates (XLSX 12 kb) Additional file 2: Table S2. Assembly statistics (XLSX 11 kb) Additional file 3: Figure S1. Bradyrhizobium spp. do not affect Arabidopsis shoot fresh weight. Plants were grown for 14 days on sterilised river sand, supplemented with the Bradyrhizobium spp. strains, or mock treated (Control). Only shoot weight was measured, roots were used as template for the qPCR in Fig. 5. Each dot represents one replicate (n = 30 for each treatment, except for RPG001, for which is n = 29). (PDF 6 kb) Additional file 1: Table S1. DNA fingerprinting of 102 isolates (XLSX 12 kb) Additional file 2: Table S2. Assembly statistics (XLSX 11 kb) Additional file 3: Figure S1. Bradyrhizobium spp. do not affect Arabidopsis shoot fresh weight. Plants were grown for 14 days on sterilised river sand, supplemented with the Bradyrhizobium spp. strains, or mock treated (Control). Only shoot weight was measured, roots were used as template for the qPCR in Fig. 5. Each dot represents one replicate (n = 30 for each treatment, except for RPG001, for which is n = 29). (PDF 6 kb) Genomic DNA isolation for sequencing of the bacterial strains Strains were inoculated in 10 ml liquid YEM medium. After 5 days of growth, cells were centrifuged at 4000 g, and the medium was discarded. For isolation of DNA, the Qiagen Blood & Tissue kit was used, according to the manufacturer’s instructions. DNA quantity and pur- ity was checked by nanodrop and gel electrophoresis. MOS003 underwent one more round of DNA isolation because the total quantity was not enough to meet the minimal requirements for sequencing. Acknowledgments We would like to thank the students of the course Molecular Toolbox #60 and #63 at the Wageningen University for executing pilot experiments, Dr. Marnix Medema for his valuable comments on the bioinformatics and Yosapol Harnvanichvech and Zhichun Yan for their help during preparation of this manuscript. Authors’ contributions MS, SP, XC and CF isolated and characterised the bacterial strains. MS and LS assembled the genomes and executed in silico analyses. MS, XC and CF executed other experiments. MS, RG and TB wrote the manuscript. All authors read and approved the final manuscript. Analyses of functional diversity The presence or absence of KEGG Orthologue (KO) groups was determined in the ORFs by employing HMM models publically published by Bai et al. [21]. The KO groups were identified using the HMMER v. 3.1b2 [43] hmmsearch tool with thresholds set at an E-value smaller than 10 × 10−5 and larger than 70% coverage. For multiple hits meeting this criteria, only the hit with the smallest E- value was retained. The data was transformed from abso- lute counts per KO group to either absent or present. Using the vegdist command from the Vegan package in R, a dis- tance matrix was created using the binary distance meas- ure. PcoA was performed by the cmdscale command of the Stats package in R, with Bray-Curtis as method for dissimilarity. Phylogenetic diversity analysis Ethics approval and consent to participate Not applicable. The 31 AMPHORA genes were found to be present in all the assemblies. The nucleotide sequence of inidividal AMPHORA genes were aligned with Clustal Omega v. 1.2.4 [40] and non-conserved segments were removed pp Not applicable. References FastQC: a quality control tool for high throughput sequence data. https://www.bioinformatics.babraham.ac.uk/projects/fastqc/. 7. van Rhijn P, Vanderleyden J. The rhizobium-plant symbiosis. Microbiol Rev. 1995;59:124–42. 33. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. 2014;30:2114–20. 8. Batut J, Andersson SGE, O’Callaghan D. The evolution of chronic infection strategies in the α -proteobacteria. Nat Rev Microbiol. 2004;2:933–45. 34. Bankevich A, Nurk S, Antipov D, Gurevich AA, Dvorkin M, Kulikov AS, et al. SPAdes: a new genome assembly algorithm and its applications to single- cell sequencing. J Comput Biol. 2012;19:455–77. 9. Sachs JL, Ehinger MO, Simms EL. Origins of cheating and loss of symbiosis in wild Bradyrhizobium. J Evol Biol. 2010;23:1075–89. 10. Hartmann M, Howes CG, Vaninsberghe D, Yu H, Bachar D, Christen R, et al. Significant and persistent impact of timber harvesting on soil microbial communities in northern coniferous forests. ISME J. 2012;6:2199–218. 35. Gurevich A, Saveliev V, Vyahhi N, Tesler G. QUAST: quality asses for genome assemblies. Bioinformatics. 2013;29:1072–5. 36. Wu Y-W, Tang Y-H, Tringe SG, Simmons BA, Singer SW. MaxBin: an automated binning method to recover individual genomes from metagenomes using an expectation-maximization algorithm. Microbiome 2014;2:26. 11. Uroz S, Buée M, Murat C, Frey-Klett P, Martin F. Pyrosequencing reveals a contrasted bacterial diversity between oak rhizosphere and surrounding soil. Environ Microbiol Rep. 2010;2:281–8. 12. VanInsberghe D, Maas KR, Cardenas E, Strachan CR, Hallam SJ, Mohn WW. Non-symbiotic Bradyrhizobium ecotypes dominate north American forest soils. ISME J. 2015;9:2435–41. 37. Hyatt D, Chen G-L, LoCascio PF, Land ML, Larimer FW, Hauser LJ. Prodigal: prokaryotic gene recognition and translation initiation site identification. BMC Bioinformatics. 2010;11:119. 13. Jones FP, Clark IM, King R, Shaw LJ, Woodward MJ, Hirsch PR. Novel European free-living, non-diazotrophic Bradyrhizobium isolates from contrasting soils that lack nodulation and nitrogen fixation genes - a genome comparison. Sci Rep. 2016;6:25858. 38. Seemann T. Prokka: rapid prokaryotic genome annotation. Bioinformatics. 2014;30:2068–9. 39. Cock PJA, Chilton JM, Grüning B, Johnson JE, Soranzo N. NCBI BLAST+ integrated into galaxy. Gigascience. 2015;4:39. 14. Wang Q, Garrity GM, Tiedje JM, Cole JR. Naïve Bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy. Appl Environ Microbiol. 2007;73:5261–7. 40. Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, et al. Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal omega. Mol Syst Biol. 2011;7:539. 15. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. 25. Yoon SH, Ha S. Min, Lim J, kwon S, Chun J. A large-scale evaluation of algorithms to calculate average nucleotide identity. Antonie van Leeuwenhoek, Int J gen. Mol Microbiol. 2017;110:1281–6. Leeuwenhoek, Int J gen. Mol Microbiol. 2017;110:1281–6. References 1. Masson-Boivin C, Giraud E, Perret X, Batut J. Establishing nitrogen-fixing symbiosis with legumes: how many rhizobium recipes? Trends Microbiol. 2009;17:458–66. 1. Masson-Boivin C, Giraud E, Perret X, Batut J. Establishing nitrogen-fixing 1. Masson-Boivin C, Giraud E, Perret X, Batut J. Establishing nitrogen-fixing symbiosis with legumes: how many rhizobium recipes? Trends Microbiol. 2009;17:458–66. 28. Parada AE, Needham DM, Fuhrman JA. Every base matters: assessing small subunit rRNA primers for marine microbiomes with mock communities, time series and global field samples. Environ Microbiol. 2016;18:1403–14. 2. Udvardi M, Poole PS. Transport and metabolism in legume-rhizobia symbioses. Annu Rev Plant Biol. 2013;64:781–805. 2. Udvardi M, Poole PS. Transport and metabolism in legume-rhizobia symbioses. Annu Rev Plant Biol. 2013;64:781–805. 29. Pérez-Jaramillo JE, Carrión VJ, Bosse M, Ferrão LFV, De Hollander M, Garcia AAF, et al. Linking rhizosphere microbiome composition of wild and domesticated Phaseolus vulgaris to genotypic and root phenotypic traits. ISME J. 2017;11:2244–57. 3. Remigi P, Zhu J, Young JPW, Masson-Boivin C. Symbiosis within symbiosis: evolving nitrogen-fixing legume symbionts. Trends Microbiol. 2016;24:63–75. 3. Remigi P, Zhu J, Young JPW, Masson-Boivin C. Symbiosis within symbiosis: evolving nitrogen-fixing legume symbionts. Trends Microbiol. 2016;24:63–75. 4. Jordan DC. NOTES: transfer of rhizobium japonicum Buchanan 1980 to Bradyrhizobium gen. Nov., a genus of slow-growing, root nodule Bacteria from leguminous plants. Int J Syst Bacteriol. 1982;32:136–9. 4. Jordan DC. NOTES: transfer of rhizobium japonicum Buchanan 1980 to Bradyrhizobium gen. Nov., a genus of slow-growing, root nodule Bacteria from leguminous plants. Int J Syst Bacteriol. 1982;32:136–9. 30. Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, et al. QIIME allows analysis of high-throughput community sequencing data. Nat Methods. 2010;7:335–6. 5. Parte AC. LPSN - list of prokaryotic names with standing in nomenclature. Nucleic Acids Res. 2014;42:D613–6. 5. Parte AC. LPSN - list of prokaryotic names with standing in 5. Parte AC. LPSN - list of prokaryotic n Nucleic Acids Res. 2014;42:D613–6. 5. Parte AC. LPSN - list of prokaryotic Nucleic Acids Res. 2014;42:D613–6. Nucleic Acids Res. 2014;42:D613–6. Nucleic Acids Res. 2014;42:D613–6. 31. Hoagland DR, Arnon DI. The water-culture method for growing plants without soil. Calif Agric Exp Stn Circ. 1950;347:1–32. 6. Trinick MJ, Hadobas PA. Biology of the Parasponia-Bradyrhizobium symbiosis. Plant Soil. 1988;110:177–85. 6. Trinick MJ, Hadobas PA. Biology of the Parasponia-Bradyrhizobium symbiosis. Plant Soil. 1988;110:177–85. 6. Trinick MJ, Hadobas PA. Biology of the Parasponia-Bradyrhizob symbiosis. Plant Soil. 1988;110:177–85. 32. Andews S. Competing interests Competing interests 2 expression to control Arabidopsis root meristem cell number. Plant cell. 2017;29:tpc. 2017:00366. The authors declare that they have no competing interests. 24. Yeoh YK, Dennis PG, Paungfoo-Lonhienne C, Weber L, Brackin R, Ragan MA, et al. Evolutionary conservation of a core root microbiome across plant phyla along a tropical soil chronosequence. Nat Commun. 2017;8:215. Received: 3 December 2017 Accepted: 26 March 2018 26. Sachs JL, Skophammer RG, Regus JU. Evolutionary transitions in bacterial symbiosis. Proc Natl Acad Sci. 2011;108(Supplement 2):10800–7. 27. Lundberg DS, Lebeis SL, Paredes SH, Yourstone S, Gehring J, Malfatti S, et al. Defining the core Arabidopsis thaliana root microbiome. Nature. 2012;488: 86–90. Consent for publication Not applicable. Consent for publication Not applicable. Consent for publication Not applicable. Page 9 of 9 Page 9 of 9 Page 9 of 9 Schneijderberg et al. BMC Plant Biology (2018) 18:61 References Kaneko T, Nakamura Y, Sato S, Minamisawa K, Uchiumi T, Sasamoto S, et al. Complete genomic sequence of nitrogen-fixing symbiotic bacterium Bradyrhizobium japonicum USDA110. DNA Res. 2002;9:189–97. 41. Castresana J. Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Mol Biol Evol. 2000;17:540–552. 42. Letunic I, Bork P. Interactive tree of life (iTOL) v3: an online tool for the display and annotation of phylogenetic and other trees. Nucleic Acids Res. 2016;44:W242–5. 16. Sugawara M, Tsukui T, Kaneko T, Ohtsubo Y, Sato S, Nagata Y, et al. Complete genome sequence of Bradyrhizobium diazoefficiens USDA 122, a nitrogen-fixing soybean symbiont. Genome Announc. 2017;5:e01743-16. 43. Eddy SR. Profile hidden Markov models. Bioinformatics. 1998;14:755–63. 17. Simão FA, Waterhouse RM, Ioannidis P, Kriventseva EV, Zdobnov EM. BUSCO: assessing genome assembly and annotation completeness with single-copy orthologs. Bioinformatics. 2015;31:3210–2. 44. Bulgarelli D, Rott M, Schlaeppi K. Ver Loren van Themaat E, Ahmadinejad N, Assenza F, et al. revealing structure and assembly cues for Arabidopsis root- inhabiting bacterial microbiota. Nature. 2012;488:91–5. 18. Wu M, Eisen JA. A simple, fast, and accurate method of phylogenomic inference. Genome Biol. 2008;9:R151. 45. Lebeis SL, Paredes SH, Lundberg DS, Breakfield N, Gehring J, McDonald M, et al. Salicylic acid modulates colonization of the root microbiome by specific bacterial taxa. Science. 2015;349:860–4. 19. Rivas R, Martens M, de Lajudie P, Willems A. Multilocus sequence analysis of the genus Bradyrhizobium. Syst Appl Microbiol. 2009;32:101–10. 46. Edwards J, Johnson C, Santos-Medellín C, Lurie E, Podishetty NK, Bhatnagar S, et al. Structure, variation, and assembly of the root-associated microbiomes of rice. Proc Natl Acad Sci. 2015;112:E911–20. 20. Zhang XX, Guo HJ, Wang R, Sui XH, Zhang YM, Wang ET, et al. Genetic divergence of Bradyrhizobium strains nodulating soybeans as revealed by multilocus sequence analysis of genes inside and outside the symbiosis island. Appl Environ Microbiol. 2014;80:3181–90. 47. Niu B, Paulson JN, Zheng X, Kolter R. Simplified and representative bacterial community of maize roots. Proc Natl Acad Sci. 2017;114:E2450–9. 21. Bai Y, Müller DB, Srinivas G, Garrido-Oter R, Potthoff E, Rott M, et al. Functional overlap of the Arabidopsis leaf and root microbiota. Nature. 2015;528:364–9. 22. Ogata H, Goto S, Sato K, Fujibuchi W, Bono H, Kanehisa M. KEGG: Kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 1999;27:29–34. 23. Li H, Torres-Garcia J, Latrasse D, Benhamed M, Schilderink S, Zhou W, et al. Plant-specific histone deacetylases HDT½ regulate GIBBERELLIN 2-OXIDASE
https://openalex.org/W3198658078	https://zenodo.org/record/4737577/files/FTS4VMC.pdf	English	null	Static analysis and family-based model checking of featured transition systems with VMC	null	2,021	cc-by	11,309	This paper is published under the Creative Commons Attribution 4.0 International (CC-BY 4.0) license. Authors reserve their rights to disseminate the work on their personal and corporate Web sites with the appropriate attribution. https://doi.org/10.5281/zenodo.4497888 ABSTRACT associated with features that condition their presence in classical Labelled Transition Systems (LTSs) that model individual variant behaviour. Proving correctness of such behavioural models through model checking or testing is challenging. Ideally, the compactness of the FTS is exploited to reason on the whole system at once. Such an all-in-one technique, by which the behaviour of all variants is examined only once simultaneously, is called family-based analysis, in contrast to an enumerative product-based analysis, by which the behaviour of each individual variant is examined one-by-one [61]. During the past decade, FTSs proved amenable to family-based model-checking and testing [10, 15, 24–26, 28, 34, 35, 37, 38, 41, 50]. In [7, 9], we tackled the automated analysis of FTSs. We defined three ambiguities for an FTS: (i) a dead transition (i.e., a transition that is unreachable, and thus cannot be executed, in any variant); (ii) a false optional transition (i.e., a transition that can be executed in all variants in which its source state is reachable); and (iii) a hidden deadlock state (i.e., a state from which a transition can be executed only in some variants, but not all). In analogy with the above mentioned anomaly detection for variability models, we developed an algorithm to detect ambiguities in FTSs, and a means to resolve them, with a proof of its correctness. The motivations were twofold. First, an ambiguous FTS is often undesired, since it gives an unclear view of the behaviour of the configurable system. Second, an unambiguous FTS paves the way for an efficient kind of family-based model checking. We implemented this algorithm exploiting the SAT solving features of the Z3 SMT solver [32]. The Python code of our implementation (analyser.py), which accepts FTSs in the format .dot as input, is publicly available [8]. associated with features that condition their presence in classical Labelled Transition Systems (LTSs) that model individual variant behaviour. Proving correctness of such behavioural models through model checking or testing is challenging. Ideally, the compactness of the FTS is exploited to reason on the whole system at once. Such an all-in-one technique, by which the behaviour of all variants is examined only once simultaneously, is called family-based analysis, in contrast to an enumerative product-based analysis, by which the behaviour of each individual variant is examined one-by-one [61]. KEYWORDS VMC, tool support, configurable systems, software product lines, variability models, featured transition systems, modal transition systems, static analysis, model checking Static Analysis and Family-based Model Checking of Featured Transition Systems with VMC Maurice H. ter Beek Franco Mazzanti ISTI–CNR, Pisa, Italy Michael Lienhardt ONERA, Palaiseau, France Ferruccio Damiani Luca Paolini Giordano Scarso Michele Valfrè University of Turin, Italy • Software and its engineering →Software product lines; For- mal methods; Model checking; Automated static analysis. • Software and its engineering →Software product lines; For- mal methods; Model checking; Automated static analysis. REFERENCE FORMAT Maurice H. ter Beek, Franco Mazzanti, Ferruccio Damiani, Luca Paolini, Giordano Scarso, Michele Valfrè, and Michael Lienhardt. 2021. Static Analy- sis and Family-based Model Checking of Featured Transition Systems with VMC. https://doi.org/10.5281/zenodo.4497888 In this paper, we present FTS4VMC, a front-end for VMC [11, 13, 14], developed to make VMC amenable to FTSs. VMC1 is a tool for the analysis of behavioural models with variability. It accepts as in- put a Modal Transition System (MTS) with a set of logical variability constraints (akin to an FTS’ feature expressions). An MTS [51, 53] is an LTS that distinguishes admissible (‘may’), necessary (‘must’), and optional (may but not must) transitions which are such that by definition all necessary and optional transitions are also admissible. VMC offers explicit-state on-the-fly model checking of properties of MTSs expressed in the dedicated variability-aware action-based and state-based branching-time temporal logic v-ACTL [5, 11] de- rived from ACTL [33], which is an action-based version of the well-known logic CTL [21]. VMC offers both product-based analy- sis, upon explicit generation of behavioural variant models (LTSs), and a kind of family-based analysis (on MTSs). The latter relies on the fact that for properties expressed in v-ACTLive2, which ABSTRACT During the past decade, FTSs proved amenable to family-based model-checking and testing [10, 15, 24–26, 28, 34, 35, 37, 38, 41, 50].i A Featured Transition System (FTS) is a formalism for modelling variability in configurable systems, such as software product lines. The behaviour of all variants (products) is modelled in a single compact FTS by associating the possibility to perform an action and transition from one state to another with features that condi- tion the execution of an action in specific variants. We present a front-end tool for the Variability Model Checker VMC such that the resulting toolchain allows a modeller to analyse an FTS for ambiguities (dead or false optional transitions and hidden deadlock states), transform an ambiguous FTS into an unambiguous one, and perform an efficient kind of family-based model checking of an FTS without hidden deadlock states. We use examples from the literature to illustrate the novelties offered by the toolchain. In [7, 9], we tackled the automated analysis of FTSs. We defined three ambiguities for an FTS: (i) a dead transition (i.e., a transition that is unreachable, and thus cannot be executed, in any variant); (ii) a false optional transition (i.e., a transition that can be executed in all variants in which its source state is reachable); and (iii) a hidden deadlock state (i.e., a state from which a transition can be executed only in some variants, but not all). In analogy with the above mentioned anomaly detection for variability models, we developed an algorithm to detect ambiguities in FTSs, and a means to resolve them, with a proof of its correctness. The motivations were twofold. First, an ambiguous FTS is often undesired, since it gives an unclear view of the behaviour of the configurable system. Second, an unambiguous FTS paves the way for an efficient kind of family-based model checking. We implemented this algorithm exploiting the SAT solving features of the Z3 SMT solver [32]. The Python code of our implementation (analyser.py), which accepts FTSs in the format .dot as input, is publicly available [8]. 1http://fmt.isti.cnr.it/vmc 1 INTRODUCTION A state s ∈S is reachable (via p) in L if there exists a path p that visits it, i.e., p(i) = s for some i ≥0; s is a deadlock if it has no outgoing transitions, i.e., ∄(s,a,s′) ∈δ, for all a ∈Σ and s′ ∈S. A state s ∈S is reachable (via p) in L if there exists a path p that visits it, i.e., p(i) = s for some i ≥0; s is a deadlock if it has no outgoing transitions, i.e., ∄(s,a,s′) ∈δ, for all a ∈Σ and s′ ∈S. A transition t = (s,a,s′) ∈δ is reachable (via p) in L if there exists a path p that visits it, i.e., p{i} = t, for some i > 0. Outline. This paper is organised as follows. Section 2 discusses related work. The necessary background for understanding the paper is provided in Section 3. The toolchain is presented and applied to examples from the literature in Section 4. Section 5 anticipates future work, while Section 6 wraps up the paper. A transition t = (s,a,s′) ∈δ is reachable (via p) in L if there exists a path p that visits it, i.e., p{i} = t, for some i > 0. Example 3.3. In Figure 1, we depict the LTSs L1 and L2, mod- elling the behaviour of two different coffee machines, adapted from [7, 9, 10, 15]. Each LTS has actions to insert coins (ins) and to pour either standard (std) or extra large (xxl) coffee upon the insertion of one or two coins, respectively. Clearly all states are reachable and there are no deadlocks. 3.1 Featured Transition Systems FTSs were introduced in [27] to concisely model the behaviour of all products of an SPL, modelled as LTSs, in one transition system by annotating transitions with conditions expressing their presence in (product) LTSs. Let B = {⊤, ⊥} denote the Boolean constants true (⊤) and false (⊥), and let B(F) denote the set of propositional formulas over a set of features F (i.e., using features as propositional variables). The elements of B(F) are also called feature expressions. An FTS is an LTS equipped with a feature model and a function that labels each transition with a feature expression. In below definition, the feature model is represented by the set of its (variant) config- urations, where each configuration is represented by a Boolean assignment to the features (i.e., selected = ⊤and unselected = ⊥). Family-based model checking of behavioural variability models provides a means to simultaneously verify multiple behavioural variant models in a single run. Properties can be verified with dedi- cated SPL model-checking tools like SNIP [24, 26], ProVeLines [28], VMC [11, 13, 14], fNuSMV [25, 40], ProFeat [20] (for probabilistic model checking), or QFLan [12, 63] (for statistical model check- ing), or—through suitable abstractions or encodings—with well- known classical model checkers like SPIN [38, 39, 41], PRISM [43] (for probabilistic model checking), Maude [55], mCRL2 [10, 15], or NuSMV [37]. The aforementioned survey [61] also discusses software model checking, operating directly on source code in Java or C [2, 60]. Definition 3.4 (FTS). A Featured Transition System (FTS) is a sex- tuple F = (S, Σ,s0,δ, F, Λ), where S is a finite (non-empty) set of states, Σ is a set of actions, s0 ∈S is the initial state, δ ⊆ S × Σ × B(F) × S is a transition relation, F is a set of features, and Λ ⊆{ λ : F →B } is a set of (variant) configurations. Given a feature expressionϕ ∈B(F), (s,a,ϕ,s′) ∈δ is called a featured tran- sition (labelled with a and limited to configurations satisfying ϕ), which we may write as s a \| ϕ −−−−→s′, and (s,a, ⊤,s′) ∈δ must transition. 1 INTRODUCTION The automated analysis of variability models, such as the detection of anomalies like dead or false optional features in feature diagrams, has a 30-year history [16, 62]. Variability in behavioural models has a shorter history [3, 4, 44–48, 52, 54]. Moreover, such behavioural models received considerable attention only during the last decade, following the seminal paper by Classen et al. [27] that introduced FTSs and an efficient means to model check them. An FTS is a formal model with variability encoding to capture the behaviour of all variants of a configurable system in a single transition system [26, 29]; its transitions, labelled with actions, are ter Beek, et a s0 s1 L1 ins std s0 s1 s2 L2 ins ins xxl Figure 1: LTSs L1 and L2 modelling coffee machines ter Beek, et al. ter Beek, et al. Static Analysis and Family-based Model Checking of FTSs with VMC is a rich fragment of v-ACTL interpreted on so-called live MTSs, validity on the (live) MTS guarantees validity of the property for all its products (cf. [11, Theorem 4]). FTS4VMC allows a modeller to check an FTS for ambiguities (dead or false optional transitions and hidden deadlock states), by calling analyser.py; to remove ambiguities and thus turn an am- biguous FTS into an unambiguous one, by calling disambiguator.py; to transform an FTS into an MTS, by calling translator.py; and to perform an efficient kind of family-based model checking of an MTS obtained in this way from an FTS without hidden deadlock states, by calling VMC via vmc_controller.py. The FTS4VMC implementation, including the Python code of disambiguator.py and translator.py, is publicly available from https://github.com/fts4vmc/FTS4VMC. We use examples from the literature to illustrate the novelties offered by this toolchain. Figure 1: LTSs L1 and L2 modelling coffee machines (si−1,ai,si ) ∈δ for all i > 0; p is said to visit states s0,s1, . . . and transitions t1,t2, . . . and we denote its ith state by p(i) and its ith transition by p{i}. (si−1,ai,si ) ∈δ for all i > 0; p is said to visit states s0,s1, . . . and transitions t1,t2, . . . and we denote its ith state by p(i) and its ith transition by p{i}. 2 RELATED WORK An encompassing overview of analysis strategies for Software Prod- uct Lines (SPLs), including static analysis and model checking, can be found in [61] and a recent empirical study on applying variability- aware static analysis techniques to real-world configurable systems is presented in [59]. As anticipated, static analysis of FTSs mim- ics the automated analysis of variability models by defining be- havioural counterparts of dead and false optional features [16, 62]. Furthermore, it is related to static (program) analysis [19, 57], such as bug detection in code (e.g., using a variable before its initialisa- tion) but also the identification redundant or unreachable code. 3.2 Anomaly Detection in FTSs When applying automated analysis of feature models, the better known analysis operations that are typically being performed con- cern the detection of anomalies (cf., e.g., [16, 62]). These anomalies reflect ambiguous or even contradictory information, such as dead and false optional features. A feature is dead if it is not contained in any configuration of the FTS, whereas it is false optional if it is contained in all configurations of the FTS even though it is not a designated mandatory feature. Such anomalies are typically due to an incorrect use of cross-tree constraints. Figure 2: FTS F modelling a product line of coffee machines ¤ $ e Mandatory Optional Alternative Excludes Requires Figure 3: Feature model of product line of coffee machines ¤ $ e In [7, 9], we formalised equivalent notions in a behavioural set- ting, by adapting these well-known notions to (featured) transitions of an FTS. Recall from Definition 3.5 that all states of a (variant) LTS of an FTS are reachable from the initial state. e Figure 3: Feature model of product line of coffee machines Definition 3.7 (dead transition [9]). We say that a transition (of an FTS) is dead to mean that in all the FTS’s variants the corresponding (LTS) transition is not reachable. Without loss of generality, in the sequel we only consider FTSs that do not contain two featured transitions q a \| ϕ −−−−→q′ and q a \| ϕ′ −−−−→q′ such that ϕ , ϕ′. Any FTS that does not satisfy this criterion can be transformed into one that does by replacing the two transitions with one featured transition q a \| ϕ ∨ϕ′ −−−−−−−→q′. Clearly, since an FTS is intended to compactly represent the behaviour of all variants of a product line, a dead transition in an FTS indicates a modelling error that must be signalled to the modeller so it can be corrected. Such correction can mean removing the transition or changing its feature expression. Definition 3.5 (variant of FTS). Let F = (S, Σ,s0,δ, F, Λ) be an FTS. The LTS specified by a particular configuration λ ∈Λ, denoted by F \|λ, is called a variant (product) of F . 3.2 Anomaly Detection in FTSs It is obtained from F by first removing all featured transitions whose feature expressions are not satisfied by λ (resulting in the LTS (S, Σ,s0,δ ′), with δ ′ = { (s,a,s′) \| (s,a,ϕ,s′) ∈δ and λ \|= ϕ}), and then removing all unreachable states and their outgoing transitions. Given a featured transition (s,a,ϕ,s′) ∈δ, we call (s,a,s′) ∈δ ′ its corresponding (LTS) transition. The set of variants of F is denoted by lts(F ). Definition 3.8 (false optional transition [9]). We say that a tran- sition (of an FTS) is false optional to mean that: (i) it is not dead, (ii) it is not annotated with the feature expression ⊤, and (iii) its corresponding (LTS) transition is present in all the FTS’s variants in which its source state is present. Definition 3.8 is a slightly revised version of that of [7, Def. 3.2], in which condition (i) was not explicitly required. A false optional transition in an FTS indicates a redundancy, in the sense that the associated feature expression can be replaced by ⊤without changing the behaviour of any of the variants of the product line. This redundancy may be intentional syntactic sugar, to underline the fact that the considered transition is part of the behaviour of those configurations that satisfy the feature expression, but otherwise it may be useful for the modeller to know. Moreover, substitution of the feature expression with ⊤allows for more efficient verification because it results in one more must transition, and thus one less feature expression to be evaluated. Note that, by construction: (i) no variant contains unreachable states or transitions, (ii) no variant contains states or actions that were not originally present in the FTS, (iii) each must transition of the FTS has a corresponding transition in the variants in which it is reachable, and (iv) each featured transition has a unique corre- sponding LTS transition when its source state is reachable. The feature (model) expression of F , denoted by fmF , is a feature expression representing Λ, i.e. for all λ : F →B, λ \|= fmF iff λ ∈Λ. Example 3.9. In Figure 4, we depict an FTS F with features f1 and f2 and feature expression fmF = f1 ⊕f2. The LTSs F \|λ1 and F \|λ2, also depicted in Figure 4, model the behaviour of its two valid configurations: λ1 = {f1} and λ2 = {f2}. 3 BACKGROUND LTSs are the underlying behavioural structure of MTSs and FTSs. LTSs are the underlying behavioural structure of MTSs and FTSs. Definition 3.1 (LTS). A Labelled Transition System (LTS) is a quadruple L = (S, Σ,s0,δ), where S is a finite (non-empty) set of states, Σ is a set of actions, s0 ∈S is an initial state, and δ ⊆S ×Σ×S is a transition relation. We call (s,a,s′) ∈δ an a-(labelled) transi- tion, which we may write as s a −−→s′. The notions from Definition 3.2 (path, reachability, deadlock) are carried over to FTSs by ignoring the feature expressions.ii A configuration λ ∈Λ satisfies a feature expression ϕ ∈B(F), denoted by λ \|= ϕ, whenever ϕ is valid in the interpretation λ, i.e., the result of substituting the value of the features occurring as variables in ϕ according to λ is ⊤. Thus, by definition, λ \|= ⊤. We recall some classical notions for LTSs. Definition 3.2 (path, reachability, deadlock). Let L = (S, Σ,s0,δ) be an LTS. A sequence p = s0t1s1t2s2 · · · is a path of L if ti = 2 Static Analysis and Family-based Model Checking of FTSs with VMC Static Analysis and Family-based Model Checking of FTSs with VMC ter Beek, et al. ter Beek, et al. s0 s1 s2 F fmF = $ ⊕e ins \|⊤ std \|e ins \|$ xxl \|⊤ Figure 2: FTS F modelling a product line of coffee machin 3.3 Modal Transition Systems MTSs were introduced in [53] to capture the refinement of a partial description into a more detailed one, reflecting increased knowl- edge on the admissible (but not necessary) behaviour. In [11], MTSs were equipped with a set of variability constraints to compactly model SPL behaviour, whose individual variant behaviour, in the form of an LTS, can be obtained by means of a special-purpose refinement relation, or by an equivalent operational derivation pro- cedure. In [6], it was shown that such MTSs are equally expressive as FTSs. We follow the definitions from [6, 11]. Figure 4: FTS F and its variant LTSs F \|λ1 and F \|λ2 Note that condition (ii) implies that hidden deadlock states of an FTS are present in one or more of its (variant) LTSs.i Definition 3.14 (MTS). A Modal Transition System (MTS) is a quintuple M = (S, Σ,s0,δ3,δ2) such that δ2 ⊆δ3 and (S, Σ,s0,δ3) is an LTS. We distinguish transition relation δ3 expressing admis- sible (may) transitions and transition relation δ2 expressing neces- sary (must) transitions, whereas the transitions in δ3\δ2 are called optional transitions. We may write s a −−→3 s′ for (s,a,s′) ∈δ3, s a −−→2 s′ for (s,a,s′) ∈δ2 and s d s′ for (s,a,s′) ∈δ3 \ δ2. A hidden deadlock in the FTS should definitely be signalled to the modeller, so it can be checked whether the deadlocks in the LTSs should be remedied. If they should not, i.e., if the deadlocks in the LTSs are intended or unavoidable, then this should be made explicit in the FTS to improve understanding. Furthermore, as was shown in [7, 9], removing them enables a kind of family-based verification. Inherited from [6, 11], in the sequel we only consider coherent MTSs, i.e. MTSs such that the set of actions labelling necessary transitions and the set of actions labelling optional transitions are disjoint: for all (p,a,p′) ∈δ2 and (s,b,s′) ∈δ3 \ δ2, we have a , b. Definition 3.11 (ambiguous FTS). We say that an FTS is ambigu- ous to mean that: (i) at least one of its states is a hidden deadlock, or (ii) at least one of its transitions is dead or false optional. Example 3.12. It is easy to see that state s2 of the FTS F depicted in Figure 4 is a hidden deadlock state, because s2 is a deadlock in the LTS F \|λ1. 3Note that ↑is the negation of conjunction (a.k.a. not and). 3.3 Modal Transition Systems ,am is reach- able in L; (3) a1 ↑a2: at most one among a1 and a2 is reachable in L;3 (4) a1 →a2: a2 is reachable in L whenever a1 is reachable in L; (5) a1 →(a2 ⊕a3 ⊕· · · ⊕an): precisely one among a2, . . . ,an is reachable in L whenever a1 is reachable in L; (6) a1 →(a2 ∨a3 ∨· · · ∨an): at least one among a2, . . . ,an is reachable in L whenever a1 is reachable in L. (1) b1 ∨b2 ∨· · · ∨bm, where bi ∈{ai, ¬ai } for all 1 ≤i ≤m (i.e. each literal bi is either equal to ai or to its negation): for at least one bi, 1 ≤i ≤m, it holds that either • bi = ai and ai is reachable in L, or • bi = ¬ai and ai is not reachable in L; (2) a1 ⊕a2 ⊕· · · ⊕am: precisely one among a1, . . . ,am is reach- able in L; (3) a1 ↑a2: at most one among a1 and a2 is reachable in L;3 (4) a1 →a2: a2 is reachable in L whenever a1 is reachable in L; (5) a1 →(a2 ⊕a3 ⊕· · · ⊕an): precisely one among a2, . . . ,an is reachable in L whenever a1 is reachable in L; (6) a1 →(a2 ∨a3 ∨· · · ∨an): at least one among a2, . . . ,an is reachable in L whenever a1 is reachable in L. (1) b1 ∨b2 ∨· · · ∨bm, where bi ∈{ai, ¬ai } for all 1 ≤i ≤m (i.e. each literal bi is either equal to ai or to its negation): for at least one bi, 1 ≤i ≤m, it holds that either • bi = ai and ai is reachable in L, or • bi = ¬ai and ai is not reachable in L; (2) a1 ⊕a2 ⊕· · · ⊕am: precisely one among a1, . . . ,am is reach- able in L; (3) a1 ↑a2: at most one among a1 and a2 is reachable in L;3 (4) a1 →a2: a2 is reachable in L whenever a1 is reachable in L; (5) a1 →(a2 ⊕a3 ⊕· · · ⊕an): precisely one among a2, . . . 2This step must be performed only for those hidden deadlock states that have not yet become explicit deadlock states upon the removal of dead transitions in step (1). 3.2 Anomaly Detection in FTSs Note that transition s2 a \| f2 −−−−→s2 is dead and transition s1 a \| f1 −−−−→s2 is false optional. Example 3.6. In Figure 2, we depict an FTS F modelling the be- haviour of the two coffee machines from Example 3.3 as a product line of coffee machines, adapted from [7, 9, 10, 15]. Imagine that extra large coffee is exclusively available for the American mar- ket, while standard coffee is exclusively available for the European market. Thus, F has transitions labelled with features $ and e, rep- resenting variants for either the American or the European market, respectively. Its feature model, depicted in Figure 3(left), can be represented by the feature expression fmF = $ ⊕e, where ⊕de- notes the exclusive disjunction operation. Hence the configurations of F are Λ = {λ1, λ2}, where λ1($) = ⊥, λ1(e) = ⊤, λ2($) = ⊤, and λ2(e) = ⊥. The LTSs F \|λ1 = L1 and F \|λ2 = L2, depicted in Figure 1, model the behaviour of the only two variants of F : configuration λ1 for the European and λ2 for the American market. An important safety property of systems concerns deadlock free- dom, i.e., the system should not reach a state in which no further action is possible, thus guaranteeing progress or liveness [1, 56]. In case of configurable systems (like FTSs) this notion can be extended to guaranteeing liveness for each variant (LTS). Recall that a state of an FTS is said to be a deadlock if it has no outgoing transitions. Definition 3.10 (hidden deadlock state [9]). We say that a state (of an FTS) is a hidden deadlock to mean that: (i) it is not a deadlock in the FTS, whereas (ii) it is a deadlock in at least one of the FTS’s variants (LTSs). 3 ter Beek, et al. tatic Analysis and Family-based Model Checking of FTSs with VMC Static Analysis and Family-based Model Checking of FTSs with VMC s0 s1 s2 F fmF = f1 ⊕f2 a \|f1 a \|f2 a \|f1 a \|f2 s0 s1 s2 F \|λ1 a a s0 F \|λ2 a Figure 4: FTS F and its variant LTSs F \|λ1 and F \|λ2 3.3 Modal Transition Systems Indeed, F is an ambiguous FTS (cf. also Example 3.9). Note that any (may) transition of an MTS is either a must transi- tion or an optional transition. When drawing MTSs, we explicitly depict their must transitions (as solid lines) and their optional tran- sitions (as dashed lines). Again, the notions from Definition 3.2 (path, reachability, deadlock) are carried over in the usual way. In [7, 9], we showed how to resolve ambiguities in an FTS. A dead transition can be removed. However, this might not always be the right thing to do, since the modeller may simply have made a mistake in the behavioural model or in the feature model. Like- wise for a false optional transition, which however might also be intentional, to make explicit that the (corresponding) transition is part of the behaviour of those configurations that satisfy the associated feature expression. Finally, a hidden deadlock should either be made explicit in the FTS, by adding a deadlock state to the FTS, or the deadlocks in the LTSs should be remedied—again by changing either the behavioural model or the feature model. Hence, based on the detailed feedback obtained by the ambiguity detection algorithm and its implementation presented in [7, 9], the modeller can iteratively improve and check the FTS until the FTS is either un- ambiguous or ambiguous, but such that it is the FTS as intended by the modeller. We moreover showed that any ambiguous FTS can be turned into an unambiguous FTS by the following transformation: Definition 3.15 (variability constraints). Let L = (S, Σ,s0,δ) be an LTS. We define the syntax and semantics of six types of vari- ability constraints on the reachability of actions of L formalised as propositional formulas over Σ interpreted as propositional variables. Let ai ∈Σ, for all i ≥1, and let m ≥2 and n ≥3. et ai Σ, for all i ≥1, and let m ≥2 and n ≥3. (1) b1 ∨b2 ∨· · · ∨bm, where bi ∈{ai, ¬ai } for all 1 ≤i ≤m (i.e. each literal bi is either equal to ai or to its negation): for at least one bi, 1 ≤i ≤m, it holds that either • bi = ai and ai is reachable in L, or • bi = ¬ai and ai is not reachable in L; (2) a1 ⊕a2 ⊕· · · ⊕am: precisely one among a1, . . . 3.3 Modal Transition Systems It is not difficult to see that this result continues to hold for MTSυs whose every state is either live or a deadlock. Figure 5: MTSυ M modellingaproductlineofcoffeemachines y In [7, 9], we noted that any FTS F can be transformed into an MTS FMTS by considering its must transitions as necessary transi- tions, its featured transitions as optional transitions, and all its tran- sitions as admissible, and by removing all feature expressions. And if F is live, then FMTS is live, with respect to the FTS’s set of variants lts(F ), because it has no hidden deadlocks. Moreover, all transi- tions of F whose corresponding (LTS) transitions are mandatorily present in all variants correspond to necessary transitions in FMTS. This demonstrates that the above result from [11] can be carried over to live FTSs, thus allowing a kind of family-based model check- ing also on such FTSs for the v-ACTL fragment v-ACTLive2. Hence, any formula ϕ of v-ACTLive 2 is preserved by live FTSs: given a live FTS F , whenever ϕ holds for FMTS, denoted by FMTS \|= ϕ, then ϕ holds for all variants F \|λ ∈lts(F ), i.e. F \|λ \|= ϕ. Thus, we may freely use any type of propositional formula over a set of actions because we know that it can always be converted into a set of disjunctive formulae (cf. type 1 from Definition 3.15). Definition 3.16 (MTSυ). AnMTSwithvariabilityconstraints(MTSυ) is a sextuple M = (S, Σ,s0,δ3,δ2, ϒ) such that (S, Σ,s0,δ3,δ2) is an MTS and ϒ is a set of variability constraints on actions from Σ. Intuitively, a variant (LTS) is derived from an MTSυ by includ- ing all must transitions of the MTSυ and a subset of its optional transitions. 3.3 Modal Transition Systems ,an is reachable in L whenever a1 is reachable in L; (2) turn the false optional transitions into must transitions; and 2 (2) turn the false optional transitions into must transitions; and (3) make explicit the hidden deadlocks (cf. [7, 9] for details).2 (6) a1 →(a2 ∨a3 ∨· · · ∨an): at least one among a2, . . . ,an is reachable in L whenever a1 is reachable in L. (3) make explicit the hidden deadlocks (cf. [7, 9] for details).2 Recall that an ambiguous FTS may be due to a mistake of the modeller in defining the feature model, in particular in the case of large feature models with many cross-tree constraints. Any propositional formula can be converted into an equivalent formula in conjunctive normal form (CNF), i.e. a conjunction of disjunctions of literals (propositional variables or their negation) by applying the laws of distribution, De Morgan’s laws, and by re- moving double negations, possibly requiring exponential time [58]. Hence, variability constraints of type 1 suffice to define all proposi- tional formulae over Σ; nevertheless, we provide as syntactic sugar the other five types of variability constraints from [6, 11], which stem directly from feature models and are all accepted by VMC. Example 3.13. Consider the FTS F depicted in Figure 4, but now with feature expression fmF = f1 →f2, i.e. the presence of feature f1 requires that of f2. In this case, the LTS F \|λ1 has two further a- transitions, viz. loops in states s0 and s2. Thus F no longer exhibits neither dead transitions nor hidden deadlock states—only the false optional transition s1 a \| f1 −−−−→s2 remains (cf. Examples 3.9 and 3.12). 2This step must be performed only for those hidden deadlock states that have not yet become explicit deadlock states upon the removal of dead transitions in step (1). 4 Static Analysis and Family-based Model Checking of FTSs with VMC ter Beek, et al. s0 s1 s2 M ϒ = {std ⊕ins2} ins1 std ins2 xxl Figure 5: MTSυ M modellingaproductlineofcoffeemachin a deadlock state in any of its variants, effectively resulting in an MTSυ in which every path is infinite. It was proved that validity of formulas expressed in v-ACTLive 2 is preserved in all variants (cf. [11, Theorem 4]), thus allowing a kind of family-based model checking of MTSυs. 3.3 Modal Transition Systems More precisely, the variant LTS has the same set of actions and the same initial state as the MTSυ, but a subset of the set of states of the MTSυ and a subset of its transitions such that the following four conditions are satisfied: (i) all states of the LTS are reachable from its initial state; (ii) all must transitions of the MTSυ are included in the LTS (except those must transitions whose source states are not reachable in the LTS); (iii) for any action a, whenever an a-transition of the MTSυ is included in the LTS, then any other (optional) a-transition in the MTSυ (from a state that is reachable in the LTS) is also included (i.e. the decision to turn one optional a-transition into a necessary a-transition must be consistently repeated for all other optional a-transitions); (iv) all variability constraints are satisfied. This is formalised next. More precisely, if (i) the FTS is live, which is the case if it has no hidden deadlocks (so, unambiguous FTSs are live), and (ii) the property ϕ to be verified is specified in v-ACTLive2, then ϕ can be verified directly on the FTS (ignoring its feature expressions) and if (iii) ϕ holds, this validity is preserved in all LTSs modelling variant behaviour, i.e. ϕ holds for all variants. If any of these three condi- tions does not hold, the property needs to be verified with classical (family-based) approaches, like those mentioned in Section 2. Example 3.19. The MTS M depicted in Figure 5 (ignoring the set ϒ of variability constraints) is exactly the MTS FMTS obtained from the FTS F of Example 3.6 according to the above transformation. It is live. In fact, v-ACTLive 2 formulas like φ = AG AFstd ∨xxl ⊤ preserve their validity for all variants. When verified on M, VMC reports that φ is true and that it holds for all variants of M. Liter- ally, φ expresses that from any state there exists a path that allows to execute std or xxl, i.e. it is always possible to eventually obtain a (standard or extra large) coffee, a prerequisite for any coffee machine. Definition 3.17 (variant of MTSυ). Let M = (S, Σ,s0,δ3,δ2, ϒ) be an MTSυ. 3.3 Modal Transition Systems The LTS L = (Sv, Σ,s0,δv ) is called a variant (product) derived from M whenever Sv ⊆S and δv ⊆δ3 are such that: (1) every s ∈Sv is reachable in L; (2) there exists no (s,a,s′) ∈δ2 \ δv such that s ∈Sv; (3) for any a ∈Σ, whenever (p,a,p′) ∈δv for some p,p′ ∈Sv, then for all s,s′ ∈Sv such that (s,a,s′) ∈δ3 it must be the case that also (s,a,s′) ∈δv;i 4 TOOLCHAIN AT WORK (4) L satisfies all variability constraints in ϒ. The methodology described in the previous section has been fully automated. With the newly developed front-end tool FTS4VMC, the toolchain constituted by (i) FTS4VMC, (ii) the Python code analyser.py from [8] (implementing the static analysis algorithm from [9], where it was shown to be more efficient than the one presented in [7]) and (iii) VMC, can assist an end user (modeller) in the following steps of the engineering and verification methodology envisioned in Figure 6 (where all green blocks are automated by the toolchain): The set of variants derived from an MTSυ M is denoted by lts(M). Note that, by construction, variants do not contain states or actions that were not already present in the MTSυ. Example 3.18. In Figure 5, we depict an MTSυ M modelling the behaviour of the two coffee machines from Example 3.3 as a product line of coffee machines. M has two must transitions and two optional transitions conditioned by the set of variability constraints ϒ = {std⊕ins2}. Indeed, the LTSs L1 and L2, depicted in Figure 1, are the only two variants of M (except that we distinguish ins1 and ins2 to make M coherent). • specify or upload an FTS in FTS4VMC or an MTSυ in VMC; h k h h h b (b • use FTS4VMC to check whether the FTS is ambiguous (by calling analyser.py); • use FTS4VMC to transform any ambiguous FTS into an un- ambiguous one (or to remove only some particular kinds of ambiguities detected, by calling disambiguator.py); Figure 6: Engineering and verification methodology Figure 6: Engineering and verification methodology Remove all ambiguities performs the next two analyses as a single operation; Remove all ambiguities performs the next two analyses as a single operation; Remove all ambiguities performs the next two analyses as a single operation; • specify a v-ACTL formula in FTS4VMC or in VMC; • decide whether the v-ACTL formula is a v-ACTLive2 for- mula with VMC; Remove false optional transitions replaces the feature ex- pression of each false optional transition of the FTS with True, thus converting these transitions into must transitions; no transitions are added or deleted; • verify a v-ACTLive2 formula on the FTS (transformed into an MTS by FTS4VMC, by calling translator.py) or directly on the MTSυ (transformed into an MTS by VMC) with VMC; Remove dead transitions and hidden deadlock states de- letes unreachable transitions if present, then resolves hidden deadlocks by adding an explicit deadlock state called DEAD and creating special transitions to DEAD starting from every state marked as hidden deadlock with a feature expression equivalent to the negation of the disjunction of every feature expression starting from the hidden deadlock state; • report validity for all variants of the FTS/MTSυ in case a v- ACTLive2 formula was verified to hold on a live FTS/MTSυ (a kind of family-based model checking) with VMC; Remove dead transitions and hidden deadlock states de- letes unreachable transitions if present, then resolves hidden deadlocks by adding an explicit deadlock state called DEAD and creating special transitions to DEAD starting from every state marked as hidden deadlock with a feature expression equivalent to the negation of the disjunction of every feature expression starting from the hidden deadlock state; • verify a v-ACTLive2/v-ACTL formula on all variants (gener- ated by VMC) of the MTSυ (product-based model checking) with VMC; The novelties are clearly indicated in the figure: the blue steps and the green steps if applied to FTSs are made possible by FTS4VMC. View MTS shows the MTS obtained from the FTS by replacing each must transition (i.e., with feature expression True) into a necessary transition of the MTS and every other transition into an admissible one; updates source and graph (cf. Fig. 9);i The FTS4VMC front-end accepts FTSs in the .dot format, a well- known graphical notation supported by the graphviz open source graph visualization software (cf. 3.4 Family-Based Model Checking In [11], a specific kind of family-based model checking was intro- duced for MTSυ. An MTSυ is defined to be live if all its states are live, where a live state of an MTSυ is such that it does not occur as • decide whether the FTS is live with FTS4VMC or whether the MTSυ is live with VMC; 5 Static Analysis and Family-based Model Checking of FTSs with VMC ter Beek, et al. initial FTS remove ambiguities with FTS4VMC is the FTS live? transform FTS into MTS with FTS4VMC is ϕ a v-ACTLive2 formula? is the MTSυ live? initial MTSυ verify product-based with VMC verify with external tool (e.g. ProVeLines) verify family-based with VMC ϕ \|= true ? ϕ holds for all variants no yes yes no no (MTSυ) no (FTS) yes no (MTSυ) no (FTS) yes FTS4VMC VMC Figure 6: Engineering and verification methodology verify family-based with VMC yes (FTS) Figure 6: Engineering and verification methodology https://www.graphviz.org).i Verify property verifies properties expressed in v-ACTL with VMC, once the analysis was completed and the FTS results live, by inserting them in the provided text area; Once the FTS has been uploaded and verified to be in the correct format, the user is offered a variety of (analysis) tasks: Show explanation shows the counterexample provided by VMC after having checked the property. Show explanation shows the counterexample provided by VMC after having checked the property.ff Full ambiguities analysis analyzes the FTS for all three types of ambiguities: once done, it outputs an updated version of the FTS with dead transitions highlighted in blue, false op- tional transitions highlighted in green, and hidden deadlock states highlighted in red (cf. Fig. 7); moreover, it enables the ambiguity removal tasks mentioned below; During execution of these tasks, the user is offered different views: During execution of these tasks, the user is offered different views: Figure 8: Full ambiguities analysis reported by FTS4VMC The minepump has to keep a mine safe from flooding by pump- ing water from a shaft while avoiding a methane explosion. The system FTS models the logic of a configurable controller, which interacts with an environment, to operate a water pump based on water and methane level sensors modelled by further FTSs. It con- tains 25 states and 41 transitions.5 The complete minepump model is the parallel composition of five FTSs, for a total of 418 states and 1255 transitions. The feature model of the minepump can be represented by the formula ϕ = (c ↔(ct ∨cp)) ∧l over the feature set F = {c, ct, cp,m,l, ll, ln, lh}, thus resulting in 64 configurations. Moreover, at any time, the user can download the displayed result (e.g., the FTS in .dot or SVG formal and with or without highlighted ambiguities, the transformed MTS, etc.). We now show how to apply our methodology to two concrete examples from the literature: the vending machine and minepump from [26, 27], both of which have become de facto FTS bench- marks [5–7, 9, 10, 15, 18, 24, 30, 34–39, 41, 42]. The vending machine (cf. Fig. 7) serves soda or tea, either for free or upon payment, in which case a compartment is opened for the customer to take the beverage before it is closed again. Its feature model is represented by the feature expression s ∨t over the features {f ,c,s,t}, thus resulting in 12 configurations. The FTS of the vending machine contains 9 states and 13 transitions. Figure 10 shows the result of the full ambiguities analysis by FTS4VMC. The system FTS contains numerous false optional tran- sitions and a hidden deadlock state: s20 is reachable in all variants upon execution of 2 must transitions (actually, the second one is false optional transition (s7, levelMsg,l,s20)) but it is a deadlock state in all 8 variants that lack a feature from the subset {ll, ln, lh}. FTS4VMC can remove this deadlock state by adding an explicit dead- lock state DEAD and a transition (s20, †, ¬ll ∧¬ln ∧¬lh, DEAD). However, a deadlock often indicates a modelling error, either in the feature model or in the behavioural model. 5Transitions with more than one label abbreviate a set of transitions, viz. one per label. 6These papers are based on the high-level specification of the minepump in fPromela as distributed with SNIP and ProVeLines or their translations for VMC or mCRL2 [17, 31]. 4Abusing notation, this concerns execution of transition (8, take, 9), not of (7, take, 1). During execution of these tasks, the user is offered different views: Console displays progress and results of the performed tasks;i Console displays progress and results of the performed tasks;i Source displays the .dot source file of the current FTS; Graph displays the rendered graph in SVG format, highlight- ing the feature model and ambiguities; Graph displays the rendered graph in SVG format, highlight- ing the feature model and ambiguities; Liveness analysis analyzes the FTS for liveness, i.e., whether it has hidden deadlock states but ignoring the detection of dead and false optional transitions; Liveness analysis analyzes the FTS for liveness, i.e., whether it has hidden deadlock states but ignoring the detection of dead and false optional transitions; Summary displays the console output after successful analysis in a more user-friendly way (cf. Fig. 8); Summary displays the console output after successful analysis in a more user-friendly way (cf. Fig. 8); Counterexample graph displays the counterexample obtain- ed upon interaction with VMC rendered as a graph. Counterexample graph displays the counterexample obtain- ed upon interaction with VMC rendered as a graph. Stop processing interrupts the current (full ambiguities or liveness) analysis; Counterexample graph displays the counterexample obtain- ed upon interaction with VMC rendered as a graph. 6 Static Analysis and Family-based Model Checking of FTSs with VMC ter Beek, et al. Figure 7: FTS with ambiguities reported by FTS4VMC Figure 8: Full ambiguities analysis reported by FTS4VMC Figure 9: MTS transformed from the FTS by FTS4VMC Figure 10: FTS with ambiguities reported by FTS4VMC Static Analysis and Family-based Model Checking of FTSs with VMC ter Beek, et al. Static Analysis and Family-based Model Checking of FTSs with VMC Static Analysis and Family-based Model Checking of FTSs with VMC tatic Analysis and Family-based Model Checking of FTSs with VMC ter Beek, et al. Figure 9: MTS transformed from the FTS by FTS4VMC Figure 7: FTS with ambiguities reported by FTS4VMC Figure 8: Full ambiguities analysis reported by FTS4VMC Figure 9: MTS transformed from the FTS by FTS4VMC Figure 7: FTS with ambiguities reported by FTS4VMC Figure 7: FTS with ambiguities reported by FTS4VMC Figure 7: FTS with ambiguities reported by FTS4VMC Figure 8: Full ambiguities analysis reported by FTS4VMC Figure 9: MTS transformed from the FTS by FTS4VMC Figure 10: FTS with ambiguities reported by FTS4VMC Figure 10: FTS with ambiguities reported by FTS4VMC Figure 10: FTS with ambiguities reported by FTS4VMC Figure 8: Full ambiguities analysis reported by FTS4VMC 9In the fPromela specification of the complete minepump model, pumpOn and methane are Booleans that are set to true when the pump is turned on or methane is detected, respectively, whereas the remaining variables are macros (e.g., readCommand defines that the minepump FTS (cf. Fig. 10) has received a commandMsg, while highWater defines that the FTS has received a levelMsg stating that the water level is high). Figure 8: Full ambiguities analysis reported by FTS4VMC In fact, the solution opted for in [10, 26, 38, 39, 41]6 is to modify the FTS by adding transitions from state s20 to initial state s6 to cover the cases in which features from the subset {ll, ln, lh} are missing, viz. (s20, highLevel, ¬lh,s6), (s20, lowLevel, ¬ll,s6), and (s20, normalLevel, ¬ln,s6). Figures 7 and 8 show the result of the full ambiguities analysis reported by FTS4VMC, while Figure 9 shows the MTS obtained by transforming the FTS. As mentioned in [7, 9], example v-ACTL formulas of branching- time properties of the vending machine FTS that can be verified in a kind of family-based manner with FTS4VMC on the corresponding MTS include the following: (1) [pay] AFtake ∨cancel ⊤: whenever a customer pays, eventually (s)he will also either take a beverage or cancel payment;4i (2) AG AFpay ∨free ⊤: infinitely often, a customer will either pay or opt for a free beverage. 7 Static Analysis and Family-based Model Checking of FTSs with VMC ter Beek, et al. Static Analysis and Family-based Model Checking of FTSs with VMC ter Beek, et al. initial FTS remove ambiguities with FTS4VMC is the FTS live? is ϕ an LTL formula? verify with external tool (e.g. ProVeLines) verify family-based with SPIN ϕ \|= true ? ϕ holds for all variants no yes no yes no yes Figure 11: LTL verification methodology is ϕ an LTL formula? verify family-based with SPIN yes yes no no no Figure 11: LTL verification methodology As mentioned in [9], example formulas of linear-time properties of the minepump FTS that can be verified in a kind of family-based manner with tools like SPIN include the following:9 As mentioned in [7, 9], example v-ACTL formulas of branching- time properties of the minepump FTS that can be verified in a kind of family-based manner with FTS4VMC on the corresponding MTS include the following (where (2) is a variant):7 (1) ( ( 23 readCommand ) ∧( 23 readAlarm ) ∧( 23 read- Level ) ) ⇒(¬32 (¬pumpOn ∧¬methane ∧highWater) ): if the controller fairly receives each of the three message types, then the pump is never indefinitely off when the water is high. 5 FUTURE WORK In [7], we considered only branching-time properties. However, in [9] we also considered linear-time properties and showed that a live FTS also enjoys the property that all valid linear-time LTL formulas are preserved by all its variants. This can be seen as follows. A path in an LTS is said to be maximal if it cannot be extended further, i.e. it is infinite or it ends in a deadlock state. Model checking LTL formulas on an LTS reduces to analysing its maximal paths: an LTL formula is valid if it holds for all maximal paths. These notions trivially carry over to FTSs by ignoring their feature expressions. Clearly, if an FTS is live, i.e. it has no hidden deadlocks, then the set of maximal paths of any variant (LTS) is a subset of the set of maximal paths of the FTS. Hence, the following result holds. Any formula ϕ of LTL is preserved by live FTSs: given a live FTS F , whenever ϕ holds for FLTS, denoted by FLTS \|= ϕ, then ϕ holds for all variants F \|λ ∈lts(F ), i.e. F \|λ \|= ϕ. The verification methodology depicted in Figure 11 can thus be envisioned. Figure 8: Full ambiguities analysis reported by FTS4VMC (1) AG [palarmMsg] ¬E [⊤¬setMethaneStopUpalarmMsg ⊤]: it is not possible that two methane palarmMsg messages occur without a setMethaneStop in between; if (2) 2 ((¬pumpOn∧lowWater∧3highWater) ⇒((¬pumpOn) U highWater )): when the pump is off and the water is low, the pump will only start once the water is high again. (2) AG ¬(level ∧EXlevelMsg ⊤): a water levelMsg message cannot be received if the water level is already known. The latter properties (1) and (2) are precisely the properties #34 and #41, respectively, as verified with both SNIP and SPIN in [26]. 7In [22, 23], the states of the FTSs constituting the complete minepump model are labelled with atomic propositions. In particular, state s20 of the system FTS (cf. Fig. 10) is labelled with the proposition level. 8In [26, 27], the states of an FTS are labelled with atomic propositions, omitted in figures to avoid clutter. For LTL model checking with SPIN, we assume that states 5 and 6 of the FTS (cf. Fig. 7) are labelled with the proposition selected, state 7 with the proposition served, and states 1 and 9 with the proposition taken. REFERENCES IEEE, 472–481. https://doi.org/10.1109/ICSE.2013.6606593 [11] Maurice H. ter Beek, Alessandro Fantechi, Stefania Gnesi, and Franco Mazzanti. 2016. Modelling and analysing variability in product families: Model checking of modal transition systems with variability constraints. J. Log. Algebr. Meth. Program. 85, 2 (2016), 287–315. https://doi.org/10.1016/j.jlamp.2015.11.006 [31] Sjoerd Cranen, Jan Friso Groote, Jeroen J. A. Keiren, Frank P. M. Stappers, Erik P. de Vink, Wieger Wesselink, and Tim A. C. Willemse. 2013. An Overview of the mCRL2 Toolset and Its Recent Advances. In Proceedings of the 19th International Conference on Tools and Algorithms for the Construction and Analysis of Systems (TACAS’13) (LNCS, Vol. 7795), Nir Piterman and Scott A. Smolka (Eds.). Springer, 199–213. https://doi.org/10.1007/978-3-642-36742-7_15fi [12] Maurice H. ter Beek, Axel Legay, Alberto Lluch Lafuente, and Andrea Vandin. 2020. A framework for quantitative modeling and analysis of highly (re)configurable systems. IEEE Trans. Softw. Eng. 46, 3 (2020), 321–345. https: //doi.org/10.1109/TSE.2018.2853726 p g [32] Leonardo Mendonça de Moura and Nikolaj Bjørner. 2008. Z3: An Efficient SMT Solver. In Proceedings of the 14th International Conference on Tools and Algorithms for the Construction and Analysis of Systems (TACAS’08) (LNCS, Vol. 4963), C. R. Ramakrishnan and Jakob Rehof (Eds.). Springer, 337–340. https://doi.org/10. 1007/978-3-540-78800-3_24 [13] Maurice H. ter Beek and Franco Mazzanti. 2014. VMC: Recent Advances and Chal- lenges Ahead. In Proceedings of the 18th International Software Product Line Con- ference (SPLC’14), Vol. 2. ACM, 70–77. https://doi.org/10.1145/2647908.2655969 [14] Maurice H. ter Beek, Franco Mazzanti, and Aldi Sulova. 2012. VMC: A Tool for Product Variability Analysis. In Proceedings of the 18th International Sympo- sium on Formal Methods (FM’12) (LNCS, Vol. 7436), Dimitra Giannakopoulou and Dominique Méry (Eds.). Springer, 450–454. https://doi.org/10.1007/978-3-642- 32759-9_36 [33] Rocco De Nicola and Frits W. Vaandrager. 1990. Action versus State based Logics for Transition Systems. In Semantics of Systems of Concurrent Processes: Proceedings of the LITP Spring School on Theoretical Computer Science (LNCS, Vol. 469), Irène Guessarian (Ed.). Springer, 407–419. https://doi.org/10.1007/3- 540-53479-2_17 [15] Maurice H. ter Beek, Sjef van Loo, Erik P. de Vink, and Tim A.C. Willemse. 2020. Family-Based SPL Model Checking Using Parity Games with Variability. In Proceedings of the 23rd International Conference on Fundamental Approaches to Software Engineering (FASE’20) (LNCS, Vol. 12076), Heike Wehrheim and Jordi Cabot (Eds.). Springer, 245–265. https://doi.org/10.1007/978-3-030-45234-6_12 [34] Xavier Devroey, Gilles Perrouin, Axel Legay, Maxime Cordy, Pierre-Yves Schobbens, and Patrick Heymans. 2014. Coverage Criteria for Behavioural Testing of Software Product Lines. REFERENCES [20] Philipp Chrszon, Clemens Dubslaff, Sascha Klüppelholz, and Christel Baier. 2018. ProFeat: feature-oriented engineering for family-based probabilistic model check- ing. Form. Asp. Comp. 30, 1 (2018), 45–75. https://doi.org/10.1007/s00165-017- 0432-4 [1] Bowen Alpern and Fred B. Schneider. 1985. Defining liveness. Inf. Process. Lett. 4, 21 (1985), 181–185. https://doi.org/10.1016/0020-0190(85)90056-0 [2] Sven Apel, Hendrik Speidel, Philipp Wendler, Alexander von Rhein, and Dirk Beyer. 2011. Detection of feature interactions using feature-aware verification. In Proceedings of the 26th IEEE/ACM International Conference on Automated Software Engineering (ASE’11). IEEE, 372–375. https://doi.org/10.1109/ASE.2011.6100075 [21] Edmund M. Clarke, E. Allen Emerson, and A. Prasad Sistla. 1986. Automatic Verification of Finite-State Concurrent Systems Using Temporal Logic Spec- ifications. ACM Trans. Program. Lang. Sys. 8, 2 (1986), 244–263. https: //doi.org/10.1145/5397.5399 [3] Patrizia Asirelli, Maurice H. ter Beek, Alessandro Fantechi, and Stefania Gnesi. 2009. Deontic Logics for Modeling Behavioural Variability. In Proceedings of the 3rd International Workshop on Variability Modelling of Software-intensive Systems (VaMoS’09) (ICB Research Report, Vol. 29), David Benavides, Andreas Metzger, and Ulrich Eisenecker (Eds.). Universität Duisburg-Essen, 71–76. g [22] Andreas Classen. 2010. Modelling with FTS: a Collection of Illustrative Examples. Technical Report P-CS-TR SPLMC-00000001. University of Namur. [23] Andreas Classen. 2011. Modelling and Model Checking Variability-Intensive Sys tems. Ph.D. Dissertation. University of Namur. [23] Andreas Classen. 2011. Modelling and Model Chec tems. Ph.D. Dissertation. University of Namur. tems. Ph.D. Dissertation. University of Namur. Ulrich Eisenecker (Eds.). Universität Duisburg-Essen, 71–76. [24] Andreas Classen, Maxime Cordy, Patrick Heymans, Axel Legay, and Pierre-Yves Schobbens. 2012. Model checking software product lines with SNIP. Int. J. Softw. Tools Technol. Transf. 14, 5 (2012), 589–612. https://doi.org/10.1007/s10009-012- 0234-1 [4] Patrizia Asirelli, Maurice H. ter Beek, Alessandro Fantechi, and Stefania Gnesi. 2010. A Logical Framework to Deal with Variability. In Proceedings of the 8th International Conference on Integrated Formal Methods (IFM’10) (LNCS, Vol. 6396), Dominique Méry and Stephan Merz (Eds.). Springer, 43–58. https://doi.org/10. 1007/978-3-642-16265-7_5 [25] Andreas Classen, Maxime Cordy, Patrick Heymans, Axel Legay, and Pierre- Yves Schobbens. 2014. Formal semantics, modular specification, and symbolic verification of product-line behaviour. Sci. Comput. Program. 80, B (2014), 416–439. https://doi.org/10.1016/j.scico.2013.09.019 [5] Maurice H. ter Beek, Ferruccio Damiani, Stefania Gnesi, Franco Mazzanti, and Luca Paolini. 2015. From Featured Transition Systems to Modal Tran- sition Systems with Variability Constraints. In Proceedings of the 13th Inter- national Conference on Software Engineering and Formal Methods (SEFM’15) (LNCS, Vol. 9276), Radu Calinescu and Bernhard Rumpe (Eds.). Springer, 344–359. REFERENCES https://doi.org/10.1007/978-3-319-22969-0_24 [26] Andreas Classen, Maxime Cordy, Pierre-Yves Schobbens, Patrick Heymans, Axel Legay, and Jean-François Raskin. 2013. Featured Transition Systems: Foundations for Verifying Variability-Intensive Systems and Their Application to LTL Model Checking. IEEE Trans. Softw. Eng. 39, 8 (2013), 1069–1089. https://doi.org/10. 1109/TSE.2012.86 p g [6] Maurice H. ter Beek, Ferruccio Damiani, Stefania Gnesi, Franco Mazzanti, and Luca Paolini. 2019. On the Expressiveness of Modal Transition Systems with Variability Constraints. Sci. Comput. Program. 169 (2019), 1–17. https://doi.org/ 10.1016/j.scico.2018.09.006 [27] Andreas Classen, Patrick Heymans, Pierre-Yves Schobbens, Axel Legay, and Jean- François Raskin. 2010. Model Checking Lots of Systems: Efficient Verification of Temporal Properties in Software Product Lines. In Proceedings of the 32nd International Conference on Software Engineering (ICSE’10). ACM, 335–344. https: //doi.org/10.1145/1806799.1806850 [7] Maurice H. ter Beek, Ferruccio Damiani, Michael Lienhardt, Franco Mazzanti, and Luca Paolini. 2019. Static Analysis of Featured Transition Systems. In Proceedings of the 23rd International Systems and Software Product Line Conference (SPLC’19). ACM, 39–51. https://doi.org/10.1145/3336294.3336295 [28] Maxime Cordy, Andreas Classen, Patrick Heymans, Pierre-Yves Schobbens, and Axel Legay. 2013. ProVeLines: A Product Line of Verifiers for Software Product Lines. In Proceedings of the 17th International Software Product Line Conference (SPLC’13), Vol. 2. ACM, 141–146. https://doi.org/10.1145/2499777.2499781 [8] Maurice H. ter Beek, Ferruccio Damiani, Michael Lienhardt, Franco Mazzanti, and Luca Paolini. 2019. Supplementary material for: “Static Analysis of Featured Transition Systems”. https://doi.org/10.5281/zenodo.2616646 SPLC’13), Vol. 2. ACM, 141–146. https://doi.org/10.1145/2499777.24997 [29] Maxime Cordy, Xavier Devroey, Axel Legay, Gilles Perrouin, Andreas Classen, Patrick Heymans, Pierre-Yves Schobbens, and Jean-François Raskin. 2019. A Decade of Featured Transition Systems. In From Software Engineering to Formal Methods and Tools, and Back, Maurice H. ter Beek, Alessandro Fantechi, and Laura Semini (Eds.). LNCS, Vol. 11865. Springer, 285–312. https://doi.org/10.1007/978- 3-030-30985-5_18 g [9] Maurice H. ter Beek, Ferruccio Damiani, Michael Lienhardt, Franco Mazzanti, and Luca Paolini. 2021. Efficient Static Analysis and Verification of Featured Transition Systems. Empir. Softw. Eng. (2021). https://doi.org/10.1007/s10664-020-09930-8 [9] Maurice H. ter Beek, Ferruccio Damiani, Michael Lienhardt, Franc [10] Maurice H. ter Beek, Erik P. de Vink, and Tim A. C. Willemse. 2017. Family-Based Model Checking with mCRL2. In Proceedings of the 20th International Conference on Fundamental Approaches to Software Engineering (FASE’17) (LNCS, Vol. 10202), M. Huisman and J. Rubin (Eds.). Springer, 387–405. https://doi.org/10.1007/978- 3-662-54494-5_23 [30] Maxime Cordy, Pierre-Yves Schobbens, Patrick Heymans, and Axel Legay. 2013. Beyond Boolean Product-Line Model Checking: Dealing with Feature Attributes and Multi-features. In Proceedings of the 35th International Conference on Software Engineering (ICSE’13). 6 CONCLUSION We presented FTS4VMC, developed specifically as a front-end for VMC with a user-friendly GUI. It has code to upload and down- load files, handle users’ session data, render graph visualization and HTML output, and communicate with VMC. It also contains Python code: analyser.py from [8], implementing the ambigui- ties analysis; disambiguator.py, implementing the ambiguities re- moval; graph.py, implementing the FTS/MTS graph rendering; translator.py, implementing the transformation of an FTS into an MTS; vmc_controller.py, handling the property verification with VMC; and process_manager.py, handling multiprocessing required for real-time output during the analysis process. The resulting toolchain allows a modeller to analyse an FTS for ambiguities, re- move them, and perform an efficient kind of family-based model checking of specific branching-time properties on the resulting FTS. As mentioned in [9], example formulas of linear-time proper- ties of the vending machine FTS that can be verified in a kind of family-based manner with tools like SPIN [49] (http://spinroot.com) include the following:8 The results, mentioned in this paper, that form the basis of the verification methodologies depicted in Figures 6 and 11 indicate spe- cific cases in which verification of live FTSs reduces to verification (with a linear complexity) of corresponding MTSs and LTSs that are obtained straightforwardly by ignoring the feature expressions (and distinguishing necessary and optional transitions in case of MTSs). However, if either (i) the property to be verified is not a v-ACTLive2 or LTL formula, or (ii) the result of the verification is false, then the formula needs to be verified with a classical family-based model checking tool or by means of product-based model checking. (1) 2 (selected ⇒3served): after a beverage is selected, the vending machine will always eventually serve a beverage; (2) 2 (served ⇒3taken): after a beverage is served, a customer will always eventually take the beverage. 8 ter Beek, et al. ter Beek, et al. Static Analysis and Family-based Model Checking of FTSs with VMC REFERENCES Principles of Program Analysis. Springer. https://doi.org/10.1007/978-3-662-03811-6 [58] B h d Pf h i 2010 C j i N l F I E l di f M hi gram Analysis. Springer. https://doi.org/10.1007/978-3-662-03811-6 g y p g p g [58] Bernhard Pfahringer. 2010. Conjunctive Normal Form. In Encyclopedia of Machine Learning, Claude Sammut and Geoffrey I. Webb (Eds.). Springer, 209–210. https: //doi.org/10.1007/978-0-387-30164-8_158 [44] Alessandro Fantechi and Stefania Gnesi. 2007. A behavioural model for prod- uct families. In Proceedings of the 6th joint meeting of the European Software Engineering Conference and the ACM SIGSOFT International Symposium on Foundations of Software Engineering (ESEC/ FSE’07). ACM, 521–524. https: //doi.org/10.1145/1287624.1287700 g [59] Alexander von Rhein, Jörg Liebig, Andreas Janker, Christian Kästner, and Sven Apel. 2018. Variability-Aware Static Analysis at Scale: An Empirical Study. ACM Trans. Softw. Eng. Methodol. 27, 4 (2018), 18:1–18:33. https://doi.org/10.1145/ 3280986 [60] Ina Schaefer, Dilian Gurov, and Siavash Soleimanifard. 2012. Compositional Algorithmic Verification of Software Product Lines. In Revised Papers of the 9th In- ternational Symposium on Formal Methods for Components and Objects (FMCO’10) (LNCS, Vol. 6957), Bernhard K. Aichernig, Frank S. de Boer, and Marcello M. Bon- sangue (Eds.). Springer, 184–203. https://doi.org/10.1007/978-3-642-25271-6_10 [45] Alessandro Fantechi and Stefania Gnesi. 2008. Formal Modeling for Product Fam- ilies Engineering. In Proceedings of the 12th International Conference on Software Product Line Engineering (SPLC’08). IEEE, 193–202. https://doi.org/10.1109/SPLC. 2008.45 [46] Dario Fischbein, Sebastián Uchitel, and Víctor A. Braberman. 2006. A Founda- tion for Behavioural Conformance in Software Product Line Architectures. In Proceedings of the ISSTA Workshop on Role of Software Architecture for Testing and Analysis (ROSATEA’06). ACM, 39–48. https://doi.org/10.1145/1147249.1147254 [61] Thomas Thüm, Sven Apel, Christian Kästner, Ina Schaefer, and Gunter Saake. 2014. A Classification and Survey of Analysis Strategies for Software Product Lines. ACM Comput. Surv. 47, 1 (2014), 6:1–6:45. https://doi.org/10.1145/2580950 [62] Thomas Thüm, Christian Kästner, Fabian Benduhn, Jens Meinicke, Gunter Saake, and Thomas Leich. 2014. FeatureIDE: An extensible framework for feature- oriented software development. Sci. Comput. Program. 79 (2014), 70–85. https: //doi.org/10.1016/j.scico.2012.06.002 y p g [47] Alexander Gruler, Martin Leucker, and Kathrin D. Scheidemann. 2008. Calculating and Modeling Common Parts of Software Product Lines. In Proceedings of the 12th International Software Product Lines Conference (SPLC’08). IEEE, 203–212. https://doi.org/10.1109/SPLC.2008.22 g [63] Andrea Vandin, Maurice H. ter Beek, Axel Legay, and Alaberto Lluch Lafuente. 2018. QFLan: A Tool for the Quantitative Analysis of Highly Reconfigurable Systems. In Proceedings of the 22nd International Symposium on Formal Methods (FM’18) (LNCS, Vol. REFERENCES In Proceedings of the 6th International Symposium on Leveraging Applications of Formal Methods, Verification and Validation (ISoLA’14) (LNCS, Vol. 8802), Tiziana Margaria and Bernhard Steffen (Eds.). Springer, 336–350. https://doi.org/10.1007/978-3-662-45234-9_24 [16] David Benavides, Sergio Segura, and Antonio Ruiz-Cortés. 2010. Automated Analysis of Feature Models 20 Years Later: a Literature Review. Inf. Syst. 35, 6 (2010), 615–636. https://doi.org/10.1016/j.is.2010.01.001 p g [35] Xavier Devroey, Gilles Perrouin, Mike Papadakis, Axel Legay, Pierre-Yves Schobbens, and Patrick Heymans. 2016. Featured Model-based Mutation Analy- sis. In Proceedings of the 38th International Conference on Software Engineering (ICSE’16). ACM, 655–666. https://doi.org/10.1145/2884781.2884821 [17] Olav Bunte, Jan Friso Groote, Jeroen J.A. Keiren, Maurice Laveaux, Thomas Neele, Erik P. de Vink, Wieger Wesselink, Anton Wijs, and Tim A.C. Willemse. 2019. The mCRL2 Toolset for Analysing Concurrent Systems: Improvements in Expressivity and Usability. In Proc. TACAS’19 (LNCS, Vol. 11428), Tomá˘s Vojnar and Lijun Zhang (Eds.). Springer, 21–39. https://doi.org/10.1007/978-3-030-17465-1_2 [36] Aleksandar S. Dimovski. 2018. Abstract Family-Based Model Checking Using Modal Featured Transition Systems: Preservation of CTL∗. In Proceedings of the 21st International Conference on Fundamental Approaches to Software Engineering (FASE’18) (LNCS, Vol. 10802), A. Russo and A. Schürr (Eds.). Springer, 301–318. https://doi.org/10.1007/978-3-319-89363-1_17 [18] Thiago M. Castro, André Lanna, Vander Alves, Leopoldo Teixeira, Sven Apel, and Pierre-Yves Schobbens. 2018. All roads lead to Rome: Commuting strategies for product-line reliability analysis. Sci. Comput. Program. 152 (2018), 116–160. https://doi.org/10.1016/j.scico.2017.10.013 p g [37] Aleksandar S. Dimovski. 2020. CTL∗family-based model checking using variabil- ity abstractions and modal transition systems. Int. J. Softw. Tools Technol. Transf. 22, 1 (2020), 35–55. https://doi.org/10.1007/s10009-019-00528-0 [19] Brian Chess and Jacob West. 2007. Secure Programming with Static Analysis. Addison-Wesley. 9 ter Beek, et al. ter Beek, et al. Static Analysis and Family-based Model Checking of FTSs with VMC [38] Aleksandar S. Dimovski, Ahmad Salim Al-Sibahi, Claus Brabrand, and An- drzej Wąsowski. 2017. Efficient family-based model checking via variabil- ity abstractions. Int. J. Softw. Tools Technol. Transf. 5, 19 (2017), 585–603. https://doi.org/10.1007/s10009-016-0425-2 [50] Chang Hwan Peter Kim, Don S. Batory, and Sarfraz Khurshid. 2011. Reducing Combinatorics in Testing Product Lines. In Proceedings of the 10th International Conference on Aspect-Oriented Software Development (AOSD’11). ACM, 57–68. https://doi.org/10.1145/1960275.1960284 p g [39] Aleksandar S. Dimovski, Ahmad Salim Al-Sibahi, Claus Brabrand, and Andrzej Wąsowski. 2015. Family-Based Model Checking Without a Family-Based Model Checker. In Proceedings of the 22nd International SPIN Symposium on Model Check- ing of Software (SPIN’15) (LNCS, Vol. 9232), Bernd Fischer and Jaco Geldenhuys (Eds.). Springer, 282–299. REFERENCES https://doi.org/10.1007/978-3-319-23404-5_18 [51] Jan Křetínský. 2017. 30 Years of Modal Transition Systems: Survey of Extensions and Analysis. In Models, Algorithms, Logics and Tools, Luca Aceto, Giorgio Bacci, Giovanni Bacci, Anna Ingólfsdóttir, Axel Legay, and Radu Mardare (Eds.). LNCS, Vol. 10460. Springer, 36–74. https://doi.org/10.1007/978-3-319-63121-9_3 Vol. 10460. Springer, 36–74. https://doi.org/10.1007/978-3-319-63121-9 p g p g [52] Kim G. Larsen, Ulrik Nyman, and Andrzej Wąsowski. 2007. Modal I/O Automata for Interface and Product Line Theories. In Proceedings of the 16th European Symposium on Programming (ESOP’07) (LNCS, Vol. 4421), Rocco De Nicola (Ed.). Springer, 64–79. https://doi.org/10.1007/978-3-540-71316-6_6 [40] Aleksandar S. Dimovski, Axel Legay, and Andrzej Wąsowski. 2019. Variability Abstraction and Refinement for Game-Based Lifted Model Checking of Full CTL. In Proceedings of the 22nd International Conference on Fundamental Approaches to Software Engineering (FASE’19) (LNCS, Vol. 11424), Reiner Hähnle and Wil van der Aalst (Eds.). Springer, 192–209. https://doi.org/10.1007/978-3-030-16722-6_11i Springer, 64–79. https://doi.org/10.1007/978-3-540-71316-6_6 [53] Kim Guldstrand Larsen and Bent Thomsen. 1988. A Modal Process Logic. In Proceedings of the 3rd Symposium on Logic in Computer Science (LICS’88). IEEE, 203–210. https://doi.org/10.1109/LICS.1988.5119 p g p g [41] Aleksandar S. Dimovski and Andrzej Wąsowski. 2017. Variability-Specific Ab- straction Refinement for Family-Based Model Checking. In Proceedings of the 20th International Conference on Fundamental Approaches to Software Engineering (FASE’17) (LNCS, Vol. 10202), Marieke Huisman and Julia Rubin (Eds.). Springer, 406–423. https://doi.org/10.1007/978-3-662-54494-5_24f p g [54] Kim Lauenroth, Klaus Pohl, and Simon Töhning. 2009. Model Checking of Domain Artifacts in Product Line Engineering. In Proceedings of the 24th International Conference on Automated Software Engineering (ASE’09). IEEE, 269–280. https: //doi.org/10.1109/ASE.2009.16 g [55] Malte Lochau, Stephan Mennicke, Hauke Baller, and Lars Ribbeck. 2016. Incre- mental model checking of delta-oriented software product lines. J. Log. Algebr. Meth. Program. 85, 1 (2016), 245–267. https://doi.org/10.1016/j.jlamp.2015.09.004 [42] Clemens Dubslaff. 2019. Compositional Feature-Oriented Systems. In Proceedings of the 17th International Conference on Software Engineering and Formal Methods (SEFM’19) (LNCS, Vol. 11724), Peter Csaba Ölveczky and Gwen Salaün (Eds.). Springer, 162–180. https://doi.org/10.1007/978-3-030-30446-1_9f g ( ) p g j j p [56] Zohar Manna and Amir Pnueli. 1995. Temporal Verification of Reactive Systems: Safety Springer https://doi org/10 1007/978 1 4612 4222 2 [43] Clemens Dubslaff, Christel Baier, and Sascha Klüppelholz. 2015. Probabilistic Model Checking for Feature-Oriented Systems. In Transactions on Aspect-Oriented Software Development XII (LNCS, Vol. 8989), Shigeru Chiba, Éric Tanter, Erik Ernst, and Robert Hirschfeld (Eds.). Springer, 180–220. https://doi.org/10.1007/978-3- 662-46734-3_5 Safety. Springer. https://doi.org/10.1007/978-1-4612-4222-2 [57] Flemming Nielson, Hanne Riis Nielson, and Chris Hankin. 2005. REFERENCES 10951), Klaus Havelund, Jan Peleska, Bill Roscoe, and Erik de Vink (Eds.). Springer, 329–337. https://doi.org/10.1007/978-3-319-95582-7_19 [48] Alexander Gruler, Martin Leucker, and Kathrin D. Scheidemann. 2008. Mod- eling and Model Checking Software Product Lines. In Proceedings of the 10th International Conference on Formal Methods for Open Object-Based Distributed Systems (FMOODS’08) (LNCS, Vol. 5051), Gilles Barthe and Frank S. de Boer (Eds.). Springer, 113–131. https://doi.org/10.1007/978-3-540-68863-1_8 [49] Gerard J. Holzmann. 2003. The Spin Model Checker: Primer and Reference Manual. Addison-Wesley Professional. 10
https://openalex.org/W4313588262	https://djm.uodiyala.edu.iq/index.php/djm/article/download/953/753	Arabic	null	The role of anti-TPO as an additional analytical marker in thyroid disease patients in Erbil City	Maǧallaẗ Diyālá al-ṭibbiyyaẗ/Maǧallaẗ diyālá al-ṭibbiyyaẗ	2,022	cc-by	6,158	The role of anti-TPO as an additional analytical marker in thyroid disease patients in Erbil City Ronia Shawkat Kawther (MSc)1, Rebaz Tahir Lak (FIBMS-CM Sahar Mohammed Zaki Abdullah (MSc)3 1,2 Kurdistan Higher Council of Medical Specialties , Erbil, Iraq 3 College of Health Sciences, Hawler Medical University, Erbil, Iraq Ronia Shawkat Kawther (MSc)1, Rebaz Tahir Lak (FIBMS-CM)2 , Sahar Mohammed Zaki Abdullah (MSc)3 1,2 Kurdistan Higher Council of Medical Specialties , Erbil, Iraq 3 College of Health Sciences, Hawler Medical University, Erbil, Iraq ORIGINAL RESEARCH Published: 25 December 2022 Doi: 10.26505/DJM.23026680717 Diyala Journal of Medicine Correspondence Address: Ronia Shawkat Kawther Patients and Methods: This is a cross-sectional study achieved in the laboratories of Rizgary Teaching Hospital and Erbil Teaching Hospital in Erbil, Iraq, within the period of May 2020 to April 2021. The study included 66 patients, both males and females included. These patients had clinical indications and were suspected of having a sort of thyroiditis. Anti-TPO electrochemiluminescence immunoassay analysis has been included by the analysts in conjunction with conventional markers immunoassay of thyroid TSH, T3, and Free T4 trusting that this would help in lessening morbidity and related wellbeing concerns. Results: There was an increase in the level of anti-TSH antibodies in the hyperthyroidism group (60.6%) which was more than in those with hypothyroidism (36.4%), significant association exist (P≤0.049). Moreover, the level of Anti-TPO Abs was higher among hypothyroidism patients (63.6%) than among hyperthyroidism cases (33.3%), this association was statistically significant (P≤0.014). Conclusion: Demonstrating the clinical importance of this antibody and the benefit of adding anti-TPO, in combination with TSH and FT4. Addition of one test could potentially save expenditure on long-term diseases such as overt thyroid disease and its attended morbidities. y Keywords:Thyroid autoantibodies, thyroid peroxidase, thyroglobulin, thyroid-stimulating hormone receptor OPEN ACCESS Correspondence Address: Ronia Shawkat Kawther Kurdistan Higher Council of Medical Specialties , Erbil, Iraq Email: roniarr99@gmail.com Copyright: ©Authors, 2022, College of Medicine, University of Diyala. This is an open access article under the CC BY 4.0 license (http://creativecommons.org/licenses/by/4.0/) Website: https://djm.uodiyala.edu.iq/index.php/djm Received: 17 July 2022 Accepted: 21 August 2022 Published: 25 December 2022 Objective: Both alone or in grouping have been utilized to expect perfection of hypo or hyperthyroidism. Correspondence Address: Ronia Shawkat Kawther Abstract Background: There is a connection among anti-TPO and anti-TG antibodies and levels of thyroid hormone and both alone or in grouping have been utilized to expect perfection of hypo or hyperthyroidism. OPEN ACCESS Introduction they have various activities including metabolism, development, protein synthesis, and the control of numerous other vital hormones. Dysfunction of the thyroid gland can influence the generation of thyroid hormones (T3 and T4) which can be Thyroid organ has one of the foremost imperative functions in individual body because it regulates common physiological activities of the body. Thyroid hormones thyroxine (T4) and triiodothyronine (T3) are hormones delivered from thyroid gland and 120 December 2022 ,Volume 23, Issue 2 ORIGINAL RESEARCH Published: 25 December 2022 Doi: 10.26505/DJM.23026680717 ORIGINAL RESEARCH Published: 25 December 2022 Doi: 10.26505/DJM.23026680717 Diyala Journal of Medicine found a connection between changed levels of anti-thyroid antibodies and TSH and the development of hypothyroidism in euthyroid patients[6,7].Actually, a study linking anti- thyroid antibodies to the development of autoimmune thyroiditis came to the conclusion that cytotoxicity caused by complement was responsible [8]. connected to several pathologies throughout the body [1]. Thyroid dysfunction, which is described as a variety of conditions related to the thyroid gland has a significant negative impact on people's ability to prosper. In the United States, approximately 20 million people have thyroid illness in some form[1]. Functional thyroid dysfunction is often categorized as either hypothyroidism or hyperthyroidism, which are further split into overt and covert conditions[2]. As a result, their existence may really occur several years before the onset of obvious thyroid disease or abnormal tests of thyroid function. In anti-thyroid antibody persons, additional thyroid profile testing is very essential for making a timely diagnosis[9]. Studies on anti-thyroid antibodies in various patient populations, including SLE subsets, have varied (anti-TPO 54.76 percent) [10], patients with hepatitis C (anti-TPO 26.8%) [11], those who experience chronic urticaria (anti-TPO 57.4 percent and anti-TG 42.6 percent)[12]. However, the incidence of anti- thyroid antibodies and their relationship to the thyroid profile (TSH, T4, and T3) in the general population remain unknown. Building a correlation between anti-thyroid antibodies and thyroid profile testing may help identifying patients with exacerbated thyroid profiles who also need thyroid autoantibody testing to rule out a more serious autoimmune condition [13]. In light of this, we proposed the following hypothesis: Patients with an unbalanced thyroid profile had a higher prevalence of positive anti-thyroid tib di [14] Th f th i Since symptoms of thyroid disease are vague and progress gradually, many sufferers go untreated. Evidence regarding the efficacy of treatment in patients found by screening to have subclinical thyroid dysfunction is inconclusive. Introduction No trials of treatment of subclinical hyperthyroidism have been done [3,4]. Antibodies against thyrocytes' transmembrane protein involved in the production of thyroid hormone, known as anti-thyroid peroxidase (TPO). The precursor to thyroid hormone thyroglobulin is attacked by anti-thyroglobulin (TG) antibodies. Since these antibodies are present in more than 90% of instances of Hashimoto's thyroiditis and more than 80% of cases of Graves' disease, anti-TPO antibodies, also known as anti-thyroid microsomal antibodies, and anti- TG antibodies are regarded as diagnostic for autoimmune thyroid diseases (AITDs) [5]. autoimmune thyroid diseases. Because of their connection to thyroid stimulating hormone (TSH) levels, anti-TPO and anti-TG antibodies have both been employed separately or in combination to predict the beginning of hypo- /hyperthyroidism. Numerous studies have antibodies[14].Therefore, their presence may actually precede progress of evident thyroid disease or disturbed thyroid function tests by several years. Observing the results of lab tests of T3, free T4 hormone, TSH, anti- TPO, anti-TG, and 121 December 2022 ,Volume 23, Issue 2 ORIGINAL RESEARCH Published: 25 December 2022 Doi: 10.26505/DJM.23026680717 Diyala Journal of Medicine 5.0 to 12.0 mU/dL, 80-220 ng/dl, and O.3-4.5 IU/ml. anti-TSH, would identify those euthyroid subjects with potential risk of developing thyroid disease, prevent associated disease and long-term morbidity. Thus the researchers combined anti-TPO with conventional thyroid indicators. To lessen morbidity and associated health concerns. Immunoassay for the in vitro quantitative measurement of antibodies to thyroglobulin in human serum and plasma is the intended usage of Elecsys Anti-TG. A tool for identifying autoimmune thyroid disorders is the anti TG determination. The Roche Elecsys and Cobas e immunoassay analyzers are intended for use with the electrochemiluminescence immunoassay, or "ECLIA." Patients and Methods Between May 2020 and April 2021, this cross-sectional study was carried out in the labs of Erbil Teaching Hospital and Rizgary Teaching Hospital in Erbil, Iraq. Sixty-six participants in age range 18-65 both males and females (16,50 respectively) participated in the trial; they all exhibited clinical symptoms and were thought to have some form of thyroid illness. Standard sampling tubes or tubes with separating gel, sodium heparin, K2 and K3 EDTA plasma are used to collect serum. Recovery within 85 to 115 percent of serum value, or slope 0.85 to 1.15 plus intercept within 0.95, is the criterion. Use of sodium citrate plasma or Li heparin is prohibited. 3 days at 28 °C and 1 month at -20 °C stable. range of 10.0-4000 IU/mL measurements (defined by the lower detection limit and the maximum of the master curve). Lower detection limit values are stated as 4000 IU/ml. anticipated values Studies using the Elecsys anti-TG assay were done at the 94th percentile level for Roche's Elecsys® Anti- TSHR, which is now the threshold value of 115 IU/mL. In individuals suspected of having thyroid- related illnesses, we looked at the results of thyroid hormone function, paying particular attention to TSH, T3, T4, anti-TG, anti-TSH, and thyroid peroxidase antibody (anti-TPO) titers. They were quantified using the (cobas e 411 analyzer), a completely automated analyzer that performs immunoassay analysis using patented electrochemiluminescence (ECL) technology. The electrochemiluminescence immunoassay, or "ECLIA," developed by Roche to be used on the Electrosys and Cobas e immunoassay analyzers, was used to estimate TSH, T3, and T4. Range of measurement: 10.0-4000 IU/mL (defined by the lower detection limit and the maximum of the master curve). The electrochemiluminescence immunoassay, or "ECLIA," is designed for use on Elecsys and Cobas e immunoassay analyzers by Roche. Elecsys Anti-TSHR (TRAK) is a fully automated test for the identification of autoantibodies to the TSH receptor. 0.8 to 40 (results below the LoD are given as 0.8 IU/L) is the measurement range. Limit: 1.75 IU/L. The reported value for values below the lower detection limit is 10.0 IU/ml. Values that are higher than the measurement range are recorded as > 4000 IU/ml. The normal ranges for TSH, T4, and T3 are respectively The electrochemiluminescence immunoassay "ECLIA" is intended for use 122 December 2022 ,Volume 23, Issue 2 ORIGINAL RESEARCH Published: 25 December 2022 Doi: 10.26505/DJM.23026680717 Diyala Journal of Medicine Information privacy and secrecy were assured. Patients and Methods on (Elecsys) and (cobas e) immunoassay analyzers made by Roche. It is an immunoassay for the in vitro quantitative detection of antibodies to thyroid peroxidase in human serum and plasma (anti- TPO). Range of measurement: 5.00–600 IU/mL (defined by the lower detection limit and the maximum of the master curve). Anti-TPO has a normal range upper limit of >30 IU/ml [15]. Statistical Analysis The Factual Bundle for Social Sciences adaptation 26 was used to analyze the data (SPSS Inc., IBM Company, Chicago, Illinois, USA). Inferential results were contrasted between the individuals with different variables using a statistical significance level of 0.05, and when appropriate, Fisher's exact or Pearson Chi-square tests were used to examine the data. Descriptive analyses were expressed as frequencies and rates. The means and standard deviation of numerical data were examined. A questioner regarding the age and weather they were taking any medications was included and filled by all patients. In this study we excluded the following groups; children, adolescents and pregnant women. The prevalence of unsuspected thyroid disease is lowest in men and highest in older women [4]. Another study should be done to assess the effects of different screening methods (and subsequent management) for thyroid dysfunction in pre-pregnancy and during pregnancy on maternal and infant outcomes. Every patient had this study explained to them, and their verbal consent— or that of their legal guardian—was obtained. Results The total number of the studied sample was 66. The mean age ± SD of the studied sample was 39.05±13.73 years, mean ± SD of T3, T4, TSH, and Anti-TPO are 2.63±1.475, 118.48±79.105, 5.46±9.742, 91.35±121.108 respectively ranging from 47 to 65 years. The median was 44 years. Around two-thirds, 50 of the patients (62%) of the sample were females Table (1). Table (10):Descriptive Statistics of the study variables Table (10):Descriptive Statistics of the study variables Around 40 patients ( 60.6% ) had normal T3 levels while26 patients (39.4%) had abnormal T3 levels and among abnormal levels,14 patients( 42.4%) had hypothyroidism while 12 patients (36.4%) had hyperthyroidism while 19 patients (57.6%) of those with normal T3 had hypothyroidism and 21 patients (63.6% )of normal T3 patients had hyperthyroidism with no significant difference and p-value was 0.614 Table (2). Variables N Range Minimum Maximum Mean Std. Deviation Age 66 47 18 65 39.05 13.73 T3 66 5.41 1 6.41 2.63 1.47 T4 66 290.765 0.735 291.5 118.48 79.10 TSH 66 45.996 0.004 46 5.46 9.742 Anti-TPO 66 496 4 500 91.35 121.10 A d 40 i ( 60 6% ) h d l T3 (57 6%) f h i h l T3 Variables N Range Minimum Maximum Mean Std. Deviation Age 66 47 18 65 39.05 13.73 T3 66 5.41 1 6.41 2.63 1.47 T4 66 290.765 0.735 291.5 118.48 79.10 TSH 66 45.996 0.004 46 5.46 9.742 Anti-TPO 66 496 4 500 91.35 121.10 (57.6%) of those with normal T3 had hypothyroidism and 21 patients (63.6% )of normal T3 patients had hyperthyroidism with no significant difference and p-value was 0.614 Table (2). Around 40 patients ( 60.6% ) had normal T3 levels while26 patients (39.4%) had abnormal T3 levels and among abnormal levels,14 patients( 42.4%) had hypothyroidism while 12 patients (36.4%) had hyperthyroidism while 19 patients 123 December 2022 ,Volume 23, Issue 2 Diyala Journal of Medicine ORIGINAL RESEARCH Published: 25 December 2022 Doi: 10.26505/DJM.23026680717 Diyala Journal of Medicine Table (2): T3 level among hyperthyroidism and hypothyroidism groups T3 Groups Total Pearson Chi- square Hypothyroidism Hyperthyroidism abnormal 14 (42.4%) 12 (36.4%) 26 (39.4%) p-value 0.614 normal 19 (57.6%) 21 (63.6%) 40 (60.6%) Total 33 (100%) 33 (100%) 66 (100%) while 11(33.3%) of them had normal T4 level. Results Regarding those with hyperthyroidism 57.6% of them had abnormal T4 levels while 42.4% of them had normal T4 levels with no significant differences (p-value: 0.447). while 11(33.3%) of them had normal T4 level. Regarding those with hyperthyroidism 57.6% of them had abnormal T4 levels while 42.4% of them had normal T4 levels with no significant differences (p-value: 0.447). Table (3) shows that 41 (62.1%) of patients with thyroid diseases had an abnormal level of T4 while 25 (37.9%)of patients had normal T4 levels and that 22 patients (66.7%) with hypothyroidism had abnormal T4 levels Table (3): T4 level among hyperth Table (3) shows that 41 (62.1%) of patients with thyroid diseases had an abnormal level of T4 while 25 (37.9%)of patients had normal T4 levels and that 22 patients (66.7%) with hypothyroidism had abnormal T4 levels Table (3): T4 level among hyperth Table (3): T4 level among hyperthyroidism and hypothyroidism groups T4 Groups Total Pearson Chi- square Hypothyroidism Hyperthyroidism abnormal 22 (66.7%) 19 (57.6%) 41 (62.1%) P-value 0.447 normal 11 (33.3%) 14 (42.4%) 25 (37.9%) Total 33 (100%) 33 (100%) 66 (100%) According to Table (4), 56 patients (84.8% ) patients with thyroid diseases had abnormal TSH level 10 patients (15.2%) of patients had normal TSH level and 30 patients (90.9%) of those had hypothyroidism had abnormal TSH level and only 3 patients (9.1%) of them had normal level of TSH level and also 26 patients (78.7%) of those had hyperthyroidism had abnormal TSH level and 7 patients (21.2%) of them had normal TSH level with no significant difference (p- value: 0.170). Table (4): TSH level among hyperthyroidism and hypothyroidism groups Table (4): TSH level among hyperthyroidism and hypothyroidism groups TSH Groups Total Pearson Chi- square Hypothyroidism Hyperthyroidism abnormal 30 (90.9%) 26 (78.8%) 56 (84.8%) P- value 0.170 normal 3 (9.1%) 7 (21.2%) 10 (15.2%) Total 33 (100%) 33 (100%) 66 (100%) Table (5) revealed that although the level of ATG was higher in patients with hypothyroidism 9 patients (27.3%) in comparison with those with hyperthyroidism 8 patients (24.2%), but there was no significant statistical difference between ATG and study groups and p-value was 0.778. Results Table (5): Distribution of ATG between study groups ATG Groups Total Pearson Chi- square Hypothyroidism Hyperthyroidism Positive 9 (27.3%) 8 (24.2%) 17 (25.8%) P- value p:0.778 Negative 24 (72.7%) 25 (75.8%) 49 (74.2%) Total 33 (100%) 33 (100%) 66 (100%) Table (5): Distribution of ATG between study groups Table (5): Distribution of ATG between study groups 124 December 2022 ,Volume 23, Issue 2 ORIGINAL RESEARCH Published: 25 December 2022 Doi: 10.26505/DJM.23026680717 Diyala Journal of Medicine Table (6) showed that the level of Anti-TSH was higher in those with hyperthyroidism 20 patients (60.6%) than those with hypothyroidism which was 12 patients(36.4%), this difference was statistically significant,chi-square test was done and the p-value was 0.049. Table (6): Distribution of anti-TSH between study groups Anti-TSH Groups Total Pearson Chi- square Hypothyroidism Hyperthyroidism positive 12 (36.4%) 20 (60.6%) 32 (48.5%) P- value 0.049 negative 21 (63.6%) 13 (39.4%) 34 (51.5%) Total 33 (100%) 33 (100%) 66 (100%) Table (6): Distribution of anti-TSH between study groups (33.3%), this difference association was statistically significant. Chi square test was done and p-value was 0.014. Table (7) showed that the level of Anti-TPO level was higher among hypothyroidism 21 patients (63.6%) than those with hyperthyroidism which was11 patients Table (7): Distribution of anti-TPO between study groups Anti-TPO Groups Total Pearson Chi- square Hypothyroidism Hyperthyroidism positive 21 (63.6%) 11 (33.3%) 32 (48.5%) P- value p:0.014 negative 12 (36.4%) 22 (66.7%) 34 (51.5%) Total 33 (100%) 33 (100%) 66 (100%) Table (7): Distribution of anti-TPO between study groups In the current study, 62% of the participants were female, and the average age (± SD) of the sample under examination was 39.05±13.73 years. This is in agreement with a study by Meng et al. (2021), which found that women had a significantly higher overall incidence of hypothyroidism and hyperthyroidism than men did and that there was a significant tendency for the rate of hypothyroidism to increase with aging [19]. In the current study, 62% of the participants were female, and the average age (± SD) of the sample under examination was 39.05±13.73 years. This is in agreement with a study by Meng et al. (2021), which found that women had a significantly higher overall incidence of hypothyroidism and hyperthyroidism than men did and that there was a significant tendency for the rate of hypothyroidism to increase with aging [19]. Discussion The most prevalent autoimmune diseases that affect the thyroid, known as autoimmune thyroid disorders (AITDs), are Hashimoto's thyroiditis and Graves' infection [16]. Despite the fact that only about 1% of people have AITDs, 15% of people with euthyroidism may have subclinical or focal thyroiditis and circulating antithyroid antibodies [17]. In the current investigation, we discovered that a normal T3 level was present in 60.6% of patients with autoimmune thyroiditis. In almost all cases of hyperthyroidism, the T3 rises and typically does so before the T4 increases. In order to confirm hyperthyroidism that a suppressed TSH has Direct assessment of blood concentration and Thyroid Stimulating Hormone (TSH) have been widely widely and appropriately used as a screening test. Unfortunately, the trend has been to rely on TSH measurements alone for the assessment of complicated thyroid disease [8]. 125 December 2022 ,Volume 23, Issue 2 ORIGINAL RESEARCH Published: 25 December 2022 Doi: 10.26505/DJM.23026680717 ORIGINAL RESEARCH Published: 25 December 2022 Doi: 10.26505/DJM.23026680717 Diyala Journal of Medicine hormone, are known as anti-thyroglobulin (TG) antibodies [24]. already shown, T3 levels are preferred, since it is a more sensitive sign of hyperthyroidism than total T4 levels. Additionally, T3 tests are useful for determining whether a patient has T3 thyrotoxicosis, a kind of hyperthyroidism that displays as extremely elevated T3 and suppressed TSH levels while maintaining (normal) T4 levels [20]. Since (anti-TPO antibodies) are present in more than 90% of cases of Hashimoto's thyroiditis and more than 80% of cases of Graves' infection, they are regarded as diagnostic of AITDs [25]. When compared to the combined control group, both subclinical/overt hypothyroidism and hyperthyroidism exhibited a significant number of participants who have anti-TPO before the onset of thyroid dysfunction [26 ,27]. The cornerstone of thyroid function testing today is the serum thyroid stimulating hormone (TSH) concentration, which can be measured with an adequate sensitivity assay. For untreated groups of people that are at risk for primary thyroid dysfunction, a normal TSH concentration almost certainly rules out an anomaly. Nevertheless, serum TSH can provide a false signal of thyroid state in a number of crucial circumstances, most notably pituitary disorders and the early therapy of thyroid dysfunction [21]. Anti-TPO antibodies were more prevalent than anti-Tg antibodies in all outcomes, according to a study by Siriwardhane et al. They investigated whether these two indicators might be employed as standalone markers [28]. Discussion Microsomal antibody methods have been replaced by the TPO antibody radioimmunoassay, which is more sensitive and precise. Although anti-thyroglobulin assays are important for the accurate interpretation of serum thyroglobulin assays, they are less useful for the diagnosis of immunological thyroid disease [20, 30]. In this study, the prevalence of anti-TPO was higher in hypothyroid subjects (63.6%) compared to hyperthyroid subjects (33.3%), with significant differences (p-value: 0.014). This finding was in agreement with a study by Bromiska et al They found that anti-TPO titers were above reference range values in individuals with hypothyroidism (60%) and euthyroid participants (31.4%) [22]. Anti-thyroid antibodies are more frequently present when TSH is out of range, according to research done by Mistry et al. These people should be tested for anti-thyroid antibodies. The fact that a considerable portion of older adults may have anti-thyroid antibodies makes screening even more crucial [31]. Despite the fact that only 1% of people have AITDs, 15% of people with normal thyroid function may have subclinical and localized thyroiditis and circulating antithyroid antibodies [23]. A transmembrane protein of thyrocytes involved in the manufacture of thyroid hormone, is the target of anti-thyroid peroxidase (TPO) antibodies. Antibodies to thyroglobulin, a precursor to thyroid References References [1]Asaad DA, Sultan AS. Changes in Freet 3; T4 Thyroxine and Thyrotropin in Hemodialysis and Non-Dialysis Chronic Renal Failure Patients. Indian Journal of Public Health Research & Development. 2019:1;10(10). Conclusions The current study verified the relationship between thyroid function test and anti-TPO antibody values, demonstrating the clinical importance of this antibody and 126 December 2022 ,Volume 23, Issue 2 ORIGINAL RESEARCH Published: 25 December 2022 Doi: 10.26505/DJM.23026680717 Diyala Journal of Medicine potential empirical research will be important for defining which conflicts need to be better addressed and how to achieve this goal. recommending additional follow-up for those with high anti-TPO antibody titers in addition to a thorough and detailed clinical examination. Recommendations [1]Asaad DA, Sultan AS. Changes in Freet 3; T4 Thyroxine and Thyrotropin in Hemodialysis and Non-Dialysis Chronic Renal Failure Patients. Indian Journal of Public Health Research & Development. 2019:1;10(10). [1]Asaad DA, Sultan AS. Changes in Freet 3; T4 Thyroxine and Thyrotropin in Hemodialysis and Non-Dialysis Chronic Renal Failure Patients. Indian Journal of Public Health Research & Development. 2019:1;10(10). Our goal in doing this study was to demonstrate the advantage of using anti-TPO in addition to TSH and FT4 as a first-tier test. One additional test could potentially reduce costs associated with chronic conditions. Nevertheless, a large study should be performed especially on women of childbearing age, pregnancy, and cardiovascular diseases. In addition, involving other variables that can affect thyroid disease status such as current thyroid treatments and smoking habits with associated dysfunctions in reproductive health. [2]Hollowell JG, Staehling NW, Flanders WD, Hannon WH, Gunter EW, Spencer CA, Braverman LE. Serum TSH, T4, and thyroid antibodies in the United States population (1988 to 1994): National Health and Nutrition Examination Survey (NHANES III). The Journal of Clinical Endocrinology & Metabolism. 2002 1;87(2):489-99. [3]U.S. Preventive Services Task Force. Guide to clinical preventive services: report of the U.S. Preventive Services Task Force. 2d ed. Washington, D.C., Office of Disease Prevention and Health Promotion, 1996: 209–18. Source of funding: Kurdistan Higher Council of Medical Specialties. Ethical clearance: This study is based on a cross-sectional analysis of identified laboratory data that had been got scientific and ethical approval from Kurdistan higher council of medical specialties. The data and materials in this manuscript have not been published elsewhere and are not under consideration by another journal. [4]Helfand M. Screening for subclinical thyroid dysfunction in nonpregnant adults: a summary of the evidence for the US Preventive Services Task Force. Annals of internal medicine. 2004 20;140(2):128-41. [5]Walsh JP, Bremner AP, Feddema P, Leedman PJ, Brown SJ, O'Leary P. Thyrotropin and thyroid antibodies as predictors of hypothyroidism: a 13-year, longitudinal study of a community-based cohort using current immunoassay techniques. The Journal of Clinical Endocrinology & Metabolism. 2010 1;95(3):1095-104. [4]Helfand M. Screening for subclinical thyroid dysfunction in nonpregnant adults: a summary of the evidence for the US Preventive Services Task Force. Annals of internal medicine. 2004 20;140(2):128-41. [4]Helfand M. Screening for subclinical thyroid dysfunction in nonpregnant adults: a summary of the evidence for the US Preventive Services Task Force. Annals of internal medicine. 2004 20;140(2):128-41. [4]Helfand M. Recommendations Screening for subclinical thyroid dysfunction in nonpregnant adults: a summary of the evidence for the US Preventive Services Task Force. Annals of internal medicine. 2004 20;140(2):128-41. [5]Walsh JP, Bremner AP, Feddema P, Leedman PJ, Brown SJ, O'Leary P. Thyrotropin and thyroid antibodies as predictors of hypothyroidism: a 13-year, longitudinal study of a community-based cohort using current immunoassay techniques. The Journal of Clinical Endocrinology & Metabolism. 2010 1;95(3):1095-104. [5]Walsh JP, Bremner AP, Feddema P, Leedman PJ, Brown SJ, O'Leary P. Thyrotropin and thyroid antibodies as predictors of hypothyroidism: a 13-year, longitudinal study of a community-based cohort using current immunoassay techniques. The Journal of Clinical Endocrinology & Metabolism. 2010 1;95(3):1095-104. [5]Walsh JP, Bremner AP, Feddema P, Leedman PJ, Brown SJ, O'Leary P. Thyrotropin and thyroid antibodies as predictors of hypothyroidism: a 13-year, longitudinal study of a community-based cohort using current immunoassay techniques. The Journal of Clinical Endocrinology & Metabolism. 2010 1;95(3):1095-104. Conflict of interest: The size of our sample dictates the amount of information we have and consequently, in part, determines the precision or level of confidence that we have in our sample estimates and therefore how valid and reliable our conclusions will be. The larger the sample size the more information we have and so our uncertainty reduces. Identifying and managing financial and non-financial conflicts is needed by the development of specific committees and [6]Bossowski A, Moniuszko M, Idźkowska E, Grubczak K, Singh P, Bossowska A, 127 December 2022 ,Volume 23, Issue 2 ORIGINAL RESEARCH Published: 25 December 2022 Doi: 10.26505/DJM.23026680717 Diyala Journal of Medicine [13] Fade JV, Franklyn JA, Cross KW, Jones SC, Sheppard M. Prevalence and follow‐up of abnormal thyrotrophin (TSH) concentrations in the elderly in the United Kingdom. Clinical endocrinology. 1991 ;34(1):77-84. Diana T, Kahaly GJ. Decreased proportions of CD4+ IL17+/CD4+ CD25+ CD127− and CD4+ IL17+/CD4+ CD25+ CD127− FoxP3+ T cells in children with autoimmune thyroid diseases. Autoimmunity. 2016 Jul 3;49(5):320-8. Diana T, Kahaly GJ. Decreased proportions of CD4+ IL17+/CD4+ CD25+ CD127− and CD4+ IL17+/CD4+ CD25+ CD127− FoxP3+ T cells in children with autoimmune thyroid diseases. Autoimmunity. 2016 Jul 3;49(5):320-8. [7]Rodríguez Y, Rojas M, Monsalve DM, Acosta-Ampudia Y, Pacheco Y, Rodríguez- Jiménez M, Ramírez-Santana C, Anaya JM. Latent autoimmune thyroid disease. Journal of Translational Autoimmunity. 2020 1;3:100038. [14]Carrion M, Ramos-Levi AM, Seoane IV, Martinez-Hernandez R, Serrano-Somavilla A, Castro D, Juarranz Y, Gonzalez-Alvaro I, Gomariz RP, Marazuela M. Vasoactive intestinal peptide axis is dysfunctional in patients with Graves’ disease. Recommendations Scientific Reports. 2020 3;10(1):1-0. [8]Mikoś H, Mikoś M, Obara-Moszyńska M, Niedziela M. The role of the immune system and cytokines involved in the pathogenesis of autoimmune thyroid disease (AITD). Endokrynologia Polska. 2014;65(2):150-5. [8]Mikoś H, Mikoś M, Obara-Moszyńska M, Niedziela M. The role of the immune system and cytokines involved in the pathogenesis of autoimmune thyroid disease (AITD). Endokrynologia Polska. 2014;65(2):150-5. [15]Meena MQ. Diagnosing Graves’ Disease and Non-Graves Hyperthyroidism Using TSH Receptor Antibody Test versus Non- TSH Receptor Antibody Test Methods of Diagnosis. Open Journal of Endocrine and Metabolic Diseases. 2020 17;10(2):7-17. [16] Dittfeld A, Gwizdek K, Michalski M, Wojnicz R. A possible link between the Epstein-Barr virus infection and autoimmune thyroid disorders. Central-European journal of immunology. 2016;41(3):297. [15]Meena MQ. Diagnosing Graves’ Disease and Non-Graves Hyperthyroidism Using TSH Receptor Antibody Test versus Non- TSH Receptor Antibody Test Methods of Diagnosis. Open Journal of Endocrine and Metabolic Diseases. 2020 17;10(2):7-17. [9]Hutfless S, Matos P, Talor MV, Caturegli P, Rose NR. Significance of prediagnostic thyroid antibodies in women with autoimmune thyroid disease. The Journal of Clinical Endocrinology & Metabolism. 2011 1;96(9):E1466-71. [10]Riaz M, Afzal N, Mahmud TH, Shahzad F, Rasheed S, Rasheed A. The high percentages of anti-thyroid antibodies positive SLE patients at Sheikh Zayed Hospital, Lahore (Pakistan). Majmaah Journal of Health Sciences. 2015 ;216(2672):1-4. [17]Tipu HN, Ahmed D, Bashir MM, Asif N. Significance of testing anti-thyroid autoantibodies in patients with deranged thyroid profile. Journal of thyroid research. 2018 11;2018. [17]Tipu HN, Ahmed D, Bashir MM, Asif N. Significance of testing anti-thyroid autoantibodies in patients with deranged thyroid profile. Journal of thyroid research. 2018 11;2018. [18]Ross DS. Serum thyroid-stimulating hormone measurement for assessment of thyroid function and disease. Endocrinology and metabolism clinics of North America. 2001 1;30(2):245-64. [11]Shafiq MI, Gauhar A, Akram M, Elahi S. Thyroid peroxidase antibodies in non- interferon treated hepatitis C patients in Pakistan. BioMed Research International. 2015 3;2015. [11]Shafiq MI, Gauhar A, Akram M, Elahi S. Thyroid peroxidase antibodies in non- interferon treated hepatitis C patients in Pakistan. BioMed Research International. 2015 3;2015. [19] Meng Z, Liu M, Zhang Q, Liu L, Song K, Tan J, Jia Q, Zhang G, Wang R, He Y, Ren X. Gender and age impacts on the association between thyroid function and metabolic syndrome in Chinese. Medicine. 2015 ;94(50): 1-9. [19] Meng Z, Liu M, Zhang Q, Liu L, Song K, Tan J, Jia Q, Zhang G, Wang R, He Y, Ren X. Recommendations Gender and age impacts on the association between thyroid function and metabolic syndrome in Chinese. Medicine. 2015 ;94(50): 1-9. [19] Meng Z, Liu M, Zhang Q, Liu L, Song K, Tan J, Jia Q, Zhang G, Wang R, He Y, Ren X. Gender and age impacts on the association between thyroid function and metabolic syndrome in Chinese. Medicine. 2015 ;94(50): 1-9. [12] Aamir IS, Tauheed S, Majid F, Atif A. Frequency of autoimmune thyroid disease in chronic urticaria. J Coll Physicians Surg Pak. 2010 1;20(3):158-61. 128 December 2022 ,Volume 23, Issue 2 ORIGINAL RESEARCH Published: 25 December 2022 Doi: 10.26505/DJM.23026680717 ORIGINAL RESEARCH Published: 25 December 2022 Doi: 10.26505/DJM.23026680717 Diyala Journal of Medicine [26] Eggertsen R, Petersen K, Lundberg P-A, et al. Screening for thyroid disease in a primary care unit with a thyroid stimulating hormone assay with a low detection limit. Br Med J 1988; 297:1586-92. [20]Stockigt J. Assessment of thyroid function: towards an integrated laboratory- clinical approach. The Clinical Biochemist Reviews. 2003 ;24(4):109. [21]Fröhlich E, Wahl R. Thyroid autoimmunity: role of anti-thyroid antibodies in thyroid and extra-thyroidal diseases. Frontiers in immunology. 2017 9;8: 521. [27]Petersen K. Lindstedt G. Lundberg PA, et al. Thyroid disease in middle-aged and elderly Swedish women; thyroid-related hormones, thyroid dysfunction and goitre in relation to age and smoking. J Intern Med 1991;229:407-13. [22] Bromińska B, Bromiński G, Owecki M, Michalak M, Czarnywojtek A, Waśko R, Ruchała M. Anti-thyroidal peroxidase antibodies are associated with thyrotropin levels in hypothyroid patients and in euthyroid individuals. Ann Agric Environ Med. 2017 21;24(3):431-4. Michalak M, Czarnywojtek A, Waśko R, Ruchała M. Anti-thyroidal peroxidase antibodies are associated with thyrotropin levels in hypothyroid patients and in euthyroid individuals. Ann Agric Environ Med. 2017 21;24(3):431-4. [23]Roldán JC, Amaya-Amaya J, Castellanos-De La Hoz J, Giraldo-Villamil J, Montoya-Ortiz G, Cruz-Tapias P, Rojas- Villarraga A, Mantilla RD, Anaya JM. Autoimmune thyroid disease in rheumatoid arthritis: a global perspective. Arthritis. 2012;2012. [24]Saravanan P, Dayan CM. Thyroid autoantibodies. Endocrinology and metabolism clinics of North America. 2001 1;30(2):315-37. [25]Jo K, Lim DJ. Clinical implications of anti-thyroglobulin antibody measurement before surgery in thyroid cancer. The Korean journal of internal medicine. 2018 ;33(6):1050. [28]Siriwardhane T, Krishna K, Ranganathan V, Jayaraman V, Wang T, Bei K, Ashman S, Rajasekaran K, Rajasekaran JJ, Krishnamurthy H. Significance of anti-TPO as an early predictive marker in thyroid disease. Autoimmune diseases. 2019 :28;2019. Recommendations [23]Roldán JC, Amaya-Amaya J, Castellanos-De La Hoz J, Giraldo-Villamil J, Montoya-Ortiz G, Cruz-Tapias P, Rojas- Villarraga A, Mantilla RD, Anaya JM. Autoimmune thyroid disease in rheumatoid arthritis: a global perspective. Arthritis. 2012;2012. Montoya-Ortiz G, Cruz-Tapias P, Rojas- Villarraga A, Mantilla RD, Anaya JM. Autoimmune thyroid disease in rheumatoid arthritis: a global perspective. Arthritis. 2012;2012. [29]Uysal HB, Ayhan M. Autoimmunity affects health-related quality of life in patients with Hashimoto's thyroiditis. The Kaohsiung Journal of Medical Sciences. 2016 Aug1;32(8):427-33. [30]Akamizu T, Amino N. Hashimoto’s thyroiditis. Endotext [Internet]. 2017 : 17. [31]Mistry N, Wass J, Turner MR. When to consider thyroid dysfunction in the neurology clinic. Practical neurology. 2009 1;9(3):145- 56. [29]Uysal HB, Ayhan M. Autoimmunity affects health-related quality of life in patients with Hashimoto's thyroiditis. The Kaohsiung Journal of Medical Sciences. 2016 Aug1;32(8):427-33. [24]Saravanan P, Dayan CM. Thyroid autoantibodies. Endocrinology and metabolism clinics of North America. 2001 1;30(2):315-37. [24]Saravanan P, Dayan CM. Thyroid autoantibodies. Endocrinology and metabolism clinics of North America. 2001 1;30(2):315-37. [30]Akamizu T, Amino N. Hashimoto’s thyroiditis. Endotext [Internet]. 2017 : 17. [31]Mistry N, Wass J, Turner MR. When to consider thyroid dysfunction in the neurology clinic. Practical neurology. 2009 1;9(3):145- 56. [25]Jo K, Lim DJ. Clinical implications of anti-thyroglobulin antibody measurement before surgery in thyroid cancer. The Korean journal of internal medicine. 2018 ;33(6):1050. 56. 129 December 2022 ,Volume 23, Issue 2 ORIGINAL RESEARCH Published: 25 December 2022 Doi: 10.26505/DJM.23026680717 Diyala Journal of Medicine دور مضادات بيروكسيداز الغدة الدرقية كعالمة تحليلية إضافية في مرضى الغدة الدرقية في مدينة اربيل رونيا شوكت كو ثر١ ، ريباز طاهر لك ٢ ، سحر محمد زكي عبد هللا3 الملخص دور مضادات بيروكسيداز الغدة الدرقية كعالمة تحليلية إضافية في مرضى الغدة الدرقية في مدينة اربيل رونيا شوكت كو ثر١ ، ريباز طاهر لك ٢ ، سحر محمد زكي عبد هللا3 الملخص :خلفية الدراسة هناك عالقة بين االجسام المضادة البيروكسيدا ز ومضادات ثايروغلوبيولين ومستويات هرمون المنبه ، الغدة الدرقية وقد تم استخدام كالهما بمفرده أو في مجموعة لتوقعات في قصور الغدة الدرقية أو فرط نشاط الغدة .الدرقية اهداف الدراسة: .تم استخدام كالهما بمفرده أو في مجموعة لتوقعات في قصور الغدة الدرقية أو فرط نشاط الغدة الدرقية :المرضى والطرائق هذه دراسة مقطعية تم أجرائه .ا في مستشفى رزكاري التعليمي و مستشفى أربيل التعليمي في أربيل العراق خالل الفترة من أيار ٢0٢0 الى نيسان ٢0١٢ .و شملت الدراسة66 مريضا من الذكور و األناث. Recommendations هؤالء المرضى لديهم مؤشرات سريرية و يشتبه في أصابتهم بنوع من ألتهاب الغدة الدرقية.تم تضمين المقايسة المن اعية المضادة للبيروكسيدايس من قبل المحللين بالتزامن مع المقايسة المناعية التقليدية للغدة الدرقية، هورمون المنبه للغدة الدرقيةTSH مع هورمون الغدةالدرقية الحرT4,T3 .و أن هذا من شأنه المساعدة في تقليل المرض و المخاوف المتعلقة بالرفاهية اهداف الدراسة: .تم استخدام كالهما بمفرده أو في مجموعة لتوقعات في قصور الغدة الدرقية أو فرط نشاط الغدة الدرقية :المرضى والطرائق هذه دراسة مقطعية تم أجرائه .ا في مستشفى رزكاري التعليمي و مستشفى أربيل التعليمي في أربيل العراق خالل الفترة من أيار ٢0٢0 الى نيسان ٢0١٢ .و شملت الدراسة66 مريضا من الذكور و األناث. هؤالء المرضى لديهم مؤشرات سريرية و يشتبه في أصابتهم بنوع من ألتهاب الغدة الدرقية.تم تضمين المقايسة المن اعية المضادة للبيروكسيدايس من قبل المحللين بالتزامن مع المقايسة المناعية التقليدية للغدة الدرقية، هورمون المنبه للغدة الدرقيةTSH مع هورمون الغدةالدرقية الحرT4,T3 .و أن هذا من شأنه المساعدة في تقليل المرض و المخاوف المتعلقة بالرفاهية :النتائج كانت هناك زيادة في مستوى األجسام المضادة لهورمون المنبه للغدة الدرقية في مجموعة فرط نشاط الغدة الدرقية %( 60.6 ( ) والتي كانت أعلى من تلك الموجودة في مجموعة قصور الغدة الدرقية34.4% ) مع وجود داللة أحصائية بفرق (معنويP≤0.049 ) . باألضافة ألى ذلك كان مستوى مضادا ت البيروكسيدايس أعلى بين مرضى قصور الغدة الدرقية63.6 % (مقارنة بحاالت فرط نشاط الغدة الدرقية و كان هذا األرتباط ذو داللة أحصائيةبفرق معنويP≤0.014 ) . االستنتاجات: أثبات األهمية السريرية لهذا الجسم المضاد و فائدة أضافة مضاد البيروكسيدايس باألشتراك مع هورمون الثايرويد الحر و يمكن أن يؤدي أضافة أختبار واحد الى توفير النفقات على أألمراض طويلة المدى مثل مرض الغدة الدرقية و األمراض اهداف الدراسة: .تم استخدام كالهما بمفرده أو في مجموعة لتوقعات في قصور الغدة الدرقية أو فرط نشاط الغدة الدرقية :المرضى والطرائق هذه دراسة مقطعية تم أجرائه .ا في مستشفى رزكاري التعليمي و مستشفى أربيل التعليمي في أربيل العراق خالل الفترة من أيار ٢0٢0 الى نيسان ٢0١٢ .و شملت الدراسة66 مريضا من الذكور و األناث. Recommendations هؤالء المرضى لديهم مؤشرات سريرية و يشتبه في أصابتهم بنوع من ألتهاب الغدة الدرقية.تم تضمين المقايسة المن اعية المضادة للبيروكسيدايس من قبل المحللين بالتزامن مع المقايسة المناعية التقليدية للغدة الدرقية، هورمون المنبه للغدة الدرقيةTSH مع هورمون الغدةالدرقية الحرT4,T3 .و أن هذا من شأنه المساعدة في تقليل المرض و المخاوف المتعلقة بالرفاهية :النتائج كانت هناك زيادة في مستوى األجسام المضادة لهورمون المنبه للغدة الدرقية في مجموعة فرط نشاط الغدة الدرقية %( 60.6 ( ) والتي كانت أعلى من تلك الموجودة في مجموعة قصور الغدة الدرقية34.4% ) مع وجود داللة أحصائية بفرق (معنويP≤0.049 ) . باألضافة ألى ذلك كان مستوى مضادا ت البيروكسيدايس أعلى بين مرضى قصور الغدة الدرقية63.6 % (مقارنة بحاالت فرط نشاط الغدة الدرقية و كان هذا األرتباط ذو داللة أحصائيةبفرق معنويP≤0.014 ) . االستنتاجات: أثبات األهمية السريرية لهذا الجسم المضاد و فائدة أضافة مضاد البيروكسيدايس باألشتراك مع هورمون الثايرويد الحر و يمكن أن يؤدي أضافة أختبار واحد الى توفير النفقات على أألمراض طويلة المدى مثل مرض الغدة الدرقية و األمراض المصاحبة له. الكلمات المفتاحية: ، األجسام المضادة للغدة الدرقية ، بيروكسيداز الغدة الدرقية ، ثيروجلوبولين مستقبل الدرقية البريد:االلكتروني roniarr99@gmail.com تاريخ :استالم البحث ١7 تموز ٢0٢٢ تاريخ قبول البحث : ١٢ آب ٢0٢٢ ي :النتائج كانت هناك زيادة في مستوى األجسام المضادة لهورمون المنبه للغدة الدرقية في مجموعة فرط نشاط الغدة الدرقية %( 60.6 ( ) والتي كانت أعلى من تلك الموجودة في مجموعة قصور الغدة الدرقية34.4% ) مع وجود داللة أحصائية بفرق (معنويP≤0.049 ) . باألضافة ألى ذلك كان مستوى مضادا ت البيروكسيدايس أعلى بين مرضى قصور الغدة الدرقية63.6 % (مقارنة بحاالت فرط نشاط الغدة الدرقية و كان هذا األرتباط ذو داللة أحصائيةبفرق معنويP≤0.014 ) . الكلمات المفتاحية: ، األجسام المضادة للغدة الدرقية ، بيروكسيداز الغدة الدرقية ، ثيروجلوبولين مستقبل الدرقية البريد:االلكتروني roniarr99@gmail.com تاريخ :استالم البحث ١7 تموز ٢0٢٢ تاريخ قبول البحث : ١٢ آب ٢0٢٢ ١،٢ المجلس األعلى للتخصصات الطبية في كردستان- اربيل - العراق 3 كلية ال علوم ال صحية - جامعة هولير الطبية- اربيل– العراق 130 December 2022 ,Volume 23, Issue 2
https://openalex.org/W4367354334	http://jurnal.unpad.ac.id/responsive/article/download/44647/19191	Indonesian	null	PENGARUH BUDAYA ORGANISASI DAN KOMPENSASI TERHADAP PRESTASI KERJA KARYAWAN PT. SAI INDONESIA CABANG JAKARTA	Responsive	2,023	cc-by	4,201	ABSTRAK Penelitian ini dilakukan di PT. SAI Indonesia cabang Jakarta dan tujuan dari penelitian ini adalah untuk mengetahui pengaruh budaya organisasi dan reward terhadap kinerja pegawai pada cabang tersebut. Dalam penelitian ini yang menjadi populasi adalah karyawan PT. SAI Indonesia cabang Jakarta sebanyak 62 orang. Penelitian ini menggunakan metode kualitatif denagn menggunakan purposive sampeling dari 62 karyawan PT. SAI Jakarta yang ada di cabang Jakarta. Berdasarkan hasil analisis korelasi berganda sebesar 0,735 menunjukkan bahwa besarnya hubungan antara variabel bebas (budaya organisasi dan kompensasi) secara bersamaan atau simultan dengan variabel terikat (kinerja karyawan) adalah kuat yakni nilai korelasi gandanya adalah antara 0,60 dan 0,80. Berdasarkan hasil koefisien determinasi yaitu. 0,54 atau 54%, maka dapat disimpulkan bahwa budaya organisasi dan penghargaan menyumbang 54% variabel kinerja karyawan, sedangkan sisanya 46% dipengaruhi oleh variabel lain yang tidak diteliti. Berdasarkan dari hasil perhitungan uji F terlihat bahwa nilai F hitung > F tabel (49,916 > 1,670), maka Ho ditolak dan Ha diterima, yang artinya bahwa terdapat pengaruh yang signifikan budaya organisasi. dan kompensasi secara bersama- sama (secara simultan). terhadap prestasi kerja karyawan pada PT.SAI Indonesia cabang Jakarta dengan nilai signifikansi. adalah 0,000, lebih kecil dari 0,05. Kata kunci: Budaya orgaanisasi, Kompensasi, Prestasi kerja Karyawan Responsive: Jurnal Pemikiran Dan Penelitian Bidang Administrasi, Sosial, Humaniora Dan Kebijakan Publik, Volume 5 Nomor 4 Bulan Desember Tahun 2022 : 189 - 196 Responsive: Jurnal Pemikiran Dan Penelitian Bidang Administrasi, Sosial, Humaniora Dan Kebijakan Publik, Volume 5 Nomor 4 Bulan Desember Tahun 2022 : 189 - 196 Responsive: Jurnal Pemikiran Dan Penelitian Bidang Administrasi, Sosial, Humaniora Dan Kebijakan Publik, Volume 5 Nomor 4 Bulan Desember Tahun 2022 : 189 - 196 Submitted: 17-01-2023; Accepted: 26-01-2023; Published : 28-01-2023 Submitted: 17-01-2023; Accepted: 26-01-2023; Published : 28-01-2023 Jatenangan Manalu Sekolah Tinggi Ilmu Ekonomi Pengembangan Bisnis dan Manajemen Jakarta, Indonesia email: jatenangan1960@gmail.com Jatenangan Manalu Sekolah Tinggi Ilmu Ekonomi Pengembangan Bisnis dan Manajemen Jakarta, Indonesia email: jatenangan1960@gmail.com Submitted: 17-01-2023; Accepted: 26-01-2023; Published : 28-01-2023 A. Prestasi kerja Karyawan 1. Pengertian Prestasi kerja Karyawan 1. Pengertian Prestasi kerja Karyawan Kinerja positif yang ditunjukan oleh seorang karyawan akan berdampak langsung terhadap prestasi. Prestasi kerja merupakan dampak lansgung yang dapat dirasakan oleh setiap karyawan dari hasil apa yang dia lakukan dalam pekerjaannya baik bersifat fisik maupun non fisik. Steers dalam Ahmad Nur Rofi (2012) mengungkapkan bahwa prestasi kerja merupakan hal yang tidak terpisahkan dari kinerja seseorang, yang dapat dilihat dari peningkatan derajat jabatan seseorang dalam organisasi tempatnya bekerja. Penentuan prestasi kerja seseorang tentunnya mengikuti rangkaian atau proses sesuai dengan ketentuan atau standar yang ada di setiap organisasi atau perusahaan. Sebagaimana yang di ungkapkan oleh Handoko, T. Hani (2009:135) bahwa penilaian kinerja merupakn suatu proses dimana kerja karyawan dinilai atau dievaluasi oleh organisasi. Senada dengan yang disampaikan oleh DS Simanjutak dkk (2017) bahwa penilaian prestasi kerja adalah sebuah sistem manajemen yang mengatur pemberian reward berdasarkan prestasi kerjanya yang tentuya prestasi setiap karyawan berbeda. Faktor lain yang mempengaruhi prestasi kerja adalah kompensasi. Sebagaimana diutarakan oleh hakim dalam Andi Yolanda dkk (2021) bahwa kompensasi merupakan suatu hal yang dapat mengaktifkan upaya dan atau keinginan serta dorongan seseorang untuk melaksanakan tugas-tugas sesuai target dalam dunia pekerjaan. Kompensasi adalah suatu imbalan atau imbalan yang diterima oleh seorang karyawan dalam bentuk materil maupun immateriil dari perusahaan atau lembaga tempatnya bekerja. Jenis, bentuk dan besaran kompensasi mencerminkan penghargaan karyawan. Namun penyerahan kompensasi tersebut harus dilakukan terstruktur, sehingga menjadi pemicu bagi karyawan lain untuk meningkatkan motivasi kerja dalam meingkatkan produktivitas kerja karyawan tersebut. Sistem penghargaan yang baik, misalkan sistem reward yang mempertimbangkan kesesuaian reward dengan kebutuhan karyawan. Setidaknya kebutuhan hidupnya terpenuhi dan prestasi kerjanya dihargai tinggi. Hal tersebut diyakini dapat meningkatkan produktivitas karyawan. Dengan demikian dapat dikatakan bahwa penilaian prestasi kerja ialah hasil yang dapat diperoleh oleh setiap karyawan setelah melakukan setiap pekerjaan yang dibebankan kepadanyan. PT. SAI Indonesia Cabang Jakarta adalah salah satu perusahan yang sudah menerapakan kedua hal tersebut yakni budaya organisasi dan pemberian kompensasi kepada karyawan yang berprestasi, namun PT.SAI Indonesia Jakarta cabang Jakarta belum tahu persis seberapa pengaruh budaya dan kopensasi berefek pada prestasi kerja. Penelitian ini bermaksud untuk menganalisa seberapa pengaruh budaya organisasi organisasi dan kompesasi terhadap presestasi kerja dengan menjadikan 65 karyawan PT. SAI indonesi Cabang Jakarta sebagai samplingnya. PENDAHULUAN kinerja baik atau yang berprestasi. Dua cara ini merupakan cara yang umum yang dilakukan perusahaan. hal ini dipengaruhi oleh beberapa teori atau penelitian yang menunjukan bahwa adanya pengaruh antara budaya organisasi dengan prestasi karyawan. Wibowo dalam Vivi Rosvita dkk (2017) mengungkapkan bahwa budaya sangat berpengaruh dalam menunjukan cicta sebuah organisiasi, citra positif dari sebuah organisasi adalah akibat dari budaya positif yang dimiliki oleh organisasi tersebut dan sebaliknya. Di dalam organisasi, Sumber Daya Manusia( SDM) merupakan salah satu bagian yang penting dalam sebuah organisasi, tanpa peran SDM roda organisasi tidakan dapat berjalan dengan baik, sekalipun factor-faktor lain sudah tersedia. Hal ini menunjukan bahwa manuasia adalah sebagai penggerak. Mengingat pentingnya SDM dalam setiap organisasi, maka pengelolaan SDM sangat perlu diperhatikan oleh perusahaan. Setiap perusahaan memiliki pendekatan yang berbeda dalam mengelola SDM nya yang bisa berdampak pada kinerja karyawan. Beberapa cara perusahaan dalam meningkatkan prestasi kerja karyawan diantaranya penguatan dalam budaya organisasi dan pemberian kompensasi kepada karyawan yang menunjukan Budaya perusahaan adalah dasar atau pondasi untuk menciptakan lingkungan kerja yang baik. Mengembangkan budaya organisasi merupakan hal yang esensian untuk menjadikan kinerja karyawan jauh lebih optimal untuk mencapai organisasi yang lebih baik. Budaya yang diciptakan oragnisasi / perusahaan 189 Pengaruh Budaya Organisasi Dan Kompensasi Terhadap Prestasi Kerja Karyawan PT. SAI Indonesia Cabang Jakarta (Jatenangan Manalu) Pengaruh Budaya Organisasi Dan Kompensasi Terhadap Prestasi Kerja Karyawan PT. SAI Indonesia Cabang Jakarta (Jatenangan Manalu) Pengaruh Budaya Organisasi Dan Kompensasi Terhadap Prestasi Kerja Karyawan PT. SAI Indonesia Cabang Jakarta (Jatenangan Manalu) TINJAUAN PUSTAKA memberikan pandangan positif bahwa oragnisasi memiliki keselarasan antara tujuan dan menaikan motivasi kerja untuk mencapai kinerja yang lebih baik. Hal tersebut menjadi alasan orang lebih nyaman terhadap organisasi. Pemberdayaan, kepercayaan diri, sikap belajar dan kerja tim adalah beberapa bentuk yang kuat dari budaya organisasi. Budaya yang baik berdampak besar pada kinerja karyawan. 2. 2. Sedarmayanti (2017:72) ada dua belas faktor yang dapat meningkatkan prestasi kerja diantaranya: a. Motivasi Kerja, Umumnya karaywan dengan motivasi kerja tinggi akan bekerja keras dengan tekun untuk mencapai hasil kerja yang tinggi. b. kedisiplinan Orang dengan yang memiliki kedisipilna yang tingi dalam bekerja akan melakukan pekerjaan yang ditugaskan dengan penuh tanu tanggung jawab. Hal ini menjadi modal utama 190 Responsive: Jurnal Pemikiran Dan Penelitian Bidang Administrasi, Sosial, Humaniora Dan Kebijakan Publik, Volume 5 Nomor 4 Bulan Desember Tahun 2022 : 189 - 196 seorang karyawan untuk mencapai target kerja yang mana berefek langsung pada kinerja karyawan tersebut. seorang karyawan untuk mencapai target kerja yang mana berefek langsung pada kinerja karyawan tersebut. 3. Manfaat Penilaian Prestasi Kerja Nurmansyah dalam Said Muamarizal dkk (2015) mengungkapkan bahwa dalam penilaian prestasi dapat diperoleh hal-hal sebagai berikut: c. Pendidikan Pendidikan sangat berpengaruh dalam prestasi kerka seseoaran karena dengan pendidikan tinggi memungkin dia memiliki pandangan yang lebih luas. a. Tujuan penilaian Ini akan menentukan kebijakan terkait dengan membadingkan kinerja semua karyawan untuk memutuskan karyawan yang berprestasi. d. Keterampilan dan kemampuan Jika pekerja lebih terampil, dalam arti mereka dapat bekerja lebih baik dan menggunakan ruang kerja dengan baik. b. Tujuan Pengembangan Tujuannnya adalah untuk mengembangkan kompetensi / keahlian karyawan dan motivasi sebagai umpan balik karyawan dalam bekerja. Hasil dari penilaian ini akan memutuskan jenis pelatihan yang tepat untuk pengembangan karyawan tersebut. e. Budaya Organisasi Budaya kerja yang baik akan menciptakan suasana yang baik myang berdampak pada motivasi kerja yang tinggi. f. Tingkat Penghasilan Ketika tingkat pendapatan seorang karyawan tinggi, hal itu menyebabkan motivasi dan kosentrasi kerja sehingga kerjapun jadi lebih efektif dan efisien. Responsive: Jurnal Pemikiran Dan Penelitian Bidang Administrasi, Sosial, Humaniora Dan Kebijakan Publik, Volume 5 Nomor 4 Bulan Desember Tahun 2022 : 189 - 196 1. Pengertian Kompensasi Henry Simamora 1. Pengertian Kompensasi Henry Simamora (2004:441) mengungkapakan bahawa kopensasi adalah segala sesuatu yang diterima oleh karyawan baik bersifat finansial maupun non finasial sebagai bentuk hubungan kepegaiwaian. Dengan kata lain, kopensasi adalah imbalan yang diberikan perusahaan sebagai ganti kontribusi yang mereka berikan kepada perusahaan. Senada yang diungkapkan panggabean dalam Usman Fauzi (2014) menjelaskan tentang definisi kompensasi adalah penghargaan dari perusahaan yang diterima oleh karyawan atas segala jenis kontribusi terhadap perusahaan. Lebih lanjut lagi Saydam dalam Usman Fauzi (2014) mendefinisikan bahawa kompensasi adalah imbalan atas segala sesuatu yang dikeluarkan oleh karyawan baik berupa tenaga, pikiran, dan waktu kepada perusahaan. R. Wayne Mondy (2008: 5) mengungkapkan kompensasi non finansial dapat berupa perhargaan yang non materil yang diterima oleh karyawan seperti kepuasan dan kenyaman lingkungan kerja yang diperoleh seorang karyawan dalam dunia kerja. B. Populasi dan Sampel Populasi ialah semua subjek yang dijadikan bahan penelitian, sedangkan sampel adalah sebagian dari populasi yang diteliti (Sunyoto, 2012:177). Dalam penentuan sample seorang peneliti dapat menggunakan semua populasi sebagai sample jika populasinnya dibawah 100 orang. Jumlah karyawan PT. SAI Indonesia cabang Jakarta adalah 62 orang, jadi dalam penelitian ini peneliti menjadikan seluruh karyawan sebagai sample peneltian. Hal ini dikuatkan oleh Suharsimi (2006:131) mengungkapkan bahwa sample adalah diambil dari Sebagian dari populasi yang ada. Namun seandainya populasi dibawah 100 orang, maka populasi tersebut dapat diambil semua sebagai sample penelitian. Pada penelitian ini pengambilan sampel dengan menggunakan sampel jenuh yaitu sampel diambil sejumlah populasi dari karyawan PT.SAI Indonesia cabang Jakarta yaitu 62 orang 2. Tujuan Kompensasi Kompensasi sangat penting diberikan karenan dengan kompesasi dapat memberikan semangat kerja yang lebih, tercipata persaingan positif dalam bekerja, dan juga dapat menarik orang- orang yang berkualitas bergabung dengan perusahaan. Sebagaimana yang dikatakan Schuler dan Jackson dalam Ninuk Muljani (2002) tentang tujuan pemberian kompesansi adalah untuk menciptakan suasana kerja yang kompetitif, menjadikan karyawan loyal terhadap perusahaan serta untuk menarik calon karyawan potential bergabung dengan perusahaan. Oleh karena itu, sangat perlu dikondisikan oleh setiap perusahaan selain sebagai penunjang kerja yang lebih maksimal bahkan karyawan akan dengan sukarela menggunakan tenaga dan pikiran untuk meningkatkan kinerjanya, kompensasi sebagai daya daya tarik karyawan orang untuk bergabung dengan perusahaan tersebut. B. Budaya organisasi 1. Definisi Budaya organisasi Menurut Colquitt dkk Dalam Nur haris Efendi dan Sudirman (2021) budaya organisasi sebagai media untuk berbagi ilmu sosial dalam sebuah organisai yang mencakup peraturan, norma, dan nilai yang akan membetuk etika dan prilaku karyawan. Hal ini diperkuat oleh peryantaan dari Edgar H. Schein yang dikutip oleh Mangkunegara, A.A Anwar Prabu (2005: 113) mengungkapkan bahwa budaya organisasi merupakan satu set keyakinan, norma-norma dan nilai sebagai pedoman prilaku dalam sebuah organisasi. Hal ini diperlukan untuk sebagai landasan karywan dalam beradaptasi terhadap lingkungan kerja. 1. Definisi Budaya organisasi Menurut Colquitt dkk Dalam Nur haris Efendi dan Sudirman (2021) budaya organisasi sebagai media untuk berbagi ilmu sosial dalam sebuah organisai yang mencakup peraturan, norma, dan nilai yang akan membetuk etika dan prilaku karyawan. Hal ini diperkuat oleh peryantaan dari Edgar H. Schein yang dikutip oleh Mangkunegara, A.A Anwar Prabu (2005: 113) mengungkapkan bahwa budaya organisasi merupakan satu set keyakinan, norma-norma dan nilai sebagai pedoman prilaku dalam sebuah organisasi. Hal ini diperlukan untuk sebagai landasan karywan dalam beradaptasi terhadap lingkungan kerja. g. Gizi dan Kesehatan Jika karyawan dapat memenuhi kebutuhan nutrisinya dan memiliki fisik yang fit maka akan semakin kuat dalam bekerja yang dapat meningkatkan prestasi kerja. h. Jaminan Sosial Ketersedian Jaminan sosial bagi setiap karyawan akan memperkuat komitmen dan moral karyawan. Jika jaminan sosial karyawan cukup, dapat menyebabkan pengisian pekerjaan.Lingkungan Kerja i. Sarana Produksi Kualitas fasilitas produksi mempengaruhi peningkatan kinerja staf, karena orang dapat bekerja dengan antusias di fasilitas produksi yang lebih berkualitas. j. Teknologi Jika teknologi yang digunakan lebih tepat, memungkinkan volume produk yang lebih tinggi dan lebih berkualitas serta mengurangi limbah bahan bekas. 2. Fungsi Budaya Organisasi Robbins dan Judge dalam Rasmulia Sembiring dan Winarto (2020) menerangkan beberapa fungsi budaya organisasi adalah sebagai identitas suatu organisasi yang dapat menstabilkan urusan sosial antar karyawan serta mewujudkan komitmen perusahaan diatas urusan pribadi atau golongan dan juga budaya organisasi dapat membetuk sikap dan prilaku setiap karyawan yang ada. k. Kesempatan Berprestasi System karyawan berprestasi akan menciptakan psikologis bersaing atau kompetisi antar karyawan sehingga setiap karyawan ingin melakukan yang terbaik dalam setiap pekkerjaannya. 191 Pengaruh Budaya Organisasi Dan Kompensasi Terhadap Prestasi Kerja Karyawan PT. SAI Indonesia Cabang Jakarta (Jatenangan Manalu) Pengaruh Budaya Organisasi Dan Kompensasi Terhadap Prestasi Kerja Karyawan PT. SAI Indonesia Cabang Jakarta (Jatenangan Manalu) A. Metode Pengumpulan data Menurut Riduwan dalam Chesley Tanujaya menerangkan bahwa teknik pengumpulan data adalah “Metode pengumpulan data merupakan cara seseorang dalam mendapatkan data-data yang dibutuhkan selama penelitian. 3 teknik pengumpulan data yang digunakan dalam tulisan ini adalah studi Pustaka, penyebaran kuesioner, dan wawancara. METODE PENELITIAN A. Metode Pengumpulan data C. Kompensasi sifatnya non materil seperti lingkungan kerja yang nyaman, kebutuhan kerja yang diinginkan atau bentuk lainnya. Garry Dessler (2007:46) kompesasi finansial langsung dapat berupa gaji, bonus, dan komisi lainnya. Kompensasi finansial yang tidak langsung dapat berupa tunjangan Kesehatan, tunjangan pension, tunjangan Pendidikan, tunjangan perumahan, Pendidikan dan lain-lain. 3. Jenis-jenis Kompensasi Variabel Penelitian dan Definisi Operasional Adapun variabel yang digunakan dalam penelitian ini adalah variabel bebas dan varibael terikat. No Variable penelitian Definisi Variabel Indikator Variable Sumber 1 Budaya Organisasi (X1) Satu set keyakina n yang impleme ntasikan oleh stiap karyawa n dan digunaka (1) memiliki Inovasi dan Keberania n dalam mengambi l risiko; (2) Sangat Chesley Tanujaya (2017) dan kontribus i terhadap perusaha an. kedisiplia nan 6. Kerjasa ma yang kuat 7. Kuintita s pekerjaan 3 Prestasi Kerja (Y) Pencapai an seseoran g dari hasil kerja yang dilakuka n dalam dunia kerja. 1. Kualita s dan Kuantit as Kerja 2. Dapat diandalk an Hasrudy Tanjung (2017) Responsive: Jurnal Pemikiran Dan Penelitian Bidang Administrasi, Sosial, Humaniora Dan Kebijakan Publik, Volume 5 Nomor 4 Bulan Desember Tahun 2022 : 189 - 196 Responsive: Jurnal Pemikiran Dan Penelitian Bidang A Volume 5 Nomor 4 Bulan Des C. Variabel Penelitian dan Definisi Operasional Adapun variabel yang digunakan dalam penelitian ini adalah variabel bebas dan varibael terikat. No Variable penelitian Definisi Variabel Indikator Variable Sumber 1 Budaya Organisasi (X1) Satu set keyakina n yang impleme ntasikan oleh stiap karyawa n dan digunaka n untuk mengatas i kendala- kendal karyawa n dalam beradapp tasi dan sebagai dasar dalam menanga ni permasal ahan eksternal. (1) memiliki Inovasi dan Keberania n dalam mengambi l risiko; (2) Sangat memperjh atikan hal- hal yang detail.(3) mengutam akan hasil. Menjadika n Manusia sebagai asset penting. (5) mengutam akan kepentina gn tim (6) Agresif dan (7) Stabil. Chesley Tanujaya (2017) 2 Kompesan si (X2) Bentuk pengharg aam yang diterima oleh karyawa n sebagai bentuk imbalan atas jasa 1. keahlia n kerja 2. Kualita s kerja 3. Rasa Tanggung Jawab 4. Prakars a 5. Memili ki Ahmad fauzi (2014) 3. Jenis-jenis Kompensasi Kompensasi terbagi dalam 2 kelompok yakni kompensasi secara finansial dan non- finansial. Kompensasi finansial dapat bersifat langsung dan tidak langsung. Sedangkan kompesasi non-finasial bisa berupa hal yang 192 Responsive: Jurnal Pemikiran Dan Penelitian Bidang Administrasi, Sosial, Humaniora Dan Kebijakan Publik, Volume 5 Nomor 4 Bulan Desember Tahun 2022 : 189 - 196 C. Variabel Penelitian dan Definisi Operasional Adapun variabel yang digunakan dalam penelitian ini adalah variabel bebas dan varibael terikat. No Variable penelitian Definisi Variabel Indikator Variable Sumber 1 Budaya Organisasi (X1) Satu set keyakina n yang impleme ntasikan oleh stiap karyawa n dan digunaka n untuk mengatas i kendala- kendal karyawa n dalam beradapp tasi dan sebagai dasar dalam menanga ni permasal ahan eksternal. (1) memiliki Inovasi dan Keberania n dalam mengambi l risiko; (2) Sangat memperjh atikan hal- hal yang detail.(3) mengutam akan hasil. Menjadika n Manusia sebagai asset penting. (5) mengutam akan kepentina gn tim (6) Agresif dan (7) Stabil. Chesley Tanujaya (2017) 2 Kompesan si (X2) Bentuk pengharg aam yang diterima oleh karyawa n sebagai bentuk imbalan 1. keahlia n kerja 2. Kualita s kerja 3. Rasa Tanggung Jawab 4. Prakars a 5 Memili Ahmad fauzi (2014) dan kontribus i terhadap perusaha an. kedisiplia nan 6. Kerjasa ma yang kuat 7. Kuintita s pekerjaan 3 Prestasi Kerja (Y) Pencapai an seseoran g dari hasil kerja yang dilakuka n dalam dunia kerja. 1. Kualita s dan Kuantit as Kerja 2. Dapat diandalk an Hasrudy Tanjung (2017) HASIL DAN PEMBAHASAN 1. Korelasi dan determinasi Model Summary Model R R Square Adjusted R Square Std. Error of the Estimate 1 .735a .540 .529 .25227 a. Predictors: (Constant), Budaya organisasi, kompensasi Dari hasil analisis korelasi ganda dari hasil ringkasan model, nilai korelasi budaya iorganisasi dan penghargaan (kompensasi) karyawan terhadap kinerja adalah 0,735. Koefisien ini menunjukkan besarnya hubungan antara dependent variable yakni budaya organisasi dan kompensasi pada waktu yang sama atau bersamaan dengan independent variable yakni kinerja karyawan. Karena nilai korelasi ganda berkisar antara 0,60 hingga 0,80, maka dapat disimpulkan bahwa ada kinerja karyawan /prestasi karyawab sangat daapt dipengaruhi oleh budaya dan kompesasi yang diberikan oleh perusahaan. . Analisis determinasi yang diaplikasikan untuk mengetahui berapa besar dampak dari variable budaya dan reward (kompensasi) porsentase Responsive: Jurnal Pemikiran Dan Penelitian Bidang Administrasi, Sosial, Humaniora Dan Kebijakan Publik, Volume 5 Nomor 4 Bulan Desember Tahun 2022 : 189 - 196 C. C. Variabel Penelitian dan Definisi Operasional Adapun variabel yang digunakan dalam penelitian ini adalah variabel bebas dan varibael terikat. HASIL DAN PEMBAHASAN 2. Regresi berganda 3. Koefisien regresi (b2) variable X2 (kompensasi) adalah sebesar 0,280, angka tersebut dapat dimaknai bahawa jika diasumsikan kompensasi naik 1 satuan maka prestasi kerja akan meningkatkan 0,280 tingkat. Analisis regresi linier berganda bertujuan untuk melihat arah hubungan masing-masing variabel apakah setiap variabel saling berhubungan positif dan atau negative. g Coefficientsa Model Unstandardize d Coefficients Standar dized Coeffici ents t Sig. B Std. Error Beta (Const ant) 1.886 .226 8.328 .000 Buday a organi sasi .295 .109 .396 2.691 .000 Komp ensasi .280 .113 .364 2.473 .001 a. Dependent Variable: Prestasi kerja Karyawan Berdasarkan tabel diatas bentuk persamaan regresi ketiga variable tersebut dapat digambarkan dengan persamaan Y = 1,886 + 0,295X1 + 0,280X2 + e. Coefficientsa Model Unstandardize d Coefficients Standar dized Coeffici ents t Sig. B Std. Error Beta (Const ant) 1.886 .226 8.328 .000 Buday a organi sasi .295 .109 .396 2.691 .000 Komp ensasi .280 .113 .364 2.473 .001 a. Dependent Variable: Prestasi kerja Karyawan Berdasarkan tabel diatas bentuk persamaan regresi ketiga variable tersebut dapat digambarkan dengan persamaan Y = 1,886 + 0,295X1 + 0,280X2 + e. Coefficientsa Model Unstandardize d Coefficients Standar dized Coeffici ents t Sig. B Std. Error Beta (Const ant) 1.886 .226 8.328 .000 Buday a organi sasi .295 .109 .396 2.691 .000 Komp ensasi .280 .113 .364 2.473 .001 HASIL DAN PEMBAHASAN 1. Korelasi dan determinasi Model Summary Model R R Square Adjusted R Square Std. Error of the Estimate 1 .735a .540 .529 .25227 a. Predictors: (Constant), Budaya organisasi, kompensasi a. Predictors: (Constant), Budaya organisasi, kompensasi Dari hasil analisis korelasi ganda dari hasil ringkasan model, nilai korelasi budaya iorganisasi dan penghargaan (kompensasi) karyawan terhadap kinerja adalah 0,735. Koefisien ini menunjukkan besarnya hubungan antara dependent variable yakni budaya organisasi dan kompensasi pada waktu yang sama atau bersamaan dengan independent variable yakni kinerja karyawan. Karena nilai korelasi ganda berkisar antara 0,60 hingga 0,80, maka dapat disimpulkan bahwa ada kinerja karyawan /prestasi karyawab sangat daapt dipengaruhi oleh budaya dan kompesasi yang diberikan oleh perusahaan. . Analisis determinasi yang diaplikasikan untuk mengetahui berapa besar dampak dari variable budaya dan reward (kompensasi) porsentase terhadap prestasi karyawan. pengaruh variabel 193 Pengaruh Budaya Organisasi Dan Kompensasi Terhadap Prestasi Kerja Karyawan PT. SAI Indonesia Cabang Jakarta (Jatenangan Manalu) Persamaan ini dapat diinterpretasikan bahwa : budaya organisasi dan reward secara simultan terhadap kinerja karyawan. Berdasarkan tabel diatas maka dapat dilihat bahwa angka koofisien determinasi adalah sebesar 0,54 atau 54%. Nilai ini menunjukkan dampak dari keberadaan budaya organisasi dan pemeberian kompensasi akan mempengaruhi prestasi karyawan yakni 54% sedangkan sisanya sebesar 46% dapat dipengaruhi oleh faktor-faktor lain yang tidak diteliti dalam penelitian ini. Namun, dengan prosenstase 54% tersebut sudah cukup sebagai dasar bahwa budaya dan kompensasi saling berkaitan dalam menunjang prestasi karyawan. 1. Konstanta (a) bernilai sebesar 1,886 yang artinya jika budaya organisasi (X1) dan kompensasi (X2) nilainya adalah 0, maka prestasi kerja karyawan (Y) nilainya sebesar 1,886 (menurut skala likert berada dalam kondisi rendah) 2. Koefisien regresi (b1) variable X1 (budaya organisasi) adalah sebesar 0,295, nilai tersebut dapat diartikan bahwa seandainya diasumsikan X1 (budaya organisasi) naik 1 satuan maka prestasi kerja akan naik 0,295 tingkat Pengaruh Budaya Organisasi Dan Kompensasi Terhadap Prestasi Kerja Karyawan PT. SAI Indonesia Cabang Jakarta (Jatenangan Manalu) d. Uji f Uji F (uji koefisien regresi secara simultan diaplikasikan untuk mengetahui apakah independent variable (budaya organisasi dan kompensasi) secara simultan menunjukan dampak yang signifikan terhadap dependent variable (prestasi kerja karyawan). Hasil uji F dapat dilihat pada hasil ANOVA dengan cara mengkomparasikan antara nilai signifikansi dengan 0,05 atau membandingkan angka F hitung dengan F tabel Dari hasil tabel Anova dapat dilihat bahwa nilai F hitung > F tabel (49,916 > 1,670) ANOVAb Model Sum of Squares df Mean Square F Sig. 1 Regression 6.354 2 3.177 49. 916 .000a Residual 5.410 85 .064 Total 11.763 87 a. Predictors: (Constant), Budaya organisasi, kompensasi b. Dependent Variable: Prestasi kerja Karyawan Berdasarkan tabel diatas bentuk persamaan regresi ketiga variable tersebut dapat digambarkan dengan persamaan Y = 1,886 + 0,295X1 + 0,280X2 + e. b. Dependent Variable: Prestasi kerja Karyawan Dari hasil tabel Anova dapat dilihat bahwa nilai F hitung > F tabel (49,916 > 1,670) 194 Responsive: Jurnal Pemikiran Dan Penelitian Bidang Administrasi, Sosial, Humaniora Dan Kebijakan Publik, Volume 5 Nomor 4 Bulan Desember Tahun 2022 : 189 - 196 Responsive: Jurnal Pemikiran Dan Penelitian Bidang Administrasi, Sosial, Humaniora Dan Kebijakan Publik, Volume 5 Nomor 4 Bulan Desember Tahun 2022 : 189 - 196 maka dapat disimpulkan bahwa Ho ditolak dan Ha diterima. dikarenakan nilai F hitung (49,916) jauh lebih besar dari F tabel (3,191) maka dapat diinterpretasikan bahwa budaya organisasi dan kompensasi memberikan pengaruh yang signifikan secara simultan terhadap presetasi karyawan PT.SAI Indonesia cabang Jakarta yakni nilai signifikansi lebih kecil dari 0,05. DAFTAR PUSTAKA Ependi, N. H., & Sudirman, S. (2021). Pengaruh Budaya Organisasi dan Kepercayaan (Trust) terhadap Komitmen Organisasi. Jurnal Ilmu Pendidikan (JIP) STKIP Kusuma Negara, 12(2), 172-181. Fauzi, U. (2014). Pengaruh kompensasi terhadap kinerja karyawan pada PT. Trakindo Utama Samarinda. Jurnal Ilmu Administrasi Bisnis, 2(3), 172-185. Pengaruh Budaya Organisasi Dan Kompensasi Terhadap Prestasi Kerja Karyawan PT. SAI Indonesia Cabang Jakarta (Jatenangan Manalu) KESIMPULAN Berdasarkan pembahasan diatas maka dapat disimpulkan sebagai berikut: Handoko, T. Hani. (2009). Manajemen Personalia dan Sumber Daya Manusia. BPFE, Yogyakarta 1. Hasil analisis korelasi ganda adalah 0,735 ini menunjukkan besarnya hubungan yang terjadi antara variable independen (budaya organisasi dan kompensasi) secara serentak atau simultan terhadap variable depended (prestasi kerja karyawan) adalah kuat, karena nilai korelasi ganda berada diantara 0,60 - 0,80. Berdasarkan hasil koofisien determinasi yaitu 0,529 atau 54% ini dapat dimpretasikan bahwa dampak budaya organisasi dan pemberian kompensasi kepada karyawan sanagt mempengaruhi kinerja atau prestasi karyawan dengan porsentase sebesar 54% sedangkan sisanya sebesar 46% di pengaruhi oleh variable lain yang yang belum tidak diteliti dalam penelitian ini. Tanjung, H. (2017). Pengaruh Disiplin Kerja Dan Motivasi Kerja Terhadap Prestasi Kerja Pegawai Pada Dinas Sosial Dan Tenaga Kerja Kota Medan. Jurnal Ilmiah Manajemen Dan Bisnis, 15(1) Lubis, A. Y. O., & Susanti, F. (2019). Pengaruh Gaya Kepemimpinan Dan Kompensasi Terhadap Prestasi Kerja Karyawan (Studi pada PT Japfa Comfeed Indonesia (JCI) Tbk Devisi Fam 1. Mangkunegara, A.A. Anwar Prabu. (2011). Manajemen Sumber Daya Manusia Perusahaan, Remaja Rosda Karya, Bandung 2. Berdasarkan hasil perhitungan uji F terlihat bahwa nilai F hitung > F tabel (49,916 > 1,670) maka Ho ditolak dan Ha diterima, yang artinya bahwa terdapat pengaruh yang signifikan budaya organisasi dan kompensasi secara bersama-sama (secara simultan) terhadap prestasi kerja karyawan pada PT.SAI Indonesia cabang Jakarta dengan nilai signifikansi adalah 0,000, lebih kecil dari 0,05 Muamarizal, S. (2015). Pengaruh Pengalaman Kerja dan Penilaian Prestasi Kerja terhadap Pengembangan Karir Karyawan pada PT. Jasaraharja Putera Cabang Pekanbaru. Jurnal Online Mahasiswa (JOM) Bidang Ilmu Ekonomi, 2(1), 1-21. Nurrofi, A. (2012). Pengaruh disiplin kerja dan pengalaman kerja terhadap prestasi kerja karyawan pada departemen produksi PT. Leo agung raya semarang. Jurnal Ilmu Manajemen dan Akuntansi Terapan (JIMAT), 3(1). 3. Hasil penelitian ini menunjukan ada pengaruh yang kuat antara dependent variable terhadap independet variable yang dapat diaplikasikan oleh pelaku usaha dalam mengambil kebijakan sebagai salah satu cara untuk meningkatkan kinerja karyawan. Rosvita, V., Setyowati, E., & Fanani, Z. (2018). Pengaruh Budaya Organisasi Terhadap Kinerja Karyawan. Indonesia Jurnal Farmasi, 2(1), 46-52. Saran untuk penelitian berkelanjutan untuk melakukan penelitian yang sama dengan populasi lebih heterogeny dan jumlah lebih banyak serta menggunakan random sampling. Sedarmayanti. (2011). Sumber Daya Manusia dan Produktivitas Kerja, CV Mandar Maju, Bandung Sembiring, R., & Winarto, W. (2020). Pengaruh Budaya Kerja dan Komitmen terhadap 195 Pengaruh Budaya Organisasi Dan Kompensasi Terhadap Prestasi Kerja Karyawan PT. KESIMPULAN SAI Indonesia Cabang Jakarta (Jatenangan Manalu) Pengaruh Budaya Organisasi Dan Kompensasi Terhadap Prestasi Kerja Karyawan PT. SAI Indonesia Cabang Jakarta (Jatenangan Manalu) Kinerja Karyawan (Studi Kasus Pada Perawat Di Rumah Sakit Milik Pemerintah). Jurnal Ilmiah METHONOMI, 6(1), 21-30. Simanjuntak, D. S., Nadapdap, K. M. N., & Winarto, W. (2017). Pengaruh Persepsi Penilaian Prestasi Kerja terhadap Kepuasan Kerja Karyawan. Jurnal Manajemen, 3(2), 6-13. Simanjuntak, D. S., Nadapdap, K. M. N., & Winarto, W. (2017). Pengaruh Persepsi Penilaian Prestasi Kerja terhadap Kepuasan Kerja Karyawan. Jurnal Manajemen, 3(2), 6-13. Sunyoto, Danang. (2012). Konsep Dasar Riset Pemasaran dan Perilaku Konsumen. Cetakan ke-2. Yogyakarta: CAPS (Center for Academic Publishing Service) Sunyoto, Danang. (2012). Konsep Dasar Riset Pemasaran dan Perilaku Konsumen. Cetakan ke-2. Yogyakarta: CAPS (Center for Academic Publishing Service) Tanujaya, C. (2017). Perancangan Standart Operational Procedure Produksi Pada Perusahaan Coffeein. Jurnal Manajemen Dan Start-Up Bisnis, 2(1), 90-95. 196
https://openalex.org/W3119457747	https://ijs.uobaghdad.edu.iq/index.php/eijs/article/download/764/484	English	null	Chloroform Vapor Sensor Based on Air-Gap of the Mach-Zehnder Interferometer	Iraqi journal of science	2,019	cc-by	2,111	Chloroform Vapor Sensor Based on Air-Gap of the Mach-Zehnder Interferometer Dhuha Nihad Issa, Hanan J. Taher Dhuha Nihad Issa, Hanan J. Taher Institute of Laser for Postgraduate Studies, University of Baghdad, Baghdad, Iraq Iraqi Journal of Science, 2019, Vol. 60, No.5, pp: 996-999 DOI: 10.24996/ijs.2019.60.5.8 Iraqi Journal of Science, 2019, Vol. 60, No.5, pp: 996-999 DOI: 10.24996/ijs.2019.60.5.8 Issa and Taher ISSN: 0067-2904 Abstract The proposal of this study is demonstrating a simple vapor sensor for chloroform (CHCI3) utilitizing air gap region of the Mach-Zehnder interferometer (MZI) by using a single mode optical fiber coupler (3 dB) structure. In the last few decades, flammable liquids such as chloroform have been highly used. This chemical liquid has some degrees of carcinogenic effects in humans in addition to acute and chronic exposure results like blurred vision and nausea. The two arms of MZI contain a free space gap utilized to serve the sensing mechanism by adding chemical liquid volumes (0.2, 0.4, 0.6, 0.8, and 1) ml and to set the phase difference with air-gap distance 0.5 mm. The small variation in the effective refractive index of chloroform is due to changing the propagation of the transmitted light in the MZI’s gap that will further shift the optical phase of the signal. The experimental results indicate that the output power decreased when the air gap distance is 0.5 mm, and the shifting of wavelength will increase toward blue shift when increasing the chloroform liquid volumes. A maximum wavelength shift of chloroform is 0.0251 nm when the liquid volume is 1 ml and sensitivity of chloroform is 0.0223 nm/ml at air gap distance 0.5 mm for different chloroform liquid volumes. Keywords: Mach-Zehnder interferometer, chloroform vapor sensor, wavelength shift, sensitivity, phase shift. حدس بخار كلورفورم بأستخدم فجوة الهواء كمقياس التداخل ماخ-زندر ضحى نهاد عيسى1 ، حنان جعفر طاهر معيج الليدر للجراسات العليا، جامعة بغجاد، بغجاد، العخاق ا ____________________________ Email:duha93nihad@gmail.com Introduction The environmental sensor has been of interest to recent research, the optical fiber sensors are being utilized for different detecting applications, e.g. optical ﬁber sensors are utilized for recognizing the nearness of air poisons in the climate [1] and in addition to being utilized in the biomedical ﬁeld [2, 3]. A lot of research in optical fibers has been devoted to sense, and then found applications in the chemical field [4]. Concoction detecting innovation has turned out to be vital in a wide assortment of regions, including mechanical plants, exploring labs, home and different military applications [5]. In recent years the chemical detection techniques have been developed. Chemical sensing with optical fibers is one of the most interesting emerging sensor techniques because of the geometrical flexibility and the small size to be transmitted encoding information between the spectrometer and a remote sample [6]. A chemical transducer and an optical fiber are the two components that constitute an optical fiber chemical sensor. The optical fiber is the medium which propagates the light from the source to the chemical transducer and guides the light again from the transducer to the photo detector. A chemical parameter is then measured in the form of light modified by the transducer [7, 8]. In this work, a kind of Mach-Zehnder interferometer of chloroform vapor sensor is fabricated based on the air gap using optical fiber coupler (3 dB). The system of wavelength is 1550 nm designed, this wavelength is of low cost and available components and. The transmission spectrum of the proposed sensor is obtained and then measures the sensitivity of chloroform for different volumes and adjusts 0.5 mm air distance. الخالصة ا الغخض من الجراسة ىه اعيار متحدذ بخار كلهروفهم(CHCI3) لسشظقة فجهة اليهاء في مقياس التجاخل ماخدنجرMZI بأستخجام أحادي الشسط للسقدم اليف البرخي dB) 3 ). ,في العقهد القليلة الساضية أستخجم سائل شجيجة االشتعال مثل الكلهروفهرم. ىحه الدائل الكيسيائية لو بعض درجات التأثيخ السدخطشة في اإلندان حيث أن نتائج التعخض الظفيفة والسدمشة تذهش الخؤية والغثيان. يحتهي ذراعيMZI على فجهة ىهاء بسدافة0.5 مليسيتخ لتعسل كآلية استذعار بهاسظة اضافة حجم البخوبانهل( 0.2 , 0.4 , 0.6 , 0.0 , 1 ) مل ( ولزبط إشارة الظهر مع مدافة الفجهة0.5) م ليسيتخ . يتغيخ ثابت االنتذا ر β للزهء السخسل في فجهة ماخدنجر .بدبب التباين الرغيخ الحي سيديح اشارة الظهر البرخي أشارت الشتائج التجخيبية إلى ان الظاقة الخارجة تشخفض عشج مدافة الفجهة اليهائية0.5 مليسيتخ، والتغيخ في الظهل السهجي سيدداد نحه السشظقة الدرقاء عشج زيادة أحجام الدائل الكلهروفهرم ، وكانت أقرى تغيخ في الظهل ا لسهجي للكلهروفهرم ىه 0.0251 نانهمتخ عشجما يكهن حجم الدائل1 مل وحداسية الكلهروفهرم0.0223 نانهمتخ / مل في السدافة الفجهة اليهائية0.5 مليسيتخ لسختلف أحجام الدائل الكلهروفهرم. ____________________________ *Email:duha93nihad@gmail.com 996 Iraqi Journal of Science, 2019, Vol. 60, No.5, pp: 996-999 Iraqi Journal of Science, 2019, Vol. 60, No.5, pp: 996-999 Issa and Taher Materials and Methods Figure-1 shows the set-up utilizing light source which is a diode laser with wavelength of 1550 nm, two optical coupler (3 dB) types of single mode fiber, two chambers are Parallel Rectangle-shape sealed chambers, fabricated from Polyvinyl chloride (PVC) plastic with dimensions (length =24 cm, height = 14 cm, and width =16 cm), and finally optical spectrum analyzer (OSA). The MZI is manufactured utilizing 3 dB couplers, the first optical fiber coupler (3 dB) connected to the diode laser and the second optical fiber coupler (3 dB) joined to the OSA (YOKOKAWA, Ando AQ6370). Figure 1-Experimental work for chloroform vapor sensor. Sensing arm Reference arm Air gap MZI Diode laser OSA Figure 1-Experimental work for chloroform vapor sensor. A created MZI with air gap distance is 0.5 mm under controlled conditions, the length of both arms are the same. Chloroform liquid, with volumes (0.2, 0.4, 0.6, 0.8, and 1) ml, is injected in the sensing chamber while the reference arm contains only air. The two arms of MZI cause the phase shift when the air gap is distant and adjusted, and then the chloroform liquid volumes are injected by a medical syringe through a small opening at the top of chamber in the sensing arm. Thus, the small variation in the effective refractive index inside the sensing chamber causes phase dissimilarity in the MZI, due to the disparate phase velocities [9]. Since the phase difference and the phase velocities are dependent on wavelength, the propagating wavelength along the two different arms of the MZI creates a superposition pattern. The separation between the two peaks in a two-mode interferometer is deﬁned as [10]:- 2 ne (1) 997 Iraqi Journal of Science, 2019, Vol. 60, No.5, pp: 996-999 Issa and Taher where represents the wavelength, is the MZI length and ne is the effective difference in the refractive index between the two MZI arms. The changes in the RI in the MZI’s sensing arm can be deﬁned as [11]:- (2) φ β. . ne (2π/ ). . ne φ β. . ne (2π/ ). . ne ( where φ is the phase shift and β is the difference between the traveling signal’s initial and instant propagation constant (2π/ ne ). The sensitivity of chloroform was calculated by using the program Origin 9, the experimental measurement of the wavelength shift responds to the chloroform liquid volumes. From Fig. 3, it can be observed that air-gap distance at 0.5 mm shows the sensitivity (0.0223 nm/ml). The relationship between the wavelength shift and chloroform volume was linear when the liquid volume increased the sensitivity increased too. Results and Discussion 60, No.5, pp: 996-999 Issa and Taher Figure 3-Sensitivity of chloroform vapor sensor at air-gap distance 0.5 mm. Figure 3-Sensitivity of chloroform vapor sensor at air-gap distance 0.5 mm. Figure 3-Sensitivity of chloroform vapor sensor at air-gap distance 0.5 mm. Results and Discussion Figure-2 shows the transmission spectrum of MZI for air- gap distance with injected chloroform liquid volumes. The black color line refers to only air in air-gap region and the red, blue, pink, green, and indigo color lines refer to liquid volume (0.2, 0.4, 0.6, 0.8, and 1) ml in air-gap region respectively. The transmission spectrum displacement shows that the chloroform liquid volumes increased. The wavelength shift increased toward the blue region because the refractive index of chloroform was 1.43 while the refractive index of air was 1. The refractive index is of an inverse proportion with the wavelength as shown in equation (1), therefore the chloroform liquid with the higher refractive index will show the wavelength shift in the blue region. Figure 2-Transmission spectrum of chloroform at air-gap 0.5 mm. Figure 2-Transmission spectrum of chloroform at air-gap 0.5 mm. To calculate the changes in effective refractive index, propagation constant, and the phase shift can be done by using equations (1) and (2). This is shown in Table-1 below: Table 1-The numerical calculation of chloroform Air gap distance Liquid volume (ml) ∆λ (nm) ∆ne ∆β (nm-1)×103 ∆φᵒ ×103 0.5 mm 0.2 0.0074 0.152 0.848 0.257 0.4 0.0110 0.108 0.570 0.123 0.6 0.0164 0.072 0.382 0.055 0.8 0.0210 0.056 0.299 0.033 1 0.0251 0.047 0.250 0.023 The sensitivity of chloroform was calculated by using the program Origin 9, the experimental measurement of the wavelength shift responds to the chloroform liquid volumes. From Fig. 3, it can be observed that air-gap distance at 0.5 mm shows the sensitivity (0.0223 nm/ml). The relationship between the wavelength shift and chloroform volume was linear when the liquid volume increased the sensitivity increased too. To calculate the changes in effective refractive index, propagation constant, and the phase shift can be done by using equations (1) and (2). This is shown in Table-1 below: Table 1-The numerical calculation of chloroform To calculate the changes in effective refractive index, propagation constant, and the phase shift can be done by using equations (1) and (2). This is shown in Table-1 below: Table 1-The numerical calculation of chloroform Air gap distance Liquid volume (ml) ∆λ (nm) ∆ne ∆β (nm-1)×103 ∆φᵒ ×103 0.5 mm 0.2 0.0074 0.152 0.848 0.257 0.4 0.0110 0.108 0.570 0.123 0.6 0.0164 0.072 0.382 0.055 0.8 0.0210 0.056 0.299 0.033 1 0.0251 0.047 0.250 0.023 998 Iraqi Journal of Science, 2019, Vol. Conclusion In this paper, a novel and simple single mode MZI chloroform vapor sensor are proposed. Experimental results of chloroform vapor sensor show the increase in the chloroform liquid volume causes the shift in wavelength (blue shift). The sensitivity of this sensor is 0.0223 nm/ml. In this paper, a novel and simple single mode MZI chloroform vapor sensor are proposed. Experimental results of chloroform vapor sensor show the increase in the chloroform liquid volume causes the shift in wavelength (blue shift). The sensitivity of this sensor is 0.0223 nm/ml. Reference 1. Hisham. H.K. 2018. Optical Fiber Sensing Technology: Basics, Classifications and Applications, American Journal of Remote Sensing, 6(1): 1-5. 1. Hisham. H.K. 2018. Optical Fiber Sensing Technology: Basics, Classifications and Applications, American Journal of Remote Sensing, 6(1): 1-5. f g 2. Tubb, A.J.C., Payne, F.P., Millington, R.B. and Lowe, C.R. 1997. Single-Mode Optical Fiber Surface Plasma Wave Chemical Sensor, Sens. Actuators B Chem., 41: 71–79. 2. Tubb, A.J.C., Payne, F.P., Millington, R.B. and Lowe, C.R. 1997. Single-Mode Optical Fiber Surface Plasma Wave Chemical Sensor, Sens. Actuators B Chem., 41: 71–79. 3. Rhines, T.D., Arnold. 1989. Fiber-Optic Biosensor for Urea Based on Sensing of Ammonia Gas, Anal. Chim. Acta., 227: 387–396. 4. Rodriguez-Rodriguez, E., Barrales, R. and Torres, R.Q. 2016. Fiber Optics Evanescent Wave Sensor, Merida-DL Tentative, International Society for Optics and Photonics, 5: 17130–17134. 5. Liu, X., Cheng, S., Liu, H., Hu, S., Zhang, D. and Ning. H. 2012. A survey on gas sensing technology, Sensors (Switzerland), 12: 9635–9665. 4. Rodriguez-Rodriguez, E., Barrales, R. and Torres, R.Q. 2016. Fiber Optics Evanescent Wave Sensor, Merida-DL Tentative, International Society for Optics and Photonics, 5: 17130–17134. 5 Liu X Cheng S Liu H Hu S Zhang D and Ning H 2012 A survey on gas sensing 5. Liu, X., Cheng, S., Liu, H., Hu, S., Zhang, D. and Ning. H. 2012. A survey on gas technology, Sensors (Switzerland), 12: 9635–9665. 6. Wang, X.D. and Wolfbeis, O.S. 2008. Fiber-optic chemical sensors and biosensors (2008 Anal. Chem. 2013, 85: 487–508. 7. Timmer, B.H., Delft, K.M.Y. Otjes, R.P., Olthuis, W. and Berg, A.y.d. 2004. A miniaturized measurement system for Ammonia in air, Anal. Chim. Acta., 507(1): 139–145. 8. Pospíšilová, M., Kuncová, G. and Trögl. J. 2015. Fiber-Optic Chemical Sensors and Fiber-Optic Bio-Sensors, Sensors, 15: 25208-25259. 9. Patel, P.B. and Hamde, S.T. 2017. Analysis of the Mach-Zehnder interferometer waveguide structure for refractive index measurement, J Opt. 46(4): 398-402. 10. Soltanian, M.R.K., Amiri, I.S., Ariannejad, M.M., Ahmad, H. and Yupapin. P. 2017. A simple humidity sensor utilizing air-gap as sensing part of the Mach–Zehnder interferometer, Opt Quant Electron, 49: 308. 11. Salman, N.A. and Taher, H. 2017. Refractive index sensor based on a solid-core photonic crystal fiber interferometer, Iraqi Journal of Physics, 15: 99-105. 999
W3093886794.txt	https://iris.uniroma1.it/bitstream/11573/1449155/1/Grassi_Caucasians-in-Turkey_2020.pdf	en		The Cultural and Political Claims of the Caucasian Minorities in Turkey	Eurasiatica	2,020	cc-by	5,226	Armenia, Caucaso e Asia Centrale Ricerche 2020 a cura di Carlo Frappi e Paolo Sorbello The Cultural and Political Claims of the Caucasian Minorities in Turkey Fabio L. Grassi Sapienza Università di Roma, Italia Abstract More than in the previous years, in 2019 the organisations and the social media groups of the Turkish citizens who are fully or partly descendants of Caucasian refugees looked active not only around 21 May, their “genocide commemoration” day, but also around 2 May, remembering the 2 May 1923, when the Kemalist government deported Eastwards many Circassian villages located in Western Anatolia. In sum, we are witnessing that now the “Circassians of Turkey” (a term which generally includes NorthEastern Caucasians like Chechens and South-Eastern Caucasians like Abkhazians) are struggling not only for a worldwide recognition of the “Circassian genocide”, but also for an open debate on what has meant and means being “Circassian” in the Republic. This paper tries to draw an updated picture of what is up within Circassian intelligencija and what Caucasians of Turkish nationality are aiming at. Keywords Circassians. Turkey. Political Debate. Identity. Abkhazians. Summary 1 The Background. – 2 Jıneps. – 3 The Abkhazians. Eurasiatica 15 Edizioni Ca’Foscari e-ISSN 2610-9433 \| ISSN 2610-8879 ISBN [ebook] 978-88-6969-453-0 \| ISBN [print] 978-88-6969-454-7 Peer review \| Open access Submitted 2020-03-23 \| Accepted 2020-05-04 \| Published 2020-10-22 © 2020 \| cb Creative Commons Attribution 4.0 International Public License DOI 10.30687/978-88-6969-453-0/011 225 Fabio L. Grassi The Cultural and Political Claims of the Caucasian Minorities in Turkey 1 The Background This contribution is an ideal sequel of my book about the Circassian ordeal (Grassi 2014, 2017, 2018).1 The 21st of May is the day when the Circassians of Turkish nationality, who are the greatest Circassian community in the World and form the ‘golden share’ of the Caucasian diaspora in Turkey, mobilize and publicly commemorate the extermination and expulsion their ancestors suffered in 1862-4 due to the expansionist policy of the Tsarist Empire. And generally the other Caucasian-origined communities join them. But recently I have noticed a secondary mobilization around the 2nd of May. On 2 May 1923, eight months after the successful military conclusion of the Independence War, the Kemalist government deported Eastwards many Circassian villages located in Western Anatolia. The communities of these villages were accused to have collaborated with the Greek occupants, because they had been in contact with the most famous and ill-famed guerrilla leader of the war, Ethem the Circassian.2 Merging Islam and bolshevism, this brave and skilled chieftain wanted the resistance of the Muslim communities of Western Anatolia against Greeks and Allies to evolve in a socialist revolution. In Ankara he had realized that the leading pashas had far different ideas, so at the end he rebelled against Mustafa Kemal’s superior authority and escaped to the Greek-occupied area, living in exile the rest of his life (Grassi 2020, 197, 206, 211, 214). Hence, he is the only famous personality of contemporary Turkey whose nickname is directly associated to a non-Turkish community and at the same time the villain par excellence in the official history of Contemporary Turkey. Naturally, the Circassians of Turkey are very sensitive about this issue. They try to contrast this deprecative association in two ways. The first one is to recall the many Circassian-born Turkish military and civilian personalities who well served the Turkish state and the Turkish nation; 1 Grassi 2014, 2017 (shortened Turkish translation), 2018 (shortened English translation). Here I refer to the English version. To tackle once and for all an annoying yet unavoidable question and go ahead with the topic of this paper, my personal opinion is as follows: the war of extermination led by the Tsarist Empire against the native populations of the North-Western Caucasus and the expulsion of nearly all the survivors are one of most terrible misdeeds of 19th century; the ferocious elimination of the Armenians from Anatolia in 1915 was one of the most terrible misdeeds of 20th century; however, let them be labelled or not ‘genocide’, it is wrong to match them with the shoah and more in general with mass murders having a decisive ideological root. 2 See for example Yelbaşı 2018. Çerkes Ethem obviously was Circassian, but in this case Çerkes had an onomastic function, to distinguish him from other Ethems in a society where regular Western-type family names were absent (Republican Turkey adopted them in 1934-5). In some documents and works he can be mentioned as Çerkez Ethem. Actually, even today the Turkish word for ‘Circassian’ swings between the forms Çerkes (prevailing) and Çerkez. To be unquestionably Circassian are the Adiges (sometimes transcribed as Adıge). They accept and use the word Çerkes but prefer to be known as Adige. Eurasiatica 15 Armenia, Caucaso e Asia Centrale. Ricerche 2020, 225-236 226 Fabio L. Grassi The Cultural and Political Claims of the Caucasian Minorities in Turkey the second one is to challenge the official history in order to partially excuse Ethem’s shocking choice. This latter is a risky way, because Atatürk’s assertions against Ethem, especially the assertions included in the 36-hours speech he delivered in six days in 1927, are the cornerstone of the official history. It means that it is hard to belie official history without belying the Great Leader himself. Moreover, the apology of Ethem has been for long time a matter of far-left intellectuals, a quite embarassing aspect for conservative Circassians. The situation has changed in the last two decades, when a wider range of intellectuals, including some religious and conservative ones, have begun to openly challenge the official history. The crucial issue in this quarrel is the battle (more exactly the battles) of Gediz, in Western Anatolia (24 October-12 November 1920). In the abovementioned megaspeech Atatürk shortly and plainly stated that the Turkish forces, including Ethem’s warriors and units of the regular army under the command of Ali Fuat Cebesoy, had been defeated by the Greeks, whereas a growing number of authors argue that around Gediz the Turkish forces scored a victory.3 I have already recently observed that Atatürk imposed silence not only on the Armenian question but on almost everything that had happened before 1923 and that Kemalist regime was concrete poured over an exploded volcano (Grassi 2018, 132). Like the other non-Turkish Muslim communities, along the single-party era the Circassians and the other Caucasus-origined communities experienced a complete denial of their identity – with the unpleasant exception of the public deprecative memory of a Circassian who officially had committed betrayal – and could start their struggle for positive visibility and their quest for self-consciousness in the multiparty era, with a strong acceleration in the AKP-era. At the same time, a formerly concealed debate came to the surface: which ones, among the Caucasian population who suffered Russian conquest, are to be considered Circassian, which ones are not? A never-ending debate, I am inclined to argue. Another misfortune of the Caucasian diaspora in Turkey is that in the last twenty years the fight for the acknowledgement of the Armenian genocide has become an identity flag of the Turkish democratic intelligencija. The historians and the intellectuals forming this cultural-political milieu does not like the policies of massacre and expulsion suffered by Muslim communities such as the Caucasians and 3 See for example Armağan 2018, 74-9. The meaningful title of the chapter is “Tarih açılımı Refet Paşa ve Çerkez Ethem’i de kapsayacak?” (Will the Revisionist Wave in Turkish Historiography Include Refet Pasha and Ethem the Circassian?). Refet pasha was one of the first companions of Mustafa Kemal in the adventure of the war of independence. Like the others, he was quickly dropped out in favour of more acquiescent executors like İsmet İnönü. Eurasiatica 15 Armenia, Caucaso e Asia Centrale. Ricerche 2020, 225-236 227 Fabio L. Grassi The Cultural and Political Claims of the Caucasian Minorities in Turkey the Balkan Muslims to be recalled, because they are afraid that the nationalists use these events as an excuse for the decision taken by the Unionist government to eradicate the Armenians from Anatolia. Indeed, too many times the cold-blooded goal to understand what happened before 1915, during 1915 and after 1915, and why – i.e. what must be the single goal of a historian – has been sacrificed to the tactical needs of a (surely well-meaning) political-cultural fight. Thus Turkish revisionist historiography, just like the Western, looks as much ideologically biased as the Turkish official nationalist historiography, with – as far as I know – one bright exception: Fikret Adanır (Grassi 2019, 157-67). 2 Jıneps Circassian and, broadly speaking, Caucasian organizations and social media in Turkey and in the other countries where a more or less relevant diaspora is present (websites, social media etc.) are a very complex archipelago.4 All the organizations and organs of this kind in the world, from the organizations of the native Americans to the Szekely communities of Romania, run the risk to be very narrow-sighted, to remain confined in dealing with the life of the community and in cherishing the memory of the (generally tragic) turning point of their own history. They are generally stuffed with news of conferences, marriages, obituaries, calls to action in occasion of the key-dates, and show widespread deference to general and local political powers. An additional feature of the Caucasian diaspora are the above-mentioned taxonomical discussions: who is Adıge? Who is Circassian but not Adıge? Who is not Circassian? But some years ago a share of the Caucasian diaspora in Turkey realized that the particular identities in Turkey are a great national matter of democracy, that the preservation of the Caucasian identities, the opening of university chairs of Circassian language and culture, the opening of TV and radio broadcasts in Circassian, a wider knowledge of the Circassian genocide must go beyond the fences of the communitarian claims to become a way for a general democratization of Turkey and decided to issue a politically-oriented monthly magazine. Its name is Jıneps (‘resin drop’ in Circassian). It was founded in December 2005, when Turkey was experiencing a positive wave of democratization. This monthly magazine is also a tipically bilingual ‘bulletin of the community’ (the pages in adige are written in Cyrillic characters); it is also a house organ, but its cover pages are the ones of a militant leftist democratic 4 A brave, uncommon, useful attempt to draw a map of Circassian/Caucasian associationism in Turkey, Germany, Jordan and USA is Mattei 2019. Eurasiatica 15 Armenia, Caucaso e Asia Centrale. Ricerche 2020, 225-236 228 Fabio L. Grassi The Cultural and Political Claims of the Caucasian Minorities in Turkey magazine. Above the name of the magazine one can read the following slogan words: “Bağımsızlık demokrasi özgürlük ve birlik” (Independence, Democracy, Freedom and Unity). Under the name one can read “Çerkeslerin özgür sesi” (The Free voice of the Circassians). As a prominent and combative Circassian intellectual, Yalçın Karadaş, personally confirmed to me, intellectuals, militants and journalists who founded Jıneps consciously followed an admirable model, that of Agos (‘furrow’ in Armenian).5 Agos is the bilingual weekly newspaper of the Armenian community of Turkey. Its founder, the great intellectual and journalist Hrant Dink, wanted it to be an authoritative voice of the whole of democratic Turkey. After Hrant Dink’s shameful murdering (19 January 2007) Agos has remained faithful to Dink’s ideals. For example, it regularly hosts the articles of a well-known great Turkish democratic intellectuals as Baskın Oran, who is not an Armenian-origined Turkish citizen like Dink but a ‘standard Turk’ (I refrain from writing ‘ethnically Turk’, being the word ‘ethnically’ pure nonsense in the giant melting pot that is Turkey). Shortly, Agos is still a common ground of common fights for all those Turkish citizens who challenge the ‘official ideology’ and discriminatory policies. Jıneps looks as Agos’ younger brother. Here, however, we must add that the core of the Armenian community in Turkey prefers low profile, just like the Jew... and the Circassian. Indeed, Jıneps is the most important and interesting voice of this community, but it is all but universally beloved inside the community itself. Let’s take as example of the inspiration of Jıneps three recent issues, starting from March 2019. Nearly the whole of the cover page is dedicated not to a whatsoever Circassian/Caucasian internal issue but to the Women’s Day, which is not addressed in a rhetorical and innocuous way. On the contrary, the full-page title is on the brave resistance of the democratic women of Istanbul who had organized a ‘feminist night stroll’ against the ferocious assaults of the police who had tried to stop them, as their demonstration had been forbidden. Such a title makes us understand that Jıneps, just like Agos, has remained bravely faithful to its original mission even in these last years, when the level of freedom and democracy in Turkey has dramatically dropped. The cover of April 2019 issue (clearly ‘closed’ before 31 March) is far more multifaceted. On the top two titles: the first one, on the left, is about how had been celebrated the Newroz, the great celebration of the Spring equinox: “Newroz’da binler buluştu” (For the Newroz 5 Yalçın Karadaş, while enquiring the complex identity of his community, has been calling for a common democratic struggle of all the ethnical and religious communities of Turkey against the structural ‘monist’ and repressive attitude of all the regimes who have taken place the Republican age (Karadaş 2009). Eurasiatica 15 Armenia, Caucaso e Asia Centrale. Ricerche 2020, 225-236 229 Fabio L. Grassi The Cultural and Political Claims of the Caucasian Minorities in Turkey thousands of people gathered). As the readers know, the celebration of the Spring equinox is particularly important and historically has taken a mass and solemn form first of all among the Iranian peoples, being later adopted by several neighbouring peoples. It was for a long time forbidden and repressed in Turkey, insofar it was a symbol of Kurdish identity. The ‘W’ character was long time forbidden as well, because it was not part of the alphabet adopted in 1928. This title is in itself a sign of solidarity towards the Kurds and a sign of dislike of the ‘official ideology’ which has dominated Turkey since the foundation of the Republic. The second title, on the right, is a “let’s start the engine” for the coming anniversary of the tsitsekun, the word Circassians adopted as equivalent of the Armenian metz yeghérn.6 Indeed, under the word tsitsekun lies in smaller characters the Turkish translation: “Çerkes soykırımı” (Circassian Genocide). The main title this time is about the local elections that were about to be held. Again, no routine: “Daha fazla demokrasi Daha az nefret dili” (More Democracy. Less Hate Speech). This time, however, the main title does not stretch from edge to edge of the page, as it leaves two columns to the mass shootings of Muslims committed on 15 March 2019 in New Zealand by a ‘white supremacist’. But what the title, the photo and the subsequent article emphasize is the strong solidarity towards the Muslim community expressed by New Zealand’s Prime Minister. Less important news is given space towards the bottom of the page. A piece of news refers to the “Çoğulcu Demokrasi Partisi” (Pluralist Democracy Party), a recently-funded party born inside the Circassian community. As its name shows, this party claims to be an enlightened and liberal organization, but many liberal and leftist Circassians are quite sceptical about it, arguing that a new little party can only cause a dangerous fragmentation of the democratic forces. This concern was particularly high about the aforementioned local elections. The decision of this party to present its own candidates in Ankara and Istanbul, where a head-to-head race between the candidate of the government and the candidate of the opposition was expected, encountered harsh criticism and arose suspects of being a calculated service in favour of the AKP regime.7 It is interesting that the word tsitsekun does not belong to the alive adige language, the language spoken by the better part of those who feel themselves Circassian, but to ubıh language, a dead language we know mainly thanks to the researches of great scholars like Georges Dumézil and Hans Vogt and thanks to Tevfik Esenç, the last ubıh speaking human being, dead in 1992, who accepted to be recorded ( Jıneps, May 2019, 1; Grassi 2018, 22-3). Thus, a dead word of a dead language has been willingly chosen to symbolize the demographic and cultural destruction committed by the Tsarist policies. 6 Indeed, in the local elections held on 31 March, 2019, a head-to-head race occurred in Istanbul, where the common candidate of the main opposition parties, Ekrem 7 Eurasiatica 15 Armenia, Caucaso e Asia Centrale. Ricerche 2020, 225-236 230 Fabio L. Grassi The Cultural and Political Claims of the Caucasian Minorities in Turkey Let’s come to the number of May 2019. Obviously, the first and many of the following pages are full with the commemoration of the tsitsekun. The first page, in particular, contains the translation of Russian sources (dispatches and souvenirs) proving the genocidal policy of the Tsarist state. But this first page is not monopolized by the Genocide Day: two ‘windows’ are devoted to other issues: the first one to the Workers’ Day (1st of May); the second one to the three farleft militants hanged on 6 May 1972, in the frame of the turn of the screw imposed by the Army with the ‘coup d’état by communiqué’ of 12 March 1971. Especially the leader of the group, Deniz Gezmiş, is the sorrowful icon of Turkish left. Other sides of this first page are devoted to the visit to Turkey of the President of the Russian Republic of Adygeia (23-26 April) and to the results of the local elections. This number includes (page 17) a short article on the metz yeghérn, that Armenians commemorate on 24 April. The article records the demonstration held in Istanbul to commemorate the Armenian genocide day. The beautiful title is “Birlikte yaşama kültürüne hançer” (A Stab to the Coexistence Culture). Just before delivering the final version of this paper I can record with pleasure that the number of May 2020 devotes larger space, more exactly the better part of page 23, to the Armenian genocide with a long article titled “Ermeni soykırımı’nın 105. yılı – ‘Artık yüzleşin’” (105th Anniversary of the Armenian Genocide – ‘Face It, Time Has Come’). I have personally and/or publicly asked to some Circassians – all people fiercely claiming what the Tsarist Empire committed was a genocide – if they acknowledged the metz yeghérn as a genocide, and all of them replied they did. The ground for a sincere reciprocal and even common recognition and remembrance is ready. Yet, once again the situation of the Circassians is ambiguous. Like the Kurds, they were part of the winning side, the Muslims of Anatolia who during the Armageddon of 1914-22 wiped out almost all the Christians; then Caucasian identity was repressed or – for the Circassians – unpleasantly quoted in reference to Ethem. However the Caucasian-born communities had more than the Kurds a debt of gratitude towards Turks, they did not rebel, they were not subject to terrible repressions like the Kurds, in a higher rate compared to Kurds they were accepted as ‘brothers’ of the Turks and individually admitted in the cadres of the state. In sum, Circassians are different from the Armenians and the Kurds. For them, putting at the centre of their identity the cult of the genocide means shadowing their being part of the winning Muslim İmamoğlu, defeated the candidate of the government, the former Prime Minister Binali Yıldırım, for a handful of votes. The losing side obtained, with specious reasons, the repetition of the elections. In this second round (23 June) İmamoğlu won again, this time with a consistent gap. Eurasiatica 15 Armenia, Caucaso e Asia Centrale. Ricerche 2020, 225-236 231 Fabio L. Grassi The Cultural and Political Claims of the Caucasian Minorities in Turkey side. For the defeated and wiped-out Armenians there is no alternative. For the Circassians there is. Consequently, even a broad-minded magazine like Jıneps refrains from openly tackling the Armenian file. As for the psychological relations with the Kurds, they were much more victimized than the Circassians but later, in the years of the democratization process, were given a TV channel and other opportunities of free expression in their native languages. In spite of the recent developments, they still enjoy what they obtained about ten years ago. As a consequence, Circassians are inclined to think Kurds as the community who has obtained what they have not obtained yet. This feeling can be a drawback on the road of harmony and solidarity among Turkey’s ‘minorities’. Indeed, the last time I went to Istanbul I was invited to attend an event in the frame of the International Day of the Native Languages. I listened to the speeches of representatives of Circassians, Hemshins, Zazas, Abazas, Lazes and Pomaks (all of them frequently switched from their native language to Turkish) and enjoyed their songs. And these representatives more or less openly expressed the desire to see their native language gaining the same position as the Kurdish language. The Kurds were absent. Jıneps is a monthly 24-page magazine printed in 900 copies. All the people who work for Jıneps are voluntary. True, it is read along one month in many circles by more than 900 people. Moreover, it reaches a broader audience with its digital edition. Its likeness is not the one of an amateur bulletin; on the contrary, its graphic is accurate, professional and elegant. But it does not regularly host contributions of prominent opinion makers. Jıneps is a first step, and if you do not take the first step you cannot arrive anywhere. But it is still far from having reached the importance of the weekly Agos. The Armenian model is all but a popular review. It is a niche organ just like Jıneps. However, Agos has got a firm place in relevant intellectual circles and an authoritative voice in Turkish political debate. The positions of these two publications symbolize the huge gap still existing between the Armenians of Turkey and the far more numerous Caucasians of Turkey on the way of becoming permanent subjects and participants of the public discourse (let’s add that at the moment this gap is even greater if the two out-of-Turkey diasporas are compared). In the Republican era the surviving Armenians of Turkey struggled on as a de facto discriminated, half-tolerated, half-bullied millet. Caucasians were denied public recognition and remembrance of their sufferings, however they could feel part and were actually allowed to be part of the dominant community, on the condition to dismiss whatsoever public extra-Turkish identity. A man like Hrant Dink knew what to do once the slightest chance to raise publicly the question of ‘the 1915’ would rise in Turkey, being aware of how dangerous such a task was. Moreover, Dink was in contact – not always in full harmony – with a powerful diaspora, who had raised awareness about the Armenian Eurasiatica 15 Armenia, Caucaso e Asia Centrale. Ricerche 2020, 225-236 232 Fabio L. Grassi The Cultural and Political Claims of the Caucasian Minorities in Turkey tragedy among historians, men of culture, parliaments. Instead, on the one hand no Circassian fighting for the acknowledgement of the Circassian genocide runs the risk to be murdered for this reason; on the other hand, the voice of the Circassians abroad is very feeble and until now unable to influence the international public opinion. This ambiguous position is one of the reasons – maybe the most important – of a persistent incertitude and weakness in the ‘politicalcultural platform’ of the organized Caucasian diaspora. 3 The Abkhazians In the Caucasian diaspora archipelago the Abkhazians occupy a special place for several reasons. Firstly, despite having ties with the Circassians, they lived in the Southern side of the Caucasus, much closer to the Russian-Ottoman boundary. Secondly, their forced migration, occurred mostly in 1865-7, was a classic migration by land which produced a far lesser rate of casualties than the extermination/expulsion of the stricto sensu Circassians. Thirdly, a self-proclaimed independent Abkhaz state, backed by Russian Federation but not recognized by the ‘international community’ (just like the self-proclaimed SouthOssetian state), has seceded from Georgia. Therefore the consistent Abkhaz diaspora in Turkey does not share the generally anti-Russian mood of the Caucasian diaspora. On the contrary, it must cope with the pro-Georgia stance of the Turkish state. Turkey records important relations with Georgia, which is a crucial partner in the distribution of Azerbaijani oil and gas. Consequently, the official position of Ankara is quite clear: Turkey strongly supports territorial integrity of Georgia and does not recognize the so-called independence of Abkhazia and South Ossetia. Turkey hopes that these conflicts will be resolved within Georgia’s territorial integrity and sovereignty through peaceful means. Turkey also supports Georgia’s efforts for integration with Euro-Atlantic organizations.8 Under the surface, however, the situation is not that simple. Together with the Mesketian Turks, Abkhazians are a thorny issue in Turkish-Georgian relations. The pure descendants of the Abkhaz refugees in Turkey are some 100,000. They are the core of a much wider ‘Abkhaz-participated familiar area’. This core has got direct family 8 Republic of Turkey. Ministry of Foreign Affairs (2011). “Political Relations between Turkey and Georgia”. http://www.mfa.gov.tr/relations-between-turkey-and-georgia.en.mfa. Eurasiatica 15 Armenia, Caucaso e Asia Centrale. Ricerche 2020, 225-236 233 Fabio L. Grassi The Cultural and Political Claims of the Caucasian Minorities in Turkey links with the Abkhazians living in Abkhazia and forms a powerful lobby. While Ankara should abide by the rule of economic embargo to the self-proclaimed independent Abkhazia, in practice an intense direct trade is run between Turkey and Abkhazia, tolerated by Turkish authorities. Moreover, many Turkish citizens go to Abkhazia passing through Russia (Göksel 2013, 4-5).9 Naturally Abkhaz authorities view the diaspora as an important political ally, as well as an economic and demographic resource, and have encouraged Turkish Abkhaz to resettle in Abkhazia [...]. As residence in Abkhazia is not a requirement for Abkhaz citizenship (which is open to all ethnic Abkhaz worldwide), the number of diaspora representatives holding Abkhaz passports is much larger. (Weiss, Zabanova 2016, 2) In sum, Circassians and Chechens want the Republic of Turkey to put in its political agenda the defence of their rights in the international arena and the recognition of 1862-4 facts as a genocide. They know that if Turkey opened these two files serious consequences in Turkish-Russian relations would occur, but they desire it and are covertly disappointed when Moscow-Ankara relations look good. On the contrary, the Abkhazians are sad when Moscow-Ankara relations are tense and are happy when they are – or look – good: The Federation of Abkhaz Associations (Abhaz Dernekleri Federasyonu, or Abhazfed), which is the leading diaspora organisation, established in 2010, has been known for its generally pro-Russian stance. In the wake of the rift between Russia and Turkey, Abhazfed publicly stated its loyalty to the Turkish government, yet refrained from criticising Russia directly. Shortly after Turkey’s downing of a Russian military jet on the Syrian border, Abhazfed representatives visited the Russian Ambassador in Ankara to discuss future relations and promote dialogue. In April 2016, diaspora activists and Turkish think-tank analysts took part in a round table in Sukhum(i) with the participation of Abkhaz officials, as well as Russian MPs, businesspeople, and pro-government experts, to discuss options for improving Russian-Turkish relations. It is likely ‘Militant’ Abkhazians affirm that the descendants of the Abkhaz refugees are some 500,000, i.e. much more than the some 125,000 Abkhazians living in the self-proclaimed independent Republic of Abkhazia. As modern Turkey is a giant melting pot, it is absolutely realistic to esteem up to 500,000 the number of Turkish citizens who have also Abkhaz ascendants, but this does not mean that the main identity of all of them is Abkhaz (they may have got either a plain Turkish identity or the identity of another particular group). Therefore it looks realistic that the correct figure of people feeling themselves mainly Abkhazian is 300,000 ca., of which no more than 25,000 are native or fluent Abkhazian-speakers. 9 Eurasiatica 15 Armenia, Caucaso e Asia Centrale. Ricerche 2020, 225-236 234 Fabio L. Grassi The Cultural and Political Claims of the Caucasian Minorities in Turkey that the Abkhaz diaspora’s conciliatory stance towards Russia protected it from harsher repercussions. (Weiss, Zabanova 2016, 4-5) Correspondently, Circassians love Georgia, until now the only state officially recognizing tsitsekun as a genocide, and look grimly at the ‘opportunistic’ choice of the Abkhazians. We can notice that backing the secessions of Abkhazia and South Ossetia Vladimir Putin has succeeded not only in punishing Georgia and in reaffirming Russian presence, if not hegemony, in ‘Transcaucasia’, but also in dividing the Caucasian diaspora. Bibliography Armağan, M. (2018). Paşaların hesaplaşması [The Revenge of the Pashas]. İstanbul: Ketebe Yayınları. Besleney, Z.A. (2014). The Circassian Diaspora in Turkey. A Political History. London: Routledge. Ferrari, A. (2012). “I Circassi in Russia. Un genocidio sconosciuto?”. Ferrari, A. (a cura di), Il grande paese. Saggi sulla storia e la cultura della Russia. Milano: Mimesis, 199-208. Göksel, D.N. (2013). “Turkey and Georgia. Zero-Problems?”. On Wider Europe. Black Sea Trust for Regional Cooperation, 19 June 2013. https://www.gmfus.org/publications/turkey-and-georgia-zero-problems. Grassi, F.L. (2014). Una Nuova Patria. L’Esodo dei Circassi verso l’Impero Ottomano. Istanbul: ISIS. Grassi, F.L. (2017). Yeni bir Vatan. Çerkeslerin Osmanlı İmparatorluğuna zorunlu Göçü (1864) )(A New Homeland. The Forced Migration of the Circassians in the Ottoman Empire). İstanbul: Tarihçi Kitabevi. Grassi, F.L. (2018). A New Homeland. The Massacre of the Circassians, Their Exodus to the Ottoman Empire, Their Place in Modern Turkey. Istanbul: Istanbul Aydın University Publications. Grassi, F.L. (2019). “On Turkish Historiography”. Ciampani, A.; Ugolini, R. (eds), The Great War. A European Commitment of Research and Reflection. Rome; Soveria Mannelli: Istituto Storico del Risorgimento Italiano; Rubbettino, 151-67. Grassi, F.L. (2020). Atatürk. Il fondatore della Turchia moderna. Roma: Salerno. Hacısalihoğlu, M. (ed.) (2014). 1864 Kafkas Tehciri. Kafkasya’da Rus Kolonizasyonu, Savaş ve Sürgün. Caucasian Exodus of 1864. Russian Colonization of Caucasia, War and Exodus. Istanbul: Yıldız Teknik Üniversitesi Balkan ve Karadeniz Araştırmaları Merkezi (BALKAR) & İslam Tarih, Sanat ve Kültür Araştırma Merkezi (IRCICA); Yıldız Technical University Center for Balkan and Black Sea Studies (BALKAR) & Research Center for Islamic History, Art and Culture (IRCICA). Karadaş, Y. (2009). Çerkes Kimliği. Türkiye’nin Sorunları (Circassian Identity. Turkey’s Troubles). İstanbul: Sorun Yayınları. Kutay, C. (1973). Çerkes Ethem Dosyası (The Ethem. The Circassian Dossier). İstanbul: Boğaziçi Yayınevi. Eurasiatica 15 Armenia, Caucaso e Asia Centrale. Ricerche 2020, 225-236 235 Fabio L. Grassi The Cultural and Political Claims of the Caucasian Minorities in Turkey Mattei, A. (2019). Ricordo, identità, rivendicazioni. Un monitoraggio della diaspora circassa [Remembrance, Identity, Claims. A Monitoring of the Circassian Diaspora]. [MA thesis]. Rome: La Sapienza University of Rome. Richmond, W. (2013). The Circassian Genocide. New Brunswick: Rutgers University Press. Sarıarslan, N. (2020). Kahramandan Haine. Çerkez Ethem. Hero to Traitor. Circassian Ethem. https://www.academia.edu/42033689/Kahramandan_Haine_Çerkez_Ethem_Hero_to_Traitor_Circassian_Ethem. Weiss, A.; Zabanova, Y. (2016). “Georgia and Abkhazia Caught between Turkey and Russia. Turkey’s Changing Relations with Russia and the West in 2015-2016 and their Impact on Georgia and Abkhazia”. SWP Comments 54. German Institute for International and Security Affairs. https://www.swpberlin.org/fileadmin/contents/products/comments/2016C54_ wis_zbv.pdf. Yelbaşı, C. (2018). “Exile, Resistance and Deportation. Circassian Opposition to Kemalists in South Marmara in 1922-1923”. Middle Eastern Studies, 54(6), 936-47. https://doi.org/10.1080/00263206.2018.1473249. Online sources A selection of the websites of Circassian organizations in Turkey: http://abhazfederasyonu.org https://www.facebook.com/cerkesethembey http://www.demokratikcerkeshareketi.org http://istanbulkafkaskultur.net http://www.kafdav.org.tr https://www.kaffed.org/en Eurasiatica 15 Armenia, Caucaso e Asia Centrale. Ricerche 2020, 225-236 236
https://openalex.org/W2969841087	https://www.zora.uzh.ch/id/eprint/186051/3/40510_2019_Article_285.pdf	English	null	Paediatricians’ awareness on orthodontic problems and related conditions—a national survey	Progress in orthodontics	2,019	cc-by	4,319	Zurich Open Repository and Archive University of Zurich University Library Strickhofstrasse 39 CH-8057 Zurich www.zora.uzh.ch RESEARCH Open Access * Correspondence: mkoufatzidou@yahoo.gr 1School of Dentistry, National and Kapodistrian University of Athens, Athens, Greece Full list of author information is available at the end of the article Abstract Background: Correction of dentofacial deformities via orthodontics is an integral part of oral health as promotes optimal function, periodontal health, aesthetics and overall oral health-related quality of life. The aim of this study was to examine whether paediatricians refer their patients to orthodontists, whether they have sufficient knowledge in basic orthodontic principles and whether they examine their patients for orthodontic abnormalities. Results: We conducted a survey study of paediatricians in Greece. Questionnaires were completed by 96 out of 123 paediatricians (response rate 78%). In the assessment of the examination of the oral cavity, a low frequency of examination of the position of the teeth (54%) and jaws (51%) was found. Reasons paediatricians referred patients to specialists varied from mouth breathing-snoring 24% (23/96) to face or teeth asymmetry 87% (84/96). In the multivariable analyses for the effect of gender, work sector or years of experience in the decision for orthodontic referral, we could not identify any significant predictors. Conclusions: The results of this study indicate that there was variability regarding orthodontic knowledge among paediatricians. Although the majority were aware of the importance of examination of the oral cavity, they did not have the appropriate knowledge to perform a full and systematic screening for orthodontic problems. The probability of referral was different for the various orthodontic anomalies. Keywords: Orthodontic knowledge paediatricians, Reference to orthodontists, Paediatricians’ role, Oral health, Orthodontic principals, Reference, Education, Paediatric residency result to a significant reduction in dental caries and peri- odontal disease [5]. © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Paediatricians’ awareness on orthodontic problems and related conditions—a national survey Marianna Koufatzidou1* , Despina Koletsi2, Eirini Iouliani Basdeki1, Nikolaos Pandis3 and Argy Polychronopoulou4 Paediatricians’ awareness on orthodontic problems and related conditions—a national survey atzidou, Marianna ; Koletsi, Despina ; Basdeki, Eirini Iouliani ; Pandis, Nikolaos ; Polychronopoulou, Arg DOI: https://doi.org/10.1186/s40510-019-0285-x DOI: https://doi.org/10.1186/s40510-019-0285-x Posted at the Zurich Open Repository and Archive, University of Zurich ZORA URL: https://doi.org/10.5167/uzh-186051 Journal Article Published Version The following work is licensed under a Creative Commons: Attribution 4.0 Inter Posted at the Zurich Open Repository and Archive, University of Zurich ZORA URL: https://doi.org/10.5167/uzh-186051 Journal Article Published Version ollowing work is licensed under a Creative Commons: Attribution 4.0 International (CC BY 4.0) License. Originally published at: Koufatzidou, Marianna; Koletsi, Despina; Basdeki, Eirini Iouliani; Pandis, Nikolaos; Polychronopoulou, Argy (2019). Paediatricians’ awareness on orthodontic problems and related conditions—a national survey. Progress in Orthodontics, 20(1):33. DOI ht //d i / / ( ) DOI: https://doi.org/10.1186/s40510-019-0285-x Koufatzidou et al. Progress in Orthodontics (2019) 20:33 https://doi.org/10.1186/s40510-019-0285-x (2019) 20:33 Koufatzidou et al. Progress in Orthodontics https://doi.org/10.1186/s40510-019-0285-x Background f Correction of dentofacial deformities via orthodontics is an integral part of oral health as it promotes optimal function, periodontal health, aesthetics, and overall oral health-related quality of life [1, 2]. Paediatricians are responsible for the health status of infants and children and as such, oral health cannot be excluded from the overall health assessment [7]. Further- more, there is evidence that young children are more likely to visit a medical office than a dental one [8, 9] Therefore, it is important that paediatricians conduct initial orthodon- tic screenings in order to diagnose abnormalities early and refer the patients accordingly. Inadequate lip coverage, increased overjet with labial proclination of maxillary incisors, and anterior open bite are examples of dentofacial deformities that may be suc- cessfully managed with orthodontic treatment, resulting in functional improvement and reduction in the risk of maxil- lary incisor and gingival tissue trauma [1–5]. Dentofacial abnormalities have been associated with speech disorders [3] and people that have undergone orthodontic treatment are able to clean their teeth more effectively, which may Paediatricians may help in early diagnosis of orthodontic problems and this may improve the treatment outcome and its stability over the years [10, 11]. For instance, unilateral posterior crossbite has been documented as one of the most frequent malocclusions of the primary teeth of Caucasian children. If left untreated or not timely treated, lateral mandibular displacement may lead to facial asymmetry due to undesirable growth modification [10]. Page 2 of 6 Koufatzidou et al. Progress in Orthodontics (2019) 20:33 Koufatzidou et al. Progress in Orthodontics (2019) 20:33 Page 2 of 6 Koufatzidou et al. Progress in Orthodontics (2019) 20:33 Koufatzidou et al. Progress in Orthodontics (2019) 20:33 It is therefore important to treat crossbites during the early mixed dentition in order to establish a physiologic transverse occlusion in addition to a skeletal equilibrium [11]. Lastly, obese as well as allergic children may develop mouth breathing and obstructive sleep apneas which are associated with types of malocclusion that need an early diagnosis and specific treatment [12]. In addition to the correction of a functional abnormality, orthodontic treat- ment at an early age may provide patients with a good aesthetic outcome which can have positive consequences on their self-esteem [3, 5, 6]. communicated to all the participants and anonymity was ensured. Results A total of 96 out of 123 paediatricians returned the ques- tionnaires completed (response rate 78%). The completion response frequency for demographic variables ranged from 80 to 99 percent, with age bearing the lowest fraction of questionnaire completion. Demographic characteristics are available in Table 1. Female responders (64/95, 67%) predominated male responders, while the mean age was 45.2 years old (SD 13.0). Fifty-five percent reported work- ing in the private sector (51/93), while only 21% (18/85) reported having obtained a subspecialty. Most of the responders reported working duration times up to 50 h per week (53/92, 58%). Statistical analysis/analytical approach y y pp Descriptive statistics were performed for responders’ demographic data. To test the association between paedia- tricians’ related characteristics and overall referral to orthodontists or otherwise, Pearson chi-square test and Fisher’s exact test were undertaken as appropriate. Cross- tabulations and frequency distributions were presented for the examination of the oral cavity and orthodontic-related problems, or reasons for referral to a specialist. Univariable and multivariable logistic regression analyses were performed to assess the effect of gender, work sector or years of experience in the decision for orthodontic referral. The variables gender, work sector, and years of experience served as the independent variables, while the decision for orthodontic referral was the dependent or outcome vari- able. Model fit was checked using the Hosmer–Lemeshow test. The level of significance was pre-specified at p < 0.05. Although paediatricians are expected to be knowledgeable about oral health-related issues in order to fulfil their responsibilities as professionals, the educational curriculum of paediatric specialty rarely includes oral health education and when it does, the devoted time is limited. [7, 15]. The study hypothesis was that paediatricians might not have sufficient knowledge to examine their patients for orthodontic-related conditions and might not refer them to an adequate level to the orthodontists. The aim of this study was to examine whether paediatricians examine their patients as far as orthodontic problems are concerned, whether they have sufficient knowledge in basic orthodontic principles and whether they refer their patients for orthodontic problems. All statistical analyses were conducted with Stata ver- sion 15.1 software (Stata Corporation, College Station, TX, USA). Methods The study was not a priori registered. Informed consent was waivered due to the anonymous and voluntary character of this survey. Survey The questionnaire consisted of four parts. First, demo- graphic characteristics such us age, gender, and work sector, were recorded. On the second part, the participants were queried about their examination ritual, whether they examine the oral cavity, the position of teeth and jaws and on their knowledge in specific orthodontic anomalies such as crowding, overjet, and the prevalence in their patients. On the third part, their referral practices to orthodontists and their reasoning were assessed. Finally, their personal orthodontic experience and the source of their orthodon- tic education, if any, was recorded. Assessment of the examination of the oral cavity is presented in Table 2. Although paediatricians examined the mucosa (95/96, 99%), the tongue (93/96, 97%) and even the teeth of their patients (83/96, 86%), the examin- ation of the position of the teeth (52/96, 54%) and jaws (49/96, 51%) was rarely performed. Regarding paediatricians’ awareness of the prevalence of common orthodontic anomalies in their patients, their responses varied from 31 to 95%. Thirty out of 96 (31%) examined their patients for crossbite, 34 out of 96 (35%) for overbite, while 84 out of 96 (87%) for prognathism, and 91 out of 96 (95%) for paranormal functional habits like finger sucking. (Table 3). Sample A questionnaire was handed to paediatricians who partici- pated in the 55th Panhellenic Congress for Pediatrics, which was held in Kos from the 2nd until the 4th of June 2017. Paediatricians from all over Greece were asked to answer the survey that was given to them. Background f The administration of the survey and the data collection procedures were conducted by two—under- graduate students of the School of Dentistry, National and Kapodistrian University of Athens—co-authors of this manuscript. Survey administration % N % Gender 0.43* Male 23 74 8 26 31 100 Female 52 81 12 19 64 100 Total 75 79 20 21 95 100 Age 0.01# 26–35 22 100 0 0 22 100 36–45 14 74 5 26 19 100 46–55 10 63 6 37 16 100 over 55 15 79 4 21 19 100 Total 61 80 15 20 76 100 Subspecialty 0.35* No 55 82 12 18 67 100 Yes 13 72 5 28 18 100 Total 68 80 17 20 85 100 Work sector 0.02# Public 33 92 3 8 36 100 Private 34 67 17 33 51 100 Both 5 83 1 17 6 100 Total 72 77 21 23 93 100 Years at work 0.05# 1 to 5 21 95 1 5 22 100 6 to 15 18 82 4 18 22 100 16 to 30 19 66 10 34 29 100 Over 30 8 73 3 27 11 100 Total 66 79 18 21 84 100 Hours per week 0.55# 1 to 25 7 70 3 30 10 100 26 to 50 32 74 11 26 43 100 51 to 75 20 77 6 23 26 100 over 75 12 92 1 8 13 100 Total 71 77 21 23 92 100 Patients per day 0.68# 1 to 10 24 73 9 27 33 100 11 to 20 28 82 6 18 34 100 21 to 30 10 77 3 23 13 100 over 30 9 90 1 10 10 100 Total 71 79 19 21 90 100 Pearson chi-squared test #Fisher’s exact test Table 2 Responses of participants in relation to examination of the oral cavity Examination of oral cavity elements No Yes Total N (%) N (%) N (%) Oral cavity 0 (0) 96 (100) 96 (100) Mucosa 1 (1) 95 (99) 96 (100) Tongue 3 (3) 93 (97) 96 (100) Teeth 13 (14) 83 (86) 96 (100) Teeth position 44 (46) 52 (54) 96 (100) Jaw position 47 (49) 49 (51) 96 (100) Table 2 Responses of participants in relation to examination of the oral cavity Their responses for the reasons for referral to special- ists differed for each condition from 24% (23/96) for mouth breathing-snoring to 87% (84/96) for face or teeth asymmetry. Pearson chi-squared test Survey administration The survey was handed to study participants in “paper and pencil” format. The purpose of the project was Koufatzidou et al. Progress in Orthodontics (2019) 20:33 Koufatzidou et al. Progress in Orthodontics Page 3 of 6 Page 3 of 6 Table 1 Demographic characteristics of paediatricians by referral pattern to orthodontist Referral to orthodontist No Yes Total p value N % No. #Fisher’s exact test Survey administration Progress in Orthodontics Page 4 of 6 Page 4 of 6 Table 4 Responses of paediatricians in relation to the reason for orthodontic referral Referral related No Yes Total N (%) N (%) N (%) Early tooth loss 47 (49) 49 (51) 96 (100) Delayed eruption 61 (64) 35 (36) 96 (100) Difficulty in biting 49 (51) 47 (49) 96 (100) Sounds from tmj 55 (57) 41 (43) 96 (100) Face/teeth asymmetry 12 (13) 84 (87) 96 (100) Jaw deviation (mouth closing) 26 (27) 70 (73) 96 (100) Mouth breathing/snoring 73 (76) 23 (24) 96 (100) Crowding 42 (44) 54 (56) 96 (100) Crossbite 62 (65) 34 (35) 96 (100) Overbite 59 (61) 37 (39) 96 (100) Missing teeth 51 (53) 45 (47) 96 (100) Grinding at sleep 60 (63) 36 (37) 96 (100) Spaces 49 (51) 47 (49) 96 (100) Prognathism 20 (21) 76 (79) 96 (100) Retrognathism 40 (42) 56 (58) 96 (100) Delayed teeth change 71 (74) 25 (26) 96 (100) tmj, temporomandibular joint Table 4 Responses of paediatricians in relation to the reason for orthodontic referral patterns from the paediatrician to the orthodontist. While conditions such as face asymmetry and prognathism were readily recognized and resulted in high referral frequen- cies, other anomalies were not common reasons for refer- rals. Orthodontic problems less likely to result in referrals included mouth breathing—snoring, delayed eruption, crossbite, overbite, and nocturnal grinding. The lack of orthodontic prevention and screening, at an early age, is manifested throughout the bibliography. In the Albanian population, 85% have been reported to present oral habits like pacifier sucking, while a severe and very severe need for orthodontic treatment was found in up to 17% [13, 14]. Studies in other counties, namely, in Austria and Croatia, manifest a great need for orthodontic treatment among children aged 8–10 and adolescents aged 12–18, respectively. It is therefore obvious that orthodontic prevention, at an early age if possible, should be reinforced [15, 16]. We could not identify any other survey regarding orthodontic screening and referral from paediatricians in the literature. Therefore, we could only compare our results with studies assessing paediatrician’s knowledge and referrals for oral hygiene and dental caries [17–19]. Survey administration More specifically, 36% (35/96) tended to refer for delayed eruption, 73% (70/96) for jaw deviation, 56% (54/96) for crowding, 35% (34/96) for crossbite, 39% (37/96) for overbite, 49% (47/96) for spaces, 79% (76/96) for prognathism, 58% (56/96) for retrognathism, and 26% (25/96) for delayed teeth change (Table 4). We examined the paediatricians’ referral patterns to orthodontists in relation to the demographic character- istics such as gender, age, work sector, subspecialty, and years of work. In the univariable logistic regression, there was a statistically significant result for the effect of the work sector. Notwithstanding, in the multivari- able regression model for the effect of gender, work sector or years of experience in the decision for ortho- dontic referral, we could not identify any significant predictors overall. More specifically, we could only detect a 12.75 fold increase in the odds for referral for those with 21 to 30 years of working experience com- pared to those with 0–10 years in practice, after adjusting for gender and work sector (OR = 12.75, 95% CIs 1.16, Table 3 Responses of paediatricians in relation to examination of the orthodontic problems Examination of orthodontic-related elements No Yes Total N (%) N (%) N (%) Crowding 42 (44) 54 (56) 96 (100) Crossbite 66 (69) 30 (31) 96 (100) Overbite 62 (65) 34 (35) 96 (100) Missing teeth 39 (41) 57 (59) 96 (100) Spaces 41 (43) 55 (57) 96 (100) Prognathism 12 (13) 84 (87) 96 (100) Retrognathism 39 (41) 57 (59) 96 (100) Habits (i.e., finger-sucking) 5 (5) 91 (95) 96 (100) Table 3 Responses of paediatricians in relation to examination of the orthodontic problems Koufatzidou et al. Progress in Orthodontics (2019) 20:33 Koufatzidou et al. Survey administration In a national survey with 1618 post-residency members of the American Academy of Pediatrics, 90% of the responders claimed that they should examine the oral cavity and teeth, while only 54% claimed to examine the teeth of half of their 0–3-year-old patients. Lack of train- ing was the most common reason for not performing an oral examination [17]. In another study including gen- eral dentists, paediatric dentists, and paediatricians, only 5% of the paediatricians recommended a dental visit by the age of 1 year old [18]. Therefore, it may be presumed that paediatricians have limited to basic dental education 140.26; p value = 0.04). However, there was great uncer- tainty in the estimate (Table 5). 140.26; p value = 0.04). However, there was great uncer- tainty in the estimate (Table 5). Wald test for the overall effect References 1. Abanto J, Carvalho TS, Mendes FM, Wanderley MT, Bo necker M, Raggio DP. Impact of oral diseases and disorders on oral health-related quality of life of preschool children. Community Dent Oral Epidemiol. 2011;39:105–14. 1. Abanto J, Carvalho TS, Mendes FM, Wanderley MT, Bo necker M, Raggio DP. Impact of oral diseases and disorders on oral health-related quality of life of preschool children. Community Dent Oral Epidemiol. 2011;39:105–14. Lastly, as there was no related previous study on the topic, we could not have used a pre- existing survey/ questionnaire as a validated guide. Thus, the question- naire used was custom- made and formal data about its validity are lacking. 2. Wagner Y, Heinrich-Weltzien R. Risk factors for dental problems: Recommendations for oral health in infancy. Early Hum Dev. 2017; 114:16–21. 3. Hassan AH, Amin HE-S. Association of orthodontic treatment needs and oral-health quality of life in young adults. Am J Orthod Dentofacial Orthop. 2010;137(1):42–7. 4. Batista KB, Thiruvenkatachari B, Harrison JE, O’Brien KD. Orthodontic treatment for prominent upper front teeth (Class II malocclusion) in children and adolescents. Cochrane Database Syst Rev. 2018;13(3): CD003452. Conclusions The results of this study indicate that there is a great variability regarding orthodontic problems and examin- ation practices among paediatricians. Although most of the practitioners are aware of the need for examining the oral cavity, they do not seem to undertake a systematic orthodontic. 5. Hunt O, Hepper P, Johnston C, Stevenson M, Burden D. Professional perception of the benefits of orthodontic treatment. Eur J Orthod. 2001;23:315–23. 6. Feu D, de Oliviera BH, de Oliviera Almeida ΜA, Kiyak HA, Miguel JAM. Oral health-related quality of life and orthodontic treatment seeking. Am J Orthod Dentofacial Orthop. 2010;138(2):152–9. There is a need for the two specialties to work together for the benefit of the patient. A possible solution, in order to establish effective cooperation, is through the inclusion of dental courses regarding orthodontics in paediatric resi- dency curriculum and through inter-professional seminars and interaction. 7. Krol DM. Children’s oral health and the role of the pediatrician. Curr Opin Pediatr. 2010;22(6):804–8. https://doi.org/10.1097/MOP.0b013e3283402e3b. 8. Quinonez RB, Kranz AM, Lewis CW, Barone L, Boulter S, O’Connor KG, Keels MA. Oral health opinions and practices of pediatricians: Updated results from a National Survey. Acad Pediatr. 2014. https://doi.org/10.1 016/j.acap.2014.07.001. 9. Nirschl RF, Kronmiller JE. Evaluating oral health needs in preschool children. Clin Pediatr. 1986;25(7):358–62. 10. bOvsenik M, Primozic J. How to push the limits in the transverse dimension? Facial asymmetry, palatal volume and tongue posture in children with unilateral posterior cross bite: a three-dimensional evaluation of early treatment. Orthod Fr. 2014;85(2):139–49. Consent for publication Not applicable Consent for publication Not applicable Consent for publication Not applicable Competing interests The authors declare that they have no competing interests Discussion The variability of the orthodontic examination practices and possibly the ability to recognize the prevalence of orthodontic problems is reflected in the patient referral Table 5 Univariable and multivariable logistic regression for the effect of sex, work sector, and years of experience (as a proxy measure of age and years at work) on orthodontic referrals (n = 90) Category Univariable Multivariable OR 95% CI p value OR 95% CI p value Gender Male Reference Female 0.66 0.24, 1.84 0.43 1.01 0.29, 3.49 0.98 Work sector 0.04 0.60* Public Reference Private 5.50 1.47, 20.54 0.01 1.96 0.40, 9.51 0.41 Both 2.20 0.19, 25.52 0.53 0.87 0.06, 12.00 0.92 Years of experience 0.10* 0.16* 0–10 Reference 11–20 9.33 1.04, 84.09 0.05 6.70 0.57, 78.50 0.13 21–30 16.00 1.82, 140.92 0.01 12.75 1.16, 140.26 0.04 Over 30 8.00 0.82, 78.47 0.07 4.81 0.34, 67.49 0.24 *Wald test for the overall effect ltivariable logistic regression for the effect of sex, work sector, and years of experience (as a proxy work) on orthodontic referrals (n = 90) Page 5 of 6 Page 5 of 6 Koufatzidou et al. Progress in Orthodontics (2019) 20:33 Koufatzidou et al. Progress in Orthodontics Page 5 of 6 which leads to low confidence for oral cavity screening and recommendations or consultation [17, 18]. Authors’ contributions MK and AP conceived the idea for the study. MK and EIB carried out the data collection and prepared the manuscript. DK performed the statistical analyses. DK, NP, and AP provided critical revisions and reviewed the manuscript. MK took the lead role as the corresponding author. All authors contributed to the study design. All authors read and approved the final manuscript. Ideally, orthodontic screenings should be performed to all children in both dental and paediatric practices as each specialty can provide care and advice for their pa- tients’ orthodontic health. Paediatricians built a relation- ship with both patients and parents from an early age. As they usually examine their patients before orthodon- tists do, they have the chance to advise, guide, and refer as deemed necessary. Abbreviations CIs: Confidence intervals; OR: Odds ratio; SD: Standard deviation Availability of data and materials h d d d l d d The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. y One limitation of this study is the relatively small sample size which included 96 participants. There was no formal assessment on how responders and non-responders might have differed. Some of the non-responders might have had different answers than those of the responders and this might have had an implication for the generalizability of the study findings. The origin of the participants of this study was not assessed. Therefore, we could not identify inter-area differences. Responder bias is a common prob- lem in studies involving questionnaires. Some of the participants may have answered more favorably regarding their examination rituals in order to appear more compre- hensive in their examination than what they actually do in practice. This is a study reflecting on the attitude of the participants on a specific time and may differ in general. Recall bias may be present in this study, since question- naire-studies retrieve their results from participants who have to recall their experiences and knowledge in order to answer. Ethics approval and consent to participate Not applicable Ethics approval and consent to participate Not applicable Author details 1 1School of Dentistry, National and Kapodistrian University of Athens, Athens, Greece. 2Clinic of Orthodontics and Paediatric Dentistry, Center of Dental Medicine, University of Zurich, Zurich, Switzerland. 3Department of Orthodontics and Dentofacial Orthopedics, Dental School/Medical Faculty, University of Bern, Bern, Switzerland. 4Department of Preventive and Community Dentistry, School of Dentistry, National and Kapodistrian University of Athens, Athens, Greece. Received: 25 April 2019 Accepted: 22 July 2019 Received: 25 April 2019 Accepted: 22 July 2019 9. Nirschl RF, Kronmiller JE. Evaluating oral health needs in preschool children. Clin Pediatr. 1986;25(7):358–62. Abbreviations CI C fid Abbreviations CIs: Confidence intervals; OR: Odds ratio; SD: Standard deviation 11. Peiro AC. Interceptive orthodontics: The need for early diagnosis and treatment of posterior crossbites. Med Oral Patol Oral Cir Bucal. 2006; 11(2):E210–4. Acknowledgements Not applicable Page 6 of 6 Koufatzidou et al. Progress in Orthodontics (2019) 20:33 Koufatzidou et al. Progress in Orthodontics (2019) 20:33 12. Favero L, Arreghini A, Cocilovo F, Favero V. Respiratory disorders in paediatric age: orthodontic diagnosis and treatment in dysmetabolic obese children and allergic slim children. Eur J Paediatr Dent. 2013;14(3):190–4. 13. Lagana G, Fabi F, Abazi Y, Brshiri Nastasi E, Vinjolli F, Cozza P. Oral habits in a population of Albanian growing subjects. Eur J Pediatr Dent. 2013;14(4):309–13. 14. Lagana G, Abazi Y, Beshiri Nastasi E, Vinjolli F, Divizia M, Cozza P. Oral health conditions in an Albanian adolescent population: an epidemiological study. BMC Oral Health. 2015;14(15):67. 15. Steinmassl O, Steinmassl PA, Schwarz A, Crismani A. Orthodontic treatment need of Austrian schoolchildren in the mixed dentition stage. Swiss Dent J. 2017;127(2):122–8. 16. Spalj S, Slaj M, Athanasiou AE, Govorko DK, Slaj M. The unmet orthodontic treatment need of adolescents and influencing factors for not seeding orthodontic therapy. Coll Antropol. 2014;38(Suppl 2):173–80. 16. Spalj S, Slaj M, Athanasiou AE, Govorko DK, Slaj M. The unmet orthodontic treatment need of adolescents and influencing factors for not seeding orthodontic therapy. Coll Antropol. 2014;38(Suppl 2):173–80. 17. Lewis CW, Boulter S, Keels MA, Kril DM, Mouradian WE, O’Connor KG, Quinonez RB. Oral health and Pediatricians: results of a national survey. Acad Pediatr. 2009. https://doi.org/10.1016/j.acap.2009.09.016. 17. Lewis CW, Boulter S, Keels MA, Kril DM, Mouradian WE, O’Connor KG, Quinonez RB. Oral health and Pediatricians: results of a national survey. Acad Pediatr. 2009. https://doi.org/10.1016/j.acap.2009.09.016. 18. Brickhouse TH, Unke JH, Kacitis I, Best AM, Davis RD. Infant oral health care: a survey of general dentists, pediatric dentists and pediatricians in Virginia. Pediatr Dent. 2008;30(2):147–53. 18. Brickhouse TH, Unke JH, Kacitis I, Best AM, Davis RD. Infant oral health care: a survey of general dentists, pediatric dentists and pediatricians in Virginia. Pediatr Dent. 2008;30(2):147–53. 19. Krol DM. Educating pediatricians on children’s oral health: past, present, and future. Pediatrics. 2004;113(5):e487–92. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
https://openalex.org/W2605683421	https://hal.archives-ouvertes.fr/hal-01604591/file/MunschGauchi2017_1.pdf	English	null	N2 Gas Flushing Limits the Rise of Antibiotic-Resistant Bacteria in Bovine Raw Milk during Cold Storage	Frontiers in microbiology	2,017	cc-by	11,669	ORIGINAL RESEARCH published: 19 April 2017 doi: 10.3389/fmicb.2017.00655 N2 Gas Flushing Limits the Rise of Antibiotic-Resistant Bacteria in Bovine Raw Milk during Cold Storage Patricia Munsch-Alatossava 1, Susanna Jääskeläinen 1, Tapani Alatossava 1 and Jean-Pierrre Gauchi 2 1 Department of Food and Environmental Sciences, University of Helsinki, University of Helsinki, Finland, 2 MaIAGE, INRA, Université Paris-Saclay, Jouy en Josas, France Antibiotic resistance has been noted to be a major and increasing human health issue. Cold storage of raw milk promotes the thriving of psychrotrophic/psychrotolerant bacteria, which are well known for their ability to produce enzymes that are frequently heat stable. However, these bacteria also carry antibiotic resistance (AR) features. In places, where no cold chain facilities are available and despite existing recommendations numerous adulterants, including antibiotics, are added to raw milk. Previously, N2 gas ﬂushing showed real potential for hindering bacterial growth in raw milk at a storage temperature ranging from 6 to 25◦C. Here, the ability of N2 gas (N) to tackle antibiotic- resistant bacteria was tested and compared to that of the activated lactoperoxidase system (HT) for three raw milk samples that were stored at 6◦C for 7 days. To that end, the mesophiles and psychrotrophs that were resistant to gentamycin (G), ceftazidime (Ce), levoﬂoxacin (L), and trimethoprim-sulfamethoxazole (TS) were enumerated. For the log10 ratio (which is deﬁned as the bacterial counts from a certain condition divided by the counts on the corresponding control), classical Analyses of Variance (ANOVA) was performed, followed by a mean comparison with the Ryan-Einot-Gabriel-Welsch multiple range test (REGWQ). If the storage “time” factor was the major determinant of the recorded effects, cold storage alone or in combination with HT or with N promoted a sample-dependent response in consideration of the AR levels. The efﬁciency of N in limiting the increase in AR was highest for fresh raw milk and was judged to be equivalent to that of HT for one sample and superior to that of HT for the two other samples; moreover, compared to HT, N seemed to favor a more diverse community at 6◦C that was less heavily loaded with antibiotic multi-resistance features. Our results imply that N2 gas ﬂushing could strengthen cold storage of raw milk by tackling the bacterial spoilage potential while simultaneously hindering the increase of bacteria carrying antibiotic resistance/multi-resistance features. Keywords: antibiotic resistance (AR), raw milk, cold storage, N2 gas, lactoperoxidase system, Ryan-Einot-Gabriel- Welsch test Edited by: Javier Carballo, University of Vigo, Spain Reviewed by: Cristiana Garofalo, Università Politecnica delle Marche, Italy David Rodriguez-Lazaro, University of Burgos, Spain Correspondence: Patricia Munsch-Alatossava patricia_munsch@yahoo.fr Specialty section: This article was submitted to Food Microbiology, a section of the journal Frontiers in Microbiology Received: 11 January 2017 Accepted: 30 March 2017 Published: 19 April 2017 Specialty section: This article was submitted to Food Microbiology, a section of the journal Frontiers in Microbiology Received: 11 January 2017 Accepted: 30 March 2017 Published: 19 April 2017 Citation: Munsch-Alatossava P, Jääskeläinen S, Alatossava T and Gauchi J-P (2017) N2 Gas Flushing Limits the Rise of ntibiotic-Resistant Bacteria in Bovine Raw Milk during Cold Storage. Front. Microbiol. 8:655. doi: 10.3389/fmicb.2017.00655 Specialty section: This article was submitted to Food Microbiology, a section of the journal Frontiers in Microbiology Received: 11 January 2017 Accepted: 30 March 2017 Published: 19 April 2017 Citation: unsch-Alatossava P, Jääskeläinen S, Specialty section: This article was submitted to Food Microbiology, a section of the journal Frontiers in Microbiology Received: 11 January 2017 Accepted: 30 March 2017 Published: 19 April 2017 INTRODUCTION 6◦C) or the activation of the lactoperoxidase system (HT) when economical or technical constraints prevent the use of cooling facilities (FAO, 1991, 2005). When raw milk is cold- stored, psychrotrophic/psychrotolerant bacteria take over within a few days and promote the spoilage of the raw milk due to the production of varied and mostly heat-stable enzymes (Chambers, 2002). On the other side, in the absence of a cold chain infrastructure, many adulterating substances, including antibiotics, are used to prevent excessive bacterial growth in raw milk (Bari et al., 2015; Botelho et al., 2015). The statement is unanimous among the highest international authorities: the increasing prevalence of antibiotic-resistant bacteria constitutes one of the most serious threats to human health; both the WHO and FAO have issued action plans to tackle antibiotic resistance (AR), which is perceived as a global and increasing threat (WHO, 2015; FAO, 2016). In Europe, it is estimated that AR already induces extra health costs and productivity losses amounting to at least 1,500 million euros, and it leads to 25,000 deaths annually (WHO, 2015). Although, the European Union already placed a ban on using antibiotics as growth promoters in 2006, some countries still sell more antibiotics for animal use than for treating humans. Due to the simultaneous presence of spoilage microorganisms in food materials, of food-borne pathogens and antibiotic-resistant bacteria, food production systems face great challenges. The warning that “the resistance problem in human medicine will not be solved if there is a constant inﬂux of resistance genes into the human microﬂora via the food chain” was already made earlier (Teuber, 1999). Food has been identiﬁed as one main direct vehicle for the transmission of antibiotic-resistant bacteria from animals to humans and AR genes that are carried by zoonotic bacteria (Perreten et al., 1997; SØrum and L’Abbé Lund, 2002; Aquilanti et al., 2007; Garofalo et al., 2007; Wang et al., 2012; Rolain, 2013; ECDC/EFSA/EMA, 2015). Wichmann et al. (2014) also demonstrated that the cow microbiome was a signiﬁcant reservoir of AR genes; however, the scientiﬁc community still remains divided regarding the contribution of antibiotic use on food production, despite the fact that more evidence points to the anthropogenic causes of increasing AR in both human or environmental microbiomes (Forslund et al., 2014). INTRODUCTION During earlier analyses of bacterial isolates that were involved in raw milk spoilage, resistance to several classes of antibiotics (or multi-resistance) seemed to increase during the cold storage of raw milk (Munsch-Alatossava and Alatossava, 2006, 2007). Further investigations revealed that psychrotrophic bacterial populations exhibited higher AR levels compared to their corresponding mesophilic populations; most importantly, based on their AR proﬁles, distinct bacterial populations succeeded one another during cold storage; strangely enough, the AR was highest at the time when the “total bacterial counts” had reached approximately 105 cfu/ml; after that, the relative AR level was lower but the milk was spoiled (Munsch-Alatossava et al., 2012a). owe but t e was spo ed ( u sc atossava et a ., 0 a). The limitations of raw milk cold storage, together with the observation that cold storage seems to promote an increase in AR (Munsch-Alatossava and Alatossava, 2007; Munsch-Alatossava et al., 2012a,b), motivated research eﬀorts to attempt to control bacterial growth in raw milk more eﬀectively. Two studies examined the use of N2 gas to prevent bacterial growth in raw milk that was kept in a “closed system” (Murray et al., 1983; Dechemi et al., 2005). By considering an “open system,” both culture-dependent investigations and DNA barcoding studies revealed that no pathogen, no spoilage bacteria or any anaerobe was clearly advantaged by the N2 gas ﬂushing treatment when it was applied to raw milk at the laboratory scale, despite the fact that 104- fold lower bacterial counts diﬀerentiated the N2-ﬂushed from non-ﬂushed cold-stored raw milk samples (Munsch-Alatossava et al., 2010a,b; Gschwendtner et al., 2016). Recently, the eﬃciency of N2 gas ﬂushing and the activated lactoperoxidase system in controlling bacterial growth in raw milk samples stored at 15 and 25◦C was compared; overall, the gas treatment showed a time-limited eﬀect that was comparable to the eﬀect of the activated lactoperoxidase system (Munsch- Alatossava et al., 2016). Because the N2 gas ﬂushing inhibited the mesophilic and psychrotrophic populations in cold-stored raw milk, we undertook an investigation as to whether the N2 gas-based treatment could also inhibit the growth of antibiotic- resistant bacteria in raw milk stored at 6◦C, and its eﬃciency was compared to that of the activated lactoperoxidase system for both mesophiles and psychrotrophs. Abbreviations: S1, S2, S3, raw milk samples; M, mesophiles; P, psychrotrophs; C, control; HT, activated lactoperoxidase system; N, N2 gas ﬂushing; AR, antibiotic resistance; AB, antibiotic; G, gentamycin; Ce, ceftazidime; L, levoﬂoxacin; TS, trimethoprim-sulfamethoxazole. Citation: Munsch-Alatossava P, Jääskeläinen S, Alatossava T and Gauchi J-P (2017) N2 Gas Flushing Limits the Rise of Antibiotic-Resistant Bacteria in Bovine Raw Milk during Cold Storage. Front. Microbiol. 8:655. doi: 10.3389/fmicb.2017.00655 April 2017 \| Volume 8 \| Article 655 Frontiers in Microbiology \| www.frontiersin.org Hindering AR Dissemination in Milk Munsch-Alatossava et al. April 2017 \| Volume 8 \| Article 655 INTRODUCTION Consequently, bacteria that were resistant to gentamycin (G), ceftazidime (Ce), levoﬂoxacine (L), and trimethoprim-sulfamethoxazole (TS) were enumerated from three cold-stored raw milk samples while they were subjected to the following three conditions: the application of N2 gas ﬂushing, the activation of the lactoperoxidase system (HT) and no additional treatment as the control (C). For the statistical treatment of the data, we deﬁned the ratio as the amounts of colonies that were recovered in the presence of one antibiotic type divided by the number of colonies enumerated from the Although milk is considered to be sterile when it is secreted from a healthy udder, various contamination sources increase its bacterial level. Depending on the production area, the farm type and the milk handling practices, the types and levels of bacteria vary greatly in raw milk (Chambers, 2002; Frank and Hassan, 2002). In developed countries, recommendations state that the bacterial load in raw milk should not exceed 105 cfu/ml and 3 × 105 cfu/ml “total bacterial counts” in the farm and dairy tanks, respectively. These counts are determined under mesophilic incubation conditions (3 d at 30◦C) (Anonymous, 2004); there are no recommendations concerning AR in raw milk. With regards to antibiotics (ABs), the dairy industry states that raw milk should not contain AB residues (which could compromise the growth of the starter strains that form the bases of yogurt or cheese production). If ABs are present in milk, then the farmer is not allowed to sell this milk, which is then used to feed cattle, with the risk that the bacteria that have become resistant may enter the food chain (Pereira et al., 2014; Andremont, 2015). Two options are recommended for the preservation of raw milk: either cold storage (at approximately 4◦C/below April 2017 \| Volume 8 \| Article 655 April 2017 \| Volume 8 \| Article 655 Frontiers in Microbiology \| www.frontiersin.org 2 Hindering AR Dissemination in Milk Munsch-Alatossava et al. corresponding control. During the whole study, the decimal logarithm of the ratio, i.e., the log10 (ratio), was analyzed with the Ryan-Einot-Gabriel-Welsch (REGWQ) multiple range test (Hsu, 1996). agar plates were 16 mg/L for G, 32 mg/L for Ce, 8 mg/L for L and 8 mg/L trimethoprim together with 152 mg/L sulfamethoxazole for TS. These concentrations correspond to 4-fold of the MIC for G, Ce and L, or 2-fold the MIC for TS, as indicated by EUCAST for pseudomonads (EUCAST, 2010). INTRODUCTION All the agar plates were stored overnight at 4◦C and protected from light before use. Following the analyses performed as described before (Munsch-Alatossava et al., 2012a,b), the plates were incubated under aerobic conditions for 2–3 days at 30◦C, or for 10 days at 7◦C to enumerate the “total” bacterial mesophilic and psychrotrophic counts, respectively. Microbiological Analyses g Antibiotic resistance was quantiﬁed by enumerating bacterial colonies on agar plates that contained one of each of the considered antibiotics (ABs) (each one is representative of one AB class) and by comparing their levels to the corresponding “no AB” (no antibiotics) control. The analyses were performed at day 0 (shortly after the samples were received) and after 3 and 7 days of cold storage. All the bacterial counts were determined mostly from triplicate if not duplicate platings on Mueller-Hinton agar (Lab M, Ltd, Lancashire, UK). The antimicrobial agents [gentamycin (aminoglycosides), ceftazidime (β-lactams, cephems), levoﬂoxacin (quinolones) and trimethoprim-sulfamethoxazole (at a ratio of 1/19, a folate pathway inhibitor) (Sigma-Aldrich, Steinheim, Germany)] which are abbreviated G, Ce, L, and TS, respectively, were added to the agar in accordance with the EUCAST guidelines (EUCAST, 2000). The AB solutions were freshly prepared by dissolving the powders into the following solvents: MilliQ water for G, 0.1M phosphate buﬀer (pH 7) for Ce, 0.1M NaOH for L, 0.1M lactic acid for T, and 95% ethanol for S (EUCAST, 2000). With the exception of S, all the AB solutions were ﬁlter-sterilized prior to their addition to adequately cooled agar. The ﬁnal AB concentrations in the Statistical Analyses y Statistical analyses were performed independently for every raw milk sample. To compare the eﬃciency of the treatments, the analyses employed the ratio that was deﬁned as “the counts obtained in the presence of one of the four antibiotic (AB) types divided by the counts enumerated on the corresponding control.” To overcome the variability of microbiological data (excessively low or high bacterial counts), the decimal logarithm of the ratio [log10 (ratio)] was analyzed. When the ratio denominator of the bacterial counts was “zero,” “zero” was replaced with “one.” To overcome the situations in which the ratios were equal to zero and to allow for log transformation, 0.0001 was added to all the values. The statistical analyses relied on a classical Analysis of Variance (ANOVA) followed by a multiple comparison of the means with the Ryan-Einot-Gabriel-Welsch test (REGWQ) (Hsu, 1996) with an alpha risk of 0.05, as applied previously (Munsch- Alatossava et al., 2016). All the calculations were performed with SAS/STAT software version 9.4/ GLM procedure (SAS Institute, NC, USA). Low vs. high values of the log10(ratio) indicate that the bacterial populations were strongly or poorly controlled by the applied treatments. The statistical analyses determined the inﬂuence of the following three factors: (1) the “time” or duration of cold storage (which addressed the trends in the counts between days 0 and 3, and between days 0 and 7) visualized by two levels “day 3” and “day 7”; (2) the ”treatment type” of either non- treated raw milk, which accounts for the control (C), or where the lactoperoxidase system was activated (HT), or for N2-ﬂushed milk (N); and (3) the “AB type,” which considered the response to the following ﬁve conditions: the absence of any antibiotic (no AB), or the presence of one of the antibiotics (whether G, Ce, L, or TS). Moreover, two-factor and three-factor interactions were evaluated for the log10 (ratio). Materials and Treatment of Raw Milk Samples Three bovine raw milk samples (S1, S2, and S3), which represent commingled lorry milk delivered to Helsinki Dairy Ltd. in Helsinki (Finland) in November and December 2015, were subjected. After the raw milk arrived, 100 ml of each sample were added to 250 ml sterile bottles and placed on a multi- place magnetic stirrer (Variomag, Oberschleißheim, Germany). The bottles were partially immersed in a refrigerated water bath (MGW Lauda MS/2), which allowed, with the help of an immersion thermostat, a constant temperature to be maintained (Munsch-Alatossava et al., 2010a). The raw milk samples were mixed continuously at 220 rpm and kept at 6 ± 0.1◦C for 7 days. The activation of the lactoperoxidase system (HT) consisted in the addition of both hydrogen peroxide and thiocyanate at 10 ppm each (FAO, 2005); the source of H2O2 was a 30% H2O2 solution (Perdrogen R 30 Gew%, Riedel de Häen, Seelze, Germany). The thiocyanate anion SCN−took the form of NaSCN (Sigma-Aldrich, Steinheim, Germany) and a 1% (w/v) stock solution was sterile-ﬁltered and cold- stored until use. The N2 gas (AGA Ltd, Riihimäki, Finland) was 99.999% pure and the ﬂow rate for the continuous N2 gas ﬂushing treatments (N) was adjusted to 120 ml/min (Munsch-Alatossava et al., 2010a). The description of the conditions (which were applied to every sample) is as follows: C, control; HT, H2O2+ SCN−(for the activated lactoperoxidase system); and N, N2 gas ﬂushing. Frontiers in Microbiology \| www.frontiersin.org Growth Trend in Resistant Bacteria during Cold Storage Bacterial Levels in Non-treated Raw Milk (C) ( ) At the beginning of the storage period, all three samples presented initial “total” mesophilic counts (S1M, S2M, and S3M) that ranged between 734 and 3,700 cfu/ml (2.86 and 3.57 log- units) (Figure 1). Despite having rather similar initial counts, the samples revealed diﬀerent bacterial growth kinetics for the “no AB” condition after 3 days of cold storage given that the counts were below 104 cfu/ml for S2M, below the limit (3 × 105 cfu/ml or 5.48 log units) for S3M and close to 106 for S1M; the additional 4 days of cold storage promoted 3 to 5 log-units of additional increases in the counts for all three samples (Figure 1). At the initial stage, all the samples showed detectable levels of mesophilic-resistant bacteria, which were highest with Ce and TS for S1M and highest with Ce for both S2M and S3M. After 3 days of cold storage for all the samples, the TS-resistant mesophiles had increased and even reached an equal level to that of “no AB” for S1M and S2M; the G-, Ce-, and L-resistant bacterial levels largely remained constant out of a 1 log-unit drop for L-resistant bacteria for S3M (Figure 1). After 7 days of cold storage, the G-, Ce-, and L-resistant counts had increased moderately for S1M compared to that of “no AB”; between 3 and 7 days, the resistant bacterial counts mostly increased for S2M and S3M. For G and L, the levels were equivalent for both S2M and S3M (slightly below or approximately 4 log-units). The Ce-resistant counts were also at this level for S2M, but they were considerably higher for S3M for which the levels were “equal” to that of the TS- resistant bacteria. For all three samples, at the end of the storage period, the levels of resistant mesophiles were below 3 × 105 cfu/ml (5.48 log units) for the controls (C), with the exception of TS-resistant bacteria, and of Ce-resistant bacteria in addition for S3M (Figure 1). Psychrotrophs that were resistant to G, Ce, and L were initially below the detection levels for all three samples, whereas TS- resistant bacteria were recovered at levels equivalent to the “no AB” counts for S1P and S2P (Figure 1). In contrast to the mesophiles, for which “day 3” was rather equivalent to “day 0,” resistant psychrotrophs increased notably in all three samples (Figure 1). Bacterial Levels in Raw Milk under the Activated Lactoperoxidase System (HT) Lactoperoxidase System (HT) After 3 days under cold storage, the “no AB” condition revealed that the mesophilic counts remained unchanged at approximately 3 log-units for S2M and they were slightly above the initial levels for S1M and S3M. Compared to the corresponding controls (C), the counts for “no AB” were approximately 2 log-units below the levels recorded for S1M and S3M, whereas no drop was recorded for S2M. After 7 days, HT had provoked a large inhibition of approximately 5 log-units of bacterial growth for S1M and S3M compared to approximately 3 log-units lower counts for S2M (Figure 1). For resistant mesophiles, rather minor changes occurred during the ﬁrst 3 days: this ﬁnding is well illustrated for S1M under the activated lactoperoxidase system (HT), when in the presence of G, Ce and L, the mesophilic counts remained unchanged during the ﬁrst 3 days and were at the level of the corresponding control (C); for S1M, day 7 was primarily characterized by a moderate increase in Ce- and TS-resistant bacteria, a greater increase in G- and L-resistant bacteria, and all the counts, including those of “no AB,” were at approximately 4 log-units. If for S1M, the counts for “no AB” and TS were clearly lower for HT compared to the control (C), the G- and L-resistant bacteria were moderately higher compared to the levels of the corresponding control (C) (Figure 1). For S2M at day 3, the counts for “no AB,” G-, Ce-, and L- resistant bacteria were at approximately their initial levels. However, the activated lactoperoxidase (HT) system triggered a notable change at day 3 concerning TS-resistant bacteria, which were reduced to a non-detectable level (Figure 1). Day 7 was characterized by a large increase (of approximately 5 log-units) of bacteria that were resistant to TS, which were only one log- unit lower than the level attained by TS-resistant bacteria for C. Although the counts that were enumerated on the G-, Ce-, and L-plates had increased by approximately 3 log-units, they were still below 3 × 105 cfu/ml, but they exceeded the levels recovered for the corresponding control (C) by approximately 1 log-unit (Figure 1). Bacterial Levels in Raw Milk under the Activated Lactoperoxidase System (HT) For S3M, HT promoted a drop in G- and TS-resistant mesophiles (to an undetectable level in the latter case) at the intermediate storage time; with the exception of the mesophilic counts recovered on Ce containing agar (for which the counts remained fairly equivalent during the course of the experiment), the bacteria that were resistant to G, L and TS greatly increased between 3 and 7 days for S3M, and they were approximately 3.5 log-units, which corresponded to the levels of G- and L-resistant bacteria that were found for the control (C) (Figure 1). At th i iti l t d th ti t d l t id (HT) that the drop ranged between 0.02 and 0.14. As expected, the non- treated raw milk samples (C) showed the largest pH decreases (from 0.3 to 0.5 unit) at the end of the storage period (Table 1). Frontiers in Microbiology \| www.frontiersin.org Impact of the Treatments on the Raw Milk pH p For all three raw milk samples (S1, S2, and S3) in which the lactoperoxidase (HT) was activated, the pH values that were measured after 7 days of cold storage were nearly equal to the initial pH values (Table 1). In addition, the 7-day-N2 gas ﬂushing (N) did not greatly alter the initial pH values of the samples, given April 2017 \| Volume 8 \| Article 655 Frontiers in Microbiology \| www.frontiersin.org 3 Hindering AR Dissemination in Milk Munsch-Alatossava et al. TABLE 1 \| pH values for raw milk samples (S1 to S3), determined at initial and ﬁnal stages (after 7d storage at 6◦C), for the three conditions: non treated raw milk (C), activated lactoperoxidase system (HT), and N2 gas ﬂushed (N) milk. over 4-fold for S2P and over 3-fold for S3P. However, out of the TS-resistant psychrotrophs that culminated in 109, 106, or above 108 cfu/ml for S1P, S2P, and S3P, respectively, all the other psychrotrophic counts remained below 3 × 105 cfu/ml at day 7 (Figure 1). Raw milk samples Initial pH C Final pH HT N S1 6.81 6.35 6.81 6.73 S2 6.74 6.46 6.76 6.60 S3 6.75 6.25 6.78 6.73 Growth Trend in Resistant Bacteria during Cold Storage The augmentation continued until the end of the storage period; by that time, the Ce-resistant counts had increased by approximately 3-fold for G, Ce and L for S1P, At the initial stage under the activated lactoperoxidase (HT), G-, Ce-, and L-resistant psychrotrophs were below the detectable levels in all three samples (Figure 1). For TS-resistant bacteria, the situation was most contrasted given that the counts were below the detectable level for S3P, and at the “no AB” level for S1P and S2P. Among the psychrotrophic populations, the April 2017 \| Volume 8 \| Article 655 4 Munsch-Alatossava et al. Hindering AR Dissemination in Milk FIGURE 1 \| Mesophilic (M) and psychrotrophic (P) bacterial counts (expressed in log cfu/ml) from raw milk samples S1, S2, and S3 that were stored for 7 days at 6◦C (C), cold-stored while the lactoperoxidase system was activated (HT) or cold-stored while ﬂushed with N2(N). The colonies were enumerated on Mueller-Hinton agar plates that contained no antibiotics (“no AB”) or one of the following ABs: G (gentamycin), Ce (ceftazidime), L (levoﬂoxacin) and TS (trimethoprim-sulfamethoxazole) (the error bars correspond to standard deviations). FIGURE 1 \| Mesophilic (M) and psychrotrophic (P) bacterial counts (expressed in log cfu/ml) from raw milk samples S1, S2, and S3 that were stored for 7 days at 6◦C (C), cold-stored while the lactoperoxidase system was activated (HT) or cold-stored while ﬂushed with N2(N). The colonies were enumerated on Mueller-Hinton agar plates that contained no antibiotics (“no AB”) or one of the following ABs: G (gentamycin), Ce (ceftazidime), L (levoﬂoxacin) and TS (trimethoprim-sulfamethoxazole) (the error bars correspond to standard deviations). quite similar patterns in the responses were observed for G-, Ce-, and L-resistant bacteria for all three raw milk samples. The level of G-resistance was lower than that of Ce, whereas L-resistant bacteria were reduced to non-detectable levels in all three samples. The TS-resistant bacteria either increased moderately for S1M and S2M, or were reduced to a non-detectable level during the ﬁrst 3 days of cold storage for S3M (Figure 1). After 7 days, the highest counts were observed in TS for S1M, with Ce and TS in the case of S2M. The maximum inhibitory eﬀect was recorded for S3M, because all of the resistant bacteria were equivalent to or below 103 cfu/ml; the resistant mesophiles were largely less numerous under N compared to HT (Figure 1). Growth Trend in Resistant Bacteria during Cold Storage sharpness of the increase in resistant bacteria was highest for S2P (Figure 1). For S1, contrary to the mesophiles, the psychrotrophs (S1P) showed lower counts at day 3 than the “no AB” condition as compared to day 0. During the 3-day storage period, the psychrotrophs that were resistant to G, Ce and L increased in all the samples with the exception of G for S1P. TS still promoted the most distinctive response. Regarding the increase in resistant counts for S2P, the counts remained at an undetectable level for S3P and showed a considerable drop in the counts between days 0 and 3 for S1P (to an undetectable level) (Figure 1). The additional 4 days of cold storage promoted a considerable increase in the AR for all three samples; irrespective of the AB type, the resistant psychrotrophs reached more or less approximately the same level as the counts from the “no AB” condition. However, HT prevented the excessive growth of TS-resistant psychrotrophs, especially for S1P and S3P (Figure 1). g y Compared to the mesophiles, more changes were observed for the psychrotrophs for all three samples under the N2 ﬂushing treatment for the “no AB” condition; at day 0, no colonies that were resistant to G, Ce or L were detected in S1P, S2P, and S3P. If TS-resistant bacteria were present in S1P and S2P (equivalent to levels of “no AB”), none were detected for S3P (Figure 1). After 3 days of storage, G-, Ce-, or L-resistant colonies remained under the detectable level for S1P; the same was observed in L-resistant bacteria for S2P and for G-resistant bacteria for S3P (Figure 1). By contrast, the TS-resistant colonies increased over the 3 days for all 3 samples, and they reached the levels that were observed for the “no AB” condition. At the end of the storage period, S1P and S2P showed a more homogeneous picture (for 3 ABs out of 4), whereas for S3P, Statistical Analyses The Ryan-Einot-Gabriel-Welsch (REGWQ) multiple range test was applied to the variable log10 (ratio), which enabled us to quantify the changes that occurred during the storage period. The results showed a decrease in the bacterial counts, depicted by “negative” values, which were indicative of an eﬀective control for mesophiles from samples S2 and S3 and for psychrotrophs from sample S1 at day 3 only (Figure 2). For all three samples, the mean log10 (ratio) values were considerably and signiﬁcantly higher at day 7 compared to day 3, which conﬁrmed that the additional four days of cold storage favored the proliferation of mesophilic and psychrotrophic bacterial populations. However, the increase in psychrotrophs (+4.73/S1, +3.54/S2, and +3.22/S3) supplanted the increase in mesophiles (+1.91/S1, +3.33/S2, and +2.75/S3) (Figure 2). FIGURE 2 \| REGWQ results depicting the impact of either 3 or 7 days of storage at 6◦C on mesophilic (M) and psychrotrophic (P) populations that were recovered from the raw milk samples S1, S2, and S3. Means with different letters indicate signiﬁcant differences (alpha risk = 0.05). FIGURE 2 \| REGWQ results depicting the impact of either 3 or 7 days of storage at 6◦C on mesophilic (M) and psychrotrophic (P) populations that were recovered from the raw milk samples S1, S2, and S3. Means with different letters indicate signiﬁcant differences (alpha risk = 0.05). FIGURE 3 \| REGWQ results describing the overall inhibitory efﬁciency of HT (the activated lactoperoxidase system) and N (N2 gas ﬂushing) treatments, compared to the control (C) for the raw milk samples S1, S2, and S3 when stored at 6◦C for 7 days. Means with the same letter are not signiﬁcantly different (alpha risk = 0.05). Bacterial Levels in N2-Flushed Raw Milk (N) Bacterial Levels in N2 Flushed Raw Milk (N) For the N2 gas ﬂushing treatment (N) at day 7, the “no AB” condition revealed a moderate increase in bacterial counts that was equivalent to HT for S3M, slightly above that of HT for S1M, and below HT for S2M, but also below the levels encountered for the controls (C) (Figure 1). The 7- day-cold storage period, together with the N2 gas ﬂushing, promoted fairly minor changes in AR- resistant mesophiles considering S1M and S3M, compared to S2M (Figure 1). After 3 days of cold storage, Frontiers in Microbiology \| www.frontiersin.org April 2017 \| Volume 8 \| Article 655 5 Munsch-Alatossava et al. Hindering AR Dissemination in Milk FIGURE 2 \| REGWQ results depicting the impact of either 3 or 7 days of storage at 6◦C on mesophilic (M) and psychrotrophic (P) populations that were recovered from the raw milk samples S1, S2, and S3. Means with different letters indicate signiﬁcant differences (alpha risk = 0.05). FIGURE 3 \| REGWQ results describing the overall inhibitory efﬁciency of HT (the activated lactoperoxidase system) and N (N2 gas ﬂushing) treatments, compared to the control (C) for the raw milk samples S1, S2, and S3 when stored at 6◦C for 7 days. Means with the same letter are not signiﬁcantly different (alpha risk = 0.05). FIGURE 2 \| REGWQ results depicting the impact of either 3 or 7 days of storage at 6◦C on mesophilic (M) and psychrotrophic (P) populations that were recovered from the raw milk samples S1, S2, and S3. Means with different letters indicate signiﬁcant differences (alpha risk = 0.05). the resistant populations were at approximately the same level with G and L on one side, and with C and TS on the other side. The resistant psychrotrophic counts from N were mostly below the AR levels recorded from the corresponding controls (C) or from the activated lactoperoxidase system (HT) (Figure 1). Efﬁciency of the Activated Lactoperoxidase System (HT) and the N2-Flushing (N) Treatments for Inhibiting Bacterial Growth FIGURE 3 \| REGWQ results describing the overall inhibitory efﬁciency of HT (the activated lactoperoxidase system) and N (N2 gas ﬂushing) treatments, compared to the control (C) for the raw milk samples S1, S2, and S3 when stored at 6◦C for 7 days. Means with the same letter are not signiﬁcantly different (alpha risk = 0.05). For the three samples (S1, S2, and S3), the REGWQ test revealed that the applied treatments (HT or N) including the cold storage alone (the control condition C) promoted rather similar responses regarding the ranking of mesophiles (M) and their corresponding psychrotrophs (P) (Figure 3). The test also highlighted a sample-dependent response based on the ranking of the treatments (HT and N), which appeared to be mostly discriminatory for sample S3 because the treatments together with the control (C) were ascribed to three distinct categories (A, B, and C), unlike the samples S1 and S2, for which only two categories (A and B) were distinguished (Figure 3). Frontiers in Microbiology \| www.frontiersin.org Ranking of the Considered Antibiotics (ABs) g ( ) For all three samples, the results from the REGWQ test showed altogether rather similar mean values for mesophiles for the “no AB” condition, which ranged between 1.44 and 1.60; by contrast, higher and more variable mean values between 2.05 and 3.8 were observed for psychrotrophs (Table 2). For sample S1, the means were mostly and moderately higher for mesophiles compared to psychrotrophs outside the conditions “no AB” and “L,” whereas the S2 and S3 samples showed the highest means for psychrotrophs irrespective of the AB type (Table 2). The means for “no AB” supplanted all the conditions in which ABs were present, with two exceptions (TS-resistant mesophiles for sample S1 and G-resistant psychrotrophs for sample S2). The same order for the AB rankings (G, L, Ce and TS, from the lowest to the highest) was observed under four conditions (Table 2). The mean value was highest for TS-resistant bacteria for mesophiles and psychrotrophs in sample S1, for mesophiles in sample S2, and for psychrotrophs in sample S3 (Table 2). For both mesophiles and psychrotrophs, the ranking of the ABs was most discriminatory for sample S1, because the REGWQ test highlighted three categories (A, B, and C). For sample S2, For sample S1, HT and N were evaluated as very eﬀective and equally inhibitory as outlined by the ranking of the treatments (and grouped into category B) compared to the control (C) (ascribed to category A) (Figure 3). If HT yielded lower mean values compared to C for mesophiles and psychrotrophs, the inhibition was variable and signiﬁcantly diﬀerent from the control (C) only for samples S1 and S3 (Figure 3). Compared to the corresponding controls (C) and to HT, all the mean values were lowest with N; for both the mesophiles and psychrotrophs, the inhibitory eﬀect was either equivalent to HT (for sample S1) or prevailed over HT (for samples S2 and S3). The highest inhibitory eﬀect was recorded for mesophiles from sample S3 which was N2-ﬂushed (Figure 3). For both samples S2 and S3, irrespective of the conditions (control, HT or N), the mean values for the psychrotrophs exceeded the corresponding levels for mesophiles (Figure 3). April 2017 \| Volume 8 \| Article 655 Frontiers in Microbiology \| www.frontiersin.org 6 Hindering AR Dissemination in Milk Munsch-Alatossava et al. Ranking of the Considered Antibiotics (ABs) TABLE 2 \| Ranking of the log10 (ratio) considering the antibiotics (G, Ce, L, TS) compared to the “no AB” condition by the REGWQ test. Samples Mesophiles Psychrotrophs Condition Mean value Signiﬁcance* Condition Mean value Signiﬁcance* S1 TS 1.81 A no AB 2.05 A no AB 1.60 A B TS 1.63 A B Ce 0.73 B C Ce 0.71 A B C L 0.34 C L 0.56 B C G 0.32 C G 0.01 C S2 no AB 1.54 A G 3.07 A TS 1.16 A no AB 2.77 A Ce 0.75 A Ce 2.12 A L 0.70 A L 1.91 A G 0.67 A TS 1.70 A S3 no AB 1.44 A no AB 3.78 A Ce 0.81 A TS 2.03 A B G –0.20 B Ce 1.04 B TS –0.38 B L 0.75 B L –0.54 B G 0.42 B Means with the same letter are not signiﬁcantly different (alpha risk = 0.05) g10 (ratio) considering the antibiotics (G, Ce, L, TS) compared to the “no AB” condition by the REGWQ test. TS) compared to the “no AB” condition by the REGWQ test. Means with the same letter are not signiﬁcantly different (alpha risk = 0.05) were below the levels encountered for HT or C (Figure 4). The greatest changes between the sampling points were observed for S2M, because 13 out of 15 means were negative at day 3 (with the lowest levels from HT-TS or N-L) (Figure 4). Compared to C, HT showed higher mean values for G, Ce, and L at day 7 and lower values for the conditions “no AB” and TS. By contrast, all the mean values were lower for N after 7 days under cold storage compared to the corresponding levels recorded for C but also for HT (Figure 4). The mean values were lower at day 3 in HT for S3M, compared to C for “no AB,” G and TS, and lower at day 7 for “no AB,” Ce, L, and TS. With N, the mean values at day 3 were, relative to the control (C), lower for “no AB,” Ce, and TS but higher for G and equivalent for L. However, at day 7, all the means were lowest for N compared to C, and also to HT when considering resistant bacteria (Figure 4). Ranking of the Considered Antibiotics (ABs) if the mean values for the diﬀerent ABs ranged between 1.70 and 3.07 for the psychrotrophs (with G-resistant psychrotrophs far above all other conditions) and between 0.67 and 1.16 for mesophiles, the statistical treatment revealed that these values were not signiﬁcantly diﬀerent from those of “no AB” for both population types (Table 2). For sample S3, if no signiﬁcant diﬀerence was recorded with three ABs, the TS condition was not judged to be signiﬁcantly diﬀerent from the “no AB” condition for psychrotrophs. For mesophiles, if the Ce condition was not signiﬁcantly diﬀerent from that of “no AB,” then the remaining three ABs (G, TS, and L) (which all yielded negative means) were ranked at the same level. That level was distinct from the “no AB” level (Table 2). The ranking based on the “AB type” illustrated sample and AB-type dependent responses. g g The results from the REGWQ test conﬁrmed the rise of psychrotrophs and of AR-resistant psychrotrophs under all the conditions with the exception of L-resistant bacteria for S3P with N (Figure 4). A high-growth dynamic was recorded for psychrotrophs (S1P) as many mean values shifted from negative to positive values during the storage period. After 3 days, the means were lower for HT compared to the control (C) for “no AB,” G and TS (Figure 4). The four additional storage days showed still lower means for “no AB” and for TS, but higher ones for G, Ce, and L compared to the control (C) (Figure 4). At day 3, all the means were lower for N compared to the control (C), and at day 7, still lower mean values were encountered for “no AB” and TS. The levels were moderately higher for L and equivalent for G and Ce (Figure 4). Compared to C, the means were lower for S2P in HT for “no AB” and TS and equivalent for G, Ce, and L at day 3. After 7 days of cold storage, the means were still lower for “no AB” and for TS, but higher for G, Ce, Frontiers in Microbiology \| www.frontiersin.org Trends in the log10 (Ratio) Mean Values for Mesophiles (M) and Psychrotrophs (P) Despite some decreases in the means at the intermediate storage time, the AR largely increased concomitantly with the increase in “total” bacterial counts which was obvious for most conditions with the exception of the TS- resistant mesophiles that were recovered from S3M with N (Figure 4). The REGWQ test revealed that the means were highest for S3M because seven conditions (ﬁve from C and two from HT) exceeded a value of 4, compared to only two conditions for S1M and one for S2M (Figure 4). All the controls (C) were characterized by high mean values with TS, which was nearly equivalent to the “no AB” condition (Figure 4). For S1M, the means were lower for HT, compared to the control (C), for the “no AB” condition and for TS, but they were higher for G and L, and moderately higher for Ce after 7 days of storage. For N, if the mean values were slightly higher for “no AB” and TS, compared to HT, all the other cases April 2017 \| Volume 8 \| Article 655 Frontiers in Microbiology \| www.frontiersin.org 7 Munsch-Alatossava et al. Hindering AR Dissemination in Milk FIGURE 4 \| REGWQ results showing the impact of the HT (the activated lactoperoxidase system) and N (N2 ﬂushed milk) treatments, compared to the control (C), on the three raw milk samples S1, S2, and S3 when considering the mesophiles (S1M, S2M, and S3M) and the psychrotrophs (S1P, S2P, and S3P) for the “no AB” condition, and in the presence of the antibiotics G (gentamycin), Ce (ceftazidime), L (levoﬂoxacin), and TS (trimethoprim-sulfamethoxazole). FIGURE 4 \| REGWQ results showing the impact of the HT (the activated lactoperoxidase system) and N (N2 ﬂushed milk) treatments, compared to the control (C), on the three raw milk samples S1, S2, and S3 when considering the mesophiles (S1M, S2M, and S3M) and the psychrotrophs (S1P, S2P, and S3P) for the “no AB” condition, and in the presence of the antibiotics G (gentamycin), Ce (ceftazidime), L (levoﬂoxacin), and TS (trimethoprim-sulfamethoxazole). interaction was highest compared to sample S3 in which the storage “time∗treatment” interaction dominated; the highest and nearly equivalent double interactions were recorded for S3M (Table 3). and L compared to the control (C) (Figure 4). DISCUSSION The trend in AR over time was evaluated here for mesophilic and psychrotrophic bacteria that were present in three bovine raw milk samples, which were subjected to a 7-day cold storage period at 6◦C. Single cold storage, which counted as the control (C), was combined with either the activated lactoperoxidase system (HT) or with a continuous N2 gas ﬂushing treatment (N). At the initial stage of the analyses, because the “total bacterial counts” (“no AB” condition) ranged between 2.86 and 3.57 log-units (Figure 1), the raw milk samples can be considered to be of very good bacteriological quality (Anonymous, 2004). This point is also corroborated by the initial pH values (Table 1) that are indicative of fresh raw milk (Tetra Pak, 2003). Trends in the log10 (Ratio) Mean Values for Mesophiles (M) and Psychrotrophs (P) For N, irrespective of the AB type, including the “no AB” condition, all the means were lower for S2P compared to the levels obtained for C and HT (out of one exception) (Figure 4). For S3P, HT showed lower mean values, compared to C, for “no AB” and for TS after 3 and 7 days of storage, in contrary to the results for G, Ce and L for which the means were equivalent to the levels recorded for the control (C) after 7 days (Figure 4). With one exception (L at day 3), all the mean values were lower for N compared to C at both sampling days; compared to HT, the means were equivalent for N considering “no AB” but lower for all the other conditions (Figure 4). Frontiers in Microbiology \| www.frontiersin.org Incidence of the Storage “Time,” “Treatment” Type and Antibiotic (“AB”) Type Factors The contribution of the storage “time” and “treatment” type factors on the recorded eﬀects was highly signiﬁcant under all conditions; whereas the storage “time” prevailed, it also displayed great heterogeneity. For example, this factor greatly aﬀected S1P and impacted more moderately S1M or S3P (Table 3). With the exception of S2M, the “AB” type factor had a minor and also variable contribution to the recorded eﬀects. The ANOVA tables did not reveal any signiﬁcant triple interactions between the tested storage “time,” “treatment” type and “AB” type factors. However, double interactions, that were signiﬁcant at an alpha risk factor of 5%, were often recorded, even though their eﬀects were lower compared to the eﬀects due to the diﬀerent factors that were considered individually (Table 3). For both population types from samples S1 and S2, the “treatment∗AB” Single cold storage largely promoted bacterial growth in a sample-dependent matter (Figures 1–4). The three samples were characterized by initial psychrotroph/mesophile ratios of 11.3, 4.5, and 1.3% for samples S1, S2, and S3, respectively, which suggests a shorter cold storage period prior to the start of the analyses for samples S3 and S2 compared to that of sample S1. The idea, that sample S1 probably underwent a longer cold storage prior to analyses is also corroborated by the observation of the growth kinetic for mesophiles (S1M) and especially psychrotrophs (S1P), which was highest for the April 2017 \| Volume 8 \| Article 655 8 Hindering AR Dissemination in Milk Munsch-Alatossava et al. TABLE 3 \| Factors and signiﬁcant double interactions that determined the recorded effects evaluated for log10 (ratio) on mesophiles (M) and psychrotrophs (P) for raw milk samples S1, S2, and S3. Incidence of the Storage “Time,” “Treatment” Type and Antibiotic (“AB”) Type Factors Sample - population F value for the factors Pr > F* Double interactions Pr > F* S1M Time F 51.46 <0.0001 TreatmentAB F 8.64 <0.0001 Treatment F 23.88 <0.0001 TimeAB F 3.15 0.0194 AB F 6.31 0.0002 S1P Time F 198.49 <0.0001 TreatmentAB F 7.63 <0.0001 Treatment F 13.21 <0.0001 TimeAB F 5.51 0.0007 AB F 4.69 0.0021 S2M Time F 134.56 <0.0001 TreatmentAB F 4.71 0.0001 Treatment F 17.24 <0.0001 TimeAB F 3.29 0.0159 S2P Time F 120.87 <0.0001 TreatmentAB F 4.14 0.0005 Treatment F 19.56 <0.0001 TimeAB F 2.16 0.0826 AB F 2.61 0.0431 S3M Time F 117.89 <0.0001 Timetreatment F12.88 <0.0001 Treatment F 43.06 <0.0001 TimeAB F 11.80 <0.0001 AB F 9.22 <0.0001 TreatmentAB F 10.87 <0.0001 S3P Time F 45.45 <0.0001 Timetreatment F 6.21 0.0033 Treatment F 10.25 0.0001 TreatmentAB F 3.77 0.0010 AB F 6.44 0.0002 p value associated with the F statistic. TABLE 3 \| Factors and signiﬁcant double interactions that determined the recorded effects evaluated for log10 (ratio) on mesophiles (M) and psychrotrophs (P) for raw milk samples S1, S2, and S3. earlier observations; shifts in the AR proﬁles over time, together with very high TS- resistance levels at the end of the storage period were also previously observed for cold-stored raw milk samples (Munsch-Alatossava et al., 2012a,b). Moreover, the highest AR increase for psychrotrophs compared to mesophiles as noted here for samples S2 and S3 (Figures 1, 4) is also in line with previous observations which showed that AR, that was evaluated from psychrotrophs, supplanted AR from the corresponding mesophiles in cold-stored raw milk (Munsch- Alatossava et al., 2012a); this point also highlights the diﬃculty of judging the quality of raw milk with the sole mesophilic bacteriological standard, in which psychrotrophs may be partially overlooked due to their growth optima as not all psychrotrophs grow under mesophilic conditions. The ranking of the ABs (Table 2) occurred in accordance with the frequency of AB usage (Kools et al., 2008). The activation of the lactoperoxidase (an oxidoreductase that is naturally present in bovine raw milk), which is recommended by the FAO to preserve raw milk in developing countries, triggers a bacteriostatic type of action against Gram-positive bacteria such as streptococci or lactobacilli and a bactericidal type of action for Gram-negative such as coliforms or Pseudomonas spp. (Björck et al., 1975; Wolfson and Sumner, 1993; FAO, 2005; Seifu et al., 2005). Incidence of the Storage “Time,” “Treatment” Type and Antibiotic (“AB”) Type Factors The temperature-dependent inhibitory eﬀect allows for an extension of the keeping quality for raw milk by 11–12 h at 25◦C or by 5–6 days at 4◦C (FAO, 2005). The described inhibitory eﬀect with HT was especially noticeable here for samples S1 and S3, where, for “no AB,” both the mesophilic and psychrotrophic counts were below the counts from the control (C) at day 3, and still below 3 × 105 cfu/ml after 7 days (Figure 1), which is consistent with the literature (FAO, 2005). A bactericidal type of eﬀect due to HT could be suspected for psychrotrophs from S1 (S1P) (Figures 1, 4). *p value associated with the F statistic. “no AB” condition for sample S1 because the counts exceeded 105 cfu/ml already after 3 days of storage at 6◦C (Figure 1). By contrast, sample S2 showed the slowest growth kinetics for both mesophiles (S2M) and psychrotrophs (S2P) because the counts were only slightly above 3 log-units for both population types after 3 days (Figure 1); however, after 7 days, both the mesophiles and psychrotrophs reached approximately 8 or 9 log- units cfu/ml and were nearly equivalent in all three samples (Figure 1). If an altogether similar growth dynamic characterized the psychrotrophs and mesophiles from sample S1 [as estimated from the mean log10 (ratio) values of approximately 2.2 for both population types (Figure 3), which further tends to conﬁrm a longer cold storage period for S1 prior to the analyses], it was distinct from the results for samples S2 and S3 in which the overall growth dynamic for the psychrotrophs was most vigorous [with mean log10 (ratio) values of 3.19 and 2.91, respectively] compared to the mesophiles [with mean log10 (ratio) values of 1.83 and 1.79, respectively; Figure 3]. After 7 days in cold storage, the control milk samples (C) were spoiled, exhibited the lowest pH values and emitted bad odors, whereas the pH values for HT- treated milk or N2-ﬂushed samples were not greatly altered (Table 1), and were in line with previous observations (Gaya et al., 1991; Dechemi et al., 2005; Munsch-Alatossava et al., 2010a). For all three samples, the activated lactoperoxidase system revealed two eﬀects that succeeded one another over time. Frontiers in Microbiology \| www.frontiersin.org Incidence of the Storage “Time,” “Treatment” Type and Antibiotic (“AB”) Type Factors Speciﬁcally, based on the “total counts” level, the initial storage phase until day 3 showed an AB type-dependent AR level; but at day 7, even though the “total counts” (for “no AB”) were still below or approximately 3 × 105 cfu/ml (for mesophiles) for all the conditions, both the psychrotrophic and mesophilic bacterial types resistant to G, Ce, L and TS were often plateauing at approximately the same level, contrary to that of the control (C) or to the N2-ﬂushed milk samples (N) (Figure 1). Surprisingly, the resistant bacterial counts enumerated under the activated lactoperoxidase system (HT) could even exceed the resistant colony levels encountered from the corresponding controls (C) as noticeable for mesophiles in sample S2 (Figure 1). All the preceding observations highlight an altogether greater growth dynamic under the activated lactoperoxidase system combined with cold storage (HT), when compared to the single cold storage (C), which resulted in a selection of bacterial populations that were recovered at similar levels from diﬀerent AB platings, implying that these bacterial types carry AB multi-resistance features. Indisputably, the rise in bacteria in the raw milk during cold storage corresponded to an increase in resistant mesophilic and psychrotrophic bacterial populations (Figures 1–4), conﬁrming The continuous N2 gas ﬂushing kept both the mesophilic and psychrotrophic “total” counts (condition “no AB”) below 105 cfu/ml over 7 days for all three samples (Figure 1) as observed April 2017 \| Volume 8 \| Article 655 Frontiers in Microbiology \| www.frontiersin.org 9 Hindering AR Dissemination in Milk Munsch-Alatossava et al. previously (Munsch-Alatossava et al., 2010a; Gschwendtner et al., 2016). The recorded inhibitory eﬀects due to N2 gas ﬂushing were also more “sample or initial microbiota than initial total bacterial level”-dependent (Figure 1); this point is well illustrated for S2M and S3M. Compared to sample S2, sample S3 presented the lowest psychrotrophic bacterial level at the initial stage and showed the highest sensitivity to N2-ﬂushing with a 5-fold lower mean for S3, despite the fact that both samples presented approximately similar initial bacterial counts (Figures 1, 3). Incidence of the Storage “Time,” “Treatment” Type and Antibiotic (“AB”) Type Factors As shown in Figure 1, the N2 gas treatment seems to trigger more moderate changes compared to HT in which large dynamic changes such as a 105-fold increase for TS-resistant bacteria could occur within 4 days. We hypothesize that the activation of the lactoperoxidase system as an enzymatic type of reaction occurs rapidly and creates a “biological emptiness” that can be exploited by certain bacterial types that clearly dominated at the end of the cold storage phase and were characterized by the ability to express resistance to four classes of antibiotics simultaneously (Figures 1–3). Cold storage promotes an increase of Gram-negative bacteria in raw milk (Chambers, 2002); the opportune action of HT that reportedly exerts primarily bactericidal action on Gram-negative representatives is therefore most appropriate for reducing the bacterial load. For all three samples considered here, the activated lactoperoxidase system clearly favored the growth of multi- resistant bacterial populations (Figures 1, 4), which suggests that the activated lactoperoxidase system may not confer full beneﬁts to all raw milk samples; whether initial good intentions are diverted by bacterial activities requires further clariﬁcations. For the ﬁrst time, the inhibitory eﬀect of N2 was “quantiﬁed” by the log10 (ratio) variable, which reﬂects the growth trend over a cold storage period of 7 days. Compared to the single cold storage, N2 gas ﬂushing was here, 81, 97, and 693-fold more eﬃcient in controlling the growth of mesophiles in raw milk for samples S1, S2 and S3, respectively. The comparative analyses of sample S3 on one side and of S1 on the other side suggest that an earlier gas application is most appropriate for preventing a massive bacterial increase and a simultaneous increase in AR (Figure 3). Some conditions yielded negative mean values (Figures 3, 4), which suggests that N2 gas ﬂushing could also trigger a bactericidal-type of eﬀect in addition to exhibiting an overall bacteriostatic type of eﬀect. To date, no clear bactericidal action was highlighted out of the observation that for Bacillus weihenstephanensis (a Gram-positive representative found in pasteurized milk), the cells were considerably damaged according to electron microscopic observations that were taken following the N2 gas ﬂushing treatment (Lechner et al., 1998; Munsch- Alatossava et al., 2013; Munsch-Alatossava and Alatossava, 2014). Microorganisms, which are present in freshly drawn raw milk, are subjected to a cold shock (from 37◦C to 3–4◦C). Incidence of the Storage “Time,” “Treatment” Type and Antibiotic (“AB”) Type Factors Notably, the study showed that N2 ﬂushing was always eﬃcient at controlling bacterial growth and the increases in AR as the mean log10 (ratio) values were judged to be signiﬁcantly diﬀerent for N compared to C for the three samples, or from HT for two samples (Figure 3); additionally, the means were mostly lowest for the N2-ﬂushed milk samples irrespective of the AB type (Figure 4). As frequently reported, Pseudomonas and Acinetobacter are key spoilers of raw milk (Cousin, 1982; Wiedmann et al., 2000; Chambers, 2002; Munsch-Alatossava and Alatossava, 2006; Rasolofo et al., 2010); isolates from both genera, that were retrieved from raw milk samples, also exhibited AB multiresistant traits (Munsch-Alatossava and Alatossava, 2007). Both of these genera host nosocomial pathogens and are well known for their intrinsic resistance to ABs together with a remarkable ability to acquire genes encoding resistance determinants (Bonomo and Szabo, 2006). The results from the present cultivation-dependent approach, together with the NGS based study which revealed that both Pseudomonas and Acinetobacter were negatively aﬀected by the N2 ﬂushing (Gschwendtner et al., 2016), are further heightening the potential of the N2 gas-based treatment. or 25◦C was somehow equivalent to that of HT (Munsch- Alatossava et al., 2016); surprisingly, at 6◦C, the N2-based treatments seemed to be more eﬃcient at inhibiting bacterial growth and also at tackling AR in bacterial populations compared to HT (Figures 1–4). At 6◦C, the “storage time” factor was preponderant over the “treatment type” under all the conditions (Table 3), in contrast to the study at 15/25◦C in which the “treatment type or condition“ factor seemed to be the major determinant of the inhibitory eﬀects for 4 of 7 samples (Munsch- Alatossava et al., 2016). A notable diﬀerence between HT and N relied on the observation that at sampling day 3, the activated lactoperoxidase system (HT) reduced TS-resistant bacterial levels to non-detectable levels under four conditions (S1P, S2M, S3M, and S3P) (Figure 1) whereas N experienced a reduction in the L- resistant bacterial types under four conditions (S1M, S2M, S2P, and S3M) (Figure 1). Because a “plateau situation” was observed less frequently with N2 compared to HT, it can be reasonably assumed that both treatments do not target the same bacterial types and that the mode of action of the two treatments is diﬀerent. Frontiers in Microbiology \| www.frontiersin.org Incidence of the Storage “Time,” “Treatment” Type and Antibiotic (“AB”) Type Factors After that, cold storage promotes considerable changes in bacterial populations with a reduction in bacterial diversity, including the decay of some genera (Lafarge et al., 2004; Raats et al., 2011; Munsch-Alatossava et al., 2012b; Gschwendtner et al., 2016). The graphs in Figure 1 showed that the N treatment seems to preserve the bacterial diversity better during cold storage because the growth trend, under the diﬀerent conditions, mostly reﬂects a “shift of a certain initial AR pattern” over time. The point that N2 gas alleviated the loss of bacterial diversity during cold storage was already highlighted by the NGS approach (Gschwendtner et al., 2016); we hypothesize that by better preserving bacterial diversity during cold storage, the N2 gas ﬂushing may better prevent the occurrence of a “favorable to a few” niche and consequently better prevent the rise of antibiotic multi-resistance in raw milk. Taken together, the trends in the mean values from both treatments (HT and N), including the cold storage alone (C), illustrated the sample or bacterial population type-, treatment type-, antibiotic type-, and time-dependent responses. As expected, the bacterial populations were best controlled at day 3 compared to day 7 for both treatments (Figures 1, 2). It was previously observed that the eﬀectiveness of N2 in controlling bacterial growth in raw milk samples stored at 15 April 2017 \| Volume 8 \| Article 655 10 Hindering AR Dissemination in Milk Munsch-Alatossava et al. Bacterial genomes carry between 1 and 15 ribosomal RNA operons (rrn) (Klappenbach et al., 2000; Vˇetrovsky and Baldrian, 2013). Pal et al. (2016) found a relative abundance of AR genes at 0.035 copies per 16S rRNA gene by investigating the human, animal and environmental resistomes; by considering a 105 cfu/ml raw milk sample and assuming an average of seven ribosomal RNA operons per bacterial cell, this level would correspond to as many as 24,500 AR genes in one milliliter of raw milk. lactoperoxidase system (HT), the quantiﬁed inhibitory eﬀect revealed equal or superior eﬃciency for N; if the bacterial levels were also kept below the 3 × 105 cfu/ml limit for HT, the treatment combined with cold storage was selective for highly multi-resistant populations in all three samples. REFERENCES EUCAST (2000). European Committee on Antimicrobial Susceptibility Testing. Determination of minimum inhibitory concentrations (MICs) of antibacterial agents by agar dilution. Clin. Microbiol. Infect. 6, 509–515. doi: 10.1046/j.1469- 0691.2000.00142.x Andremont, A. (2015). What to do about resistant bacteria in the food chain? Bull. World Health Organ. 93, 217–218. doi: 10.2471/BLT.15.030415 EUCAST (2010). AST of Bacteria: EUCAST Breakpoint Tables v1.1. Available online at: www.eucast.org Anonymous (2004). Regulation (EC) No 853/2004 of the European parliament and of the council of 29 April 2004 laying down speciﬁc hygiene rules for on the hygiene of foodstuﬀs. Oﬃcial J. Eur. Union L 139/55, 151. ECDC (European Centre for Disease Prevention and Control), EFSA (European Food Safety Authority), EMA (European Medicines Agency). (2015). ECDC/EFSA/EMA First Joint Report on the Integrated Analysis of the Consumption of Antimicrobial Agents and Occurrence of Antimicrobial Resistance in Bacteria from Humans and Food-Producing Animals. Stockholm; Parma; London: ECDC/EFSA/EMA, 2015. EFSA J. 13 (1), 4006. Aquilanti, L., Garofalo, C., Osimani, A., Silvestri, G., Vignaroli, C., and Clementi, F. (2007). Isolation and molecular characterization of antibiotic-resistant lactic acid bacteria from poultry and swine meat products. J. Food Prot. 70, 557–565. doi: 10.4315/0362-028X-70.3.557 Bari, L., Hoque, M. R., Reza, M. S. A., Hossain, M. A., and Islam, A. (2015). Adulteration of raw milk in selected regions of Tangail district of Bangladesh. J. Environ. Sci. 8, 41–44. doi: 10.3329/jesnr.v8i1.24662 FAO (1991). Guidelines for the Preservation of Raw Milk by the Use of the Lactoperoxidase System. Rome: CAC/GL 13-1991. Björck, L., Rosén, C. G., Marshall, V., and Reiter, B. (1975). Antibacterial activity of the lactoperoxidase system in milk against Pseudomonads and other Gram- negative bacteria. App. Microbiol. 30, 199–204. FAO (2016). Antimicrobial Resistance. FAO. Available online at: http://www.fao. org/antimicrobial-resistance/en/ Bonomo, R. A., and Szabo, D. (2006). Mechanisms of multidrug resistance in Acinetobacter species and Pseudomonas aeruginosa. Clin. Infect. Dis. 43(Suppl. 2), S49–S56. doi: 10.1086/504477 FAO (2005). “Beneﬁts and potential risks of the lactoperoxidase system of raw milk preservation,”in Report of an FAO/WHO Technical Meeting. FAO Headquarters (Rome). Forslund, K., Sunagawa, S., Coelho, L. P., and Bork, P. (2014). Metagenomic insights into the human gut resistome and the forces that shape it. Bioessays 36, 316–329. doi: 10.1002/bies.201300143 Botelho, B. G., Reis, N., Oliveira, L. S., and Sena, M. M. (2015). Development and analytical validation of a screening method for simultaneous detection of ﬁve adulterants in raw milk using mid-infrared spectroscopy and PLS-DA. Food Chem. 181, 31–37. Incidence of the Storage “Time,” “Treatment” Type and Antibiotic (“AB”) Type Factors By contrast, if N2 gas ﬂushing conferred beneﬁts in terms of keeping quality of raw milk by maintaining bacterial “total counts” under the 3 × 105 cfu/ml limit, it did not favor that frequent and massive proliferation of antibiotic multi-resistant bacteria. Further studies are needed to examine whether “most fresh raw milk” is under conditions that call for N2 gas ﬂushing to maximize the eﬃciency of blocking the increase in both “total counts” and in AR bacteria in raw milk during its cold storage. ACKNOWLEDGMENTS We gratefully acknowledge Antti Alavuotunki/Helsingin Meijeriliike Ltd. for the gift of the raw milk samples. We thank both reviewers for their valuable comments and suggestions to improve our manuscript. PMA dedicates this work to all misfortune companions who were swept away by restructuring at Helsinki university. AUTHOR CONTRIBUTIONS PMA and TA conceived and designed the experiments; PMA and SJ performed the experiments; JPG analyzed the data; TA and JPG contributed reagents/materials/analysis tools; PMA, TA, and JPG wrote the paper. All authors approved the ﬁnal version. CONCLUSIONS Bovine raw milk as a highly perishable food material is the target of many diﬀerent bacterial types with both spoilage and AR features. Here, N2 gas ﬂushing (N) kept the “total counts” of bacteria below 3 × 105 cfu/ml for three samples over 7 days of storage at 6◦C, but also demonstrated real potential in preventing high increases in mesophiles or psychrotrophs carrying AR or multi-resistant features. The eﬃciency, evaluated by the log10 (ratio), was maximal for the milk sample that carried low mesophilic counts and the lowest psychrotrophic initial counts. Compared to the activated REFERENCES doi: 10.1016/j.foodchem.2015.02.077 Frank, J. F., and Hassan, A. N. (2002). “Microorganisms associated with milk,” in Encyclopedia of Dairy Sciences, Vol 3, eds H. Roginski, J. W. Fuquay, and P. F. Fox (London: Academic Press), 1786–1796. Chambers, J. V. (2002). “The microbiology of raw milk,” in Dairy Microbiology Handbook, ed R. K. Robinson (New York, NY: Wiley-Interscience), 39–89. Cousin, M. A. (1982). Presence and activity of psychrotrophic microorganisms in milk and dairy products: a review. J. Food Prot. 45, 172–207. doi: 10.4315/0362-028X-45.2.172 Garofalo, C., Vignaroli, C., Zandri, G., Aquilanti, L., Bordoni, D., Osimani, A., et al. (2007). Direct detection of antibiotic resistance genes in specimens of chicken and pork meat. Int. J. Food Microbiol. 113, 75–83. doi: 10.1016/j.ijfoodmicro.2006.07.015 Dechemi, S., Benjelloun, H., and Lebeault, J. M. (2005). Eﬀect of modiﬁed atmospheres on the growth and extracellular enzymes of psychrotrophs in raw milk. Eng. Life Sci. 5, 350–356. doi: 10.1002/elsc.2005 20082 Gaya, P., Medina, M., and Nuñez, M. (1991). Eﬀect of the lactoperoxidase system on Listeria monocytogenes behavior in raw milk at refrigeration temperatures. Appl. Environ. Microbiol. 57, 3355–3360. April 2017 \| Volume 8 \| Article 655 Frontiers in Microbiology \| www.frontiersin.org 11 Munsch-Alatossava et al. Hindering AR Dissemination in Milk Gschwendtner, S., Alatossava, T., Kublik, S., Fuka, M. M., Schloter, M., and Munsch-Alatossava, P. (2016). N2 gas ﬂushing alleviates the loss of bacterial diversity and inhibits psychrotrophic Pseudomonas during the cold storage of bovine raw milk. PLoS ONE 11:e0146015. doi: 10.1371/journal.pone.0146015 Pereira, R. V., Siler, J. D., Bicalho, R. C., and Warnick, L. D. (2014). In vivo selection of resistant E. coli after ingestion of milk with added drug residues. PLoS ONE 9:e115223. doi: 10.1371/journal.pone.0115223 j p Perreten, V., Schwarz, F., Cresta, L., Boeglin, M., Dasen, G., and Teuber, M. Perreten, V., Schwarz, F., Cresta, L., Boeglin, M., Dasen, G., and Teuber, M. (1997). Antibiotic resistance spread in food. Nature 389, 801–802. doi: 10.1038/ 39767 Hsu, J. C. (1996). Multiple Comparisons: Theory and Methods. London: Chapman and Hall. (1997). Antibiotic resistance spread in food. Nature 389, 801–802. doi: 10.1038/ 39767 Raats, D., Oﬀek, M., Minz, D., and Halpern, M. (2011). Molecular analysis of bacterial communities in raw cow milk and the impact of refrigeration on its structure and dynamics. Food Microbiol. 28, 465–471. doi: 10.1016/j.fm.2010.10.009 Klappenbach, J. A., Dunbar, S. N., and Schmidt, T. M. (2000). rRNA operon copy number reﬂects ecological strategies of bacteria. Appl. Environ. Microbiol. 66, 1328–1333. REFERENCES doi: 10.1128/AEM.66.4.1328-1333.2000 Kools, S. A. E., Moltmann, J. F., and Knacker, T. (2008). Estimating the use of veterinary medicines in the European Union. Regul. Toxicol. Pharmacol. 50, 59–65. doi: 10.1016/j.yrtph.2007.06.003 Rasolofo, E. A., St-Gelais, D., La Pointe, G., and Roy, D. (2010). Molecular analysis of bacterial population structure and dynamics during cold storage of untreated and treated milk. Int. J. Food Microbiol. 138, 108–118. doi: 10.1016/j.ijfoodmicro.2010.01.008 Lafarge, V., Ogier, J. C., Girard, V., Maladen, V., Leveau, J. Y., Gruss, A., et al. (2004). Raw cow milk bacterial populations shifts attributable to refrigeration. App. Environ. Microbiol. 70, 5644–5650. doi: 10.1128/AEM.70.9.5644-5650.2004 Rolain, J. M. (2013). Food and human gut as reservoirs of transferable antibiotic resistance encoding genes. Front. Microbiol. 4:173. doi: 10.3389/fmicb.2013.00173 Lechner, S., Mayr, R., Francis, K. P., Prüβ, B. M., Kaplan, T., and Wieβner-Gunkel et al. (1998). Bacillus weihenstephanensis sp. nov. is a new psychrotolerant species of the Bacillus cereus group. Int. J. Syst. Bacteriol. 48, 1373–1382. doi: 10.1099/00207713-48-4-1373 Seifu, E., Buys, E. M., and Donkin, E. F. (2005). Signiﬁcance of the lactoperoxidase system in the dairy industry and its potential applications: a review. Trends Food Sci. Technol. 16, 137–154. doi: 10.1016/j.tifs.2004. 11.002 Munsch-Alatossava, P., and Alatossava, T. (2006). Phenotypic characterization of raw-milk associated psychrotrophic bacteria. Microbiol. Res. 161, 334–346. doi: 10.1016/j.micres.2005.12.004 SØrum, H., and L’Abbé Lund, T. M. (2002). Antibiotic resistance in food- related bacteria- a result of interfering with the global web of bacterial genetics. Int. J. Food Microbiol. 78, 43–56. doi: 10.1016/S0168-1605(02) 00241-6 Munsch-Alatossava, P., and Alatossava, T. (2007). Antibiotic resistance of raw-milk associated psychrotrophic bacteria. Microbiol. Res. 162, 115–123. doi: 10.1016/j.micres.2006.01.015 Tetra Pak (2003). Dairy Processing Handbook, 2nd Revised Edn. Lund: Tetra Pak Processing Systems AB. Munsch-Alatossava, P., and Alatossava, T. (2014). Nitrogen gas ﬂushing can be bactericidal: the temperature-dependent destiny of Bacillus weihenstephanensis KBAB4 under a pure N2 atmosphere. Front. Microbiol. 5:619. doi: 10.3389/fmicb.2014.00619 Teuber, M. (1999). Spread of antibiotic resistance with food-borne pathogens. Cell Mol. Life Sci. 56, 755–763. doi: 10.1007/s000180050022 Vˇetrovsky, T., and Baldrian, P. (2013). The variability of the 16S rRNA gene in bacterial genomes and its consequences for bacterial community analyses. PLoS ONE 8:e557923. doi: 10.1371/journal.pone.0057923 Munsch-Alatossava, P., Gauchi, J. P., Chamlagain, B., and Alatossava, T. (2012a). Trends of antibiotic resistance in mesophilic and psychrotrophic bacterial populations during cold storage of raw milk. ISRN Microbiol. 2012:918208. doi: 10.5402/2012/918208 Wang, H., McEntire, J. C., Zhang, L., Li, X., and Doyle, M. (2012). REFERENCES The transfer of antibiotic resistance from food to humans: facts, implications and future directions. Rev. Sci. Tech. 31, 249–260. doi: 10.20506/rst.31.1.2117 rections. Rev. Sci. Tech. 31, 249–260. doi: 10.20506/rst.31.1.2117 Munsch-Alatossava, P., Ghafar, A., and Alatossava, T. (2013). Potential of nitrogen gas (N2) ﬂushing to extend the shelf life of cold stored pasteurised milk. Int. J. Mol. Sci. 14, 5668–5685. doi: 10.3390/ijms14035668 Wichmann, F., Udikovic-Kolic, N., Andrew, S., and Handelsmann, J. (2014). Diverse antibiotic resistance genes in dairy cow manure. mBio 5, e01017– e01013. doi: 10.1128/mBio.01017-13 Munsch-Alatossava, P., Gursoy, O., and Alatossava, T. (2010a). Potential of nitrogen gas (N2) to control psychrotrophs and mesophiles in raw milk. Microbiol. Res. 165, 122–132. doi: 10.1016/j.micres.2009.02.002 Wiedmann, M., Weilmeier, D., Dineen, S. S., Ralyea, R., and Boor, K. J. (2000). Molecular and phenotypic characterization of Pseudomonas spp. isolated from milk. App. Environ. Microbiol. 66, 2085–2095. doi: 10.1128/AEM.66.5.2085-2095.2000 Munsch-Alatossava, P., Gursoy, O., and Alatossava, T. (2010b). Exclusion of phospholipases (PLs)-producing bacteria in raw milk ﬂushed with nitrogen gas (N2). Microbiol. Res. 165, 61–65. doi: 10.1016/j.micres.2008.07.001 Wolfson, L., and Sumner, S. S. (1993). Antibacterial activity of the lactoperoxidase system: a review. J. Food Prot. 56, 887–892. doi: 10.4315/0362-028X-56.10.887 Wolfson, L., and Sumner, S. S. (1993). Antibacterial activity of the lactoperoxidase system: a review. J. Food Prot. 56, 887–892. doi: 10.4315/0362-028X-56.10.887 WHO (2015). Global Action Plan on Antimicrobial Resistance. Geneva: World Health Organisation. Munsch-Alatossava, P., Ikonen, V., Alatossava, T., and Gauchi, J. P. (2012b). “Trends of antibiotic resistance (AR) in mesophilic and psychrotrophic bacterial populations during cold storage of raw milk, produced by organic and conventional farming systems,” in Antibiotic Resistant Bacteria, A Continuous Challenge in the New Millenium, ed M. Pana (Rijeka: Intech), 105–124. WHO (2015). Global Action Plan on Antimicrobial Resistance. Geneva: World Health Organisation. Conﬂict of Interest Statement: The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Munsch-Alatossava, P., Quintyn, R., De Man, I., Alatossava, T., and Gauchi, J. P. (2016). Eﬃciency of N2 gas ﬂushing compared to the lactoperoxidase system at controlling bacterial growth in bovine raw milk stored at mild temperatures. Front. Microbiol. 7:839. doi: 10.3389/fmicb.2016.00839 Copyright © 2017 Munsch-Alatossava, Jääskeläinen, Alatossava and Gauchi. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). Frontiers in Microbiology \| www.frontiersin.org REFERENCES The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Murray, S. K., Kwan, K. K. H., Skura, B. J., and Mc Kellar, R. C. (1983). Eﬀect of nitrogen ﬂushing on the production of proteinase by psychrotrophic bacteria in raw milk. J. Food Sci. 48, 1166–1169. doi: 10.1111/j.1365-2621.1983.tb09183.x Pal, C., Bengtsson-Palme, J., Kristiansson, E., and Larsson, D. G. (2016). The structure and diversity of human, animal and environmental resistomes. Microbiome 4, 54. doi: 10.1186/s40168-016-0199-5 April 2017 \| Volume 8 \| Article 655 Frontiers in Microbiology \| www.frontiersin.org 12

Subsets and Splits

No community queries yet

The top public SQL queries from the community will appear here once available.